Compare commits

...
99 Commits
Author SHA1 Message Date
Alex Corrado 3f71a4bfc2 Update fsharp.py 2019-11-24 14:06:21 -05:00
Alex Corrado 394ed52275 Update fsharp.py 2019-11-23 21:12:53 -05:00
Alex Corrado 8a2d0cb8a7 Update fsharp.py 2019-11-22 23:25:41 -05:00
Alex Corrado 670e1ea3cb [fsharp] Remove failing patch 2019-11-22 18:35:53 -05:00
Alex Corrado c759092277 [fsharp] Try to build with F# 4.7 2019-11-22 14:58:32 -05:00
Zoltan Varga 93e46c687b [runtime] Treat calling a non-virtual method through an open delegate the same as the normal case instead of the OPEN_VIRTUAL case. (#17832)
* [runtime] Treat calling a non-virtual method through an open delegate the same as the normal case instead of the OPEN_VIRTUAL case.

* [amd64] Fix the return value mapping in gsharedvt calls, the return address might be in different registers in the caller and callee.

* [runtime] Always initialize del->method_ptr even if interp_method is set, JITted code makes calls through it in mixed mode.

* Add a test for open delegates and gsharedvt.

* Increase nrgctx trampoline count, its needed by dynamic-method-churn.exe in full-aot-interp mode.
2019-11-22 20:34:14 +01:00
Jonathan Chambers cc137237d8 Replace embedded libgc with Unity fork of recent Boehm (bdwgc) (#16832)
* Initial commit to migrate libgc to bdwgc.

* Fix submodule url

* Don't force Boehm as runtime wrapper. Was just for testing locally.

* Build bdwgc as single object file. Helps with performance.

* Bump bdwgc submodule.

* Attempt to fix windows MSVC based build.

* Remove any references to GC_INSIDE_DLL as we manually managed threads on Windows.

* Allow building boehm on Windows amd64.

* Allow building boehm with msvc.

* Bump bdwgc to fix line endings

* Use LF for sh, am, m4, and ac files in bdwgc

* Bump bdwgc & libatomic_ops with fix for cygwin

* Bump bdwgc to fix mono-boehm being generated as libtool wrapper script.

* Link the static boehm library.

* Revert "Link the static boehm library."

This reverts commit 1f90d81d08308a61c890743cd79d8f6bb791bc99.

* Use --export-all-symbols on cygwin/mingw to fix issues with boehm causing mono symbols not to be exported.

* Fix
2019-11-22 18:03:23 +01:00
Filip Navara 0fed03ed63 Fix InternalGetHashCode for boehm/null GC (#17859)
It was unintentionally shifting the address to left instead of right. It should be shift right to get rid of zeroes from the alignment.
2019-11-22 07:51:43 -05:00
Alexander Köplinger 5fedc42ccf Fix automake warning 2019-11-22 13:18:10 +01:00
Kenneth Pouncey 9e1d59b8cc Remove unused GetHttpMessageHandler from WebAssembly.Net.Http.
- Remove linker.xml from project
2019-11-22 13:12:17 +01:00
Zoltan Varga ecd67a91c4 [wasm] Add aot+netcore support. (#17798)
* [wasm] Compile corelib as 32 bit.

* [wasm] Add a netcore cross compiler.

* [wasm] Add netcore+aot support.

* [aot] Load the dedup module lazily instead of at startup.

The previous version didn't work on netcore because it couldn't resolve references to corlib etc.
since they were not yet registered in the root domain at the time of loading.

* Add more icall trampoline signatures.

* Add more netcore pinvokes.

* Fix the build.

* Add more icall trampoline signatures.

* Add an --embed-file option to the packager.

* Add more netcore pinvokes.

* [interp] Return FALSE for IsDynamicCodeCompiled.

* Add a netcore project and targets for running BenchmarkDotNet.

* Fix pinvoke tables.

* Add missing files.

* Fix make clean.

* Add a comment.

* Revert "Fix pinvoke tables."

This reverts commit 998065e77614b7e621d2873510aabc3930669ca9.

* Make xunit-runner into a dotnet project. Add a netcore version.
2019-11-22 12:42:55 +01:00
Egor Bogatov 6a5c44b15e [netcore] Run tests with Interpreter (#17862)
* Run tests with interp and llvm

* fix build

* Run tests with llvm

* Add a few tests to rsp for LLVM
2019-11-22 09:14:23 +01:00
Kenneth Pouncey 202cb6a648 Add more descriptive message to the PlatformNotSupportedException (#17864) 2019-11-22 08:40:30 +01:00
Jay Krell dadc3d3117 Remove handles/MonoError from Mono.RuntimeGPtrArrayHandle. (#17806) 2019-11-21 14:15:29 -05:00
Jo Shields 9e00330af3 Stop referencing files from Mono BCL in Netcore System.Private.CoreLib (#17866) 2019-11-21 19:48:00 +01:00
Bernhard Urban-Forster 058d157648 [interp] fix array_element_size intrinsic (#17857)
[interp] fix array_element_size intrinsic

`mono_class_array_element_size` gives us the size of the provided MonoClass if it would be an array element. But here we want the element size of a given array's MonoClass. That's what `mono_array_element_size` is returning.

This leads to all kind of weird crashes otherwise, specifically here:
https://github.com/mono/mono/blob/2c20649539ac16e069a65f2a750c793eb341e50f/netcore/System.Private.CoreLib/src/System/Array.Mono.cs#L64

Which is then used later to compute the size to memset in order to clear the content of an array. If it's larger than it should be, then this will cause memory corruption. For example this would crash eventually:

```csharp
using System;
using System.Collections.Generic;

namespace HelloWorld {
    class Program {
        static void Main(string[] args) {
            for (int j = 0; j < 0x1000; j++) {
                var d = new Dictionary<string, string> ();
                for (int i = 0; i < 197; i++)
                    d.Add (i + "", i + "foo");
                d.Clear ();
            }
        }
    }
}
```

Thanks to @EgorBo for reporting.
2019-11-21 11:56:03 +01:00
Alexander Köplinger b6e4ddd73c [metadata] Remove native threadpool from netcore build (#17856)
We use the managed threadpool there.
2019-11-21 10:01:39 +01:00
Egor Bogatov 2c20649539 [llvm] Recognize GEP (#17844)
* initial impl

* improve jit

* fix copy-paste

* disable for non-netcore

* clean up
2019-11-20 21:25:24 +01:00
Zoltan Varga 4961d411e7 [interp] Move the freeing of interp method info into a callback. (#17852)
[interp] Move the freeing of interp method info into a callback.
2019-11-20 19:56:09 +01:00
Egor Bogatov 27e6e812f1 enable gss on android (#17838) 2019-11-20 10:53:14 -05:00
Alexander Köplinger 0396037d31 [debugger-agent] Fix CMD_VM_ALL_THREADS returning wrong value due to unitialized variable (#17847)
`remove_gc_finalizing` wasn't initialized so its value is undefined.
The `count_thread_check_gc_finalizer()` function only sets it in some cases so we could end up with the variable having an undefined (and probably not 0) value.
This resulted in the returned thread count from the debuggee being wrong because we'd decrement `count` even though we shouldn't.

Regression from https://github.com/mono/mono/pull/15618
2019-11-20 08:28:58 +01:00
dotnet-maestro[bot] a6db160596 [master] Update dependencies from dotnet/arcade dotnet/core-setup dotnet/corefx (#17690)
* Update dependencies from https://github.com/dotnet/arcade build 20191102.1

- Microsoft.DotNet.Arcade.Sdk - 5.0.0-beta.19552.1
- Microsoft.DotNet.Helix.Sdk - 5.0.0-beta.19552.1

* Update dependencies from https://github.com/dotnet/core-setup build 20191104.1

- Microsoft.NETCore.App - 5.0.0-alpha.1.19554.1

* Update dependencies from https://github.com/dotnet/corefx build 20191103.8

- Microsoft.Private.CoreFx.NETCoreApp - 5.0.0-alpha.1.19553.8

* Update dependencies from https://github.com/dotnet/arcade build 20191108.11

- Microsoft.DotNet.Arcade.Sdk - 5.0.0-beta.19558.11
- Microsoft.DotNet.Helix.Sdk - 5.0.0-beta.19558.11

* Update dependencies from https://github.com/dotnet/core-setup build 20191109.5

- Microsoft.NETCore.App - 5.0.0-alpha.1.19559.5

* Update dependencies from https://github.com/dotnet/corefx build 20191109.6

- Microsoft.Private.CoreFx.NETCoreApp - 5.0.0-alpha.1.19559.6

* Update dependencies from https://github.com/dotnet/arcade build 20191117.2

- Microsoft.DotNet.Arcade.Sdk - 5.0.0-beta.19567.2
- Microsoft.DotNet.Helix.Sdk - 5.0.0-beta.19567.2

* Update dependencies from https://github.com/dotnet/core-setup build 20191114.1

- Microsoft.NETCore.App - 5.0.0-alpha.1.19564.1

* Update dependencies from https://github.com/dotnet/corefx build 20191113.6

- Microsoft.Private.CoreFx.NETCoreApp - 5.0.0-alpha.1.19563.6

* Use NETCoreApp version 5.0.0-alpha.1.19563.3

Later versions have an issue and the new dotnet/runtime assets aren't ready yet.

* Disable failing tests

* Fix merge error
2019-11-19 16:47:57 +01:00
Steve Pfister 0c3ffb9cc9 Fixes https://github.com/mono/mono/issues/15805 at least in terms of what may be making it crash. (#17270)
Adds a null check around the sslStream before trying to dispose.
2019-11-19 14:23:51 +01:00
Zoltan Varga b8fe43da3e [interp] Add some missing netcore intrinsics. (#17782) 2019-11-18 13:55:15 +01:00
Steve Pfister 0269f7b479 Bump corefx to pick up https://github.com/mono/corefx/pull/370 (#17758)
Backport of https://github.com/dotnet/corefx/pull/34560

Added HAVE_LCHFLAGS & HAVE_STAT_FLAGS to configure.
2019-11-18 12:13:16 +01:00
Filip Navara e2ccdac3f4 [llvm] Add support for LLVM JIT on ARM64 (#17827)
* [llvm] Add support for LLVM JIT on ARM64

* Fix tabs
2019-11-18 00:24:11 +01:00
Jérôme Laban 965efade84 Update emscripten to 1.39.3 (#17816)
Includes updates for https://github.com/emscripten-core/emscripten/issues/9562 and side modules.
2019-11-16 13:59:28 +01:00
Dotnet-GitSync-Bot 6785b27567 Mirror changes from mono/runtime (#17803)
It looks like the mirror didn't fully quiesce prior to snapping coreclr/corefx for the runtime consolidation.  As a result the libraries build is breaking.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-16 11:27:32 +01:00
Nikolay Sivov 18dd48cad8 [mini] Fix NULL pointer crash in mono_decompose_vtype_opts(). (#17801)
* [mini] Fix NULL pointer crash in mono_decompose_vtype_opts().

All other call sites for mono_type_size() are already passing non-null pointer there.

* [mini] Rename variable dummy -> align.
2019-11-15 17:56:55 +01:00
Marcos Cobeña Morián 92dd4d3883 [Wasm] Forced filesystem creation (#17789)
* Forced filesystem creation

* [Wasm] Added new hello-io sample
- Goal is testing file preloading through System.IO
- Filesystem is now always forced to support later file preloading
- build-io-sample target showcases such

* Added --enable-fs flag to packager.exe

* Update packager.md
2019-11-15 15:24:50 +01:00
Jan Kotas 3ceecb8db0 Fix build breaks
Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Carlos Sanchez Lopez d3fbf4a1d3 Fix NetStandard issue in System.IO.FileSystem.AccessControl in 5.0 (#42546)
Fix NetStandard issue in System.IO.FileSystem.AccessControl

We built the fix incorrectly so that it would only apply for netcoreapp3.0 but customer needs it on netstandard2.0 (desktop + core).  Customer impact is that they’ll see a PlatformNotSupported exception rather than the new API.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Stephen Toub 00b0170bdc Fix StyleCop rule SA1129 ("Do not use default value type constructor") (#42531)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Tanner Gooding c93801d458 Adding tests for the new apis to convert between S.Numerics and S.R.Intrinsic vectors (#27481)
* Updating Vector128.AsVector3 to have a functioning implementation

* Adding test template metadata covering the new methods for converting between S.Numerics and S.R.Intrinsics

* Regenerating the existing tests

* Generating the new Vector conversion tests

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Emmanuel AndréandStephen Toub 5645032fc6 Add span-based CopyTo and CopyToAsync methods (dotnet/coreclr#27639)
* Add span-based CopyTo and CopyToAsync methods

* Update according to feedback

* Add span-based CopyTo overrides for MemoryStream

* Improve span-based CopyTo arguments validation

To avoid code duplication

* Update according to second review

Stream API is changed

* Resolve InternalReadSpan/Memory inlining

* Refactor ValidateCopyToArgs

* Update according to third review

* Update after fourth review

* Override span CopyTo for UnmanagedMemoryStream

* Apply suggestions from code review

Co-Authored-By: Stephen Toub <[email protected]>

* Update after fifth review

* Add cross sync/async support for span-based CopyTo

* Call sync action directly in async context

* Rework cross sync/async support for span-based CopyTo

Co-Authored-By: Stephen Toub <[email protected]>

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
John adcaf88db0 Fix Vector128 and Vector256 WithUpper documentation (dotnet/coreclr#27274)
* Update Vector128.cs

* Update Vector256.cs

* Update Vector128.cs

* Update Vector128.cs

* Update Vector256.cs

* Update Vector128.cs

* Update Vector256.cs

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Elinor Fung 1f2f9b929b Add tracing for handling of dependencies for Assembly.LoadFrom (#27833)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Elinor Fung 3242b5f354 Add tracing for Resolving/AssemblyResolve event handler invocation (#27788)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
ts2do 35d94dc8ff Address bugs in BigInteger (dotnet/coreclr#27280)
* Method Add(ref BigInteger lhs, uint value, ref BigInteger result) would store most of the result blocks into lhs instead of result.
* Method ShiftLeft(ulong input, uint shift, ref BigInteger output) with a shift argument exceeding 32 would generally compute the higher blocks incorrectly.
* Multiply(ref BigInteger lhs, uint value, ref BigInteger result) would not set result._length in some cases.
* IsZero() would incorrectly return false for non-canonical zeros with _length > 0.

Fix:
* Inline Add(ref BigInteger, uint, ref BigInteger) into Add(uint).
* Inline ShiftLeft(ulong, uint, ref BigInteger) into Pow2.
* Inline ExtendBlock and ExtendBlocks into Pow2.
* Properly handle 0 in SetUInt32 and SetUInt64.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Stephen Toub 2a108e958f Use Stream's BeginReadInternal from FileStream (dotnet/coreclr#27737)
FileStream has two modes, decided upon at construction time.  When it's created in non-async mode, the Read/WriteAsync methods end up queueing work items to invoke the synchronous Read/Write methods.  To do this, the base methods on Stream delegate to Begin/EndRead/Write (since they were around first) and then the resulting IAsyncResult is wrapped.  However, Stream has an optimization that checks to see whether the derived stream actually overrides Begin/EndXx, and if it doesn't, then it skips using those and goes straight to queueing a work item to Read/Write.  However, FileStream does override those, but when it delegates to the base implementation because it's in non-async mode (rather than because it's a type derived from FileStream), going through Begin/EndXx is just unnecessary overhead.  So, in the right circumstances, we can call to Stream's special helper instead.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-15 13:29:04 +01:00
Zoltan Varga 25b7e6bd42 [runtime] Add MONO_EMPTY_SOURCE_FILE to a couple of files. (#17795)
* [runtime] Add MONO_EMPTY_SOURCE_FILE to a couple of files.

* Fix a typo.

* Revert unrelated changes.
2019-11-15 09:04:35 +01:00
Vsevolod Kukol 805c7f5436 [MacSDK] Bump xamarin-gtk-theme.py to latest revision from private bockbuild (#17785) 2019-11-14 09:27:33 +01:00
Jay Krell 19a44cd32b Remove handles from System.Diagnostics.Debugger. (#17713)
* Remove handles from System.Diagnostics.Debugger.
memset (&x, 0, sizeof (type)) to memset (&x, 0, sizeof (x)).

* Bump API snapshot submodule
2019-11-14 08:18:02 +01:00
Zoltan Varga 1d81534d27 [interp] Add some inline checks from the JIT. (#17784) 2019-11-13 22:34:01 +01:00
Zoltan Varga a0fcc29d5d [runtime] Fix some handle usage in native-library.c (#17781) 2019-11-13 18:12:09 +01:00
Jay Krell 73431403c7 Remove handles from ves_icall_System_Array_CanChangePrimitive. (#17708)
Remove handles from ves_icall_System_Array_CanChangePrimitive.
2019-11-13 17:55:52 +01:00
Thays Grazia dd18ec40dd [debugger] Assert when async debug a generic method (#17727)
* When we try to call a method to get the async_id to do an async debug and we are trying to do this in a generic method like this:

async Task<T> ExecuteAsync_Broken<T>()
 {
            await Task.Delay(2);
            return default;
  }

We need to inflate the generic type before call the method or we will get the error:  Could not execute the method because the containing type 'System.Runtime.CompilerServices.AsyncTaskMethodBuilder1[T_REF]’, is not fully instantiated.

Fixes #17549
Fixes #17569
2019-11-13 16:29:33 +01:00
imhameed fe0f824db5 [llvm] Use explicit null checks with LLVM. (#16954)
* [llvm] Use explicit null checks with LLVM.

When the LLVM backend is used for a method, explicit null checks will
unconditionally be used at the mini IR level. We had previously used
implicit null checks at the mini IR level combined with llvm-volatile
loads and stores to avoid mis-optimization of the generated code.

During AOT compilation, LLVM's `ImplicitNullChecks` pass is enabled.
This conservatively elides these explicit null checks. This pass only
modifies branches tagged with `make.implicit` metadata.

Null check branches are now tagged with `make.implicit` metadata, with
three exceptions:

1. For platforms with backends that do not support zero page fault
handling, null check branches are not tagged with `make.implicit`.

2. This commit also adds a `MONO_DEBUG` option:
`llvm-disable-implicit-null-checks`. When it is enabled, null check
branches are not tagged with `make.implicit`.

3. This commit alters the behavior of the `explicit-null-checks`
`MONO_DEBUG` option. Enabling `explicit-null-checks` also implies
`llvm-disable-implicit-null-checks`, because this option is documented
(e.g. in https://www.mono-project.com/docs/debug+profile/debug/) to
completely disable signal/seh-based null checks.

* Add llvm-disable-implicit-null-checks to the MONO_DEBUG help string.

* Add llvm-disable-implicit-null-checks to mono.1.

* Don't tag null checks in EH regions with `make.implicit`.

This relocates the logic formerly used to guard individual LLVM-level
loads and stores with explicit null checks inside EH regions.

* Use atomic/volatile loads and stores for the terimation flag in thread-suspend-suspended.

LLVM was hoisting the load of and branch against `finished` out of
thread t1's loop, resulting in an infinite spin.

* Use ImplicitNullChecks with LLVM JIT.

* Suppress emission of the .llvm_faultmaps section.

Bump external/llvm to release_60
(7a8dc89adbe7e123220e070a527e096ee91e66d5).

* [ci] use LLVM built from repository
2019-11-13 01:35:44 +01:00
Zoltan Varga eaf895f2d0 [wasm] Change netcore support to use a prebuilt corefx runtime. (#17773)
* [wasm] Change netcore support to use a prebuilt corefx runtime.

* [wasm] Build and test netcore support on CI.

* [wasm] Disable netcore tests on CI for now.

* [wasm] Add an error check.

* [wasm] Add a linker exclusion.

* [wasm] Add netcore runtimes to the nupkg.
2019-11-13 01:30:09 +01:00
Filip Navara bd278dd00d Mirror changes from mono/corefx,coreclr (#17777)
* Rewrite the rest of the array non-generic fast paths in C# (#27703)

* Rewrite the rest of the array non-generic fast paths in C#

* Add missing resource string

* Use CorElementType and handle native ints to match the unmanaged implementation

* Avoid try pattern for Sort and Reverse

* Match the native implementation more closely

Signed-off-by: dotnet-bot <[email protected]>

* Rewrite Enum.CompareTo in C# (dotnet/coreclr#27792)

* Rewrite Enum.CompareTo in C#

* Handle floating point too for match the existing native implementation

Signed-off-by: dotnet-bot <[email protected]>

* Remove unused argument from GetRegionDisplayName (dotnet/coreclr#27806)

* Remove unused argument from GetRegionDisplayName

* Address race conditions in CultureData

While most of the race conditions in this file are benign, in these cases, a thread could end up returning an intermediate value due to multiple potential writes to the same field.

Signed-off-by: dotnet-bot <[email protected]>

* Fix StyleCop rule SA1129 ("Do not use default value type constructor") (#27716)

Will enable the rule in corefx once these changes have propagated.

Signed-off-by: dotnet-bot <[email protected]>

* Remove non-shared Enum.CompareTo/InternalCompareTo implementation, add new Array icalls used by shared code
2019-11-12 18:46:00 +01:00
Bernhard Urban-Forster 4728cec01f [llvm] use multiple cores to build llvm if ninja is not available (#17778) 2019-11-12 18:45:45 +01:00
Jo Shields 8bceb40878 Move zlib to its own directory, and re-enable subdir-objects
This has the following benefits:

* Clearer structure for mirroring
* Slight build time improvement on bundled zlib use (as it only builds once)
* Removes pages of subdir-objects spew from autogen
2019-11-12 09:31:08 +01:00
abrevet-dev 8683fa6db7 [WinForms] Fix #10559 In MaskedTextBox wrong Lines value when Mask setted (#17660)
[WinForms] Fix #10559 In MaskedTextBox wrong Lines value when Mask se…





<!--
Thank you for your Pull Request!

If you are new to contributing to Mono, please try to do your best at conforming to our coding guidelines http://www.mono-project.com/community/contributing/coding-guidelines/ but don't worry if you get something wrong. One of the project members will help you to get things landed.

Does your pull request fix any of the existing issues? Please use the following format: Fixes #issue-number
-->
2019-11-12 01:55:53 +01:00
abrevet-dev 2b4801d9f1 [WinForms]: Fix #17546 Fix TabPage position when enabling MultiLine (#17670)
[WinForms] Fix TabPage position when enabling MultiLine

Fixes #17546
2019-11-12 00:15:51 +01:00
abrevet-dev 61811075ca [WinForms]: Fix #16632 special values (-1 and -2) of ListView Column header Width was not handled when using List.View. (#17669) 2019-11-11 23:00:45 +01:00
abrevet-dev 9eb0bcae57 [WinForms] Fix #12249 scroll orientation was not defined in ScrollEventArgs (#17662)
* [WinForms] Fix #12249 scroll orientation was not defined in ScrollEventArgs of ScrollBar.cs

* [WinForms] Fix 12249 scroll orientation was not defined in ScrollEventArgs of DataGridView.cs
2019-11-11 22:59:53 +01:00
Jo Shields c256fc86b6 Delete some LLVM test cases from make dist (#17772)
Many of the tests from the llvm-project monorepo cause problems to `make dist`
2019-11-11 17:42:51 +01:00
Alexis Christoforides 722c6f7dd5 [netcore] Complete Monitor.LockContentionCount implementation (#17719)
* [netcore] Complete Monitor.LockContentionCount implementation

Fixes https://github.com/mono/mono/pull/16538
2019-11-11 13:03:03 +01:00
Jan Kotas 594bbbef25 Relax visibility
Signed-off-by: dotnet-bot <[email protected]>
2019-11-10 09:29:18 +01:00
Marek Safar d90030d273 Fixes build 2019-11-09 20:45:42 +01:00
Ryan Lucia 9664df7eb3 Move AssemblyLoadContext.GetResolvedUnmanagedDll to shared (#27775)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-09 20:45:42 +01:00
rob313663 e46f29313d Fixed typos in comments. (dotnet/coreclr#27789)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-09 20:45:42 +01:00
Joseph Musser 5bf661d7f9 StackTrace.GetFrames should return an array of non-nullable frames (#27787)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-09 09:37:25 +01:00
Levi Broderick ae02b6677b Improve Encoding.UTF8.GetString / GetChars performance for small inputs (#27268)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-09 08:13:11 +01:00
Alexander Köplinger f9b3b5276d Fix llvm path in CODEOWNERS file 2019-11-09 00:44:57 +01:00
Alexander Köplinger ed2d5f69e0 [netcore] Fix RuntimePropertyInfo.GetValue() in FullAOT scenarios (#17757)
It tried to use dynamic codegen which is not available in FullAOT.
We can use the `RuntimeFeature.IsDynamicCodeSupported` to check for this, added a JIT intrinsic which returns false in FullAOT.

Used a linker friendly way to check for TypeBuilder by adding an `internal virtual IsTypeBuilder` method in the `Type` class which returns false, and overwrote it in `TypeBuilder`.
2019-11-09 00:33:00 +01:00
Zoltan Varga 73b50890cc [wasm] Bump emscripten. (#17748)
* [wasm] Bump emscipten to 1.39.2.

* Disable some code in libtest.c which causes emscripten to fail.

* [wasm] Source emscripten env before make install as well.
2019-11-08 14:43:32 -05:00
Nathan Ricci 5b0a822afb Implement GC.GetGCMemoryInfo (#17607)
Implement the GC.GetGCMemoryInfo API.
2019-11-08 13:58:37 -05:00
Ryan Lucia d5b6cf33b3 [loader] Unmanaged library refactoring and NativeLibrary implementation (#17369) 2019-11-08 13:42:56 -05:00
Ryan Lucia 68d142912d [netcore] Run individual CoreCLR test suites (#17753)
This is implemented in a lazy way for convenience when manually testing until the tests are wired up to xunit
2019-11-08 13:14:06 -05:00
Alexis Christoforides 23f2baa71e [merp] Remove extraneous waitpid invocation (#17723)
The extraneous call basically guarantees that the call right after will fail. Fixes merp stating that its process has failed when it hasn't.
2019-11-08 11:40:04 -05:00
Jo Shields 3948a065af Merge pull request #17749 from directhex/dotnet-llvm-monorepo
[LLVM] Change llvm submodule to dotnet-org fork of official LLVM git repo
2019-11-08 10:42:20 -05:00
lateralusX fad91a1c1e Fix Windows LLVM build using new repository structure. 2019-11-08 09:21:30 -05:00
Dotnet-GitSync-Bot 759a8c1847 Mirror changes from mono/corefx,coreclr,corert (#17761)
* Add EventWaitHandle creation extension method that takes an ACL (#42213)

* Add EventWaitHandle creation extension method that takes an ACL

* Call OpenExisting, add basic unit tests, move MAX_PATH in csproj for all Windows platforms, ensure same exceptions are thrown in both netcore and netfx.

* Address suggestions: using for the created handle, simplify try finally, merge methods, remove an unnecessary Debug.Assert

* simplify using, reorganize PInvokes a bit, let VS format resx as it expects it, remove BOM from csproj

* Add more unit tests

* Save basic access rights in a constant, generate name randomly

* Use FullControl (EVENT_ALL_ACCESS); update unit tests to use EventWaitHandleRights

* Apply suggestion of creating an EWH, then replacing its SWH.

* Remove using causing chaos, add more unit tests, remove null string unit test.

* spacing

* Address comments

* Remove unnecessary checks/exceptions, update unit tests

* Fix netfx x86 ut failure - different exception for mode validation.

* Small documentation fix

* Remove documentation for exception thrown with null security

* more documentation fixes

* New file for unit tests

* Dispose and remove windows only attributes in tests

* suggestion to not use var

* Remove duplicate resx after merge

* Readd cs files lost during merge

* Remove resx modification since it was untouched for this PR

* Address test comments

Signed-off-by: dotnet-bot <[email protected]>

* Move NativeLibrary.Free null check to managed (dotnet/coreclr#27744)

Signed-off-by: dotnet-bot <[email protected]>

* Arm64: Add S.P.CoreLib instrinsics UnzipEven and UnzipOdd (#27731)

Signed-off-by: dotnet-bot <[email protected]>
2019-11-08 07:11:13 -05:00
Tamar Christina 771b5fb18d Arm64: Update S.P.CoreLib for ZipLow and ZipHigh (dotnet/coreclr#27730)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-08 11:01:17 +01:00
Carlos Sanchez Lopez 229d3cc097 Add Semaphore creation extension methods that take an ACL (#42377)
Approved API Proposal: #41662

Description
We don't currently have a way to create a Semaphore with a given ACL in .NET Core. We can modify the ACL, but it would be more secure to have the proper ACL on the object from the start.

Customer impact
Before this change, customers had to create a Semaphore, then set its ACLs. This presents a few problems:
- Potential security hole as semaphores can be accessed between creation and modification.
- Porting difficulties as there isn't a 1-1 API replacement
This change addresses those problems by adding a new extension method that allows creating a Semaphore and ensuring the provided ACLs are set during creation.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-08 01:59:07 +01:00
Zoltan Varga c0a9d6bf76 [aot] Improve the aot mangler a bit, handle bool/char as a primitive type and avoid emitting a System prefix. (#17739) 2019-11-07 18:18:12 -05:00
Tanner Gooding 8637912a99 Updating NumberToFloatingPointBitsSlow to handle the remaining parsed fractional digits being zero. (#27688)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-07 23:39:21 +01:00
Jo Shields 4d579d00ff Bump LLVM submodule 2019-11-07 15:52:56 -05:00
Tamar Christina b924ffa5e0 Arm64: Add S.P.CoreLib for AddV. (dotnet/coreclr#27663)
Signed-off-by: dotnet-bot <[email protected]>
2019-11-07 21:47:22 +01:00
Ryan Lucia 2c22d0de2e [netcore] Propagate ALCs through reflection functions (#16949)
* [reflection] Pass assembly load context through mono_reflection_get_type_with_rootimage

The entry points are mono_reflection_type_from_name_checked and
mono_reflection_get_type_checked which for now pass in an ambient ALC, but
should be updated to take an ALC argument.

Internally mono_reflection_get_type_with_rootimage is recursive through a half
dozen helper functions which all get an ALC argument too.

The principle place where the ALC is used is in
_mono_reflection_get_type_from_info which passes it to
mono_assembly_loaded_internal and to replace the call to mono_assembly_load by
a direct call to mono_assembly_request_byname with an explicit ALC set in the
request.

* Mark mono_assembly_load as external only

* Pass an alc through non-static reflection functions

* Fix usage in wasm debugger

* Relax assertions due to wasm
2019-11-07 15:02:23 -05:00
Marcos Cobeña Morián 8cc0402967 Enabled --preload-files without AOT (#17732) 2019-11-07 15:00:36 -05:00
Jo Shields fe69f5b699 Fix .gitmodules typo 2019-11-07 14:03:06 -05:00
Jo Shields c1c483cef8 LLVM changes for SDKs 2019-11-07 12:16:29 -05:00
Jo Shields d0d779b8d9 Update AzDo pipeine for new LLVM location 2019-11-07 12:09:58 -05:00
Alexander Köplinger 71b78536ad [sdks] Add xunit to iOS test runner and add results reporting (#17744)
- Allows the on-device runner to work with the xunit framework in preparation for netcore.

- Added a simple UI to show the test run progress

- Report the test result XML back to the host so we can process it on Jenkins/AzDO
2019-11-07 18:09:51 +01:00
Jo Shields b61263e994 [llvm] redirect from mono llvm-mirror fork to dotnet llvm-project fork 2019-11-07 12:03:29 -05:00
Jo Shields 9e6b57ce5a delete old llvm 2019-11-07 11:47:32 -05:00
Aleksey Kliger (λgeek) 2c56ac54c0 [profiler] Fix log profiling of native to managed wrappers (#17740)
* [profiler] Add test case for pinvokes and reverse pinvokes

* [profiler] Fix crash in native to managed wrappers

When we call back to managed from a pinvoke, we're in GC Safe mode and the
profiler instrumentation is inserted in the wrapper before it does the
transition to GC Unsafe.  So we need an extra transition before the profiler
calls its locking functions

Fixes https://github.com/mono/mono/issues/17687

* [coop] Mark internal GC Unsafe functions with MONO_PROFILER_API

Allow the profiler to transition to GC Unsafe mode

* [profiler] ptestrunner.pl fix profiler modules directory

we used to specify mono/mini/.libs but that's not where the profiler plugins live, they're in mono/profiler/.libs.   We probably picked up the profiler plugins from the system Mono install.
2019-11-07 11:42:52 -05:00
Thays Grazia 4bdacea237 [embed] Assert when call mono_runtime_object_init (#17589)
* Fix assert when calling externally call mono_runtime_object_init, mono_object_get_domain, mono_string_to_utf8
* Adding samples/embed as unit-test
2019-11-07 11:22:25 -03:00
Carlos Sanchez Lopez 3c84633c7c Add Mutex creation extension methods that take an ACL (#42281)
Approved API Proposal: #41662

Description
We don't currently have a way to create a Mutex with a given ACL in .NET Core. We can modify the ACL, but it would be more secure to have the proper ACL on the object from the start.

Customer impact
Before this change, customers had to create a Mutex, then set its ACLs. This presents a few problems:

Potential security hole as mutexes can be accessed between creation and modification.
Porting difficulties as there isn't a 1-1 API replacement
This change addresses those problems by adding a new extension method that allows creating a Mutex and ensuring the provided ACLs are set during creation.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-07 12:27:50 +01:00
Elinor Fung 0886ddb3a0 Populate properties of AssemblyLoadStart/Stop events (dotnet/coreclr#27558)
* Populate properties of AssemblyLoadStart/Stop events
* Add tests for basic start/stop binder tracing events
* Add property for ALC of requesting assembly

Signed-off-by: dotnet-bot <[email protected]>
2019-11-07 09:54:23 +01:00
Jay Krell 7b4b5e72d9 Remove handles from ves_icall_System_Array_InternalCreate. (#17681)
Remove handles from ves_icall_System_Array_InternalCreate.

Also combine two allocas.
2019-11-07 07:36:07 +01:00
Zoltan Varga 0680073b87 [runtime] Fix locking in mono_get_seq_points (). (#17738)
Fixes https://github.com/mono/mono/issues/17737.
2019-11-06 19:50:59 -05:00
Johan Lorensson bda239f44c Switch away from Start-Process, Wait-Process in build.ps1. (#17730)
Running Start-Process, Wait-Process on msbuild.exe have showed a couple
of hangs in the build script. This is most likely due to msbuild.exe
spawned child processes (like compile servers) running longer than
spawned msbuild.exe process. Since there is no specific requirements
when spawning the process from build.ps1, switch to regular process
exec and LastExitCode.
2019-11-06 16:51:31 +01:00
Johan Lorensson 1a6c092019 Explicit update/init only LLVM BTLS repro on external MSVC build. (#17691)
Reduce the scope of git submodule update --init to only LLVM and/or BTLS
repositories when building external MSVC dependencies.
2019-11-06 16:51:14 +01:00
Marek Safar 965beadc1e [System.Private.CoreLib] Remove unused Array methods 2019-11-06 12:46:30 +01:00
Stephen Toub e506008ebb Add Span.Sort, and make Array.Sort span-based (dotnet/coreclr#27700)
* Add Span<T>.Sort

Shares existing Array-based implementation by changing that implementation to be span based.

* Additional work to enable span-based sorts

- Cleans up changes from the previous commit, e.g. corrects nullable annotations, removes TODOs, using GetRawSzArrayData instead of GetRawArrayData, etc.
- Passes spans around rather than spans+lo+hi.
- Deletes the native TrySZSort, preferring to use the managed implementation in all cases.

Signed-off-by: dotnet-bot <[email protected]>
2019-11-06 12:46:30 +01:00
Bernhard Urban-Forster 29ba75c154 Bump Corefx (#17694)
* Bump Corefx

New commits in mono/corefx:

* mono/corefx@c44efe7297 [mono] limit DegreeOfParallelism to 16 (#369)

Diff: https://github.com/mono/corefx/compare/6e65509a17da898933705899677c22eae437d68a..c44efe7297f8e11cab157c7e96b998dc936d53b7

* another bump
2019-11-06 07:36:17 +01:00
542 changed files with 9575 additions and 65096 deletions

No files matched your search

+1 -1
View File
@@ -34,7 +34,7 @@
/external/corefx @akoeplinger @marek-safar @egorbo @steveisok
/external/corert @akoeplinger @marek-safar @egorbo @steveisok
/external/linker @marek-safar
/external/llvm @SamMonoRT @vargaz @imhameed
/external/llvm-project @SamMonoRT @vargaz @imhameed
/external/nuget-buildtasks @radical @akoeplinger
/external/roslyn-binaries @kg @marek-safar @akoeplinger @lewing
+7 -4
View File
@@ -58,13 +58,16 @@
[submodule "external/api-snapshot"]
path = external/api-snapshot
url = git://github.com/mono/api-snapshot.git
[submodule "external/llvm"]
path = external/llvm
url = git://github.com/mono/llvm.git
branch = release_60
[submodule "external/helix-binaries"]
path = external/helix-binaries
url = git://github.com/mono/helix-binaries.git
[submodule "external/illinker-test-assets"]
path = external/illinker-test-assets
url = git://github.com/mono/illinker-test-assets.git
[submodule "external/llvm-project"]
path = external/llvm-project
url = git://github.com/dotnet/llvm-project.git
branch = release/6.x
[submodule "external/bdwgc"]
path = external/bdwgc
url = git://github.com/Unity-Technologies/bdwgc.git
+3
View File
@@ -67,6 +67,9 @@ dist-hook:
rm -f `find $(top_distdir)/external -path '*\.exe' -not -path '*/roslyn-binaries/*'`
rm -f `find $(top_distdir)/external -path '*\.dll' -not -path '*/binary-reference-assemblies/*' -not -path '*/roslyn-binaries/*' -not -path '*/helix-binaries/*'`
rm -rf "$(top_distdir)/external/linker/test"
rm -rf "$(top_distdir)/external/llvm-project/lldb/test"
rm -rf "$(top_distdir)/external/llvm-project/libcxx/test"
rm -rf "$(top_distdir)/external/llvm-project/clang/test"
pkgconfigdir = $(libdir)/pkgconfig
noinst_DATA = mono-uninstalled.pc
+4 -4
View File
@@ -150,10 +150,10 @@ automake $am_opt ||
echo "Running autoconf ..."
autoconf || { echo "**Error**: autoconf failed."; exit 1; }
if test -d $srcdir/libgc; then
echo Running libgc/autogen.sh ...
(cd $srcdir/libgc ; NOCONFIGURE=1 ./autogen.sh "$@")
echo Done running libgc/autogen.sh ...
if test -d $srcdir/external/bdwgc; then
echo Running external/bdwgc/autogen.sh ...
(cd $srcdir/external/bdwgc ; NOCONFIGURE=1 ./autogen.sh "$@")
echo Done running external/bdwgc/autogen.sh ...
fi
if test x$MONO_EXTRA_CONFIGURE_FLAGS != x; then
+27 -20
View File
@@ -17,7 +17,7 @@ AC_CANONICAL_HOST
# The extra brackets are to foil regex-based scans.
m4_ifdef([_A][M_PROG_TAR],[_A][M_SET_OPTION([tar-ustar])])
AM_INIT_AUTOMAKE([1.9 dist-xz tar-ustar no-dist-gzip foreign]
AM_INIT_AUTOMAKE([1.9 dist-xz tar-ustar no-dist-gzip foreign subdir-objects]
m4_esyscmd([case `automake --version | head -n 1` in # parallel-tests is default in automake 1.13+, we need to explicitly enable it
*1.11*|*1.12*) echo parallel-tests;; # for 1.11 and 1.12 but not below as those versions don't recognize the flag
esac])) # TODO: remove this hack once we require automake 1.11+
@@ -63,7 +63,7 @@ MONO_VERSION_BUILD=`echo $VERSION | cut -d . -f 3`
# This line is parsed by tools besides autoconf, such as msvc/mono.winconfig.targets.
# It should remain in the format they expect.
#
MONO_CORLIB_VERSION=4e2d4257-1f2b-418c-9829-dd006fe09df5
MONO_CORLIB_VERSION=423e7794-9279-49a3-a477-f1cb2432e9f4
#
# Put a quoted #define in config.h.
@@ -148,7 +148,7 @@ case "$host" in
platform_wasm=yes
# FIXME:
RID="osx-x64"
COREARCH="x64"
COREARCH="wasm32"
CORETARGETS="-p:TargetsUnix=true -p:TargetsOSX=true"
build_darwin=yes
;;
@@ -173,7 +173,6 @@ case "$host" in
# Boehm not supported on 64-bit Windows.
case "$host" in
x86_64-*-* | amd64-*-*)
support_boehm=no
with_gc=sgen
RID="win-x64"
COREARCH="x64"
@@ -192,9 +191,6 @@ case "$host" in
with_tls=pthread
with_sgen_default_concurrent=yes
LN_S=cp
# This forces libgc to use the DllMain based thread registration code on win32
libgc_configure_args="$libgc_configure_args --enable-win32-dllmain=yes"
;;
*-*-*netbsd*)
CPPFLAGS="$CPPFLAGS -D_REENTRANT -DGC_NETBSD_THREADS -D_GNU_SOURCE"
@@ -1351,7 +1347,7 @@ enable_cooperative_suspend_default=no
enable_hybrid_suspend_default=no
# For the sake of clearer error messages, these numbers should all be different from each other.
INVARIANT_AOT_OPTIONS=nimt-trampolines=2000,ntrampolines=10000,nrgctx-fetch-trampolines=256,ngsharedvt-trampolines=4400,nftnptr-arg-trampolines=4000
INVARIANT_AOT_OPTIONS=nimt-trampolines=2000,ntrampolines=10000,nrgctx-fetch-trampolines=256,ngsharedvt-trampolines=4400,nftnptr-arg-trampolines=4000,nrgctx-trampolines=21000
AOT_BUILD_ATTRS=$INVARIANT_AOT_OPTIONS
@@ -2039,14 +2035,14 @@ if test "x$support_boehm" = "xyes"; then
case "x$libgc" in
xincluded)
if test "x$support_boehm" = "xyes"; then
libgc_dir=libgc
libgc_dir=external/bdwgc
fi
LIBGC_CPPFLAGS='-I$(top_srcdir)/libgc/include'
LIBGC_LIBS='$(top_builddir)/libgc/libmonogc.la'
LIBGC_STATIC_LIBS='$(top_builddir)/libgc/libmonogc-static.la'
LIBGC_CPPFLAGS='-I$(top_srcdir)/external/bdwgc/include -I$(top_srcdir)/external/bdwgc/libatomic_ops/src'
LIBGC_LIBS='$(top_builddir)/external/bdwgc/libgc.la'
LIBGC_STATIC_LIBS='$(top_builddir)/external/bdwgc/libgc-static.la'
BOEHM_DEFINES="-DHAVE_BOEHM_GC"
BOEHM_DEFINES="-DHAVE_BOEHM_GC -DGC_THREADS"
if test x$target_win32 = xyes; then
BOEHM_DEFINES="$BOEHM_DEFINES -DGC_NOT_DLL"
@@ -3409,6 +3405,15 @@ if test x$host_win32 = xno; then
AC_CHECK_DECL(ICANON, [AC_DEFINE(HAVE_ICANON, 1, [ICANON])], [], [[#include <termios.h>]])
AC_CHECK_DECL(TCSANOW, [AC_DEFINE(HAVE_TCSANOW, 1, [TCSANOW])], [], [[#include <termios.h>]])
AC_CHECK_DECL(lchflags, [AC_DEFINE(HAVE_LCHFLAGS, 1, [lchflags])], [],
[[#include <sys/types.h>
#include <sys/stat.h>]])
AC_CHECK_MEMBER(struct stat.st_flags,
[AC_DEFINE(HAVE_STAT_FLAGS, 1, [struct stat.st_flags])],
[],
[#include <sys/types.h>
#include <sys/stat.h>])
AC_CHECK_MEMBER(struct stat.st_birthtimespec,
[AC_DEFINE(HAVE_STAT_BIRTHTIME, 1, [struct stat.st_birthtimespec])],
[],
@@ -5144,7 +5149,12 @@ AC_SEARCH_LIBS(dlopen, dl)
# Apple -bind_at_load
# The rest GNU.
#
AX_APPEND_LINK_FLAGS([-Wl,--export-dynamic -Wl,-Bsymbolic -bsymbolic -Wl,-z,now -Wl,-bind_at_load])
export_link_flags=--export-dynamic
case $host_os in
mingw*|cygwin*)
export_link_flags=--export-all-symbols
esac
AX_APPEND_LINK_FLAGS([-Wl,$export_link_flags -Wl,-Bsymbolic -bsymbolic -Wl,-z,now -Wl,-bind_at_load])
AC_CHECK_HEADERS(getopt.h sys/select.h sys/time.h sys/wait.h pwd.h iconv.h sys/types.h sys/resource.h)
dnl giconv.c will check on HAVE_ICONV_H but we need this for link time
@@ -5596,9 +5606,9 @@ case "x$libgc" in
# if these are instrumented it will be very bad news
# (infinite recursion, undefined parking behavior, etc)
TMP_CPPFLAGS=`echo $TMP_CPPFLAGS | sed -e 's/-finstrument-for-thread-suspension//g'`
ac_configure_args="$ac_configure_args --disable-embed-check --with-libgc-threads=$libgc_threads $libgc_configure_args \"CPPFLAGS_FOR_LIBGC=$TMP_CPPFLAGS\" \"CFLAGS_FOR_LIBGC=$CFLAGS_FOR_LIBGC\""
ac_configure_args="$ac_configure_args --enable-single-obj-compilation --disable-embed-check --with-libgc-threads=$libgc_threads $libgc_configure_args \"CPPFLAGS_FOR_LIBGC=$TMP_CPPFLAGS\" \"CFLAGS_FOR_LIBGC=$CFLAGS_FOR_LIBGC\""
if test "x$support_boehm" = "xyes"; then
AC_CONFIG_SUBDIRS(libgc)
AC_CONFIG_SUBDIRS(external/bdwgc)
fi
;;
esac
@@ -6034,10 +6044,6 @@ fi
AM_CONDITIONAL(ENABLE_MSVC, test x$enable_msvc = xyes)
if test "x$enable_msvc" = "xyes"; then
if test "x$support_boehm" = "xyes"; then
echo "Error, Boehm GC is not supported when using --enable-msvc."
exit 1
fi
mono_msvc_build_dir='$(top_srcdir)'/msvc/build/sgen
if test "x$host_cpu" = "xi686"; then
mono_msvc_build_lib_dir='$(mono_msvc_build_dir)'/Win32/lib/Release
@@ -6707,6 +6713,7 @@ mono/native/Makefile
mono/utils/Makefile
mono/utils/jemalloc/Makefile
mono/metadata/Makefile
mono/zlib/Makefile
mono/dis/Makefile
mono/cil/Makefile
mono/arch/Makefile
+8 -8
View File
@@ -1,13 +1,13 @@
<?xml version="1.0" encoding="utf-8"?>
<Dependencies>
<ProductDependencies>
<Dependency Name="Microsoft.Private.CoreFx.NETCoreApp" Version="5.0.0-alpha.1.19527.5">
<Dependency Name="Microsoft.Private.CoreFx.NETCoreApp" Version="5.0.0-alpha.1.19563.3">
<Uri>https://github.com/dotnet/corefx</Uri>
<Sha>17c14aa2b50cfb5920d03ea8835c9a4932b5359c</Sha>
<Sha>5cee7c97d602f294e27c582d4dab81ec388f1d7b</Sha>
</Dependency>
<Dependency Name="Microsoft.NETCore.App" Version="5.0.0-alpha1.19527.2">
<Dependency Name="Microsoft.NETCore.App" Version="5.0.0-alpha.1.19564.1">
<Uri>https://github.com/dotnet/core-setup</Uri>
<Sha>b3749864f47f38599a94070497225c665124fe82</Sha>
<Sha>c77948d92a2f950140f09384f057cb893ec3955a</Sha>
</Dependency>
<Dependency Name="Microsoft.Net.Compilers" Version="3.3.1-beta4-19476-06">
<Uri>https://github.com/dotnet/roslyn</Uri>
@@ -15,13 +15,13 @@
</Dependency>
</ProductDependencies>
<ToolsetDependencies>
<Dependency Name="Microsoft.DotNet.Arcade.Sdk" Version="5.0.0-beta.19527.3">
<Dependency Name="Microsoft.DotNet.Arcade.Sdk" Version="5.0.0-beta.19567.2">
<Uri>https://github.com/dotnet/arcade</Uri>
<Sha>5ce0aaed2f8df9cd2fe0221a426c021efe37f655</Sha>
<Sha>928efaa081ffd17b37f6c662fee44775f2d0b90f</Sha>
</Dependency>
<Dependency Name="Microsoft.DotNet.Helix.Sdk" Version="5.0.0-beta.19527.3">
<Dependency Name="Microsoft.DotNet.Helix.Sdk" Version="5.0.0-beta.19567.2">
<Uri>https://github.com/dotnet/arcade</Uri>
<Sha>5ce0aaed2f8df9cd2fe0221a426c021efe37f655</Sha>
<Sha>928efaa081ffd17b37f6c662fee44775f2d0b90f</Sha>
</Dependency>
</ToolsetDependencies>
</Dependencies>
+2 -2
View File
@@ -5,9 +5,9 @@
<PreReleaseVersionLabel>prerelease</PreReleaseVersionLabel>
<!-- Package versions -->
<!-- corefx -->
<MicrosoftPrivateCoreFxNETCoreAppVersion>5.0.0-alpha.1.19527.5</MicrosoftPrivateCoreFxNETCoreAppVersion>
<MicrosoftPrivateCoreFxNETCoreAppVersion>5.0.0-alpha.1.19563.3</MicrosoftPrivateCoreFxNETCoreAppVersion>
<!-- core-setup -->
<MicrosoftNETCoreAppVersion>5.0.0-alpha1.19527.2</MicrosoftNETCoreAppVersion>
<MicrosoftNETCoreAppVersion>5.0.0-alpha.1.19564.1</MicrosoftNETCoreAppVersion>
<!-- roslyn -->
<MicrosoftNetCompilersVersion>3.3.1-beta4-19476-06</MicrosoftNetCompilersVersion>
</PropertyGroup>
+4 -2
View File
@@ -37,6 +37,8 @@
<PropertyGroup>
<DotNetSymbolExpirationInDays Condition="'$(DotNetSymbolExpirationInDays)' == ''">3650</DotNetSymbolExpirationInDays>
<PublishToSymbolServer>true</PublishToSymbolServer>
<PublishToSymWeb Condition="'$(PublishToSymWeb)' == ''">true</PublishToSymWeb>
<PublishToMSDL Condition="'$(PublishToMSDL)' == ''">true</PublishToMSDL>
<PublishToSymbolServer Condition="'@(FilesToPublishToSymbolServer)' == '' and '@(PackagesToPublishToSymbolServer)' == ''">false</PublishToSymbolServer>
</PropertyGroup>
@@ -56,7 +58,7 @@
DryRun="false"
ConvertPortablePdbsToWindowsPdbs="false"
PdbConversionTreatAsWarning=""
Condition="$(PublishToSymbolServer)"/>
Condition="$(PublishToSymbolServer) and $(PublishToMSDL)"/>
<!--
Symbol Uploader: SymWeb
@@ -73,7 +75,7 @@
DryRun="false"
ConvertPortablePdbsToWindowsPdbs="false"
PdbConversionTreatAsWarning=""
Condition="$(PublishToSymbolServer)"/>
Condition="$(PublishToSymbolServer) and $(PublishToSymWeb)"/>
</Target>
<ItemGroup>
+127
View File
@@ -0,0 +1,127 @@
# This file is a temporary workaround for internal builds to be able to restore from private AzDO feeds.
# This file should be removed as part of this issue: https://github.com/dotnet/arcade/issues/4080
#
# What the script does is iterate over all package sources in the pointed NuGet.config and add a credential entry
# under <packageSourceCredentials> for each Maestro managed private feed. Two additional credential
# entries are also added for the two private static internal feeds: dotnet3-internal and dotnet3-internal-transport.
#
# This script needs to be called in every job that will restore packages and which the base repo has
# private AzDO feeds in the NuGet.config.
#
# See example YAML call for this script below. Note the use of the variable `$(dn-bot-dnceng-artifact-feeds-rw)`
# from the AzureDevOps-Artifact-Feeds-Pats variable group.
#
# - task: PowerShell@2
# displayName: Setup Private Feeds Credentials
# condition: eq(variables['Agent.OS'], 'Windows_NT')
# inputs:
# filePath: $(Build.SourcesDirectory)/eng/common/SetupNugetSources.ps1
# arguments: -ConfigFile ${Env:BUILD_SOURCESDIRECTORY}/NuGet.config -Password $Env:Token
# env:
# Token: $(dn-bot-dnceng-artifact-feeds-rw)
[CmdletBinding()]
param (
[Parameter(Mandatory = $true)][string]$ConfigFile,
[Parameter(Mandatory = $true)][string]$Password
)
$ErrorActionPreference = "Stop"
Set-StrictMode -Version 2.0
[Net.ServicePointManager]::SecurityProtocol = [Net.SecurityProtocolType]::Tls12
. $PSScriptRoot\tools.ps1
# Add source entry to PackageSources
function AddPackageSource($sources, $SourceName, $SourceEndPoint, $creds, $Username, $Password) {
$packageSource = $sources.SelectSingleNode("add[@key='$SourceName']")
if ($packageSource -eq $null)
{
$packageSource = $doc.CreateElement("add")
$packageSource.SetAttribute("key", $SourceName)
$packageSource.SetAttribute("value", $SourceEndPoint)
$sources.AppendChild($packageSource) | Out-Null
}
else {
Write-Host "Package source $SourceName already present."
}
AddCredential -Creds $creds -Source $SourceName -Username $Username -Password $Password
}
# Add a credential node for the specified source
function AddCredential($creds, $source, $username, $password) {
# Looks for credential configuration for the given SourceName. Create it if none is found.
$sourceElement = $creds.SelectSingleNode($Source)
if ($sourceElement -eq $null)
{
$sourceElement = $doc.CreateElement($Source)
$creds.AppendChild($sourceElement) | Out-Null
}
# Add the <Username> node to the credential if none is found.
$usernameElement = $sourceElement.SelectSingleNode("add[@key='Username']")
if ($usernameElement -eq $null)
{
$usernameElement = $doc.CreateElement("add")
$usernameElement.SetAttribute("key", "Username")
$sourceElement.AppendChild($usernameElement) | Out-Null
}
$usernameElement.SetAttribute("value", $Username)
# Add the <ClearTextPassword> to the credential if none is found.
# Add it as a clear text because there is no support for encrypted ones in non-windows .Net SDKs.
# -> https://github.com/NuGet/Home/issues/5526
$passwordElement = $sourceElement.SelectSingleNode("add[@key='ClearTextPassword']")
if ($passwordElement -eq $null)
{
$passwordElement = $doc.CreateElement("add")
$passwordElement.SetAttribute("key", "ClearTextPassword")
$sourceElement.AppendChild($passwordElement) | Out-Null
}
$passwordElement.SetAttribute("value", $Password)
}
function InsertMaestroPrivateFeedCredentials($Sources, $Creds, $Password) {
$maestroPrivateSources = $Sources.SelectNodes("add[contains(@key,'darc-int')]")
Write-Host "Inserting credentials for $($maestroPrivateSources.Count) Maestro's private feeds."
ForEach ($PackageSource in $maestroPrivateSources) {
Write-Host "`tInserting credential for Maestro's feed:" $PackageSource.Key
AddCredential -Creds $creds -Source $PackageSource.Key -Username $Username -Password $Password
}
}
if (!(Test-Path $ConfigFile -PathType Leaf)) {
Write-Host "Couldn't find the file NuGet config file: $ConfigFile"
ExitWithExitCode 1
}
# Load NuGet.config
$doc = New-Object System.Xml.XmlDocument
$filename = (Get-Item $ConfigFile).FullName
$doc.Load($filename)
# Get reference to <PackageSources> or create one if none exist already
$sources = $doc.DocumentElement.SelectSingleNode("packageSources")
if ($sources -eq $null) {
$sources = $doc.CreateElement("packageSources")
$doc.DocumentElement.AppendChild($sources) | Out-Null
}
# Looks for a <PackageSourceCredentials> node. Create it if none is found.
$creds = $doc.DocumentElement.SelectSingleNode("packageSourceCredentials")
if ($creds -eq $null) {
$creds = $doc.CreateElement("packageSourceCredentials")
$doc.DocumentElement.AppendChild($creds) | Out-Null
}
# Insert credential nodes for Maestro's private feeds
InsertMaestroPrivateFeedCredentials -Sources $sources -Creds $creds -Password $Password
AddPackageSource -Sources $sources -SourceName "dotnet3-internal" -SourceEndPoint "https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal/nuget/v2" -Creds $creds -Username "dn-bot" -Password $Password
AddPackageSource -Sources $sources -SourceName "dotnet3-internal-transport" -SourceEndPoint "https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-transport/nuget/v2" -Creds $creds -Username "dn-bot" -Password $Password
$doc.Save($filename)
+117
View File
@@ -0,0 +1,117 @@
#!/usr/bin/env bash
# This file is a temporary workaround for internal builds to be able to restore from private AzDO feeds.
# This file should be removed as part of this issue: https://github.com/dotnet/arcade/issues/4080
#
# What the script does is iterate over all package sources in the pointed NuGet.config and add a credential entry
# under <packageSourceCredentials> for each Maestro's managed private feed. Two additional credential
# entries are also added for the two private static internal feeds: dotnet3-internal and dotnet3-internal-transport.
#
# This script needs to be called in every job that will restore packages and which the base repo has
# private AzDO feeds in the NuGet.config.
#
# See example YAML call for this script below. Note the use of the variable `$(dn-bot-dnceng-artifact-feeds-rw)`
# from the AzureDevOps-Artifact-Feeds-Pats variable group.
#
# - task: Bash@3
# displayName: Setup Private Feeds Credentials
# inputs:
# filePath: $(Build.SourcesDirectory)/eng/common/SetupNugetSources.sh
# arguments: $BUILD_SOURCESDIRECTORY/NuGet.config $Token
# condition: ne(variables['Agent.OS'], 'Windows_NT')
# env:
# Token: $(dn-bot-dnceng-artifact-feeds-rw)
ConfigFile=$1
CredToken=$2
NL='\n'
TB=' '
source="${BASH_SOURCE[0]}"
# resolve $source until the file is no longer a symlink
while [[ -h "$source" ]]; do
scriptroot="$( cd -P "$( dirname "$source" )" && pwd )"
source="$(readlink "$source")"
# if $source was a relative symlink, we need to resolve it relative to the path where the
# symlink file was located
[[ $source != /* ]] && source="$scriptroot/$source"
done
scriptroot="$( cd -P "$( dirname "$source" )" && pwd )"
. "$scriptroot/tools.sh"
if [ ! -f "$ConfigFile" ]; then
echo "Couldn't find the file NuGet config file: $ConfigFile"
ExitWithExitCode 1
fi
if [[ `uname -s` == "Darwin" ]]; then
NL=$'\\\n'
TB=''
fi
# Ensure there is a <packageSources>...</packageSources> section.
grep -i "<packageSources>" $ConfigFile
if [ "$?" != "0" ]; then
echo "Adding <packageSources>...</packageSources> section."
ConfigNodeHeader="<configuration>"
PackageSourcesTemplate="${TB}<packageSources>${NL}${TB}</packageSources>"
sed -i.bak "s|$ConfigNodeHeader|$ConfigNodeHeader${NL}$PackageSourcesTemplate|" NuGet.config
fi
# Ensure there is a <packageSourceCredentials>...</packageSourceCredentials> section.
grep -i "<packageSourceCredentials>" $ConfigFile
if [ "$?" != "0" ]; then
echo "Adding <packageSourceCredentials>...</packageSourceCredentials> section."
PackageSourcesNodeFooter="</packageSources>"
PackageSourceCredentialsTemplate="${TB}<packageSourceCredentials>${NL}${TB}</packageSourceCredentials>"
sed -i.bak "s|$PackageSourcesNodeFooter|$PackageSourcesNodeFooter${NL}$PackageSourceCredentialsTemplate|" NuGet.config
fi
# Ensure dotnet3-internal and dotnet3-internal-transport is in the packageSources
grep -i "<add key=\"dotnet3-internal\">" $ConfigFile
if [ "$?" != "0" ]; then
echo "Adding dotnet3-internal to the packageSources."
PackageSourcesNodeFooter="</packageSources>"
PackageSourceTemplate="${TB}<add key=\"dotnet3-internal\" value=\"https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal/nuget/v2\" />"
sed -i.bak "s|$PackageSourcesNodeFooter|$PackageSourceTemplate${NL}$PackageSourcesNodeFooter|" NuGet.config
fi
# Ensure dotnet3-internal and dotnet3-internal-transport is in the packageSources
grep -i "<add key=\"dotnet3-internal-transport\">" $ConfigFile
if [ "$?" != "0" ]; then
echo "Adding dotnet3-internal-transport to the packageSources."
PackageSourcesNodeFooter="</packageSources>"
PackageSourceTemplate="${TB}<add key=\"dotnet3-internal-transport\" value=\"https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-transport/nuget/v2\" />"
sed -i.bak "s|$PackageSourcesNodeFooter|$PackageSourceTemplate${NL}$PackageSourcesNodeFooter|" NuGet.config
fi
# I want things split line by line
PrevIFS=$IFS
IFS=$'\n'
PackageSources=$(grep -oh '"darc-int-[^"]*"' $ConfigFile | tr -d '"')
IFS=$PrevIFS
PackageSources+=('dotnet3-internal')
PackageSources+=('dotnet3-internal-transport')
for FeedName in ${PackageSources[@]} ; do
# Check if there is no existing credential for this FeedName
grep -i "<$FeedName>" $ConfigFile
if [ "$?" != "0" ]; then
echo "Adding credentials for $FeedName."
PackageSourceCredentialsNodeFooter="</packageSourceCredentials>"
NewCredential="${TB}${TB}<$FeedName>${NL}<add key=\"Username\" value=\"dn-bot\" />${NL}<add key=\"ClearTextPassword\" value=\"$CredToken\" />${NL}</$FeedName>"
sed -i.bak "s|$PackageSourceCredentialsNodeFooter|$NewCredential${NL}$PackageSourceCredentialsNodeFooter|" NuGet.config
fi
done
+15 -34
View File
@@ -31,6 +31,10 @@ else()
message(FATAL_ERROR "Arch is ${TARGET_ARCH_NAME}. Only armel, arm, arm64 and x86 are supported!")
endif()
if(DEFINED ENV{TOOLCHAIN})
set(TOOLCHAIN $ENV{TOOLCHAIN})
endif()
# Specify include paths
if(TARGET_ARCH_NAME STREQUAL "armel")
if(DEFINED TIZEN_TOOLCHAIN)
@@ -39,48 +43,25 @@ if(TARGET_ARCH_NAME STREQUAL "armel")
endif()
endif()
# add_compile_param - adds only new options without duplicates.
# arg0 - list with result options, arg1 - list with new options.
# arg2 - optional argument, quick summary string for optional using CACHE FORCE mode.
macro(add_compile_param)
if(NOT ${ARGC} MATCHES "^(2|3)$")
message(FATAL_ERROR "Wrong using add_compile_param! Two or three parameters must be given! See add_compile_param description.")
endif()
foreach(OPTION ${ARGV1})
if(NOT ${ARGV0} MATCHES "${OPTION}($| )")
set(${ARGV0} "${${ARGV0}} ${OPTION}")
if(${ARGC} EQUAL "3") # CACHE FORCE mode
set(${ARGV0} "${${ARGV0}}" CACHE STRING "${ARGV2}" FORCE)
endif()
endif()
endforeach()
endmacro()
set(CMAKE_SYSROOT "${CROSS_ROOTFS}")
set(CMAKE_C_COMPILER_EXTERNAL_TOOLCHAIN "${CROSS_ROOTFS}/usr")
set(CMAKE_CXX_COMPILER_EXTERNAL_TOOLCHAIN "${CROSS_ROOTFS}/usr")
set(CMAKE_ASM_COMPILER_EXTERNAL_TOOLCHAIN "${CROSS_ROOTFS}/usr")
# Specify link flags
add_compile_param(CROSS_LINK_FLAGS "--sysroot=${CROSS_ROOTFS}")
add_compile_param(CROSS_LINK_FLAGS "--gcc-toolchain=${CROSS_ROOTFS}/usr")
add_compile_param(CROSS_LINK_FLAGS "--target=${TOOLCHAIN}")
add_compile_param(CROSS_LINK_FLAGS "-fuse-ld=gold")
if(TARGET_ARCH_NAME STREQUAL "armel")
if(DEFINED TIZEN_TOOLCHAIN) # For Tizen only
add_compile_param(CROSS_LINK_FLAGS "-B${CROSS_ROOTFS}/usr/lib/gcc/${TIZEN_TOOLCHAIN}")
add_compile_param(CROSS_LINK_FLAGS "-L${CROSS_ROOTFS}/lib")
add_compile_param(CROSS_LINK_FLAGS "-L${CROSS_ROOTFS}/usr/lib")
add_compile_param(CROSS_LINK_FLAGS "-L${CROSS_ROOTFS}/usr/lib/gcc/${TIZEN_TOOLCHAIN}")
add_link_options("-B${CROSS_ROOTFS}/usr/lib/gcc/${TIZEN_TOOLCHAIN}")
add_link_options("-L${CROSS_ROOTFS}/lib")
add_link_options("-L${CROSS_ROOTFS}/usr/lib")
add_link_options("-L${CROSS_ROOTFS}/usr/lib/gcc/${TIZEN_TOOLCHAIN}")
endif()
elseif(TARGET_ARCH_NAME STREQUAL "x86")
add_compile_param(CROSS_LINK_FLAGS "-m32")
add_link_options(-m32)
endif()
add_compile_param(CMAKE_EXE_LINKER_FLAGS "${CROSS_LINK_FLAGS}" "TOOLCHAIN_EXE_LINKER_FLAGS")
add_compile_param(CMAKE_SHARED_LINKER_FLAGS "${CROSS_LINK_FLAGS}" "TOOLCHAIN_EXE_LINKER_FLAGS")
add_compile_param(CMAKE_MODULE_LINKER_FLAGS "${CROSS_LINK_FLAGS}" "TOOLCHAIN_EXE_LINKER_FLAGS")
# Specify compile options
add_compile_options("--sysroot=${CROSS_ROOTFS}")
add_compile_options("--target=${TOOLCHAIN}")
add_compile_options("--gcc-toolchain=${CROSS_ROOTFS}/usr")
if(TARGET_ARCH_NAME MATCHES "^(arm|armel|arm64)$")
set(CMAKE_C_COMPILER_TARGET ${TOOLCHAIN})
@@ -103,7 +84,7 @@ elseif(TARGET_ARCH_NAME STREQUAL "x86")
add_compile_options(-Wno-error=unused-command-line-argument)
endif()
# Set LLDB include and library paths
# Set LLDB include and library paths for builds that need lldb.
if(TARGET_ARCH_NAME MATCHES "^(arm|armel|x86)$")
if(TARGET_ARCH_NAME STREQUAL "x86")
set(LLVM_CROSS_DIR "$ENV{LLVM_CROSS_HOME}")
@@ -131,7 +112,7 @@ if(TARGET_ARCH_NAME MATCHES "^(arm|armel|x86)$")
endif()
endif()
set(CMAKE_FIND_ROOT_PATH "${CROSS_ROOTFS}")
set(CMAKE_FIND_ROOT_PATH_MODE_PROGRAM NEVER)
set(CMAKE_FIND_ROOT_PATH_MODE_LIBRARY ONLY)
set(CMAKE_FIND_ROOT_PATH_MODE_INCLUDE ONLY)
+21 -7
View File
@@ -6,6 +6,11 @@ parameters:
# This can also be remedied by the caller (post-build.yml) if it does not use a nested parameter
sdlContinueOnError: false # optional: determines whether to continue the build if the step errors;
dependsOn: '' # Optional: dependencies of the job
artifactNames: '' # Optional: patterns supplied to DownloadBuildArtifacts
# Usage:
# artifactNames:
# - 'BlobArtifacts'
# - 'Artifacts_Windows_NT_Release'
jobs:
- job: Run_SDL
@@ -18,13 +23,22 @@ jobs:
steps:
- checkout: self
clean: true
- task: DownloadBuildArtifacts@0
displayName: Download Build Artifacts
inputs:
buildType: current
downloadType: specific files
matchingPattern: "**"
downloadPath: $(Build.ArtifactStagingDirectory)\artifacts
- ${{ if ne(parameters.artifactNames, '') }}:
- ${{ each artifactName in parameters.artifactNames }}:
- task: DownloadBuildArtifacts@0
displayName: Download Build Artifacts
inputs:
buildType: current
artifactName: ${{ artifactName }}
downloadPath: $(Build.ArtifactStagingDirectory)\artifacts
- ${{ if eq(parameters.artifactNames, '') }}:
- task: DownloadBuildArtifacts@0
displayName: Download Build Artifacts
inputs:
buildType: current
downloadType: specific files
itemPattern: "**"
downloadPath: $(Build.ArtifactStagingDirectory)\artifacts
- powershell: eng/common/sdl/extract-artifact-packages.ps1
-InputPath $(Build.ArtifactStagingDirectory)\artifacts\BlobArtifacts
-ExtractPath $(Build.ArtifactStagingDirectory)\artifacts\BlobArtifacts
@@ -2,21 +2,28 @@ parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stageName: ''
channelName: ''
channelId: ''
transportFeed: ''
shippingFeed: ''
symbolsFeed: ''
stages:
- stage: NetCore_30_Internal_Servicing_Publishing
- stage: ${{ parameters.stageName }}
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Core 3.0 Internal Servicing Publishing
displayName: ${{ parameters.channelName }} Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
- job: publish_symbols
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.InternalServicing_30_Channel_Id))
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', ${{ parameters.channelId }} ))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
@@ -57,19 +64,23 @@ stages:
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
/p:PublishToMSDL=false
${{ parameters.symbolPublishingAdditionalParameters }}
- template: ../../steps/publish-logs.yml
parameters:
StageLabel: '${{ parameters.stageName }}'
JobLabel: 'SymbolPublishing'
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.InternalServicing_30_Channel_Id))
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', ${{ parameters.channelId }}))
pool:
vmImage: 'windows-2019'
steps:
@@ -78,12 +89,14 @@ stages:
inputs:
buildType: current
artifactName: PackageArtifacts
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
@@ -128,14 +141,20 @@ stages:
/p:ChecksumsAzureAccountKey=$(InternalChecksumsBlobFeedKey)
/p:InstallersTargetStaticFeed=$(InternalInstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(InternalInstallersBlobFeedKey)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeed='${{ parameters.shippingFeed }}'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-transport/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeed='${{ parameters.transportFeed }}'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeed='${{ parameters.symbolsFeed }}'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:PublishToMSDL=false
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/publish-logs.yml
parameters:
StageLabel: '${{ parameters.stageName }}'
JobLabel: 'AssetsPublishing'
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.InternalServicing_30_Channel_Id }}
ChannelId: ${{ parameters.channelId }}
@@ -4,20 +4,26 @@ parameters:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stageName: ''
channelName: ''
channelId: ''
transportFeed: ''
shippingFeed: ''
symbolsFeed: ''
stages:
- stage: NetCore_Release31_Publish
- stage: ${{ parameters.stageName }}
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Core 3.1 Release Publishing
displayName: ${{ parameters.channelName }} Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
- job: publish_symbols
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicRelease_31_Channel_Id))
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', ${{ parameters.channelId }} ))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
@@ -35,6 +41,18 @@ stages:
artifactName: 'PDBArtifacts'
continueOnError: true
# This is necessary whenever we want to publish/restore to an AzDO private feed
# Since sdk-task.ps1 tries to restore packages we need to do this authentication here
# otherwise it'll complain about accessing a private feed.
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish
inputs:
@@ -48,17 +66,21 @@ stages:
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- template: ../../steps/publish-logs.yml
parameters:
StageLabel: '${{ parameters.stageName }}'
JobLabel: 'SymbolPublishing'
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicRelease_31_Channel_Id))
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', ${{ parameters.channelId }}))
pool:
vmImage: 'windows-2019'
steps:
@@ -67,12 +89,14 @@ stages:
inputs:
buildType: current
artifactName: PackageArtifacts
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
@@ -118,14 +142,19 @@ stages:
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeed='${{ parameters.shippingFeed }}'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-transport/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeed='${{ parameters.transportFeed }}'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeed='${{ parameters.symbolsFeed }}'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/publish-logs.yml
parameters:
StageLabel: '${{ parameters.stageName }}'
JobLabel: 'AssetsPublishing'
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.PublicRelease_31_Channel_Id }}
ChannelId: ${{ parameters.channelId }}
@@ -1,96 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
stages:
- stage: NetCore_3_Tools_Validation_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET 3 Tools - Validation Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NETCore_3_Tools_Validation_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetValidationArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.NETCore_3_Tools_Validation_Channel_Id }}
@@ -1,131 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stages:
- stage: NetCore_3_Tools_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET 3 Tools Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_3_Tools_Channel_Id))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
artifactName: 'BlobArtifacts'
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download PDB Artifacts
inputs:
artifactName: 'PDBArtifacts'
continueOnError: true
- task: PowerShell@2
displayName: Publish
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishToSymbolServers -restore -msbuildEngine dotnet
/p:DotNetSymbolServerTokenMsdl=$(microsoft-symbol-server-pat)
/p:DotNetSymbolServerTokenSymWeb=$(symweb-symbol-server-pat)
/p:PDBArtifactsDirectory='$(Build.ArtifactStagingDirectory)/PDBArtifacts/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_3_Tools_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.NetCore_3_Tools_Channel_Id }}
@@ -1,131 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stages:
- stage: NetCore_Dev31_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Core 3.1 Dev Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicDevRelease_31_Channel_Id))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
artifactName: 'BlobArtifacts'
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download PDB Artifacts
inputs:
artifactName: 'PDBArtifacts'
continueOnError: true
- task: PowerShell@2
displayName: Publish
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishToSymbolServers -restore -msbuildEngine dotnet
/p:DotNetSymbolServerTokenMsdl=$(microsoft-symbol-server-pat)
/p:DotNetSymbolServerTokenSymWeb=$(symweb-symbol-server-pat)
/p:PDBArtifactsDirectory='$(Build.ArtifactStagingDirectory)/PDBArtifacts/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicDevRelease_31_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-transport/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.PublicDevRelease_31_Channel_Id }}
@@ -1,131 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stages:
- stage: NetCore_Dev5_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Core 5 Dev Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_5_Dev_Channel_Id))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
artifactName: 'BlobArtifacts'
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download PDB Artifacts
inputs:
artifactName: 'PDBArtifacts'
continueOnError: true
- task: PowerShell@2
displayName: Publish
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishToSymbolServers -restore -msbuildEngine dotnet
/p:DotNetSymbolServerTokenMsdl=$(microsoft-symbol-server-pat)
/p:DotNetSymbolServerTokenSymWeb=$(symweb-symbol-server-pat)
/p:PDBArtifactsDirectory='$(Build.ArtifactStagingDirectory)/PDBArtifacts/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_5_Dev_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5-transport/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.NetCore_5_Dev_Channel_Id }}
@@ -1,131 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stages:
- stage: NetCore_Release30_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Core 3.0 Release Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicRelease_30_Channel_Id))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
artifactName: 'BlobArtifacts'
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download PDB Artifacts
inputs:
artifactName: 'PDBArtifacts'
continueOnError: true
- task: PowerShell@2
displayName: Publish
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishToSymbolServers -restore -msbuildEngine dotnet
/p:DotNetSymbolServerTokenMsdl=$(microsoft-symbol-server-pat)
/p:DotNetSymbolServerTokenSymWeb=$(symweb-symbol-server-pat)
/p:PDBArtifactsDirectory='$(Build.ArtifactStagingDirectory)/PDBArtifacts/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.PublicRelease_30_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3-transport/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.PublicRelease_30_Channel_Id }}
@@ -1,131 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
symbolPublishingAdditionalParameters: ''
stages:
- stage: NetCore_Tools_Latest_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Tools - Latest Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job:
displayName: Symbol Publishing
dependsOn: setupMaestroVars
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_Tools_Latest_Channel_Id))
variables:
- group: DotNet-Symbol-Server-Pats
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
artifactName: 'BlobArtifacts'
continueOnError: true
- task: DownloadBuildArtifacts@0
displayName: Download PDB Artifacts
inputs:
artifactName: 'PDBArtifacts'
continueOnError: true
- task: PowerShell@2
displayName: Publish
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishToSymbolServers -restore -msbuildEngine dotnet
/p:DotNetSymbolServerTokenMsdl=$(microsoft-symbol-server-pat)
/p:DotNetSymbolServerTokenSymWeb=$(symweb-symbol-server-pat)
/p:PDBArtifactsDirectory='$(Build.ArtifactStagingDirectory)/PDBArtifacts/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:SymbolPublishingExclusionsFile='$(Build.SourcesDirectory)/eng/SymbolPublishingExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.symbolPublishingAdditionalParameters }}
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_Tools_Latest_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.NetCore_Tools_Latest_Channel_Id }}
@@ -1,96 +0,0 @@
parameters:
artifactsPublishingAdditionalParameters: ''
dependsOn:
- Validate
publishInstallersAndChecksums: false
stages:
- stage: PVR_Publish
dependsOn: ${{ parameters.dependsOn }}
variables:
- template: ../common-variables.yml
displayName: .NET Tools - Validation Publishing
jobs:
- template: ../setup-maestro-vars.yml
- job: publish_assets
displayName: Publish Assets
dependsOn: setupMaestroVars
variables:
- group: DotNet-Blob-Feed
- group: AzureDevOps-Artifact-Feeds-Pats
- name: BARBuildId
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.BARBuildId'] ]
- name: IsStableBuild
value: $[ dependencies.setupMaestroVars.outputs['setReleaseVars.IsStableBuild'] ]
condition: contains(dependencies.setupMaestroVars.outputs['setReleaseVars.InitialChannels'], format('[{0}]', variables.NetCore_Tools_Validation_Channel_Id))
pool:
vmImage: 'windows-2019'
steps:
- task: DownloadBuildArtifacts@0
displayName: Download Package Artifacts
inputs:
buildType: current
artifactName: PackageArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Blob Artifacts
inputs:
buildType: current
artifactName: BlobArtifacts
- task: DownloadBuildArtifacts@0
displayName: Download Asset Manifests
inputs:
buildType: current
artifactName: AssetManifests
- task: NuGetToolInstaller@1
displayName: 'Install NuGet.exe'
# This is necessary whenever we want to publish/restore to an AzDO private feed
- task: NuGetAuthenticate@0
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
displayName: Enable cross-org publishing
inputs:
filePath: eng\common\enable-cross-org-publishing.ps1
arguments: -token $(dn-bot-dnceng-artifact-feeds-rw)
- task: PowerShell@2
displayName: Publish Assets
inputs:
filePath: eng\common\sdk-task.ps1
arguments: -task PublishArtifactsInManifest -restore -msbuildEngine dotnet
/p:ArtifactsCategory=$(_DotNetValidationArtifactsCategory)
/p:IsStableBuild=$(IsStableBuild)
/p:IsInternalBuild=$(IsInternalBuild)
/p:RepositoryName=$(Build.Repository.Name)
/p:CommitSha=$(Build.SourceVersion)
/p:NugetPath=$(NuGetExeToolPath)
/p:AzdoTargetFeedPAT='$(dn-bot-dnceng-universal-packages-rw)'
/p:AzureStorageTargetFeedPAT='$(dotnetfeed-storage-access-key-1)'
/p:BARBuildId=$(BARBuildId)
/p:MaestroApiEndpoint='$(MaestroApiEndPoint)'
/p:BuildAssetRegistryToken='$(MaestroApiAccessToken)'
/p:ManifestsBasePath='$(Build.ArtifactStagingDirectory)/AssetManifests/'
/p:BlobBasePath='$(Build.ArtifactStagingDirectory)/BlobArtifacts/'
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts/'
/p:Configuration=Release
/p:PublishInstallersAndChecksums=${{ parameters.publishInstallersAndChecksums }}
/p:InstallersTargetStaticFeed=$(InstallersBlobFeedUrl)
/p:InstallersAzureAccountKey=$(dotnetcli-storage-key)
/p:ChecksumsTargetStaticFeed=$(ChecksumsBlobFeedUrl)
/p:ChecksumsAzureAccountKey=$(dotnetclichecksums-storage-key)
/p:AzureDevOpsStaticShippingFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticShippingFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticTransportFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
/p:AzureDevOpsStaticTransportFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
/p:AzureDevOpsStaticSymbolsFeed='https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
/p:AzureDevOpsStaticSymbolsFeedKey='$(dn-bot-dnceng-artifact-feeds-rw)'
${{ parameters.artifactsPublishingAdditionalParameters }}
- template: ../../steps/promote-build.yml
parameters:
ChannelId: ${{ variables.NetCore_Tools_Validation_Channel_Id }}
@@ -1,7 +1,9 @@
variables:
- group: Publish-Build-Assets
- group: AzureDevOps-Artifact-Feeds-Pats
- group: DotNet-Blob-Feed
- group: DotNet-DotNetCli-Storage
- group: DotNet-MSRC-Storage
- group: Publish-Build-Assets
# .NET Core 3.1 Dev
- name: PublicDevRelease_31_Channel_Id
@@ -39,6 +41,14 @@ variables:
- name: PublicRelease_31_Channel_Id
value: 129
# General Testing
- name: GeneralTesting_Channel_Id
value: 529
# .NET Core 3.1 Blazor Features
- name: NetCore_31_Blazor_Features_Channel_Id
value: 531
# Whether the build is internal or not
- name: IsInternalBuild
value: ${{ and(ne(variables['System.TeamProject'], 'public'), contains(variables['Build.SourceBranch'], 'internal')) }}
+141 -28
View File
@@ -8,6 +8,7 @@ parameters:
enable: false
continueOnError: false
params: ''
artifactNames: ''
# These parameters let the user customize the call to sdk-task.ps1 for publishing
# symbols & general artifacts as well as for signing validation
@@ -50,7 +51,6 @@ stages:
displayName: Signing Validation
variables:
- template: common-variables.yml
- group: AzureDevOps-Artifact-Feeds-Pats
pool:
vmImage: 'windows-2019'
steps:
@@ -64,7 +64,6 @@ stages:
# Since sdk-task.ps1 tries to restore packages we need to do this authentication here
# otherwise it'll complain about accessing a private feed.
- task: NuGetAuthenticate@0
condition: eq(variables['IsInternalBuild'], 'true')
displayName: 'Authenticate to AzDO Feeds'
- task: PowerShell@2
@@ -80,9 +79,13 @@ stages:
arguments: -task SigningValidation -restore -msbuildEngine dotnet
/p:PackageBasePath='$(Build.ArtifactStagingDirectory)/PackageArtifacts'
/p:SignCheckExclusionsFile='$(Build.SourcesDirectory)/eng/SignCheckExclusionsFile.txt'
/p:Configuration=Release
${{ parameters.signingValidationAdditionalParameters }}
- template: ../steps/publish-logs.yml
parameters:
StageLabel: 'Validation'
JobLabel: 'Signing'
- ${{ if eq(parameters.enableSourceLinkValidation, 'true') }}:
- job:
displayName: SourceLink Validation
@@ -113,63 +116,173 @@ stages:
parameters:
additionalParameters: ${{ parameters.SDLValidationParameters.params }}
continueOnError: ${{ parameters.SDLValidationParameters.continueOnError }}
artifactNames: ${{ parameters.SDLValidationParameters.artifactNames }}
- template: \eng\common\templates\post-build\channels\netcore-dev-5.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
- template: \eng\common\templates\post-build\channels\netcore-dev-31.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Dev5_Publish'
channelName: '.NET Core 5 Dev'
channelId: 131
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5-transport/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet5-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\netcore-tools-latest.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Dev31_Publish'
channelName: '.NET Core 3.1 Dev'
channelId: 128
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-transport/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\netcore-tools-validation.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Tools_Latest_Publish'
channelName: '.NET Tools - Latest'
channelId: 2
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\netcore-3-tools-validation.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
- template: \eng\common\templates\post-build\channels\netcore-3-tools.yml
parameters:
dependsOn: ${{ parameters.publishDependsOn }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
- template: \eng\common\templates\post-build\channels\netcore-release-30.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'PVR_Publish'
channelName: '.NET Tools - Validation'
channelId: 9
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\netcore-release-31.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_3_Tools_Validation_Publish'
channelName: '.NET 3 Tools - Validation'
channelId: 390
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\netcore-internal-30.yml
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_3_Tools_Publish'
channelName: '.NET 3 Tools'
channelId: 344
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-eng-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Release30_Publish'
channelName: '.NET Core 3.0 Release'
channelId: 19
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3-transport/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Release31_Publish'
channelName: '.NET Core 3.1 Release'
channelId: 129
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-transport/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_Blazor31_Features_Publish'
channelName: '.NET Core 3.1 Blazor Features'
channelId: 531
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-blazor/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-blazor/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet3.1-blazor-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-internal-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NetCore_30_Internal_Servicing_Publishing'
channelName: '.NET Core 3.0 Internal Servicing'
channelId: 184
transportFeed: 'https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-transport/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/_packaging/dotnet3-internal-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'General_Testing_Publish'
channelName: 'General Testing'
channelId: 529
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/general-testing/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/general-testing/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/general-testing-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NETCore_Tooling_Dev_Publishing'
channelName: '.NET Core Tooling Dev'
channelId: 548
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
- template: \eng\common\templates\post-build\channels\generic-public-channel.yml
parameters:
artifactsPublishingAdditionalParameters: ${{ parameters.artifactsPublishingAdditionalParameters }}
dependsOn: ${{ parameters.publishDependsOn }}
publishInstallersAndChecksums: ${{ parameters.publishInstallersAndChecksums }}
symbolPublishingAdditionalParameters: ${{ parameters.symbolPublishingAdditionalParameters }}
stageName: 'NETCore_Tooling_Release_Publishing'
channelName: '.NET Core Tooling Release'
channelId: 549
transportFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
shippingFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools/nuget/v3/index.json'
symbolsFeed: 'https://pkgs.dev.azure.com/dnceng/public/_packaging/dotnet-tools-symbols/nuget/v3/index.json'
@@ -0,0 +1,23 @@
parameters:
StageLabel: ''
JobLabel: ''
steps:
- task: Powershell@2
displayName: Prepare Binlogs to Upload
inputs:
targetType: inline
script: |
New-Item -ItemType Directory $(Build.SourcesDirectory)/PostBuildLogs/${{parameters.StageLabel}}/${{parameters.JobLabel}}/
Move-Item -Path $(Build.SourcesDirectory)/artifacts/log/Debug/* $(Build.SourcesDirectory)/PostBuildLogs/${{parameters.StageLabel}}/${{parameters.JobLabel}}/
continueOnError: true
condition: always()
- task: PublishBuildArtifacts@1
displayName: Publish Logs
inputs:
PathtoPublish: '$(Build.SourcesDirectory)/PostBuildLogs'
PublishLocation: Container
ArtifactName: PostBuilLogs
continueOnError: true
condition: always()
+1 -4
View File
@@ -446,10 +446,7 @@ function GetSdkTaskProject([string]$taskName) {
}
function InitializeNativeTools() {
if ($env:DisableNativeToolsetInstalls) {
return
}
if (Get-Member -InputObject $GlobalJson -Name "native-tools") {
if (-Not (Test-Path variable:DisableNativeToolsetInstalls) -And (Get-Member -InputObject $GlobalJson -Name "native-tools")) {
$nativeArgs= @{}
if ($ci) {
$nativeArgs = @{
+3 -3
View File
@@ -4,7 +4,7 @@
# CI mode - set to true on CI server for PR validation build or official build.
ci=${ci:-false}
disable_configure_toolset_import=${disable_configure_toolset_import:-null}
disable_configure_toolset_import=${disable_configure_toolset_import:-}
# Set to true to use the pipelines logger which will enable Azure logging output.
# https://github.com/Microsoft/azure-pipelines-tasks/blob/master/docs/authoring/commands.md
@@ -273,7 +273,7 @@ function GetNuGetPackageCachePath {
}
function InitializeNativeTools() {
if [[ -z "${DisableNativeToolsetInstalls:-}" ]]; then
if [[ -n "${DisableNativeToolsetInstalls:-}" ]]; then
return
fi
if grep -Fq "native-tools" $global_json_file
@@ -437,7 +437,7 @@ Write-PipelineSetVariable -name "Temp" -value "$temp_dir"
Write-PipelineSetVariable -name "TMP" -value "$temp_dir"
# Import custom tools configuration, if present in the repo.
if [[ "$disable_configure_toolset_import" != null ]]; then
if [[ -z "$disable_configure_toolset_import" ]]; then
configure_toolset_script="$eng_root/configure-toolset.sh"
if [[ -a "$configure_toolset_script" ]]; then
. "$configure_toolset_script"
Vendored Submodule
+1
Submodule external/bdwgc added at 6e8db07c8d.
+1 -1
-1
Submodule external/llvm deleted from 29f36fd315.
Vendored Submodule
+1
Submodule external/llvm-project added at 0a3f49c40d.
+2 -2
View File
@@ -3,7 +3,7 @@
"dotnet": "5.0.100-alpha1-014854"
},
"msbuild-sdks": {
"Microsoft.DotNet.Arcade.Sdk": "5.0.0-beta.19527.3",
"Microsoft.DotNet.Helix.Sdk": "5.0.0-beta.19527.3"
"Microsoft.DotNet.Arcade.Sdk": "5.0.0-beta.19567.2",
"Microsoft.DotNet.Helix.Sdk": "5.0.0-beta.19567.2"
}
}
-4
View File
@@ -1,4 +0,0 @@
ltconfig crlf=input
install-sh crlf=input
/BCC_MAKEFILE -crlf
/digimars.mak -crlf
-25
View File
@@ -1,25 +0,0 @@
/aclocal.m4
/config.cache
/config.guess
/config.h
/config.h.in
/config.log
/config.status
/config.sub
/configure
/libtool
/ltmain.sh
/ltconfig
/Makefile
/Makefile.in
/stamp-h
/stamp-h.in
/*.lo
/.deps
/.libs
/*.la
/*.o
/autom4te.cache
/libgc-mono.pc
/libgc-mono-*.tar.gz
dolt*
-623
View File
@@ -1,623 +0,0 @@
/******************************************************************
AmigaOS-spesific routines for GC.
This file is normally included from os_dep.c
******************************************************************/
#if !defined(GC_AMIGA_DEF) && !defined(GC_AMIGA_SB) && !defined(GC_AMIGA_DS) && !defined(GC_AMIGA_AM)
# include "gc_priv.h"
# include <stdio.h>
# include <signal.h>
# define GC_AMIGA_DEF
# define GC_AMIGA_SB
# define GC_AMIGA_DS
# define GC_AMIGA_AM
#endif
#ifdef GC_AMIGA_DEF
# ifndef __GNUC__
# include <exec/exec.h>
# endif
# include <proto/exec.h>
# include <proto/dos.h>
# include <dos/dosextens.h>
# include <workbench/startup.h>
#endif
#ifdef GC_AMIGA_SB
/******************************************************************
Find the base of the stack.
******************************************************************/
ptr_t GC_get_stack_base()
{
struct Process *proc = (struct Process*)SysBase->ThisTask;
/* Reference: Amiga Guru Book Pages: 42,567,574 */
if (proc->pr_Task.tc_Node.ln_Type==NT_PROCESS
&& proc->pr_CLI != NULL) {
/* first ULONG is StackSize */
/*longPtr = proc->pr_ReturnAddr;
size = longPtr[0];*/
return (char *)proc->pr_ReturnAddr + sizeof(ULONG);
} else {
return (char *)proc->pr_Task.tc_SPUpper;
}
}
#if 0 /* old version */
ptr_t GC_get_stack_base()
{
extern struct WBStartup *_WBenchMsg;
extern long __base;
extern long __stack;
struct Task *task;
struct Process *proc;
struct CommandLineInterface *cli;
long size;
if ((task = FindTask(0)) == 0) {
GC_err_puts("Cannot find own task structure\n");
ABORT("task missing");
}
proc = (struct Process *)task;
cli = BADDR(proc->pr_CLI);
if (_WBenchMsg != 0 || cli == 0) {
size = (char *)task->tc_SPUpper - (char *)task->tc_SPLower;
} else {
size = cli->cli_DefaultStack * 4;
}
return (ptr_t)(__base + GC_max(size, __stack));
}
#endif
#endif
#ifdef GC_AMIGA_DS
/******************************************************************
Register data segments.
******************************************************************/
void GC_register_data_segments()
{
struct Process *proc;
struct CommandLineInterface *cli;
BPTR myseglist;
ULONG *data;
int num;
# ifdef __GNUC__
ULONG dataSegSize;
GC_bool found_segment = FALSE;
extern char __data_size[];
dataSegSize=__data_size+8;
/* Can`t find the Location of __data_size, because
it`s possible that is it, inside the segment. */
# endif
proc= (struct Process*)SysBase->ThisTask;
/* Reference: Amiga Guru Book Pages: 538ff,565,573
and XOper.asm */
if (proc->pr_Task.tc_Node.ln_Type==NT_PROCESS) {
if (proc->pr_CLI == NULL) {
myseglist = proc->pr_SegList;
} else {
/* ProcLoaded 'Loaded as a command: '*/
cli = BADDR(proc->pr_CLI);
myseglist = cli->cli_Module;
}
} else {
ABORT("Not a Process.");
}
if (myseglist == NULL) {
ABORT("Arrrgh.. can't find segments, aborting");
}
/* xoper hunks Shell Process */
num=0;
for (data = (ULONG *)BADDR(myseglist); data != NULL;
data = (ULONG *)BADDR(data[0])) {
if (((ULONG) GC_register_data_segments < (ULONG) &data[1]) ||
((ULONG) GC_register_data_segments > (ULONG) &data[1] + data[-1])) {
# ifdef __GNUC__
if (dataSegSize == data[-1]) {
found_segment = TRUE;
}
# endif
GC_add_roots_inner((char *)&data[1],
((char *)&data[1]) + data[-1], FALSE);
}
++num;
} /* for */
# ifdef __GNUC__
if (!found_segment) {
ABORT("Can`t find correct Segments.\nSolution: Use an newer version of ixemul.library");
}
# endif
}
#if 0 /* old version */
void GC_register_data_segments()
{
extern struct WBStartup *_WBenchMsg;
struct Process *proc;
struct CommandLineInterface *cli;
BPTR myseglist;
ULONG *data;
if ( _WBenchMsg != 0 ) {
if ((myseglist = _WBenchMsg->sm_Segment) == 0) {
GC_err_puts("No seglist from workbench\n");
return;
}
} else {
if ((proc = (struct Process *)FindTask(0)) == 0) {
GC_err_puts("Cannot find process structure\n");
return;
}
if ((cli = BADDR(proc->pr_CLI)) == 0) {
GC_err_puts("No CLI\n");
return;
}
if ((myseglist = cli->cli_Module) == 0) {
GC_err_puts("No seglist from CLI\n");
return;
}
}
for (data = (ULONG *)BADDR(myseglist); data != 0;
data = (ULONG *)BADDR(data[0])) {
# ifdef AMIGA_SKIP_SEG
if (((ULONG) GC_register_data_segments < (ULONG) &data[1]) ||
((ULONG) GC_register_data_segments > (ULONG) &data[1] + data[-1])) {
# else
{
# endif /* AMIGA_SKIP_SEG */
GC_add_roots_inner((char *)&data[1],
((char *)&data[1]) + data[-1], FALSE);
}
}
}
#endif /* old version */
#endif
#ifdef GC_AMIGA_AM
#ifndef GC_AMIGA_FASTALLOC
void *GC_amiga_allocwrapper(size_t size,void *(*AllocFunction)(size_t size2)){
return (*AllocFunction)(size);
}
void *(*GC_amiga_allocwrapper_do)(size_t size,void *(*AllocFunction)(size_t size2))
=GC_amiga_allocwrapper;
#else
void *GC_amiga_allocwrapper_firsttime(size_t size,void *(*AllocFunction)(size_t size2));
void *(*GC_amiga_allocwrapper_do)(size_t size,void *(*AllocFunction)(size_t size2))
=GC_amiga_allocwrapper_firsttime;
/******************************************************************
Amiga-spesific routines to obtain memory, and force GC to give
back fast-mem whenever possible.
These hacks makes gc-programs go many times faster when
the amiga is low on memory, and are therefore strictly necesarry.
-Kjetil S. Matheussen, 2000.
******************************************************************/
/* List-header for all allocated memory. */
struct GC_Amiga_AllocedMemoryHeader{
ULONG size;
struct GC_Amiga_AllocedMemoryHeader *next;
};
struct GC_Amiga_AllocedMemoryHeader *GC_AMIGAMEM=(struct GC_Amiga_AllocedMemoryHeader *)(int)~(NULL);
/* Type of memory. Once in the execution of a program, this might change to MEMF_ANY|MEMF_CLEAR */
ULONG GC_AMIGA_MEMF = MEMF_FAST | MEMF_CLEAR;
/* Prevents GC_amiga_get_mem from allocating memory if this one is TRUE. */
#ifndef GC_AMIGA_ONLYFAST
BOOL GC_amiga_dontalloc=FALSE;
#endif
#ifdef GC_AMIGA_PRINTSTATS
int succ=0,succ2=0;
int nsucc=0,nsucc2=0;
int nullretries=0;
int numcollects=0;
int chipa=0;
int allochip=0;
int allocfast=0;
int cur0=0;
int cur1=0;
int cur10=0;
int cur50=0;
int cur150=0;
int cur151=0;
int ncur0=0;
int ncur1=0;
int ncur10=0;
int ncur50=0;
int ncur150=0;
int ncur151=0;
#endif
/* Free everything at program-end. */
void GC_amiga_free_all_mem(void){
struct GC_Amiga_AllocedMemoryHeader *gc_am=(struct GC_Amiga_AllocedMemoryHeader *)(~(int)(GC_AMIGAMEM));
struct GC_Amiga_AllocedMemoryHeader *temp;
#ifdef GC_AMIGA_PRINTSTATS
printf("\n\n"
"%d bytes of chip-mem, and %d bytes of fast-mem where allocated from the OS.\n",
allochip,allocfast
);
printf(
"%d bytes of chip-mem were returned from the GC_AMIGA_FASTALLOC supported allocating functions.\n",
chipa
);
printf("\n");
printf("GC_gcollect was called %d times to avoid returning NULL or start allocating with the MEMF_ANY flag.\n",numcollects);
printf("%d of them was a success. (the others had to use allocation from the OS.)\n",nullretries);
printf("\n");
printf("Succeded forcing %d gc-allocations (%d bytes) of chip-mem to be fast-mem.\n",succ,succ2);
printf("Failed forcing %d gc-allocations (%d bytes) of chip-mem to be fast-mem.\n",nsucc,nsucc2);
printf("\n");
printf(
"Number of retries before succeding a chip->fast force:\n"
"0: %d, 1: %d, 2-9: %d, 10-49: %d, 50-149: %d, >150: %d\n",
cur0,cur1,cur10,cur50,cur150,cur151
);
printf(
"Number of retries before giving up a chip->fast force:\n"
"0: %d, 1: %d, 2-9: %d, 10-49: %d, 50-149: %d, >150: %d\n",
ncur0,ncur1,ncur10,ncur50,ncur150,ncur151
);
#endif
while(gc_am!=NULL){
temp=gc_am->next;
FreeMem(gc_am,gc_am->size);
gc_am=(struct GC_Amiga_AllocedMemoryHeader *)(~(int)(temp));
}
}
#ifndef GC_AMIGA_ONLYFAST
/* All memory with address lower than this one is chip-mem. */
char *chipmax;
/*
* Allways set to the last size of memory tried to be allocated.
* Needed to ensure allocation when the size is bigger than 100000.
*
*/
size_t latestsize;
#endif
/*
* The actual function that is called with the GET_MEM macro.
*
*/
void *GC_amiga_get_mem(size_t size){
struct GC_Amiga_AllocedMemoryHeader *gc_am;
#ifndef GC_AMIGA_ONLYFAST
if(GC_amiga_dontalloc==TRUE){
// printf("rejected, size: %d, latestsize: %d\n",size,latestsize);
return NULL;
}
// We really don't want to use chip-mem, but if we must, then as little as possible.
if(GC_AMIGA_MEMF==(MEMF_ANY|MEMF_CLEAR) && size>100000 && latestsize<50000) return NULL;
#endif
gc_am=AllocMem((ULONG)(size + sizeof(struct GC_Amiga_AllocedMemoryHeader)),GC_AMIGA_MEMF);
if(gc_am==NULL) return NULL;
gc_am->next=GC_AMIGAMEM;
gc_am->size=size + sizeof(struct GC_Amiga_AllocedMemoryHeader);
GC_AMIGAMEM=(struct GC_Amiga_AllocedMemoryHeader *)(~(int)(gc_am));
// printf("Allocated %d (%d) bytes at address: %x. Latest: %d\n",size,tot,gc_am,latestsize);
#ifdef GC_AMIGA_PRINTSTATS
if((char *)gc_am<chipmax){
allochip+=size;
}else{
allocfast+=size;
}
#endif
return gc_am+1;
}
#ifndef GC_AMIGA_ONLYFAST
/* Tries very hard to force GC to find fast-mem to return. Done recursively
* to hold the rejected memory-pointers reachable from the collector in an
* easy way.
*
*/
#ifdef GC_AMIGA_RETRY
void *GC_amiga_rec_alloc(size_t size,void *(*AllocFunction)(size_t size2),const int rec){
void *ret;
ret=(*AllocFunction)(size);
#ifdef GC_AMIGA_PRINTSTATS
if((char *)ret>chipmax || ret==NULL){
if(ret==NULL){
nsucc++;
nsucc2+=size;
if(rec==0) ncur0++;
if(rec==1) ncur1++;
if(rec>1 && rec<10) ncur10++;
if(rec>=10 && rec<50) ncur50++;
if(rec>=50 && rec<150) ncur150++;
if(rec>=150) ncur151++;
}else{
succ++;
succ2+=size;
if(rec==0) cur0++;
if(rec==1) cur1++;
if(rec>1 && rec<10) cur10++;
if(rec>=10 && rec<50) cur50++;
if(rec>=50 && rec<150) cur150++;
if(rec>=150) cur151++;
}
}
#endif
if (((char *)ret)<=chipmax && ret!=NULL && (rec<(size>500000?9:size/5000))){
ret=GC_amiga_rec_alloc(size,AllocFunction,rec+1);
// GC_free(ret2);
}
return ret;
}
#endif
/* The allocating-functions defined inside the amiga-blocks in gc.h is called
* via these functions.
*/
void *GC_amiga_allocwrapper_any(size_t size,void *(*AllocFunction)(size_t size2)){
void *ret,*ret2;
GC_amiga_dontalloc=TRUE; // Pretty tough thing to do, but its indeed necesarry.
latestsize=size;
ret=(*AllocFunction)(size);
if(((char *)ret) <= chipmax){
if(ret==NULL){
//Give GC access to allocate memory.
#ifdef GC_AMIGA_GC
if(!GC_dont_gc){
GC_gcollect();
#ifdef GC_AMIGA_PRINTSTATS
numcollects++;
#endif
ret=(*AllocFunction)(size);
}
#endif
if(ret==NULL){
GC_amiga_dontalloc=FALSE;
ret=(*AllocFunction)(size);
if(ret==NULL){
WARN("Out of Memory! Returning NIL!\n", 0);
}
}
#ifdef GC_AMIGA_PRINTSTATS
else{
nullretries++;
}
if(ret!=NULL && (char *)ret<=chipmax) chipa+=size;
#endif
}
#ifdef GC_AMIGA_RETRY
else{
/* We got chip-mem. Better try again and again and again etc., we might get fast-mem sooner or later... */
/* Using gctest to check the effectiviness of doing this, does seldom give a very good result. */
/* However, real programs doesn't normally rapidly allocate and deallocate. */
// printf("trying to force... %d bytes... ",size);
if(
AllocFunction!=GC_malloc_uncollectable
#ifdef ATOMIC_UNCOLLECTABLE
&& AllocFunction!=GC_malloc_atomic_uncollectable
#endif
){
ret2=GC_amiga_rec_alloc(size,AllocFunction,0);
}else{
ret2=(*AllocFunction)(size);
#ifdef GC_AMIGA_PRINTSTATS
if((char *)ret2<chipmax || ret2==NULL){
nsucc++;
nsucc2+=size;
ncur0++;
}else{
succ++;
succ2+=size;
cur0++;
}
#endif
}
if(((char *)ret2)>chipmax){
// printf("Succeeded.\n");
GC_free(ret);
ret=ret2;
}else{
GC_free(ret2);
// printf("But did not succeed.\n");
}
}
#endif
}
GC_amiga_dontalloc=FALSE;
return ret;
}
void (*GC_amiga_toany)(void)=NULL;
void GC_amiga_set_toany(void (*func)(void)){
GC_amiga_toany=func;
}
#endif // !GC_AMIGA_ONLYFAST
void *GC_amiga_allocwrapper_fast(size_t size,void *(*AllocFunction)(size_t size2)){
void *ret;
ret=(*AllocFunction)(size);
if(ret==NULL){
// Enable chip-mem allocation.
// printf("ret==NULL\n");
#ifdef GC_AMIGA_GC
if(!GC_dont_gc){
GC_gcollect();
#ifdef GC_AMIGA_PRINTSTATS
numcollects++;
#endif
ret=(*AllocFunction)(size);
}
#endif
if(ret==NULL){
#ifndef GC_AMIGA_ONLYFAST
GC_AMIGA_MEMF=MEMF_ANY | MEMF_CLEAR;
if(GC_amiga_toany!=NULL) (*GC_amiga_toany)();
GC_amiga_allocwrapper_do=GC_amiga_allocwrapper_any;
return GC_amiga_allocwrapper_any(size,AllocFunction);
#endif
}
#ifdef GC_AMIGA_PRINTSTATS
else{
nullretries++;
}
#endif
}
return ret;
}
void *GC_amiga_allocwrapper_firsttime(size_t size,void *(*AllocFunction)(size_t size2)){
atexit(&GC_amiga_free_all_mem);
chipmax=(char *)SysBase->MaxLocMem; // For people still having SysBase in chip-mem, this might speed up a bit.
GC_amiga_allocwrapper_do=GC_amiga_allocwrapper_fast;
return GC_amiga_allocwrapper_fast(size,AllocFunction);
}
#endif //GC_AMIGA_FASTALLOC
/*
* The wrapped realloc function.
*
*/
void *GC_amiga_realloc(void *old_object,size_t new_size_in_bytes){
#ifndef GC_AMIGA_FASTALLOC
return GC_realloc(old_object,new_size_in_bytes);
#else
void *ret;
latestsize=new_size_in_bytes;
ret=GC_realloc(old_object,new_size_in_bytes);
if(ret==NULL && GC_AMIGA_MEMF==(MEMF_FAST | MEMF_CLEAR)){
/* Out of fast-mem. */
#ifdef GC_AMIGA_GC
if(!GC_dont_gc){
GC_gcollect();
#ifdef GC_AMIGA_PRINTSTATS
numcollects++;
#endif
ret=GC_realloc(old_object,new_size_in_bytes);
}
#endif
if(ret==NULL){
#ifndef GC_AMIGA_ONLYFAST
GC_AMIGA_MEMF=MEMF_ANY | MEMF_CLEAR;
if(GC_amiga_toany!=NULL) (*GC_amiga_toany)();
GC_amiga_allocwrapper_do=GC_amiga_allocwrapper_any;
ret=GC_realloc(old_object,new_size_in_bytes);
#endif
}
#ifdef GC_AMIGA_PRINTSTATS
else{
nullretries++;
}
#endif
}
if(ret==NULL){
WARN("Out of Memory! Returning NIL!\n", 0);
}
#ifdef GC_AMIGA_PRINTSTATS
if(((char *)ret)<chipmax && ret!=NULL){
chipa+=new_size_in_bytes;
}
#endif
return ret;
#endif
}
#endif //GC_AMIGA_AM
-88
View File
@@ -1,88 +0,0 @@
# Makefile for Borland C++ 5.5 on NT
# If you have the Borland assembler, remove "-DUSE_GENERIC"
#
bc= c:\Borland\BCC55
bcbin= $(bc)\bin
bclib= $(bc)\lib
bcinclude= $(bc)\include
gcinclude1 = $(bc)\gc6.2\include
gcinclude2 = $(bc)\gc6.2\cord
cc= $(bcbin)\bcc32
rc= $(bcbin)\brc32
lib= $(bcbin)\tlib
link= $(bcbin)\ilink32
cflags= -O2 -R -v- -vi -H -H=gc.csm -I$(bcinclude);$(gcinclude1);$(gcinclude2) -L$(bclib) \
-w-pro -w-aus -w-par -w-ccc -w-rch -a4 -D__STDC__=0
#defines= -DSILENT
defines= -DSILENT -DALL_INTERIOR_POINTERS -DUSE_GENERIC -DNO_GETENV -DJAVA_FINALIZATION -DGC_OPERATOR_NEW_ARRAY
.c.obj:
$(cc) @&&|
$(cdebug) $(cflags) $(cvars) $(defines) -o$* -c $*.c
|
.cpp.obj:
$(cc) @&&|
$(cdebug) $(cflags) $(cvars) $(defines) -o$* -c $*.cpp
|
.rc.res:
$(rc) -i$(bcinclude) -r -fo$* $*.rc
XXXOBJS= XXXalloc.obj XXXreclaim.obj XXXallchblk.obj XXXmisc.obj \
XXXmach_dep.obj XXXos_dep.obj XXXmark_rts.obj XXXheaders.obj XXXmark.obj \
XXXobj_map.obj XXXblacklst.obj XXXfinalize.obj XXXnew_hblk.obj \
XXXdbg_mlc.obj XXXmalloc.obj XXXstubborn.obj XXXdyn_load.obj \
XXXtypd_mlc.obj XXXptr_chck.obj XXXgc_cpp.obj XXXmallocx.obj
OBJS= $(XXXOBJS:XXX=)
all: gctest.exe cord\de.exe test_cpp.exe
$(OBJS) test.obj: include\private\gc_priv.h include\private\gc_hdrs.h include\gc.h include\private\gcconfig.h MAKEFILE
gc.lib: $(OBJS)
del gc.lib
$(lib) $* @&&|
$(XXXOBJS:XXX=+)
|
gctest.exe: tests\test.obj gc.lib
$(cc) @&&|
$(cflags) -W -e$* tests\test.obj gc.lib
|
cord\de.obj cord\de_win.obj: include\cord.h include\private\cord_pos.h cord\de_win.h \
cord\de_cmds.h
cord\de.exe: cord\cordbscs.obj cord\cordxtra.obj cord\de.obj cord\de_win.obj \
cord\de_win.res gc.lib
$(cc) @&&|
$(cflags) -W -e$* cord\cordbscs.obj cord\cordxtra.obj \
cord\de.obj cord\de_win.obj gc.lib
|
$(rc) cord\de_win.res cord\de.exe
gc_cpp.obj: include\gc_cpp.h include\gc.h
gc_cpp.cpp: gc_cpp.cc
copy gc_cpp.cc gc_cpp.cpp
test_cpp.cpp: tests\test_cpp.cc
copy tests\test_cpp.cc test_cpp.cpp
test_cpp.exe: test_cpp.obj include\gc_cpp.h include\gc.h gc.lib
$(cc) @&&|
$(cflags) -W -e$* test_cpp.obj gc.lib
|
scratch:
-del *.obj *.res *.exe *.csm cord\*.obj cord\*.res cord\*.exe cord\*.csm
clean:
del gc.lib
del *.obj
del tests\test.obj
-141
View File
@@ -1,141 +0,0 @@
#
# OS/2 specific Makefile for the EMX environment
#
# You need GNU Make 3.71, gcc 2.5.7, emx 0.8h and GNU fileutils 3.9
# or similar tools. C++ interface and de.exe weren't tested.
#
# Rename this file "Makefile".
#
# Primary targets:
# gc.a - builds basic library
# c++ - adds C++ interface to library and include directory
# cords - adds cords (heavyweight strings) to library and include directory
# test - prints porting information, then builds basic version of gc.a, and runs
# some tests of collector and cords. Does not add cords or c++ interface to gc.a
# cord/de.exe - builds dumb editor based on cords.
CC= gcc
CXX=g++
# Needed only for "make c++", which adds the c++ interface
CFLAGS= -O -DALL_INTERIOR_POINTERS -DSILENT
# Setjmp_test may yield overly optimistic results when compiled
# without optimization.
# -DSILENT disables statistics printing, and improves performance.
# -DCHECKSUMS reports on erroneously clear dirty bits, and unexpectedly
# altered stubborn objects, at substantial performance cost.
# -DFIND_LEAK causes the collector to assume that all inaccessible
# objects should have been explicitly deallocated, and reports exceptions
# -DSOLARIS_THREADS enables support for Solaris (thr_) threads.
# (Clients should also define SOLARIS_THREADS and then include
# gc.h before performing thr_ or GC_ operations.)
# -DALL_INTERIOR_POINTERS allows all pointers to the interior
# of objects to be recognized. (See gc_private.h for consequences.)
# -DSMALL_CONFIG tries to tune the collector for small heap sizes,
# usually causing it to use less space in such situations.
# Incremental collection no longer works in this case.
# -DDONT_ADD_BYTE_AT_END is meaningful only with
# -DALL_INTERIOR_POINTERS. Normally -DALL_INTERIOR_POINTERS
# causes all objects to be padded so that pointers just past the end of
# an object can be recognized. This can be expensive. (The padding
# is normally more than one byte due to alignment constraints.)
# -DDONT_ADD_BYTE_AT_END disables the padding.
AR= ar
RANLIB= ar s
# Redefining srcdir allows object code for the nonPCR version of the collector
# to be generated in different directories
srcdir = .
VPATH = $(srcdir)
OBJS= alloc.o reclaim.o allchblk.o misc.o mach_dep.o os_dep.o mark_rts.o headers.o mark.o obj_map.o blacklst.o finalize.o new_hblk.o dyn_load.o dbg_mlc.o malloc.o stubborn.o checksums.o typd_mlc.o ptr_chck.o mallocx.o
CORD_OBJS= cord/cordbscs.o cord/cordxtra.o cord/cordprnt.o
CORD_INCLUDE_FILES= $(srcdir)/gc.h $(srcdir)/cord/cord.h $(srcdir)/cord/ec.h \
$(srcdir)/cord/cord_pos.h
# Libraries needed for curses applications. Only needed for de.
CURSES= -lcurses -ltermlib
# The following is irrelevant on most systems. But a few
# versions of make otherwise fork the shell specified in
# the SHELL environment variable.
SHELL= bash
SPECIALCFLAGS =
# Alternative flags to the C compiler for mach_dep.c.
# Mach_dep.c often doesn't like optimization, and it's
# not time-critical anyway.
all: gc.a gctest.exe
$(OBJS) test.o: $(srcdir)/gc_priv.h $(srcdir)/gc_hdrs.h $(srcdir)/gc.h \
$(srcdir)/gcconfig.h $(srcdir)/gc_typed.h
# The dependency on Makefile is needed. Changing
# options such as -DSILENT affects the size of GC_arrays,
# invalidating all .o files that rely on gc_priv.h
mark.o typd_mlc.o finalize.o: $(srcdir)/include/gc_mark.h $(srcdir)/include/private/gc_pmark.h
gc.a: $(OBJS)
$(AR) ru gc.a $(OBJS)
$(RANLIB) gc.a
cords: $(CORD_OBJS) cord/cordtest.exe
$(AR) ru gc.a $(CORD_OBJS)
$(RANLIB) gc.a
cp $(srcdir)/cord/cord.h include/cord.h
cp $(srcdir)/cord/ec.h include/ec.h
cp $(srcdir)/cord/cord_pos.h include/cord_pos.h
gc_cpp.o: $(srcdir)/gc_cpp.cc $(srcdir)/gc_cpp.h
$(CXX) -c -O $(srcdir)/gc_cpp.cc
c++: gc_cpp.o $(srcdir)/gc_cpp.h
$(AR) ru gc.a gc_cpp.o
$(RANLIB) gc.a
cp $(srcdir)/gc_cpp.h include/gc_cpp.h
mach_dep.o: $(srcdir)/mach_dep.c
$(CC) -o mach_dep.o -c $(SPECIALCFLAGS) $(srcdir)/mach_dep.c
mark_rts.o: $(srcdir)/mark_rts.c
$(CC) -o mark_rts.o -c $(CFLAGS) $(srcdir)/mark_rts.c
cord/cordbscs.o: $(srcdir)/cord/cordbscs.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c $(srcdir)/cord/cordbscs.c -o cord/cordbscs.o
cord/cordxtra.o: $(srcdir)/cord/cordxtra.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c $(srcdir)/cord/cordxtra.c -o cord/cordxtra.o
cord/cordprnt.o: $(srcdir)/cord/cordprnt.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c $(srcdir)/cord/cordprnt.c -o cord/cordprnt.o
cord/cordtest.exe: $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a
$(CC) $(CFLAGS) -o cord/cordtest.exe $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a
cord/de.exe: $(srcdir)/cord/de.c $(srcdir)/cord/cordbscs.o $(srcdir)/cord/cordxtra.o gc.a
$(CC) $(CFLAGS) -o cord/de.exe $(srcdir)/cord/de.c $(srcdir)/cord/cordbscs.o $(srcdir)/cord/cordxtra.o gc.a $(CURSES)
clean:
rm -f gc.a tests/test.o gctest.exe output-local output-diff $(OBJS) \
setjmp_test mon.out gmon.out a.out core \
$(CORD_OBJS) cord/cordtest.exe cord/de.exe
-rm -f *~
gctest.exe: tests/test.o gc.a
$(CC) $(CFLAGS) -o gctest.exe tests/test.o gc.a
# If an optimized setjmp_test generates a segmentation fault,
# odds are your compiler is broken. Gctest may still work.
# Try compiling setjmp_t.c unoptimized.
setjmp_test.exe: $(srcdir)/setjmp_t.c $(srcdir)/gc.h
$(CC) $(CFLAGS) -o setjmp_test.exe $(srcdir)/setjmp_t.c
test: setjmp_test.exe gctest.exe
./setjmp_test
./gctest
make cord/cordtest.exe
cord/cordtest
-154
View File
@@ -1,154 +0,0 @@
/*
MacOS.c
Some routines for the Macintosh OS port of the Hans-J. Boehm, Alan J. Demers
garbage collector.
<Revision History>
11/22/94 pcb StripAddress the temporary memory handle for 24-bit mode.
11/30/94 pcb Tracking all memory usage so we can deallocate it all at once.
02/10/96 pcb Added routine to perform a final collection when
unloading shared library.
by Patrick C. Beard.
*/
/* Boehm, February 15, 1996 2:55 pm PST */
#include <Resources.h>
#include <Memory.h>
#include <LowMem.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "gc.h"
#include "gc_priv.h"
// use 'CODE' resource 0 to get exact location of the beginning of global space.
typedef struct {
unsigned long aboveA5;
unsigned long belowA5;
unsigned long JTSize;
unsigned long JTOffset;
} *CodeZeroPtr, **CodeZeroHandle;
void* GC_MacGetDataStart()
{
CodeZeroHandle code0 = (CodeZeroHandle)GetResource('CODE', 0);
if (code0) {
long belowA5Size = (**code0).belowA5;
ReleaseResource((Handle)code0);
return (LMGetCurrentA5() - belowA5Size);
}
fprintf(stderr, "Couldn't load the jump table.");
exit(-1);
return 0;
}
/* track the use of temporary memory so it can be freed all at once. */
typedef struct TemporaryMemoryBlock TemporaryMemoryBlock, **TemporaryMemoryHandle;
struct TemporaryMemoryBlock {
TemporaryMemoryHandle nextBlock;
char data[];
};
static TemporaryMemoryHandle theTemporaryMemory = NULL;
static Boolean firstTime = true;
void GC_MacFreeTemporaryMemory(void);
Ptr GC_MacTemporaryNewPtr(size_t size, Boolean clearMemory)
{
static Boolean firstTime = true;
OSErr result;
TemporaryMemoryHandle tempMemBlock;
Ptr tempPtr = nil;
tempMemBlock = (TemporaryMemoryHandle)TempNewHandle(size + sizeof(TemporaryMemoryBlock), &result);
if (tempMemBlock && result == noErr) {
HLockHi((Handle)tempMemBlock);
tempPtr = (**tempMemBlock).data;
if (clearMemory) memset(tempPtr, 0, size);
tempPtr = StripAddress(tempPtr);
// keep track of the allocated blocks.
(**tempMemBlock).nextBlock = theTemporaryMemory;
theTemporaryMemory = tempMemBlock;
}
# if !defined(SHARED_LIBRARY_BUILD)
// install an exit routine to clean up the memory used at the end.
if (firstTime) {
atexit(&GC_MacFreeTemporaryMemory);
firstTime = false;
}
# endif
return tempPtr;
}
extern word GC_fo_entries;
static void perform_final_collection()
{
unsigned i;
word last_fo_entries = 0;
/* adjust the stack bottom, because CFM calls us from another stack
location. */
GC_stackbottom = (ptr_t)&i;
/* try to collect and finalize everything in sight */
for (i = 0; i < 2 || GC_fo_entries < last_fo_entries; i++) {
last_fo_entries = GC_fo_entries;
GC_gcollect();
}
}
void GC_MacFreeTemporaryMemory()
{
# if defined(SHARED_LIBRARY_BUILD)
/* if possible, collect all memory, and invoke all finalizers. */
perform_final_collection();
# endif
if (theTemporaryMemory != NULL) {
long totalMemoryUsed = 0;
TemporaryMemoryHandle tempMemBlock = theTemporaryMemory;
while (tempMemBlock != NULL) {
TemporaryMemoryHandle nextBlock = (**tempMemBlock).nextBlock;
totalMemoryUsed += GetHandleSize((Handle)tempMemBlock);
DisposeHandle((Handle)tempMemBlock);
tempMemBlock = nextBlock;
}
theTemporaryMemory = NULL;
# if !defined(SILENT) && !defined(SHARED_LIBRARY_BUILD)
fprintf(stdout, "[total memory used: %ld bytes.]\n",
totalMemoryUsed);
fprintf(stdout, "[total collections: %ld.]\n", GC_gc_no);
# endif
}
}
#if __option(far_data)
void* GC_MacGetDataEnd()
{
CodeZeroHandle code0 = (CodeZeroHandle)GetResource('CODE', 0);
if (code0) {
long aboveA5Size = (**code0).aboveA5;
ReleaseResource((Handle)code0);
return (LMGetCurrentA5() + aboveA5Size);
}
fprintf(stderr, "Couldn't load the jump table.");
exit(-1);
return 0;
}
#endif /* __option(far_data) */
-886
View File
@@ -1,886 +0,0 @@
(This file must be converted with BinHex 4.0)
:$deKBe"bEfTPBh4c,R0TG!"6594%8dP8)3#3"&)e!!!"4UiT8dP8)3!(!!"50A*
-BA8#ZJ#3!aB"#3d0#'GM,MBi5bkjBf038%-ZZ3#3%)Zi!*!8"@`!N!6rN!4069"
568e$3`%!UbqAD+X`19S!!!Ba!!!,*J!!!F%!!!-PfTmj1`#3"PET)d31)LTH6H4
#*AqG5b5HI*)QjY$IIb00%ReTJSi6rG$jG(bZ,"Rc,9Umf[IRj)6FZ-j`GfGR)#!
m-#qLqB#cj'G%46qffB3q8AppLXKc+P&*il4FMJMq3N32r[U,(PlSNdrQm-J(4!p
jK)NHmKJSHY!,&chS$4)pk%8mL3I)B0'$AU6S3'q)k%%[5[5J&ffa#68)0ZM&#T!
!*fHC-2dFZ3i83[Vr[4Xh'+DNQrm'J)rrpqe%ST`,FeVi6b,*qHH")4eQc28NFMN
ZT*m,L"Y%-`pdAk6RLHDaeVV0a,,@P(4UUK66rUM'8bf91llS("lTh81)MBQ+4*q
rfHENEhD)Ke#3!09'M%bL[P1+G88fa$3e)5Gpf0kARpBf*6eIH*0`ZBHR%ii"PbN
+D&*)688M)Sm$Bm[cCdDjh2YIjmAc`(TVpi*Vka((A*&Yl@'LTSH1M*AMP#,2[A$
(FHA@S"dL4dER#3b!EfBYem(C9P5iGH"a-bb-AL(F"bb-AL,F6)%a9pJUL,(hf%B
TeQb["X5ib4DQXV!-fa6&mZf&3,(C&UDd-((SpeMBEIB`8Zc,BcZR3A5'X+jYj$'
6)6HVV+R[!`#3!`X!(E@*MFQ%R4d"))`m[3JM[c)bBS54Tj'M(AP+MK&f%VD5SdG
SANFB@3Rqc$Am83(+)`"G(D%A'9!bBQ6!b)b4Sq3SH8D1NDGNX$)bBi54!51--$*
Kj0L!M"KKK"dC--,)-h+-6#KKC-$)-F)NamL!!Z06#X!!b&%bBUHp8RcN'%%6!b,
i!!kV"`"DLHFaK*!!"Ym4K,,2i2X4c[,`c5!GIPf!ZcNi'8'VfJFpSfdpq+CY$8j
-V'f-DZr2[36#1(ael5hmfT@1cSU66D5pqDSA89pdTP-`Z[jj6T&!PmZBFZjal"&
5iG6#blE$+&kLh#QZ118&(0T1J(hZ,9)5MJ9ic*qPI!ac'RJ96QMZjSbkMq()Ui6
B+f,,#'N1icbM4N"aaBr1`3Z9U'8RY'XAiVXFKp#&k2D5Be%VCdh4%,+2QS'b"Q2
%0PNT4rE#%kTUFqYDM56bVjfe!p8MqmL)1VmjVkJY`U[*$&*L3AMSpB@LCQ*U&l%
T+3890rL,V9klFN*4@f0UTf8Z&&afN!"4GC6G8p3fN9$4+4[-@DAeK%lej"@eAAL
eU@&4[Tm28%mqqUkS(F+VDa#lB&'rlRAllRP&l460Qc,)MHR$jMh@$8Y4Xc'e`cd
ZE2AUUiH+fK96feb$epq&'RAQeLG&lCDjmP+"Kr8k9#qp'eI8RPf[6R$dS+$UcqI
ELYSV[*ETFL&j[@lr803qd9I2A#bi4Vei3*d[+@Urk*!!&abe0HTVm%44"i4A6JN
c(2I!kjRl6a9e813DK"A6p(LjRZZGaGH+1L5SiBT[(6ekd2*ILMSXU(l)#m3QMDB
V+QTG!r*NG#RQai#DNh4,l0&!Ie`dYi98Y1%1A$5hKP4,`d9cHdKP'LkD@q4hYC*
%dfdLeCCNN@i9UIBNLh5l5(8N68qhM&4R`d9cfdKP'bkD@dHU+qe&XRfNZSqc10j
#8Me*&ZNfNZT0hSYd+dP&ri-FGM6G6P,p5D,rPNT0`dQLk5+6'NLb5"HDe'$L)Pe
X8N2bj-Z'$r$6-$NZjLGC)1lB-"jQSff@[ak%LJ[rI#%p2ddAGREN(@"V+,S6CI!
I!!!0$3KRBbj38%-ZZ@0M8&"$,VN!N"#$BJ#3%4B!!!d'!*!%rj!%68e38Ne33d-
"!+X[PfqV-$P*!!!'-3!!&UB!!!(&!!!&C80(jji!N!BMM#0%$L)UANhN3L9rV@9
B`f#c2p$XpAVVCc-[`k20Y5bJ+CTHPScj`Z'!lmr*#EPaRH(ZcR!J!!rqMKG"q)#
cj'G%46qffB3q8Aqp4R6FA83PM6`KUjaYD&IlZ@jDrY"pk[b&AZrdH*kFbb9PM*S
`4Kh$c8Lf0bVe+Y`Q$amM6mc%*C1(jF&1bFSdGIlLpc*04b#X&D8[&6R%+-#6HhJ
kX"#A+Bp6%6RGkB&kM%'jh$ZLmam[1Irq,r82rGM"5H4bh1ZB+b"Z%&-pD)5CL9(
AP(4UUK6$!(lkH+UPFXFARF-MIHHMXf!5Nd%SZYRQj'pfL)G3N!$94X#(q25G8U`
VXL'QU3Njk8[phV2@0Q92J#d6rA2N1["[!%c(M4X-8p,0IcYJf2lRBmD2c)*RQEF
68m'9jqq*MjHTji&GqDp$kh501r9fqVPJe4iQDRS)L!)ELqiX08i#@40jpP1+F@p
iC&))L)Qq4Bk-cK-i*h`cDlN1cMBUbZA3+beKhX*-&UD`X%ME%F91fHB3BaCC''Y
KNba-C@(,"-40Yl"l,#c8`YCDf%#"XGD%F4m3'*i'k"iah[Ddam+k"Xd3eV@02'B
bj'D90I9p!!!-q)[jAU2HhQ[NiCQC&f(Ne`JR!hlN1''4Sjc`)hcL5IK+f(@8(q&
(1&Nj2XreTBI[M!0dGB4'MK01#CFF2c,JK"*1MNZ1(q&(11@5ii5EKimF*ja``Np
#bA(#bBL6BpQ6jq5imT-m2mQ!dq2N'H&2RT2M%Nii'6$J,PF!#N#jGS3IS9Uba%G
'A-)*8[#%!j-9'#r3@EpUPQ9+NL6$ldj*kVS6INIK@`*q'q$hGRJCPb,`pUJm(fQ
3!#mGrdQqe$Nm22hkJ2cerNp"i3$m4Z62S5YA40V([V`MbHF@)QPT2IN@3@$ceHm
I&dT3GqF9K,'&&8[6LKMTbQ6@-*%bJE#4RM,b'FA*'VC5`0BBdTa"@aNXM#)mU'"
N@d@XSIKMMiMh#RbbLSjLT49GG9"F84)Q8QfN&![N1hK"A'V5F,,dJIF@+`iNJEb
H-(5Nar84j!"*Q54MH+j&08dYQc,(ipT9I+aFqIQc-XP313&803UUPPD4*+UAIlj
$U+jMAP1QUSfEYV2Qp4HKfZ#TYQTCT)hEaCbp+ZXH0"m5USfHDV1HbL4cCT@41rr
5+d+eL4&+'hR90)iLRp$LYcm)e5McQN@UMR#&$kKqr%eHU-DBejbUCC-k+P4N5r%
Iha+Uc5aj)kVfm*'ej*8Dali5ULfHDLah-l$Zfer1#G9@6l8TTf*r,RKTZ2#Q8'h
MA2&i%MYq(0aCicHKfPlfDYLeJ3*FFEG3l@"HmfJbqFrdHU&IU+jRHE95BmQFkJF
29)qp)93hX!aCGLfYP0!jSEU4HF9)-e8M9rADGfC4U(BbVVC66+8XR2Hj2RAmGk'
kLDNk8`@p0[6F"hrG,e3h`kmm(BhDMQjBm@`ejDH1pG)YbUXYM'Y'5aD`-H(VPZ)
,*i6A,Nqe)D1Y'5@UV@HM3VAE)a3$3MT+9jAGa)HI#%*E@9ie+jmf-PA9dY#66`Z
[fkMA!l&$eZ3)bP996crcal6`ZRdT$9NG0S#+V([`rRZ&eae,A%dMGB2V4H%9YPL
LfZ3B194,NC[ik!QKZSYlaE"deVc1$3[9(XVeFJIG0T,9**@'AVXJZ2Db$%'!,$a
e+d2+8SES`Z&RD1(C`m,VlM*Aj)cP#M@ZlJI#Djp(U28`fl)VL9dKY+IXeFM!HRJ
MVc0#YCpj6@!,M0VrHYh,CMQN!FBjl1ZVEPhjaCK)``"6,6JiU@@ekMjdmEEPI@M
3DpXKj3pi+f`LFFpIUPrF058)N4X)f4ZQ*P5c1[&!pGhC4i@Ue2BCE"bRL&haLRk
Thb#ZUK&ZK-Kc9k4Z-[QKhdaf&1KhN!#*#IdZ-XfJhdPQ)I6l#![SYjD'HXp$hdA
f$1LhNlN-r4DbV8$I8iS[RSEqj#URqY@$9b3dJG1XG))%khUHJMX,Vh896Z%"I%B
PFK1MejpP2[@,$LpbTe[Q%h#[hhai0BBHF+r-MrTeL9G6k!!IKHa1rmf2qMf,9c6
d)%I[5Hq$1hVVq60(`H@-9fb&cfkb$BBDc1-Ck@@#jrVH%0cXH$@cIK[C#F&2Q9X
[qpl(HTpEQ9F`KqVA3&iYS3Pl6#ARpIXMVpCP6[+ma`PkbJPkbJPkbJPkbJPkbJP
kbJPkbJPkbJPk1MHKTlbJTlbJpqGlF2RNe4CD`1XDTfUZEYjDHE@[F0T$,KbK"Vc
mA!9AAPiGS3Qjm[HQi+l-LraVj'p1i3&mcNKce1@eZ4pFX(PY@1(66rD18)Im"eF
YAJ1K#AYcK92peXpVBfM#AZAIKi*r&r$U$"h)dkhp2[JI!kp0S3GjhdZZV))A!43
jH4kk(TLQKF4pTXhHI!ITRb%hcX3KfeN#**1EI54a"'@Z8(9Dm%D@b"Y#qhm!N!-
0!!PRBfaTBLda,VPM8&"$,VN!N"#ah3#3%!9X!!!I``#3"2q3"&"56dT,38K-!3#
TY1))Uc!eD!!!@F-!N!B563#3"2$I!*!)22J1`2KbNQaPEr+hGEX``Jk!Vpa0&eT
RDl*eSGZ&%EEAc@iGG+hAYBDRapHZd6ETQH'lV2AbpMVJ4lN,ck0G4lMb)fcKAQi
*AeLhm1)VRfPGM,"Zi8pBG1%a3VYZi@m,@rM#2'iAfhjHacE,K"[bJGYB,ZcNP&#
"$cqJ[fRG`SmXR'aMC-H6r-)AXTaNHE+Fj"HkN!"0"R[G!H4jITB&`!(!dKX"PZ#
Z+PX+S(dCS&YGZI3,cN3L+P4H)V5R@D3p,54$JD"3'!j')mhRcl%mUJ)9e2PVUaF
j[6lNX)ll!4,jajb6UrZK!hSTX[caD`$ZIHl,pdeVm&EaLeKG-YjQB6AKT)84pF,
kB$+55%ID`b-4QF0T19ckfSl,d['15$X-4cTr0"2!dIR5%1j[S4JQa0,J4lT!pkc
"EjcQ2ZmmNDF36,1DH)X!8($N3ihbR+mcX1GC!E!0fi)+ra)rCUL`#HU&V9)ke`6
IhTB!b&RK%B!&4fA8Ecr8+8IBcr)4Z8L+$bmVaA0$-Lr)$3+SMf0Xkh!%1L(hiM$
H56i!P'Q(V3ZXrmCRE,f[6f'0N!"Z$E6%fl(AqCL20Ka-#kRdjh`qA&CRACe[!5i
+PSiKjh)6PJM4H$#5%&U%HF#GqF0F$MM6fH)T68dFSQ!hQ*["e3hGME'TS#e`Fmq
Sl`'0qRTZMfEcM@b8M`(hV,a,kqB4N8iZ[4Sh5b!9ddQpT9YP#5UK!NX`BDbr,"E
!TME)X#08Bm,*$)fP2Ci@G1bTGUbETe@@q%4QL60h[2d5)BQGX-U5,*6)q)99'NX
bP3a1pJZTH#BC&"!P%4'5XP`!Fm82LidDE@#h&eejC#m'cSQd"k1C&S(CD`*"Va"
S%C+TmmkE6aJ*6S3kTd8)4GS&PNjQ"#DY1419T&!JQT+cV-0*5@'9$$5+K-58Y"%
N8Ea'&)q3!*!!UeBZ'qd'!&14D",LQVJ'$qTI1DUU3$%0cAD!e9HMkl`KaGAASBj
TJ#pMhSb5Rq0c+LJ3l3LJkD2dcrJM2Q%3Kh&mZL-JR(&m+L$L-)j29b,%B4br8)j
X!Y$j4ZUh`)[eI!A!R(d!4AHG`LH[d[f@re6*b2mAI`)H5F0aI+2XYq2iC)+N`6M
qC$b5"Z2ij,N%KHI*24K!$k@Plm*Hm'Rd8-bci0h@*rK6m%JDM[-[aZ1Nhq+IKNH
UJA&mE-V&'KM(2a129!2Mq2,5(2qIrSHmNfTSR2rTH+3D'XHRfL81irM8FE,Ep4r
eTUeM[5Ra8bilkJJ6f!)lF0e(0'p*Cke+2Nq9ccEjh#UIZq6c&[RmM(3ZV*!!cL0
k&5l"Jp4$Ilc)-m$9BDMqeV0m$l6LhM(EAX9A,10lG,aR)2GNb6Sm29&b0@CfmMd
&Mr!pHLh'hX&p"qiPVV#h)jIcaN(YAHVY!-im,lH&lp&Fc$pX!KD$+,qKqbMQh",
@BjDAX[M-KFF0&bH!le%r'GC@E`LVXP9mKXdeG)3QcED[U18Vq4jY2c-fD8XFl$a
Jb0pEdXPRCYXVR!e1c(f%qF`GKAUQcPT3T6E-YjCF2GYHhq#[aqa0'*p@XJl4r*8
qM(Fa(e1(MAb2DUZDVTq-SD2mJ+kFAj*ldAQmX-KFQf"C5i,E1fA&P2jHj`!8*c4
Cbq,eU+LUqmriLrQ-H$8"RJ(GXC,YKXYCKk(M!EcN!3MV-HG3b@DB@MEAd"P5,9[
2CjDYplkH1ckr$1D5aNf'jH[,p0ehXaPCKe@(eI0#11SC',UQT)X9K3qD(G8hK#c
C@GQUfADhU*AQPE#2X"A&i-9KaAUdDe$"bpQU)@mfJNfL,U61YQ4RBFiKFac+[hC
Y@49Fi(Ye4UjKII9Fl[b`UM[(Ca+6ZhF[@mq`0Seer)R3*#Y$$IcK`pPc%EI6FKZ
I`IV"'%bLZK'Mdl!5jqQ+3J!feU'k*f(FZf(EGY@@N!!CGAmMqd9@CrDD68d'jf(
3TlQV6AYhAEJlGh4$epjV3bSqBiDXKA!BPjeTVUYp1pI,DPfESAK1"2eSD[B-elh
H#"KCEIFl0K-Um0E-CFr[,$HC6Hhc`fDr-eb-HmN5*`iSE-8)!#TL+mfKpUV"jrc
$X6fMXIlRYZ5'5$I94YXX-&C(`""L$Dkf)VmVe*%)GZr'mh(#3i3EqlYKNKblRf*
'9fi`h"aV43`ejERI0DPfA"MDB``XX)HHa#bYS3h1c!hCcPlQ0+mDh0Yr`mEU8Hk
YrAmUXCIMj8SFBkA%6iNVCjRI%C(IMj&E3@l3G[C&a#hGId-rBQbXrT)c0e6q'2p
eC)89`[fJmPd62,qrh"5fBCA-$%rb1d1R5hbj`ddQ1G,60%Q1l'T#EqB1)110@)h
%i!95M+ekEiM0HfqSHM1k9UQY&%V$jTQPB&VZFVm*4FmG"[Acbff$#qbZ,a3IKUr
B"VZ2A1J-[B%elK$paa&k8Z63JaakNVNdL$c1fP%+A`QGIJ'bm6iH0ZklkX(0S"E
8jP*3Mb,[3pbE@&fLD'2RS@ZY1`pG"kj1X1j#2R9*X*QX*TAMbYcVef*YX2)T6FA
Q@D$Hf'AE5@VBGSP+2*elSqN#9T4Gc"`I)"SMr!P3K8hPL)Se--@E+!*#j8qBAdA
F)f`H'*JMT!TSH@V*`'V2IZI1K@DpeEljYRXA2YJ9eU,IcfjLaVQJjXS%LTUELM'
UNU1Q*M@HTVX(FV[-AA`QqadqFr3i9[JU81PlSB$r%d$A3iqhZfXV+KG!GjBeeU(
[-cfI+9deX0(XqqDqeeCrEqGcqm6iUPf$i$#AQd`B@p0rSjJ6NR2d'hX'fX5-"MQ
MU,pRS%(-F-NCDZeUk[$*BA*h$2XG9RaZHj-D6bq3!1YJC6AD61@QEFZ@lXi09,[
#3r`40LMRE"V0'C!!FecYKJh1Q(D[`hN%90BLbX@@Y!c8C8j3QmY!ApD)[GhVGTJ
**CcApF6MTA!ZjkemqUrh9AKG,PI[cVeVI+q#h6`$QIm$kKcXmZ"@c&ph+[pbaRf
+-2[6I1-)JqV1YQR9UpZ-&Cd9Uc'6i5P6JCdV6"8c-TKV%$1eQ*@af2(L22GJCe"
VaTDFcfaEffcXh1Pef-$Pm$Vic)0VQmqbL$(+mRVQJpGcr8kVcZZakIJ-9F5"VJ2
A)XVacTfpDfd&ZhSY"9l2XleH6rpD3Epa6E1D10FlQJjH!G34SPGS&qM3*fC3Pe2
L`2L%lVY,CV!*T39qcpXH[fHHVQRU'%UAhk2&Qk`VKaD[,i2ZHk`cX2[6K&iQRrQ
lbPXmS@QX)1Y!&RH`da"Y"8BfPYDc4GPC#3lV4AhlG+E(2&HTGaMM!VD)&65CaPL
Dr4lQB&J09`k9kE(,mhf[0f[T[[2#[mfpH2-6*6k4bk,U5Z`kcd%Ia$UcfEZ2Z!G
1&'%PEF2B1aKl$'0hBH`R',X1BjX`pP1-h6AD-aHa8TJD0Z"T@[KdIJ$5L*0!R+1
)NmCi#mDEj(J5i`fS4KaV[49[Y[ASjjGJCfSIkdaR)f+)e-#cLpMMH4iTJQFE+B$
RFiN4RXfXNFpBZGXAc[3QM,G2Yh*CMh@3!(q8lFE6#ID-P'YZ"AefKT9M99N2Re%
Z5UJ[cKd0UjR$Y@%N5eQr[bVdDANH1X3[2[#XjcJ0%Se1!jKa'U#f[M%BE`p&`TC
@-mfEF*1J""c`J'Sc4b0!`0Q1cH9X!e(3aCl!)H`k4qIhpfYS1)*',+EMMLJR'JM
*XAVRp4,L3*6EFHJLENI+bThcfZ@BBX$BV8U1Sr-@+@iljX&F'M+D6*J-'5#(%1k
[1&EhlT'("@L3!%(&RA-a6V0,2#9X9%3D8*&8fT'k`V(k5V),NCZX$kh*MY@GDYV
4Y-8%c[bAlh!l-U6&69c*e@N4Mj-C)C2d+XbiMLZjUSJ3--Aq8HQ-$[R0RcMaPa8
e&lLqlpUj[TGS[iMVqri'VZr9AUl[KhZi[J-YA0r"GUl[d&eFhq'YA0rr0h*pEml
RqYlHa2Ap"212)[Ba!pGh2-6e$Gc+p3dqbr80[FMe`hbZAjA&I4IA2aN0'##DQ-I
F0B%8$M1bX*!!6V&dUi!$KD&N2-DNDAZFBic&F2BrKF2r6-!j%"D+4)8c'q,aD,f
3!-3j51B9SJP@RdlLA(j+(8X++A@L25E3BD9ki@,HV9l@i1F0$6KDbP$RC(bL'2*
%ikP8)(QCZL15MXe30%"dDAVbI)DMURqBCV&i5b4dfDrbrk!LN!!@@#SGL#9B+*j
N3JH#Y3HLV#@5r"fhhq@IS5Jp9LM&BLQF6+PSMTk2cbS%9c)KQ@5a90K#Sf4N5PN
S5M[3da4hiQK)k+XiA(ND$YpSYSe-m)LIZ,6N5rL%!p$M"e)Z2G@JJJ8FXU,((EM
pQ)@$C4*&(*ZN6`SqKSGP)q02Q+F@[iqA@RaFJFBHbCM4qfMF%h!%89`D('LN6e`
k'KDkIh4i5)XM8r4*4)JcM9hKZ+)%Kcj2Rl4%aj+pAcSALTmN,qQmF&6[3Z`$k*0
%H%M18RJEF-b22R&0qM&+6,@P[&-a!BIik*1U!BGKe64B611lY)`iBNHI9"S+Ab9
l)JjKd5HT3V25,H+!P%`9Z`rkT%9kNCS1THY!pHQ6Q&%@$8)T99L%Sfhd5H*hI$J
64C28Y,C`Djl#m$6b!XGfTmrR*X8$d@L`Y6QkdK+%4i(E8[b59GP&,"cqQPC3ih4
MlA''N6k&X1iVfl4IfC%6%hNG3kaD8[4Nmd+LGcpXR+[Xb-XNFZZYEkLS`Q4G+Yd
5L413!'S-T`$1NR'U9P55`+R)+U%aM8!K9-"b-+[Xk$GR5FTkh)hN*rJB5@-L'EP
%j(6IK+GdbSlH-e9"XT!!TkM$335*3-%BFqd`miD+#P4)M`VKJ,5STAS-5DFJ,A9
lRF6mdQ"V)#Q+K-c,[YUNl&M9XNEZ@PkXmY(k8'eCj+P3G[5T%69*)e+cY5@CqV"
#$%SP0969B)9`fR3N*L#-jAfF#50kqURL8%pU-)M3+FmipZBILqkTH!E9YJip)aj
%`mKhi"GMeDhkeqSZq1IU*VIi[,SeRcM3"dM$M['C$j!!BhcZ!m11mCN2&2k,$aK
qi32[Hr5%Rh[d,hX-I&T(k6&F2UIBBc4(!m'9d93k(d+2NBr*-djj`D*SpBJAZ,f
9j!86F'3iZ$+9LDAqShqJf[jh,cLPbr2V[SPKZ8BUA*j'UT'@jR"M,2UIAFerUC*
hbU&Hqqk24KaUB492qKV`$C4!&+Z"V#$rQ"GJ24rmKPrCa6X4KAZ0c$d@5+lmTal
hVejS(qNI[*91V#iSP&p#b,2@2paR1A6E52mJe6FBBMJ1dGJL*2+9p3qIhj!![Bp
M('C8fB"h)XK)5,I&%TpfThIZ`BHa&(9Vm2+9kL#QA,kQIZdYiIaLYrARRVV2f2q
YNG[k'UGr%8DeBN-EK0EmEAlarTd(p5,rIHIa&j&hIpETLXk#R@jbC@-b,9jkj$[
SG20dc3jaep#MG,*Rm*9,kClGd#jFfLM2Qq@TmibVrRcNcU2@95h1CX5Efl"&%5r
8mURGV@U5ZdHGS,k4EYRemG4[EPCrFjZ4PqYQYFV$Li`LB4cI%5Ak4CIabTc4cV5
Z`5pfTSPdXM(B'Xb,d*RQlCVl-6rbfNK(iUpddhemB9))4J14@"k%hM42efh'efl
%*i192U1qBE',qSa81Y2F(%qfjbIV-mbRlM2Dk!QiiGN-X@CeBXhQjHJG2R%#l)P
%*m$r!"'46R)DGS+2k[XNTp(qiGGq@r81$FI)IYZ`[)lZM!cTba)YbQKh2VHq(T'
iYATPahXMf583L9i#-b!5'SA3JP$LMk5FV"eL5P&e,)!2AM(fqq[&rAqqJEX3ZJ0
4GUAcq1#I[$MlrpXrj3jb$ZiY+2BkkdRM@qKR3r"mcb,mia%m2lM89dZ[Vqh!-,f
QqNbpVjjZ29qJCq04M`2d!b+N'UT5MqGLqX832%q[Aej$mA2Gr%)2D,J,T!VQVUK
`%6jhAB9V+HAI4,rjJHFl+Pb,m4eQEZZ5@KrPp5aF@N9GqC2+ql1S&YkPdTmG6Gr
!qEV`09U+&4c&223NLQNk-DpALZNdR1mDqVXNM'QAB`crlBKL%mp(M*G"*FCZ`&J
DZ&cZG*Ki-f,J@mmLMhX`*R29E-FB[Qe,XDNr4DlPFZc[1GrDKlkqQYkKeBBaYUl
YEqK(@E3aM+N[HKM14ThU%2X*Hb(-`McNHXhpB"3j2BDaPJB6I!Ne%&qEaD`r`V`
YU-G"k"3ar)MaKKaEKl'$NQC6hd1-Lq4B$Q0G-XB+e-BRajCJ,+'*V3bd4NrqAp,
B[bJT[kddmXG*R(e#AIa5)9RRT[cr!`!!$3!*Cf0XD@)Y-LkjBe"33bkj!*!3qL)
!N"!0"J!!,h3!N!6rN!438Np+5d&)6!%!UE6L#+X`0A!!!#*k!*!'$d%!N!43[J#
3#1j"$F$iCXbcEQ9ffFS2dS@*jbZl63NYVcACZY$0##1XPDZ$V[@ke[$dmVQ6K5h
FYGEmE+(Rmc@246PGf0D9hF)@VNAi`VhS`KGM(GQA+lmmdfiI)f`c`Tq`63P23V[
Y`VEH`KHqX)9f(@(E*!Zrf-)@IZi)AhKXi3[E,M3j*432"&!HrHaD@&$M#f(,qq3
@XL1hN!$"3Rk6AcKCb%+1%di@J&@""TeG+a&(42abSQ*m9@@VL(4[%29TUPEGj%S
NfN09'd1a&"q0T8,*F(-`0#85E)pZZ-eZrEB+Z[80G6A,A6ir2'5jYd$i*mlPdrI
-@8-1XA6I6r6dUG[h&cAjUSAPI(dbhQEPDb0*+mqX6fN-*U1*9$3@'8GN$c0%(%0
GelfTH&Fd4Q0)jLrR%MNc2aM&pcf8d``Y,Ak!B(cHb*GQH1E2Phb'JLQq0Yi5)P*
IZ&DMccNrDX`mDiN1BLbSE&MC!)B+3p!!(FM4Z3"pmf##5,64Fd39&fA9Eck6N4(
q-Kr+TK`qGQ`-&dGPAb51%'Q'J"dB3bK$iZYMHPIm%$'QJ`j8f2l6cq5j@TmTYD&
8Dh0,2)CCjkGqG*&J+Y5CqU@IDmIQUUrh9q!`X*4GG$59b(1#DBYLrXT3Hc`B6B4
D3NZ)Zr'(SNLFq4ETPX+0#01J@-c9Mci&E"ETe"lZK'B2D682F5pVpcl#6cM0`cF
VIh2RdI%LA6N'$6l@jXi1I@kfp+LX3395@i-*Bq1p(FdBDS-m*N)0#&FB@QXXRJV
TqHr&d$F[UDca!YiDjchaf-C3%T1`bTUFNM26%1V@@T1GbH#dKP"R2*d-KU#5L)D
5FVQ)&NXr0"XEY)Prh,6j`NN!Fk+aB(Zk*F3lDTZ$[P"c5bMC1Arq8UD4i#5T15f
KF$3@iP2*G)M2RB8&#LRFh0iTXfaMT'5S@aDD8))aK6DZ*"9[2BV(P+51c4hG,L+
c53S*k44Xa8Acmd49U9R$Xk-p6,4P'e,Rh4bZH3"e6"(G$Pjab5Ikh&MNk*3JKBH
am`[rd,p4KJ)IdrpGAkQ!SYrdArSB+K6p(4q-kaYR%DeiK@MHTTrT+airpFpf(!c
C6D6hMrH[fSGq[SpSi@NLdj2ApC8!q05rrM0pH5A%p,FGr*AqP!RpYPrTjl,kIr)
Mrc0p)kiXJcl9Cb(1%'6hP`BRQ0MP'EU4U`lF@CCrSLp0(%#3!"HAp98B52*lSGq
&ZrfkrM3CD5@kEp'%2R+m!*ldPFM#f(9p0R-`C#rdT5&)cLr`#Kk#rMULrlIXZ[j
d'6P$Y0N+!(Y!54rDdc&h'$"brDYqB3l4$[hhr$0$4PE$2eXNb2ieb2fErJLM)1T
RZCa*(rQIH68r2Xk[*I+#iKreEj!!r52r-kc1XRmYjSpI3ai@B(RaKIqI,BSqG$#
E'MkH69X[ckB'iJEe$Qi`RhhAFB-&cq&lKKZFKRc"-D9m50)#'Z6Fp%2+jFLffS0
N5Tj%4@C5"GI&cC(ZFcD,h$e838lFZmM*m-eX'F$dP%A,,mqff[SF8$&N-KPiM91
9NF2XSa0J@f1fH(J8"hGPCVYkTSRLJ,V55r6R486P'%J,"U5PdFrVi(p*UM20Z#1
AjGIGE[0r"EdLeqdcjp[mNSplX,Y)hCYJ5aj0I@@G*jb-Gm65lHf-'iiR1d+aG!I
M4Q-YACfKpTEfZ,40CpQLY-XkZ5B+lNFp6BS(cVppFXHLm)JE3biI%jRZ4TD29iR
SY!R1P$QEBbjeBD*lqi'1GccMbIje'bEC1H@a56dI1a@*I@9pEqBF-qYcdaaAM`b
5FjP9B(QLVT*e4Aa$'kXN*T*FX[j[jrbLXcJ8Me@X&Eh%AL-JTT!!Gd4B3#S&rjI
6(0UBDSje*M'BT4+G-9BhC9*@-5jcH$[1@!XpJKl'$ZGDCHXmRb03ICB4reapCC!
!(Mqj("6&rGSNfp+B@FQGKfZV'cfXb6ZLR8&V%2h"l5[mJ8hjJPR%eT0&kPUA"r-
MPcHq*D-)FI[,GTp4[[$$5jiqJ&BGP+G#UkjaI6!H#dFM9NbNa28pDebXI1(,,(N
ED'bUV!CChjPULFDCN!"U8NG00mXke@ZV@1Ge4VY$ke-3#PpeT"PAmJT`"+9)V,N
pTl6IHLkVI,'RZ6PAIkpR2HXM[+GCRdK'0dVZpqGr6kpmXC'CT5KCd3'NL33K%LA
eT(2pQ21Q5[3dR+GDX116UUkC9$)S5UXm2KGcINq`Y6NTP421bhiMS(ba5j&Vj+N
6f#aTQ1JNeElPhNVPLj`GVbDV%DYQDdZbmeS[j5Xpee4GLelLG+PS4`JbeUXka[&
k0V$H4$f6H2FMHFHjNP0bI"Sd(Fh4'2DERk5`R-%10TmaEFjrI`$I68b$mrG)kq6
aHBBP*&LlQC0%8Xl9HQQfr9b!L@&XcMHPT*eJ*QI3,1Ibj`$iNqZ&q@YbPJ1Ha&!
Tc3P+,rc(E-IjIaGE%9QEH@4l"'92bccba&FiN!#)&l6[jHikPAbI*GrYmVe9[[I
)phhbr86Z2U8bGeIk!)'b%TGV)mAiNDCMGeGHc9GI%IUT&GqZ"BjUSA+ed+mA[-2
LXC)(FAZaC"ZB'D&IrCc3Ep!"HarI&r!YF8GmAD,SLj2'YmVA4CaPLEK2k0IH*6a
V*Vk$fS9GI4I"H5aL!-[(@%*ka9$HA3N5qMA()VUDA4&9YPT)mi[cZX*6&cM@eJP
93VpZN!!h"R3P6RiqmI$[+mN)k3@15PH6#pcRH,qPD`T@&9NVUY3'[UeNf`)(%Um
4l0h!LdSHK&T$P4pi$qrR04'Md+mkS'(0E3aI&)EejF*+mAAAd"56T5l"Ckd*lZ6
dYG-("ec$9*M3CUehlN4&9Aer+0`PT+AR#H3GeRp3FMK[%pq9er8Y223JLKM!HEY
N,mdU@jbA#DY@la65UhIkhK'(PTE4BPEM30kDR@@'[UIiiUc6TNIh["CTp`k2hPr
5`jXLjbc1QSI$eZbmE28#KdHUPIB[)RkQV95-AKqV@,pZ+bUiLHmHp@@M''(eB8f
f*6X2R,FYF5Vrc4ePeE6)rfDaf,5cCM&h@d69*`VTa,5qikYhmZK0Ble`+6c9aU-
'$C(cf9ZKQl&q68LMIi$490Bh%PU%6PbL0f'aB1Hl9(X5aT1l$Kj@l3YE82GhXer
JkbdqLcQ3!1Fk6iB8YmemmZL+iq,&A6dRGi493YT#@5[6iERXA%YphBr&!El1[CF
+&dD44l1b0lLIpNA*b0Ie[@mhS`,[c9hpkT&bXm8F@aUa0,JLKIL@V(3KLJm!)8*
&l+8LDUmD1G8`KVdmJ3fHfLH1XVUTHZhcb&J6TE``hq4Z-c@i`ef*B0pah)HB(K3
H'HbMU6,f$BBChH*)C%0(+c3dM1IjL9Re`SV`bmEQ#NIi'&Lk[$Dk84behl,DCHN
H16RiF'r0K2I@`Gr,ZCIaFJ8(9XVm+EKbPreGN!$mr6@mUF84qbhVQ,I8i-1$d1L
YqD*,(#erAVJEVY!Kh&Y92c(6UfI+c4%lZQ4ZC'U$+c`cjjFl(c$,5(pJUS`F$5#
EZE0`h)YZC!jHBaAMZcmFjCGm1&U$M9+Ne&j+T4(,h&)bVh&lrSC-Tmk6jY8epT%
+KrZQ`[0dKhfNlm)+9rKGp,K6bKpRq*MNS4mHqT0LLL3I0lp35RH%Cbk#'pph)mE
6[h0S,fP#'NXTD5D86d2hbhap`Y5EHAZ(lFME$j!!1d1fSr"6Rb5lf@C@BB2jcJl
d"Pmq29"SQ8HDhKll%9B0qe'T%Lq*l`B@mDEXREcc)d9M9,K%USLj(+VSJHQqK)Q
BUR$*mLCd,r",+)phKPA01S'YCFRQb(lRkmXX"TYMlpHHARDS*k*$hLm)m'`$`C@
&''S*&!*9bDJjS-&YYQGB2'VT%G,Cl`MTLd2Sm'j5'3C),I`f)I@3!2%1,)HU+UJ
[bkq[4qlc"L&GfMhFDr(rrZQrf[,p)kG15hMhd4&b@XV0CQ"E"aq41''CBqMY(fk
6'%db`c6B2p`N-G`b3k2E`LC4PM$L%f0jKiiA$`FdZ,h'8JHGYGjZ,MFIA,hUZ$K
Fiik-#KIi%CQcHi)c,(2FXEaGVJlG5DIV!UPX*XE&5&T'QM)AD5aPC#KEMpRZ(3F
@d#@FcrhLGd[T9XjApG)IRkldZGhZJ5-RYrVI*)HP'-lr3A8KTMck#[J2AZG[`VV
Jha3@r)a[((G3NfNVUYR5CUc-9'i"NmFYABR*P@C*M$5iH4*6"eEDLVfl+"l+"(8
@M14#qZ$f$FE-%Cr66QkRcbQN$fhIF,09`KM,jee+2Zp$4fakRpHZ&p+X)mlfR0d
"PD(-NB(YG[A4!D[DjheP`1FGh"ibp'lGS''H'jf"FrF4Q`L4&ES+2A+LQ%dj*8l
JqAe2P46cqDAU"Zq2[3hH*IV!V%Q9RJD[$Y[IcD0hlLbM[MffBNarf[!E,'IqV1S
aElL)9fHGF2%%2`0UDi(dPMEbbl2c%Kck4I2iE0i!RV[80kDaL&r1U`2Q5CH@"Lr
[j0%0QdI,$*Mbr0mIb&Vl[VlL6mAA(hfaa#pj@9j6KDPc$R)3I@Chp&h`$&mbSC-
1!RXIf22!RJ6fYm!H!,BEf0m"Hh*LCMEaT63VNSGE8@5Q-%`Tk#5JFa%k+H!Y`!-
bRJ6HK'V%dHZYf,SBN!$R'c'C1LBRd`93$,0Ui1jQlR&I`LU#Zje9!2GEQ52F,Ia
k)@hM(PmfejF`2MlEaQ@pYK(Kfraah#la*h*F5bXCXX8fMUr1HS@dXLKKFl&i-D,
KRHjGikbVar'Y9la$l2RB6pmR,LdS'+0CVLaC,H`"dT@r%Z!F2cScr3P3LVMhU0$
RDQ6lXmIBIJ6h2FZaT-(pd#Tr(GX$[`!BEfIS4+1rNEepHBe0*1LCXfaR!QFkYKh
"[C!!E89`RpfiTTEKYhU%C9l5FSYb1eVZ[NShdqFHU(5[B[`[Xmd%lNp8ZZr%``V
Z`-Sk2q2e,eY9c6DeamCH2MPq""hf),AJ0Z`'mAk4BHU,`2"fN@(D$$6B3eKJHLe
ijh+BEJhfCmrNX"X@BR0iMP35pJI3b"!RLM2TKUm#`jj4mR%B@%X1Qrhh`&k8X3q
"I82'4(M5h,f&[F[64H#l[1e2f"XKA3FdhPMh,0f#,XX(PR*-SARJ23cXC6*+rTj
($GBeQHQ,U+Ad,JkXA`G[(hJpP*%d'S#PC1a"B'rNDPDX"RC'a[6!hT)eeX&I3XE
f-%rDMYpUEQfrmLafmJQYmYTfr+%XjmL[Mpm65YCl'2rr!!d!#'GMG'9cG#kjZ@0
38%-ZZ3#3%%0D!*!3(m-!!%+&!*!%rj!%8&*25NY"5%`"!+QdiJLV-$9B!!"5l3#
3"K+K!*!%$I3!N!Me"!i!pCQCc1abX2*Ef-,&mj8EA@KjV4fRQfkf--,fZP@[Eld
Z$dq2VmN'A5Bp-hbAY9lHAJFXfQdl+AG,Z2)ME*&GEJRrA-libQIDl@-,fic`*fc
6K5HKhAEKE`YIq-)mEQiRK(pXXmb@iapGq-+kKCfFELT3q1c,IZ&ZXPf1@pl#b%)
ffjdZC,)F@FK#&m,)B+r,!D4[CPq-FBbaqZ@-eH&@A,@%-I9,M(@V+THFE3i'I@,
PFV%p`R[E)f,)lA5*'SmV)SBMaKm`"H(DkkSAQQdeb1%*lP8%I"Kcj(3rX&H6m0M
IZTkaqjrj`UCT$PZ9X*!!V`m&fSamV5GNj#ReR!CAb"Z-H0XpDBqF`ePa(%eGaiT
)S-2EcP+HcTr1B+bXmm9Kh'q$6Mf`X[$"KF4R$RhYV2*CXk3m49H%V`fdL)`T"cl
J+-2j13Fpcq@-E8&E8'&IE%H%!Ne3,pZF#1HDf2Hf""Q,&l1('*Yr8%EphJ1GXSF
r%JrNr)3rGBV*(aq@mf,a)FC8Kq$ER2+`6KCr)B9h0"r'+0,%0Xm[rQdqSqFB2cQ
eBU69f4*S4krcbhc8LClZG$iIR'*cIAh0I"abUXM3iXkAEq$(ilQ,49r!j3f+,H)
maNhp56c112ejNK@"P6JkPXIB&fjK8aKcR!drZX6iG+jqq&li[TdQiqM4U(!CR@&
rGU+(,&FBA8QAdZJ+kKT@q*eSAPdm1Mm9!Sj'C"RE!a%aQhqm(IAaK-)B'-FE!ha
jS(fj'%,(Uc#'FK,*f-@9@FC3113DEaI$J@M)*3)Pk"9$i'!+Qm`pccf[0,(*#J2
h%ZcNS8*JE#k(6ij38,[0q$[cVaRB"FIjhRDA,pSLmUCDTmXQ1P[%8(M@V%X))mK
*81HhL'j[ZmK(3P'46jb,ab@$h%jI@)iU6J@&a*8bd!J5%NZ'TC%NDKY",5%K9lA
%%1kQ%f8Z9IE(4kQ5X*9Mq!UPK%dirih2+53-k[E(m!QELQ!-Rl#ccq$6B)6Z-I`
FQ(52iC0Hd6f'2a&QlKPm`YDG`5GX%V)aI-*'%r+rq)3prJ`qB9260)C2f"21i"-
feI!B2QRI@@I`#A[5'Ic*-1NH`dIV+GeMrFY8Q(52j8mG(mdXar#TGUKe(X1R`pq
T1G'EYSlfTT4IFZ446jL-RfpLA2G!eYX*@kf3!1dTXPdLfkfbh5AE'fAlbB5G8j'
`4rJkCZFXKT(SUhpj-0jKc0+KVIl1dd)2DmAG-GY8*93X&AUb"HYJr,'#0E!H,EJ
1NCe#Mr)KS8HMKZmGh)rJ,V"iE"haZ#h!9,BPYJl''HE&0`Sp@9F+$qSClfFqB9h
h3F6FlY%JbNC43[653pSVJdcS86hQ89H[mbKL98+8Rk[YF1I00PeH*e3+2HTqAYH
N,LMMCc%HqGX+1SASE&1&f@&'l%0mMD%M4m1VBND`e)EiiS,VCTXD(2B'40m'rl5
#08#c9pE!hmAAm#U26ZK4E&E48%VR2LJ-CTF+Lq-[Q!rPj"[UJRc-'14f6EKm3Rq
[HC!!63aQaBb,eS*44IHY`T9#9"TN-1YJpRX&fl4AmahDMZpMp-1B4i1Br38Ef*5
LZGT1Yf,T@L'kG+hYpILK5iVBA1+i5A[CfL*0plhmp&KCF6DUCir(CadF[VkJLmr
hl$189GrN0XCQaUTQQmSPVV*HpY33GT)apN++X4le+M"i0Epbf"EcSZR0GUYL,E'
CL0P[#,$5,pp39-AQe,`b2HjB@cfAZmLMk)i,dH$ilTe,er+S69fpF0LG9mb$!l[
R31a#i(BDla#LU"ri@"l9MH5GKNUFPjh[CUb%le$F&p6Y@VGPQf+Mf`$HhiaG`0F
EE!CpNpCmJ'NLh(AkA6XZh4NrZ+jVe`eZK4!eX*L4F(JZ0X03ArHcH#pICpR!*Pl
XK4j0L8ffh'rc-KeIere1L4i-[$eMkE2E5r8'IIXP(S2Gl*Q)Zf#a'@X,Qq&K$)b
8&-E"[@,S'A[+pp5)VrqCMI&KiNfa[Q3Qde9lQGE01baYqAD,Zb2SkYi*qa$K!H(
QrQk@*rZq5ckG*6lNDIDh!N0&FHA[kK@2A1Tq5ZHFEh)rKLLeYSe0M3qAR,I8E&J
jY+[rT[A9)lQhp[p4)R[CAjVd`eG)q5Ap59[1Ed$+lfq3!*Xb2P4bhK@8@k6rTRj
JV+rq[$NqA2U`m"9NK3VKAUem9mqHIDj8lbP"PFc`j0R0lNQ*I,N$6AVCdp18*hY
f0%'EZEh)H$fUN6,B3ica+pmIjZHp2ebp!DT9@&,)#Mf''B9-IjQPr#f@rm`"TRV
fXT+Kq5E,f4-2X#q@$(82A'Tf[iND,j2dTmcpQ*4$$h,S#F8M6-VMR%F+f4IGNqB
J'pZ22,VGhpLkJDP%PD'3!+P'N!"h!rF@[MkB[ljcr`h&frIIb#bGV(J(mUN2X4*
pX9j4GNhmp4Y3'hcTK+D*KTP-YEkVC$Za8E*$BZ+*q*Y0FrMmf#+ql$LLcLXFCJU
2[K5SU)%*YQ!q)e6KX1%9i!l`mjL@,h-VR'U"@M4@E)Vpm1i&"NfaDF-GpbrBfZ9
43qpR0r'kZ8c&&BRN0640K&FKHr90+PMRPJr'GaLkK'MXKd,di#&8q%UQd23bTI"
9"Y@$aT[+kbSUjl2Z'0pB$phR08+dF1AJHN20YhDrGZhcfjrC,IPAlKKLCBC5[4k
q9Idh5c&Z18Dc[QH`6BT`b"(jr6f$$LR#)NHSe0H#a(a5Q2KG+Ee$aFHh0DPJl5(
93@8ePZK,p9Z@,YNC(kbfH)D&!Aj)MVPY*'C3MV'dDpHCrHTGCHB"TLM1TeLdU%9
-9@4Q+N-4da3eSVGlhF4QX!,1CRRd4iAX3Xj@qF4Il+k`@5b@hZfl9Y@m`Nb'kFM
m(e%[4TI(rJ6aDdl'AmecRb,-rM4HPmkJZV0Y@[@eEEU+cSTV%FR$LPDJFf96T)J
SBV95T"T4851Qcr(ieNkAfS!@ABKZ@GfXkpaZ+bYKPM*EQ4$GZVVj(+2NSbLEp4*
QXhjcHh'fc9U5,85T)[CflEd"+)FkYrHZ,P(Zk$8UEGDRHfh@rY@LC[fUCKAPh&$
@Y1rVM$T#D)9kIMCdBMTe139Pm1GfheX`RFmY90UY2l2DVI1bQkD-SR6CVHVV',Y
QH0(D)YCpAr&dG(pClTG)CrkkmRDVHaU[M*8KLl[iXi"f16cV#a[iKE'C33leSVV
cA&k$1%ZK,B8aKer)+j[dSeNDl&DqM%FeA$0FT%'A9r0mEmcBIIHPIa9riGZ2&Y4
)Z5bXVN6AH6jd%(9@BZSH+"mmR)p+fJ,I1r!p$0mpm2dGI$I#GaYmI`rI25-pFcj
Ib+CiY,#QH5B*Jb`#R#"`$J)R!Rm,r%fb2`5r!f`%81ZYQ*CVS1I,dCQD4M[6f8"
d%aZ`,C3pl(R%#1`5BJ$fKC34E!2I+%5,Z6XAc,!&GAHH@mc&V-9$`JriRE!1mdm
QBJfY6"1EAXca96'V%%d15UJ[MKrdU2JbblTde+I(r2fRV)GU*0F[GKFZ'6FZ&@C
!@&e$S`1V*BfZ3,[Ekc'f'QM#1TGaI6mfFAd[dRd&lTYa2mhe[DcQqPkGarAYVFD
pRq[EGj!!kh[Gb2@pdFVerHebVZqYjlLqJ6bZladIehI`(Ul[(a4Fhf(J[@rMqRk
qJHZ,jh2ph!,FAqIkPGrNqY@YA,rQDG`$A2piD5R$)dE#I+49a0+%1a6`miQp3Qa
bq2hBFJaMcC%A-H[Lh9kI1084#2JDa"!f3ALEk![b$C%30K$$+Rp)$+Z#lAk4M'@
U"BZ%FY95Keh3%Y-m5!m&aNNZUbm3$MY$+e3GhSKrHRQY-ib9%UaRb2XM&r&Bb[Q
$#1m2Y(MG+riPr[FUR"'4$dHFrL$[$S4iX30Jl8iIhq)0r5khhm926M)p@LJ6T9)
i'P,4l,[)jI1kP[&L+-6l`aiMMHaaP!k@(kR(!$5jIF64)2HV9c"fkm2Bb8M[NA,
5*ahe$KKB9T9'TSPBKI4**`H4UR2Kk*+M&9J[`FHC*Q&NUD#pVUA83F[45Jadk'0
F3Yf1$dpTM65,Hfl&AGM3!#1U'a&eQabGKF82I&eA%c-D$%HjjT%"U4TMFAb*[&A
h)@)HETXFRBf&$h`V0NVHj1U3!,`K#cY(qL511H*j`3MI14L%iN0H')LU%pY@kEb
e@+I!ap@!&jDr$K6[395bNR+a,%&ISM6!LST@Uj*V5MUX3Y#A)"$4+kM@NKY`il$
S30pF$R`T#q@S*(BHeKMSieHp#Flf)`,0AQTaDcb@&2)PHQQ)5fb5Xdb1cXF+!Vj
N8DB2,Ic5f4Kjid'T!M!XRlE0,$48%8&NcjVeLhiPLG[pfVbedR#BF'qX0CFl+(-
SP#2N$)DCki1*FLTMEYAMF%qMfLlECUkT+5IZR$kIUlACYmcS)YhC12(&iZ3YB9'
@5Q5*+ZHdkID)X$BCAmp+hXKTKT6AHm#U3r4C*hSQB(BrU*ZE[*&EJ[hH"NF&f1H
b`j%@Ei"`&+-i5TRYhSDUbbZ*lE"hTGJB!9#%@0JA5pj3Yh-5l&V,'fQFRq0a03C
$hZ956TYb(mp1hP#k+8NN)bQBbZ-#L*FT4c0ATc*h9&5!)3dB`XSCTF08SdMC5D3
Pj6BcCAk9Up8CNNK#jN9IDNVH8!QCSr)k39+0G(N`aFD&eSVN$99-XdNF%CZY,D(
`"a@L69D5SkS@&F+T)ekr#"MM-CcF0*pfUMM`5Hd-*A450pjlk`mPT8VU"Y9h0R3
Mi#,4b)#J'D-9V[Mh#PIqZX**-8jAH0BrUp"aT*4UR0)#8Sh6@T!!8Se6@T!!maX
Yd(kN"FGd1[HIG2TA[3DH,8Mf'TBDXp4V02ZFVQ8q2,U3!#'KemM%T"XRp@#KVcU
Y"q@f5Y+$A#aMZCD&Srj`4S3qiL3hckljPY445pa8@+b09#FYcCj'[bpc@BGcr'Q
!%69iq@)m[C*8URU(RG4!'ib%'PfYVS`*8j,-6"h[aReIXbG[D8k5c,e@cYh[$#h
lT)pilFFr65[(JLU"+N',p`QF2Y40KM[Pq2-plHN1e&CT4R@a((P61@0C"rU4'Q`
blVmMh8FNDTaTr9MRD@`4JjR-qSM6-pGM1,T84T8160L3!*%BDI-(2jh'hIh8YR5
r8BZ42Y@"2cR5GhfQ,m$+0,B(FZ(*qFCchdR[JG5Dl3[K98[0EFBhc6Jf!k'Hj$p
R)(rUIIG)ebZT#lVHd,,'8%3DJQ5UfdlEP"@LKiU5A8P9!ff@U2hH-(@biF`FQ[(
KV+6++NJeiI9JS(a#A@K@FPTGe,p@Pj4QR&)AdSc6kT,5M&2U3T15dqU5QT4mULl
T5FPrl#eaeipXJ`L95k4YN!"fmDV'M(FlXp`hrMJpBDZc9%XlCB(Q0M6#dJJhdpT
%2bZdFd30'KTT[d-6#2rA22prCQFCZHEjar[pNj2C69PYp)K@DM)V+8'fT!3C%RU
0$!Sc%%F&0K8NII&jQb@NScQPp1@%DKc0DD4,rDbV-ccd@PV(lCAPY$H4%a*G2UI
ARl'MdM)(c3+5MpDF8)f1Rr4*kNc)faB*9I4DMcVDlZfJPej1UXfAEck8RMde1"C
Ci0@')p(QjN#S(A*Mr%a[J*8"E)T3G!%pL5YhHBl+"RVj4bhpa)5,Y@G#d)*M[FH
rp@3IGap(N9*kF+TlbrUSQrlA5IIaD[aidXeYj&CVNMH83&CM+!&9RaC+%&Q"[`%
!PM5C'9(,)ph(*fUTr9!YMqT9DV2iP&iGfErj4+r'r8D[mMkHFibb02iMPNjf1PA
[d("$VLh(CI8d(p1LX&VN*cJbP(8k[pfF2kE#ZPqTX(51-%LC%ZXU[a22)[*i8[E
rZJ[cIcUGL4G#pHMBk,e2kCF0VX,2PP#E5Iik[#T1$qmHrqXJc[6'Fa2`XLUETTM
$*YV-$D3cYp12%m#qEb(qhJ$feL8eGE5PqJMF0!YqXU&'QZAY39+9b(8[r8`"-MX
Ah$6![T!!ITF!pTb'bfV*EbNA&PMaKL[H#UA+i@kTX"!qGeH&C3R&EkCI&X"$k6d
9PN9@f#m[VUY"R%+aB%N90%@4PhahPUZj([c3IkY-$A%eUr''+[Q8"m(LQS3[kcE
1G+!PiF[1j8b6mBiYqG4I![EZK'rFji"Ab"55leDmdYV+9*,[$[MHa&2kj,XIH(K
90KkIa-Ep'I$!Tj5(&h&2b4cN`,G2pSf$$kqZ5Vi*m(hh+pHLCV(B#pqMEAp*2`L
K$S-ce482X[1!F4&mDd`jE#EL`-(e-DD6q,X(FCd12IXm1+#IdU#-2SFi1q)HB*d
54KI`ANVie'C`8jVJFZTNa%85A%ip'ebqP1"bkZr$jj-acJ0'8-Di!,i@'@-Q-2E
*q68KTiMXZ`ja[9RqCFj@hp%rG"RpQjINMlqNrpQ&-qA@"ki53rAP&2rr!!!0$3p
YGbpRBh4PFh3Z0MK,,VN!N""453#3%#pd!!"+8`#3"2q3"%e08&*0680$!3#V,jH
ZUc!jB!!!"M%!!"R%!!!"V3!!"E(*MaZS!*!'[VXM4!iL+Pj0j%)PIdhl9fbRBC!
!DR1(JAFp3hUJ2KNcZ@(k&LeHlIYc*cMM1X2GRCf"!*`N(81C&iAQNTm4&Ifii1"
EpGII4h6#PiP+'R-jb[e$&IeM12rA3hh-XBk+D2XK9#@U!P9e!@eRU22XRT!!%ar
%6jaP3[FjFKhiIjQ@hidE$&25cAm$`-IrIXai*1U*jZd88q%pXX1%F$M`RNJbAQS
ih%%N0J*@A""6p[pE#%1,cL9X%K8j[Z%i38$F)*'R%8!QpTQQT&06TCMf4amme9+
jii[1iC(HE43E%aa#QlrCjZ4[GSL(8*!!e8D-E"#r6LR@&GN3aF6F'028K*cdTGk
aT$fkUhhK6F,P(Tj11!CFTLJ+QQSXDINp,M$RL-+Cm9q6j"VK+Hr'rhrjXB16b1@
iec&AC&Z,)bAP)A[QZNkT`brFF9bj0@L(b*(4H3)$i*YCbh9`YK90aj%$0a!Gm&!
,de[B3!XlC'%$"-Eme,D0'(Z229-8DlB`9Q$FC!Y6@9L'KA%@PQm[")V0YM#PKBP
$[mI#m!L#i#MfjAH50i4eE512Q3bj@@90I4m!N!--!'XcXfpJlh2Ij$4lRaZHF-P
a`Tr-D)4&@%FjIAiV9hi5rZ3i@3NqRhV5`hI'm8m[3MNjENHi%AjN`!NMR"`rbB$
bTrc)FA,m$%r*F51Fm*03FTa`FTa`-Q#%%hlN'4R`Pa`RA(+FF+mMamRa)mq2m$2
#bB!#GjN8B'@Y6-+0iUpN*rl)-F)*2m)*8[#%!j-9H"9SN!!()1QkKK#+`Hm@K$S
HJ&m,rN[#E`hmIJLEJ,q0bk)PQTCS@&q4J@q@4d"9U,FU)md-(0Yrf-'kLSC3Ech
QTZ6PDfM!,6kXTJh48"8c3%-B$Af2ZR8CG9Ip2$-35k-p#&9[4Zd)$4`EE%%G46!
,R0"9-23T99CN34j4,-#2%@HJ4P(6T'aDQa#N[iMDX5G2a3J5j8hqU`G8AI)J-HU
[2pc+8DXTel3Q5K1DDDe`rC'MeMLS#5QV5"2QC-jFKV@(Y,XiDUf$'TI6Q941+fY
NIrEXmabeMLSdTZC&6Ae8m48krm8h(,@HFXdUSU`BRMk!q[lRHBlD3,RQ4#QENT@
#"cXRI2X+4ie6jif)dMfM+mkEUadrc9%E(G5'h+TKlGFqRHHS#3He,LFDrPe`h($
QCBlDa(3e*P+'jG["RP9riDM0,PI9V"`8d09SikJYP'YH1C5kHVfHlZ'SDkKIpI4
i+LIkaJ28)bpbe,88e9!N694cCG6ZNqFjkMUUN!"T6DE6ZT(h&AViKGmikRVU"NX
TAdR(H9q1FY4@bY@D,XL9SfF2rY6286HiPp,*+'9G,aJIFG50p#Uce14Gj3Y'd81
Ek"h5cFV&)blrQ+1f8B8b8UTJU&0#eN-9cVh+8GXGe*U-j!-kU)P6p4b9*UB'dj*
PCDb-E#IIrF$K4qBkCkfIRK)eFi@ZrFEXr4ae-h@$T1I(e%`C&K,!AUi3T&L#1U`
I'P&bCG3h(rRp#Fje+d8&50fBrKHeFp&j@4Q5M3GV$pea1eGSfk+(0$9pa80R1GF
ZCkfce*a5FDbGI1mKMRSpifUSq482fFRj!BlD6Id+#UPkaDr(MfcMU0YGVSSeRLY
8Z0V[F05H43q4)19lk0aM"lL(GMKViS"LkT1'T(MH+rPeTkZ3!*U"!([&H8FjkLl
+0@RS306mKfX[64ZJ+`31D"5@fGUaCaUiVRd8Y@!C+5NVP42Ef6h&a0E[S,D5e*Z
k$e*4k[,4R"1qUq@S0cKV-k$Hk86c@fiEqT2V*rYSlLHcfePEppip1YM9Hl2Del9
2!&`"@TQ,U#F1&[Z''jdelZ4b1(ZHmdimH"0(45eR)(&!*q9f)f6q6PCX0VTUBad
IAd$pf!@`[ik1Br'KUlR)+fakrN"cHF(H36)2h%jb&H(+NrX0&jMF9VMIj$*$&L)
T"p)0cLf`Yq1%"AXR6JQ`Yq'FKMf0GB,GdbRXPYLiZ+lq4#IBL8k`%jeJ*cV"6R5
#RHJ%1p%*GU)6l%3Rf)P1h%qc#+[@Y15RS-eL8qhT&"fJcd&k4dVkK,dC'pb'AVi
MRZjKXmB'HccD3(IrcJ8G(KYmfk)&p1R"5Hkrqa'fKQc`$Bdfm0&Ek'dF5*Cm&25
6E"T+qQc(16M5i"iI4FpKHCCb3p#-XSR6I3[1YF$(e@dVrAm(hAhGA,f#1a4fVQ`
D)a0bM1IcX19PNiJXd-QrQrjp$rTP0Nh4$ljDEE6C0*GdfSPEQNJ$[AaI"9dkQjE
)"&rjZ5PSlpQXL6c)65I42'&jkHi((6HE659pGY(F%GhJrk#CBp-AQC!!QcfG`RF
BE0C'2GbTm18(Qh@4"hI+cbI"'a-fkb-2I05,Qq*VI86`ZS90Dq6"IEUNPpZrZ6d
IkmP@hp@`f9$5UmK,"LjZ2dGjKIdd'pTRSrf,Re6[[[HdcbYXX0R3aK[KcVI)#mr
A-dm"R8jJFcLjAc2T0r!1Xr%Ph(NRKdhm"Y1PM9qd9#9(PFc#![X)[SNKr!e@jAm
!N!-0$3pYGbpRBh4PFh3Z8&"$,VN!N"!4c!#3%%+&!*!)rj!%68e38Ne33d-"!+X
[PkkV-$P&!!!'-3!!'Z!!!!'T!!!&bE5F%03!N!B"fL0%$L)UANhN3L9r6IYAE+G
KN!"UFiH"Gce$HU!q'61jBIV#iB$[cjhJM1X-GhH'!`%ib6Q'-Lm+c58r)bVkFF(
"YqU[[irS4$#9MENFjIkKL[iaR2rVS6lQ@%G&Y2d3UK*9JDUkJ,Bce(Pf6fJm&6R
b2Z8HRJiXa'A+ir""h#2TreqK*11PKX-G4'@dI[MrP@fl(cXiL9b1Haec4BbeKmP
aeJj"iNA$iL1d#Y1J+HR89#QQrG%86l98l[LLFhLNlhad)NaL2JK&0pZFr-d1m4!
+XYS)fcSm[diTeKAC%-A8h"M6e)5Fp+AHXD3p1ZNm1FY%rabj$[`E!$0bi`E$P26
rG@!p"$aQr-JXH*CjLX,-Um9UPGj1-5VH)fY@`*(4VHaDSf,&r6CPrlq&--R1K6X
*#r!9a`Q#"HZ@0$hdcLR&Z$Fm-LN%a%6I)NG'j`NF&EkCY9`(CaX9iFL4(fpK!IC
B8#c-*P,XP1dG-@D4KE%@0XR#9"C'PcdhhF,ZXE"3#eYVB3-&a[CDHNU"FB-@YXI
#PPJD!bcX5f0T(aH0)DaV'hR-C-M0+Q[Uq``!Da0l'f3fmMSr"jhCCQZ%N3NRNdf
14LJRP"rPR[a@3Sqr%8D1NjAJmk5Hp2#G-Ic6Le"1MJm)Pachb(2###I(6c*J%8k
j%8k1RiHRj,J46[K*+$P11$P11"P3`JNrmS`-q)!-Z'6!D6eKj2L4C`f-F$+J`(8
Q"D$m9QE4e,T1r"&qK,q%%k6J#3FQ+c!qS%%HJ+LU#N)S",rE%'S`i2Fjq"D$ha,
iI4qf+2P[K53BJQi)Q['&0I#IjBQL)Y4CP"42pjcUHm,'ZSf'8'HBF--Ck@qdLS0
b3K-d'HXH'L*+S#ZS9C93Dp(hThY##E32SH*'Y!@KRP2p0@MV!TJ"6QM*DZUi,'%
T+JeJ!r"$PM03TD!SBLUKM%E&Qd60d0-c)3Z*mVDqK3&9&I13!!6eTfr[iUM&P'Y
#%F4446G@Z(l(88YXe)LB`Z+S2TE@Pf(0!mTp(,A84Uf3!",*H&STD'4qmZcc(,@
-+M3XC`4&IJbl#Phql%Z1'UCF8eL3!,#@e`G3hrdd`e(,+GHd)+EL%XlQBDHLAlh
-85ZSm`B%mB'K&HG0PBjFj+L90QTjHXf`jUXI6h28L)eDPKBdpblBE[Mm*BjDaA6
94Z1DiGV"R*4rj+M9$PGCAcdSS+ZfQD2@8+iCI$4qqhSpdmj4DkPIeF4)2#fiaJ2
8Bbpbe(889F1L)XMT!QVVKFXFYBiUT"YMLB5UC9b&(RhKCikkRVV"`)8VDEQKZf-
V4kfRA)f*V,4kp-cqUedFGB0c+Hf-8Y$eQ[B"4pe)Vc*,6IQVI%eEm0!QHSG8IIA
L@5lrN!#MEUCQhhrr8(p,Ec3@kie#4,V"pIbK)a`9-T!!GBlk-`E@KJ84,f%LG'i
f[T!!!'KLUKNie$XiPM(N0&lQH[KU'dGYS"j+L['X*Sp(8hPGSl0R1'UMM9U5&&e
!'c8b%qDSN!"L,3rTKL6Ki3+b'A[l2CZI0G[Y06`a,LMk#PhcpFQ(1'S6pDZSCSE
PC!&fUR[r&Uj3-eASS(Td!+F,U1H1r2)8jpT#83&5e5EZS1kBGej+JZb9Kc82h(-
h9kKehN1+R,MPS8ZFDjZpPPPABF@aCZbG`abeRA(9j-b+KmcBG!p(lD"q"B9NGG@
[CimeFp5G$PGXM+5cUec0YcMUVRN2@9(2pG$Xii2F3jhf'KR%ZMUQL6M[[CaIGcX
+b8Q)f,HFGj+MGP'Z-8d&S[SrA2I32!5k3L5#cQ1CV4NkAXjer4pehS"JT*BMJmh
eq+jHMVUAFXdD@Pa-LB8NHQRI3K)PI3p-0D6jHqhb!-,0lkJIrAq#kpTYIkZh1S$
iJj!!%H0,"hSUqR8TjiTU6d$LH!3qd"l'QVp5(*Z0MQj%N5IR8$IK#2YVk#b4%AU
KAhRjVG*[D*cA0T*HB1mJp`hf9R+*B@mR9a,f0R*MBGp1mJVX655"`0j)XK,X1mL
pKlf"*+irG2*l,$B1E[#"6T2S$,#X@[56ejba+FlV&"bJcm2dMZ6dm6Xk0U4jAES
MHGhp&Sp0DH#"lZkGmrT#0Q@"!rVX)TRXhr[K0j4X`S%(2RS$[3RXDKCpj(@KE-T
cqZ`NL6E3i"kI4160319LE["D@$B9G'mQ#4Ai1,Ued1qGG(GeFr6blT!!hqqbU3a
-b$&jRrZ0-TY)B)&1lYedll[ACE1T#Rl`e9TlcUBkTp0ZdVF%-H4Z[lGR8a1Bi#X
h0hN["GM8"KlNTJYSfQ*jrHjlI6UE66PpZQMZ#'l`[pH@XGNEQ*!!Qr-kq@mqf+`
,HVK6rLX60R@""hI+c5IHHaBfk`-2I,5(G,jrpK(H5aSfpB%(pqQkANlrj[4mV#G
EHm2$CN01V`9H%R"aqMR+bhpj`iDqe%&p8bAIR!qTj%[4$kpFY(MK'lcmYcPXk&Z
H1lcmlTi0lIT[mPVbJIFUL!elGjRM4BM8c8"+#$@"@kr%qK5GrJGH8d5JeDSp%6Z
S`aY94TZmpLQ+$H(Nh"cl%r`RK-KrL#Vr!3#3!aq$!!!"!*!$!43!N!-8!*!$-Tr
lRLe!rr#`!,K[$#eZd!6rm2rdd"lm`FAKdkSV8FY+$deKBe"bEfTPBh4c,R0TG!)
!N!06594%8dP8)3#3"P0*9%46593K!*!BUc!jI3!!8M8!!!&'"1"2l'mDG@6JrHc
K@5U#NI*HN@GK!Z"2kQ`FG&2UN!"S!!,L@5[48(adA`CdC!EJ6qj[8hJS!!EJEHl
LEe5!)D$!FJC1ANl!*IrX51FI-#D`jL63G!*&0K!+1Li!&Ri!)VX-S"lbUKQJ(Z`
3!+SDI!$!#3ZT8,aIE!!!Q$!'8!6"aG!!N!-3!#X!"3%B!J#3"`-!N!-"!*!$!43
!N!-8!*!$-J$j(l!@#J#3!a`!-J!!8f9dC`#3!`S!!2rr!*!&q@G%'@B:
-91
View File
@@ -1,91 +0,0 @@
/*
MacOS_Test_config.h
Configuration flags for Macintosh development systems.
Test version.
<Revision History>
11/16/95 pcb Updated compilation flags to reflect latest 4.6 Makefile.
by Patrick C. Beard.
*/
/* Boehm, November 17, 1995 12:05 pm PST */
#ifdef __MWERKS__
// for CodeWarrior Pro with Metrowerks Standard Library (MSL).
// #define MSL_USE_PRECOMPILED_HEADERS 0
#include <ansi_prefix.mac.h>
#ifndef __STDC__
#define __STDC__ 0
#endif
#endif
// these are defined again in gc_priv.h.
#undef TRUE
#undef FALSE
#define ALL_INTERIOR_POINTERS // follows interior pointers.
//#define SILENT // want collection messages.
//#define DONT_ADD_BYTE_AT_END // no padding.
//#define SMALL_CONFIG // whether to a smaller heap.
#define NO_SIGNALS // signals aren't real on the Macintosh.
#define USE_TEMPORARY_MEMORY // use Macintosh temporary memory.
// CFLAGS= -O -DNO_SIGNALS -DALL_INTERIOR_POINTERS -DSILENT
//
//LIBGC_CFLAGS= -O -DNO_SIGNALS -DSILENT \
// -DREDIRECT_MALLOC=GC_malloc_uncollectable \
// -DDONT_ADD_BYTE_AT_END -DALL_INTERIOR_POINTERS
// Flags for building libgc.a -- the last two are required.
//
// Setjmp_test may yield overly optimistic results when compiled
// without optimization.
// -DSILENT disables statistics printing, and improves performance.
// -DCHECKSUMS reports on erroneously clear dirty bits, and unexpectedly
// altered stubborn objects, at substantial performance cost.
// Use only for incremental collector debugging.
// -DFIND_LEAK causes the collector to assume that all inaccessible
// objects should have been explicitly deallocated, and reports exceptions.
// Finalization and the test program are not usable in this mode.
// -DSOLARIS_THREADS enables support for Solaris (thr_) threads.
// (Clients should also define SOLARIS_THREADS and then include
// gc.h before performing thr_ or GC_ operations.)
// This is broken on nonSPARC machines.
// -DALL_INTERIOR_POINTERS allows all pointers to the interior
// of objects to be recognized. (See gc_priv.h for consequences.)
// -DSMALL_CONFIG tries to tune the collector for small heap sizes,
// usually causing it to use less space in such situations.
// Incremental collection no longer works in this case.
// -DLARGE_CONFIG tunes the collector for unusually large heaps.
// Necessary for heaps larger than about 500 MB on most machines.
// Recommended for heaps larger than about 64 MB.
// -DDONT_ADD_BYTE_AT_END is meaningful only with
// -DALL_INTERIOR_POINTERS. Normally -DALL_INTERIOR_POINTERS
// causes all objects to be padded so that pointers just past the end of
// an object can be recognized. This can be expensive. (The padding
// is normally more than one byte due to alignment constraints.)
// -DDONT_ADD_BYTE_AT_END disables the padding.
// -DNO_SIGNALS does not disable signals during critical parts of
// the GC process. This is no less correct than many malloc
// implementations, and it sometimes has a significant performance
// impact. However, it is dangerous for many not-quite-ANSI C
// programs that call things like printf in asynchronous signal handlers.
// -DGC_OPERATOR_NEW_ARRAY declares that the C++ compiler supports the
// new syntax "operator new[]" for allocating and deleting arrays.
// See gc_cpp.h for details. No effect on the C part of the collector.
// This is defined implicitly in a few environments.
// -DREDIRECT_MALLOC=X causes malloc, realloc, and free to be defined
// as aliases for X, GC_realloc, and GC_free, respectively.
// Calloc is redefined in terms of the new malloc. X should
// be either GC_malloc or GC_malloc_uncollectable.
// The former is occasionally useful for working around leaks in code
// you don't want to (or can't) look at. It may not work for
// existing code, but it often does. Neither works on all platforms,
// since some ports use malloc or calloc to obtain system memory.
// (Probably works for UNIX, and win32.)
// -DNO_DEBUG removes GC_dump and the debugging routines it calls.
// Reduces code size slightly at the expense of debuggability.
-89
View File
@@ -1,89 +0,0 @@
/*
MacOS_config.h
Configuration flags for Macintosh development systems.
<Revision History>
11/16/95 pcb Updated compilation flags to reflect latest 4.6 Makefile.
by Patrick C. Beard.
*/
/* Boehm, November 17, 1995 12:10 pm PST */
#ifdef __MWERKS__
// for CodeWarrior Pro with Metrowerks Standard Library (MSL).
// #define MSL_USE_PRECOMPILED_HEADERS 0
#include <ansi_prefix.mac.h>
#ifndef __STDC__
#define __STDC__ 0
#endif
#endif /* __MWERKS__ */
// these are defined again in gc_priv.h.
#undef TRUE
#undef FALSE
#define ALL_INTERIOR_POINTERS // follows interior pointers.
#define SILENT // no collection messages.
//#define DONT_ADD_BYTE_AT_END // no padding.
//#define SMALL_CONFIG // whether to use a smaller heap.
#define NO_SIGNALS // signals aren't real on the Macintosh.
#define USE_TEMPORARY_MEMORY // use Macintosh temporary memory.
// CFLAGS= -O -DNO_SIGNALS -DSILENT -DALL_INTERIOR_POINTERS
//
//LIBGC_CFLAGS= -O -DNO_SIGNALS -DSILENT \
// -DREDIRECT_MALLOC=GC_malloc_uncollectable \
// -DDONT_ADD_BYTE_AT_END -DALL_INTERIOR_POINTERS
// Flags for building libgc.a -- the last two are required.
//
// Setjmp_test may yield overly optimistic results when compiled
// without optimization.
// -DSILENT disables statistics printing, and improves performance.
// -DCHECKSUMS reports on erroneously clear dirty bits, and unexpectedly
// altered stubborn objects, at substantial performance cost.
// Use only for incremental collector debugging.
// -DFIND_LEAK causes the collector to assume that all inaccessible
// objects should have been explicitly deallocated, and reports exceptions.
// Finalization and the test program are not usable in this mode.
// -DSOLARIS_THREADS enables support for Solaris (thr_) threads.
// (Clients should also define SOLARIS_THREADS and then include
// gc.h before performing thr_ or GC_ operations.)
// This is broken on nonSPARC machines.
// -DALL_INTERIOR_POINTERS allows all pointers to the interior
// of objects to be recognized. (See gc_priv.h for consequences.)
// -DSMALL_CONFIG tries to tune the collector for small heap sizes,
// usually causing it to use less space in such situations.
// Incremental collection no longer works in this case.
// -DLARGE_CONFIG tunes the collector for unusually large heaps.
// Necessary for heaps larger than about 500 MB on most machines.
// Recommended for heaps larger than about 64 MB.
// -DDONT_ADD_BYTE_AT_END is meaningful only with
// -DALL_INTERIOR_POINTERS. Normally -DALL_INTERIOR_POINTERS
// causes all objects to be padded so that pointers just past the end of
// an object can be recognized. This can be expensive. (The padding
// is normally more than one byte due to alignment constraints.)
// -DDONT_ADD_BYTE_AT_END disables the padding.
// -DNO_SIGNALS does not disable signals during critical parts of
// the GC process. This is no less correct than many malloc
// implementations, and it sometimes has a significant performance
// impact. However, it is dangerous for many not-quite-ANSI C
// programs that call things like printf in asynchronous signal handlers.
// -DGC_OPERATOR_NEW_ARRAY declares that the C++ compiler supports the
// new syntax "operator new[]" for allocating and deleting arrays.
// See gc_cpp.h for details. No effect on the C part of the collector.
// This is defined implicitly in a few environments.
// -DREDIRECT_MALLOC=X causes malloc, realloc, and free to be defined
// as aliases for X, GC_realloc, and GC_free, respectively.
// Calloc is redefined in terms of the new malloc. X should
// be either GC_malloc or GC_malloc_uncollectable.
// The former is occasionally useful for working around leaks in code
// you don't want to (or can't) look at. It may not work for
// existing code, but it often does. Neither works on all platforms,
// since some ports use malloc or calloc to obtain system memory.
// (Probably works for UNIX, and win32.)
// -DNO_DEBUG removes GC_dump and the debugging routines it calls.
// Reduces code size slightly at the expense of debuggability.
-9
View File
@@ -1,9 +0,0 @@
/*
dataend.c
A hack to get the extent of global data for the Macintosh.
by Patrick C. Beard.
*/
long __dataend;
-9
View File
@@ -1,9 +0,0 @@
/*
datastart.c
A hack to get the extent of global data for the Macintosh.
by Patrick C. Beard.
*/
long __datastart;
-107
View File
@@ -1,107 +0,0 @@
#-----------------------------------------------------------------------------#
# Makefile.DLLs, version 0.4.
# Contributed by Fergus Henderson.
# This Makefile contains rules for creating DLLs on Windows using gnu-win32.
#-----------------------------------------------------------------------------#
# This rule creates a `.def' file, which lists the symbols that are exported
# from the DLL. We use `nm' to get a list of all the exported text (`T')
# symbols and data symbols -- including uninitialized data (`B'),
# initialized data (`D'), read-only data (`R'), and common blocks (`C').
%.def: %.a
echo EXPORTS > $@
nm $< | grep '^........ [BCDRT] _' | sed 's/[^_]*_//' >> $@
# We need to use macros to access global data:
# the user of the DLL must refer to `foo' as `(*__imp_foo)'.
# This rule creates a `_globals.h' file, which contains macros
# for doing this.
SYM_PREFIX = $(firstword $(SYM_PREFIX-$*) $*)
DLL_MACRO = $(SYM_PREFIX)_USE_DLL
IMP_MACRO = $(SYM_PREFIX)_IMP
GLOBAL_MACRO = $(SYM_PREFIX)_GLOBAL
%_globals.h: %.a
echo "/* automatically generated by Makefile.DLLs */" > $@
echo "#if defined(__GNUC__) && defined(_WIN32) \\" >> $@
echo " && defined($(DLL_MACRO))" >> $@
echo "# define $(IMP_MACRO)(name) __imp_##name" >> $@
echo "# define $(GLOBAL_MACRO)(name) (*$(IMP_MACRO)(name))" >> $@
echo "#else" >> $@
echo "# define $(GLOBAL_MACRO)(name) name" >> $@
echo "#endif" >> $@
echo "" >> $@
for sym in `nm $< | grep '^........ [BCDR] _' | sed 's/[^_]*_//'`; do \
echo "#define $$sym $(GLOBAL_MACRO)($$sym)" >> $@; \
done
# This rule creates the export object file (`foo.exp') which contains the
# jump table array; this export object file becomes part of the DLL.
# This rule also creates the import library (`foo_dll.a') which contains small
# stubs for all the functions exported by the DLL which jump to them via the
# jump table. Executables that will use the DLL must be linked against this
# stub library.
%.exp %_dll.a : %.def
dlltool $(DLLTOOLFLAGS) $(DLLTOOLFLAGS-$*) \
--def $< \
--dllname $*.dll \
--output-exp $*.exp \
--output-lib $*_dll.a
# The `sed' commands below are to convert DOS-style `C:\foo\bar'
# pathnames into Unix-style `//c/foo/bar' pathnames.
CYGWIN32_LIBS = $(shell echo \
-L`dirname \`gcc -print-file-name=libgcc.a | \
sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \
-L`dirname \`gcc -print-file-name=libcygwin.a | \
sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \
-L`dirname \`gcc -print-file-name=libkernel32.a | \
sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \
-lgcc -lcygwin -lkernel32 -lgcc)
RELOCATABLE=yes
ifeq "$(strip $(RELOCATABLE))" "yes"
# to create relocatable DLLs, we need to do two passes
%.dll: %.exp %.a dll_fixup.o dll_init.o
$(LD) $(LDFLAGS) $(LDFLAGS-$*) --dll -o $*.base \
-e _dll_entry@12 dll_init.o \
dll_fixup.o $*.exp $*.a \
$(LDLIBS) $(LDLIBS-$*) \
$(CYGWIN32_LIBS)
$(LD) $(LDFLAGS) $(LDFLAGS-$*) --dll --base-file $*.base -o $@ \
-e _dll_entry@12 dll_init.o \
dll_fixup.o $*.exp $*.a \
$(LDLIBS) $(LDLIBS-$*) \
$(CYGWIN32_LIBS)
rm -f $*.base
else
%.dll: %.exp %.a dll_fixup.o dll_init.o
$(LD) $(LDFLAGS) $(LDFLAGS-$*) --dll -o $@ \
-e _dll_entry@12 dll_init.o \
dll_fixup.o $*.exp $*.a \
$(LDLIBS) $(LDLIBS-$*) \
$(CYGWIN32_LIBS)
endif
# This black magic piece of assembler needs to be linked in in order to
# properly terminate the list of imported DLLs.
dll_fixup.s:
echo '.section .idata$$3' > dll_fixup.s
echo '.long 0,0,0,0, 0,0,0,0' >> dll_fixup.s
# This bit is necessary to provide an initialization function for the DLL.
dll_init.c:
echo '__attribute__((stdcall))' > dll_init.c
echo 'int dll_entry(int handle, int reason, void *ptr)' >> dll_init.c
echo '{return 1; }' >> dll_init.c
dont_throw_away: dll_fixup.o dll_init.o
-190
View File
@@ -1,190 +0,0 @@
# Copyright (c) 1999-2001 by Red Hat, Inc. All rights reserved.
#
# THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
# OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
#
# Permission is hereby granted to use or copy this program
# for any purpose, provided the above notices are retained on all copies.
# Permission to modify the code and to distribute modified code is granted,
# provided the above notices are retained, and a notice that the code was
# modified is included with the above copyright notice.
#
# Original author: Tom Tromey
# Severely truncated by Hans-J. Boehm
## Process this file with automake to produce Makefile.in.
## FIXME: `make dist' in this directory will not currently work. Many
## files that should be in the distribution are not mentioned in this
## Makefile.am.
ACLOCAL_AMFLAGS = -I m4
SUBDIRS = m4 include doc
AM_CPPFLAGS = -I$(top_builddir)/.. -I$(top_srcdir)/..
#
# libtool is not capable of creating shared/static versions of a convenience
# library, so we have to do it ourselves
#
noinst_LTLIBRARIES = libmonogc.la libmonogc-static.la
EXTRA_DIST =
## more items will be succesively added below
if POWERPC_DARWIN
asm_libgc_sources = powerpc_darwin_mach_dep.s
endif
if TARGET_IA64
asm_libgc_sources = ia64_save_regs_in_stack.s
endif
libmonogc_la_SOURCES = allchblk.c alloc.c blacklst.c checksums.c dbg_mlc.c \
dyn_load.c finalize.c gc_dlopen.c gcj_mlc.c headers.c \
malloc.c mallocx.c mark.c mark_rts.c misc.c new_hblk.c \
obj_map.c os_dep.c pcr_interface.c ptr_chck.c real_malloc.c reclaim.c \
solaris_pthreads.c solaris_threads.c specific.c stubborn.c typd_mlc.c \
backgraph.c win32_threads.c \
pthread_support.c pthread_stop_world.c darwin_stop_world.c \
mach_dep.c $(asm_libgc_sources)
# Include THREADDLLIBS here to ensure that the correct versions of
# linuxthread semaphore functions get linked:
libmonogc_la_LIBADD = $(THREADDLLIBS) $(UNWINDLIBS)
libmonogc_la_DEPENDENCIES =
libmonogc_la_LDFLAGS = -version-info 1:2:0
EXTRA_libmonogc_la_SOURCES = alpha_mach_dep.S \
mips_sgi_mach_dep.s mips_ultrix_mach_dep.s powerpc_darwin_mach_dep.s \
rs6000_mach_dep.s sparc_mach_dep.S sparc_netbsd_mach_dep.s \
sparc_sunos4_mach_dep.s ia64_save_regs_in_stack.s
libmonogc_static_la_SOURCES = $(libmonogc_la_SOURCES)
libmonogc_static_la_LIBADD = $(libmonogc_la_LIBADD)
libmonogc_static_la_LDFLAGS = -static
EXTRA_DIST += alpha_mach_dep.S mips_sgi_mach_dep.s sparc_mach_dep.S
AM_CFLAGS = $(WERROR_CFLAGS) @GC_CFLAGS@
if CPLUSPLUS
extra_checks = test_cpp
else
extra_checks =
endif
#check_PROGRAMS = gctest $(extra_checks)
#test.o: $(srcdir)/tests/test.c
# $(COMPILE) -c $(srcdir)/tests/test.c
# Using $< in the above seems to fail with the HP/UX on Itanium make.
#test_cpp.o: $(srcdir)/tests/test_cpp.cc
# $(CXXCOMPILE) -c $(srcdir)/tests/test_cpp.cc
## FIXME: this is probably the reason why some files from BUILT_SOURCES
## are included in the distribution
# gctest_OBJECTS = test.o
#gctest_SOURCES = tests/test.c
#gctest_LDADD = ./libgc.la $(THREADDLLIBS) $(UNWINDLIBS) $(EXTRA_TEST_LIBS)
#test_cpp_SOURCES = tests/test_cpp.cc
#test_cpp_LDADD = ./libgc.la ./libgccpp.la $(THREADDLLIBS) $(UNWINDLIBS) $(EXTRA_TEST_LIBS)
#TESTS = gctest $(extra_checks)
## FIXME: relies on internal code generated by automake.
all_objs = $(libgc_la_OBJECTS)
$(all_objs) : include/private/gcconfig.h include/private/gc_priv.h \
include/private/gc_hdrs.h include/gc.h include/gc_gcj.h include/gc_mark.h
## FIXME: we shouldn't have to do this, but automake forces us to.
if COMPILER_XLC
## XLC neither requires nor tolerates the unnecessary assembler goop
ASM_CPP_OPTIONS =
else
## We use -Wp,-P to strip #line directives. Irix `as' chokes on
## these.
ASM_CPP_OPTIONS = -Wp,-P -x assembler-with-cpp
endif
.s.lo:
## We use -Wp,-P to strip #line directives. Irix `as' chokes on
## these.
$(LTCOMPILE) $(ASM_CPP_OPTIONS) -c $<
## We have our own definition of LTCOMPILE because we want to use our
## CFLAGS, not those passed in from the top level make.
LTCOMPILE = $(LIBTOOL) --mode=compile --tag=CC $(CC) $(DEFS) \
-I$(top_srcdir)/include $(AM_CPPFLAGS) $(CPPFLAGS) \
$(AM_CFLAGS) $(MY_CFLAGS) $(GC_CFLAGS)
LINK = $(LIBTOOL) --mode=link $(CC) $(AM_CFLAGS) $(MY_CFLAGS) $(LDFLAGS) -o $@
# Work around what appears to be a GNU make bug handling MAKEFLAGS
# values defined in terms of make variables, as is the case for CC and
# friends when we are called from the top level Makefile.
AM_MAKEFLAGS = \
"AR_FLAGS=$(AR_FLAGS)" \
"CC_FOR_BUILD=$(CC_FOR_BUILD)" \
"CFLAGS=$(CFLAGS)" \
"CXXFLAGS=$(CXXFLAGS)" \
"CFLAGS_FOR_BUILD=$(CFLAGS_FOR_BUILD)" \
"CFLAGS_FOR_TARGET=$(CFLAGS_FOR_TARGET)" \
"INSTALL=$(INSTALL)" \
"INSTALL_DATA=$(INSTALL_DATA)" \
"INSTALL_PROGRAM=$(INSTALL_PROGRAM)" \
"INSTALL_SCRIPT=$(INSTALL_SCRIPT)" \
"LDFLAGS=$(LDFLAGS)" \
"LIBCFLAGS=$(LIBCFLAGS)" \
"LIBCFLAGS_FOR_TARGET=$(LIBCFLAGS_FOR_TARGET)" \
"MAKE=$(MAKE)" \
"MAKEINFO=$(MAKEINFO) $(MAKEINFOFLAGS)" \
"PICFLAG=$(PICFLAG)" \
"PICFLAG_FOR_TARGET=$(PICFLAG_FOR_TARGET)" \
"SHELL=$(SHELL)" \
"EXPECT=$(EXPECT)" \
"RUNTEST=$(RUNTEST)" \
"RUNTESTFLAGS=$(RUNTESTFLAGS)" \
"exec_prefix=$(exec_prefix)" \
"infodir=$(infodir)" \
"libdir=$(libdir)" \
"prefix=$(prefix)" \
"tooldir=$(tooldir)" \
"AR=$(AR)" \
"AS=$(AS)" \
"CC=$(CC)" \
"CXX=$(CXX)" \
"LD=$(LD)" \
"LIBCFLAGS=$(LIBCFLAGS)" \
"NM=$(NM)" \
"PICFLAG=$(PICFLAG)" \
"RANLIB=$(RANLIB)" \
"DESTDIR=$(DESTDIR)"
MAKEOVERRIDES=
# files used by makefiles other than Makefile.am
#
EXTRA_DIST += add_gc_prefix.c gcname.c if_mach.c if_not_there.c hpux_test_and_clear.s pc_excludes gc.mak MacOS.c MacProjects.sit.hqx mach_dep.c setjmp_t.c threadlibs.c AmigaOS.c version.h Mac_files/datastart.c Mac_files/dataend.c Mac_files/MacOS_config.h Mac_files/MacOS_Test_config.h
# cord package
#
EXTRA_DIST += cord cord/cordbscs.c cord/cordtest.c cord/de.c cord/de_win.c cord/de_win.ICO cord/cordprnt.c cord/cordxtra.c cord/de_cmds.h cord/de_win.h cord/de_win.RC
# this is an auxiliary shell file used by Makefile and Makefile.direct
#
CONFIG_STATUS_DEPENDENCIES = $(srcdir)/configure.host
# :FIXME: why do we distribute this one???
#
EXTRA_DIST += configure.host
#
# :GOTCHA: GNU make rule for making .s out of .S is flawed,
# it will not remove dest if building fails
.S.s:
if $(CPP) $< >$@ ; then :; else rm -f $@; fi
test:
test-bundle:
-685
View File
@@ -1,685 +0,0 @@
# This is the original manually generated Makefile. It may still be used
# to build the collector.
#
# Primary targets:
# gc.a - builds basic library
# c++ - adds C++ interface to library
# cords - adds cords (heavyweight strings) to library
# test - prints porting information, then builds basic version of gc.a,
# and runs some tests of collector and cords. Does not add cords or
# c++ interface to gc.a
# cord/de - builds dumb editor based on cords.
ABI_FLAG=
# ABI_FLAG should be the cc flag that specifies the ABI. On most
# platforms this will be the empty string. Possible values:
# +DD64 for 64-bit executable on HP/UX.
# -n32, -n64, -o32 for SGI/MIPS ABIs.
AS_ABI_FLAG=$(ABI_FLAG)
# ABI flag for assembler. On HP/UX this is +A64 for 64 bit
# executables.
CC=cc $(ABI_FLAG)
CXX=g++ $(ABI_FLAG)
AS=as $(AS_ABI_FLAG)
# The above doesn't work with gas, which doesn't run cpp.
# Define AS as `gcc -c -x assembler-with-cpp' instead.
# Redefining srcdir allows object code for the nonPCR version of the collector
# to be generated in different directories.
srcdir= .
VPATH= $(srcdir)
CFLAGS= -O -I$(srcdir)/include -DATOMIC_UNCOLLECTABLE -DNO_SIGNALS -DNO_EXECUTE_PERMISSION -DSILENT -DALL_INTERIOR_POINTERS
# To build the parallel collector on Linux, add to the above:
# -DGC_LINUX_THREADS -DPARALLEL_MARK -DTHREAD_LOCAL_ALLOC
# To build the parallel collector in a static library on HP/UX,
# add to the above:
# -DGC_HPUX_THREADS -DPARALLEL_MARK -DTHREAD_LOCAL_ALLOC -D_POSIX_C_SOURCE=199506L -mt
# To build the thread-safe collector on Tru64, add to the above:
# -pthread -DGC_OSF1_THREADS
# HOSTCC and HOSTCFLAGS are used to build executables that will be run as
# part of the build process, i.e. on the build machine. These will usually
# be the same as CC and CFLAGS, except in a cross-compilation environment.
# Note that HOSTCFLAGS should include any -D flags that affect thread support.
HOSTCC=$(CC)
HOSTCFLAGS=$(CFLAGS)
# For dynamic library builds, it may be necessary to add flags to generate
# PIC code, e.g. -fPIC on Linux.
# Setjmp_test may yield overly optimistic results when compiled
# without optimization.
# These define arguments influence the collector configuration:
# -DSILENT disables statistics printing, and improves performance.
# -DFIND_LEAK causes GC_find_leak to be initially set.
# This causes the collector to assume that all inaccessible
# objects should have been explicitly deallocated, and reports exceptions.
# Finalization and the test program are not usable in this mode.
# -DGC_SOLARIS_THREADS enables support for Solaris (thr_) threads.
# (Clients should also define GC_SOLARIS_THREADS and then include
# gc.h before performing thr_ or dl* or GC_ operations.)
# Must also define -D_REENTRANT.
# -DGC_SOLARIS_PTHREADS enables support for Solaris pthreads.
# (Internally this define GC_SOLARIS_THREADS as well.)
# -DGC_IRIX_THREADS enables support for Irix pthreads. See README.irix.
# -DGC_HPUX_THREADS enables support for HP/UX 11 pthreads.
# Also requires -D_REENTRANT or -D_POSIX_C_SOURCE=199506L. See README.hp.
# -DGC_LINUX_THREADS enables support for Xavier Leroy's Linux threads.
# see README.linux. -D_REENTRANT may also be required.
# -DGC_OSF1_THREADS enables support for Tru64 pthreads.
# -DGC_FREEBSD_THREADS enables support for FreeBSD pthreads.
# Appeared to run into some underlying thread problems.
# -DGC_DARWIN_THREADS enables support for Mac OS X pthreads.
# -DGC_AIX_THREADS enables support for IBM AIX threads.
# -DGC_DGUX386_THREADS enables support for DB/UX on I386 threads.
# See README.DGUX386.
# -DGC_WIN32_THREADS enables support for win32 threads. That makes sense
# for this Makefile only under Cygwin.
# -DGC_THREADS should set the appropriate one of the above macros.
# It assumes pthreads for Solaris.
# -DALL_INTERIOR_POINTERS allows all pointers to the interior
# of objects to be recognized. (See gc_priv.h for consequences.)
# Alternatively, GC_all_interior_pointers can be set at process
# initialization time.
# -DSMALL_CONFIG tries to tune the collector for small heap sizes,
# usually causing it to use less space in such situations.
# Incremental collection no longer works in this case.
# -DLARGE_CONFIG tunes the collector for unusually large heaps.
# Necessary for heaps larger than about 500 MB on most machines.
# Recommended for heaps larger than about 64 MB.
# -DDONT_ADD_BYTE_AT_END is meaningful only with -DALL_INTERIOR_POINTERS or
# GC_all_interior_pointers = 1. Normally -DALL_INTERIOR_POINTERS
# causes all objects to be padded so that pointers just past the end of
# an object can be recognized. This can be expensive. (The padding
# is normally more than one byte due to alignment constraints.)
# -DDONT_ADD_BYTE_AT_END disables the padding.
# -DNO_SIGNALS does not disable signals during critical parts of
# the GC process. This is no less correct than many malloc
# implementations, and it sometimes has a significant performance
# impact. However, it is dangerous for many not-quite-ANSI C
# programs that call things like printf in asynchronous signal handlers.
# This is on by default. Turning it off has not been extensively tested with
# compilers that reorder stores. It should have been.
# -DNO_EXECUTE_PERMISSION may cause some or all of the heap to not
# have execute permission, i.e. it may be impossible to execute
# code from the heap. Currently this only affects the incremental
# collector on UNIX machines. It may greatly improve its performance,
# since this may avoid some expensive cache synchronization.
# -DGC_NO_OPERATOR_NEW_ARRAY declares that the C++ compiler does not support
# the new syntax "operator new[]" for allocating and deleting arrays.
# See gc_cpp.h for details. No effect on the C part of the collector.
# This is defined implicitly in a few environments. Must also be defined
# by clients that use gc_cpp.h.
# -DREDIRECT_MALLOC=X causes malloc to be defined as alias for X.
# Unless the following macros are defined, realloc is also redirected
# to GC_realloc, and free is redirected to GC_free.
# Calloc and strdup are redefined in terms of the new malloc. X should
# be either GC_malloc or GC_malloc_uncollectable, or
# GC_debug_malloc_replacement. (The latter invokes GC_debug_malloc
# with dummy source location information, but still results in
# properly remembered call stacks on Linux/X86 and Solaris/SPARC.
# It requires that the following two macros also be used.)
# The former is occasionally useful for working around leaks in code
# you don't want to (or can't) look at. It may not work for
# existing code, but it often does. Neither works on all platforms,
# since some ports use malloc or calloc to obtain system memory.
# (Probably works for UNIX, and win32.) If you build with DBG_HDRS_ALL,
# you should only use GC_debug_malloc_replacement as a malloc
# replacement.
# -DREDIRECT_REALLOC=X causes GC_realloc to be redirected to X.
# The canonical use is -DREDIRECT_REALLOC=GC_debug_realloc_replacement,
# together with -DREDIRECT_MALLOC=GC_debug_malloc_replacement to
# generate leak reports with call stacks for both malloc and realloc.
# This also requires the following:
# -DREDIRECT_FREE=X causes free to be redirected to X. The
# canonical use is -DREDIRECT_FREE=GC_debug_free.
# -DIGNORE_FREE turns calls to free into a noop. Only useful with
# -DREDIRECT_MALLOC.
# -DNO_DEBUGGING removes GC_dump and the debugging routines it calls.
# Reduces code size slightly at the expense of debuggability.
# -DJAVA_FINALIZATION makes it somewhat safer to finalize objects out of
# order by specifying a nonstandard finalization mark procedure (see
# finalize.c). Objects reachable from finalizable objects will be marked
# in a sepearte postpass, and hence their memory won't be reclaimed.
# Not recommended unless you are implementing a language that specifies
# these semantics. Since 5.0, determines only only the initial value
# of GC_java_finalization variable.
# -DFINALIZE_ON_DEMAND causes finalizers to be run only in response
# to explicit GC_invoke_finalizers() calls.
# In 5.0 this became runtime adjustable, and this only determines the
# initial value of GC_finalize_on_demand.
# -DATOMIC_UNCOLLECTABLE includes code for GC_malloc_atomic_uncollectable.
# This is useful if either the vendor malloc implementation is poor,
# or if REDIRECT_MALLOC is used.
# -DHBLKSIZE=ddd, where ddd is a power of 2 between 512 and 16384, explicitly
# sets the heap block size. Each heap block is devoted to a single size and
# kind of object. For the incremental collector it makes sense to match
# the most likely page size. Otherwise large values result in more
# fragmentation, but generally better performance for large heaps.
# -DUSE_MMAP use MMAP instead of sbrk to get new memory.
# Works for Solaris and Irix.
# -DUSE_MUNMAP causes memory to be returned to the OS under the right
# circumstances. This currently disables VM-based incremental collection.
# This is currently experimental, and works only under some Unix,
# Linux and Windows versions.
# -DMMAP_STACKS (for Solaris threads) Use mmap from /dev/zero rather than
# GC_scratch_alloc() to get stack memory.
# -DPRINT_BLACK_LIST Whenever a black list entry is added, i.e. whenever
# the garbage collector detects a value that looks almost, but not quite,
# like a pointer, print both the address containing the value, and the
# value of the near-bogus-pointer. Can be used to identifiy regions of
# memory that are likely to contribute misidentified pointers.
# -DKEEP_BACK_PTRS Add code to save back pointers in debugging headers
# for objects allocated with the debugging allocator. If all objects
# through GC_MALLOC with GC_DEBUG defined, this allows the client
# to determine how particular or randomly chosen objects are reachable
# for debugging/profiling purposes. The gc_backptr.h interface is
# implemented only if this is defined.
# -DGC_ASSERTIONS Enable some internal GC assertion checking. Currently
# this facility is only used in a few places. It is intended primarily
# for debugging of the garbage collector itself, but could also
# -DDBG_HDRS_ALL Make sure that all objects have debug headers. Increases
# the reliability (from 99.9999% to 100% mod. bugs) of some of the debugging
# code (especially KEEP_BACK_PTRS). Makes -DSHORT_DBG_HDRS possible.
# Assumes that all client allocation is done through debugging
# allocators.
# -DSHORT_DBG_HDRS Assume that all objects have debug headers. Shorten
# the headers to minimize object size, at the expense of checking for
# writes past the end of an object. This is intended for environments
# in which most client code is written in a "safe" language, such as
# Scheme or Java. Assumes that all client allocation is done using
# the GC_debug_ functions, or through the macros that expand to these,
# or by redirecting malloc to GC_debug_malloc_replacement.
# (Also eliminates the field for the requested object size.)
# occasionally be useful for debugging of client code. Slows down the
# collector somewhat, but not drastically.
# -DSAVE_CALL_COUNT=<n> Set the number of call frames saved with objects
# allocated through the debugging interface. Affects the amount of
# information generated in leak reports. Only matters on platforms
# on which we can quickly generate call stacks, currently Linux/(X86 & SPARC)
# and Solaris/SPARC and platforms that provide execinfo.h.
# Default is zero. On X86, client
# code should NOT be compiled with -fomit-frame-pointer.
# -DSAVE_CALL_NARGS=<n> Set the number of functions arguments to be
# saved with each call frame. Default is zero. Ignored if we
# don't know how to retrieve arguments on the platform.
# -DCHECKSUMS reports on erroneously clear dirty bits, and unexpectedly
# altered stubborn objects, at substantial performance cost.
# Use only for debugging of the incremental collector.
# -DGC_GCJ_SUPPORT includes support for gcj (and possibly other systems
# that include a pointer to a type descriptor in each allocated object).
# Building this way requires an ANSI C compiler.
# -DUSE_I686_PREFETCH causes the collector to issue Pentium III style
# prefetch instructions. No effect except on X86 Linux platforms.
# Assumes a very recent gcc-compatible compiler and assembler.
# (Gas prefetcht0 support was added around May 1999.)
# Empirically the code appears to still run correctly on Pentium II
# processors, though with no performance benefit. May not run on other
# X86 processors? In some cases this improves performance by
# 15% or so.
# -DUSE_3DNOW_PREFETCH causes the collector to issue AMD 3DNow style
# prefetch instructions. Same restrictions as USE_I686_PREFETCH.
# Minimally tested. Didn't appear to be an obvious win on a K6-2/500.
# -DUSE_PPC_PREFETCH causes the collector to issue PowerPC style
# prefetch instructions. No effect except on PowerPC OS X platforms.
# Performance impact untested.
# -DGC_USE_LD_WRAP in combination with the old flags listed in README.linux
# causes the collector some system and pthread calls in a more transparent
# fashion than the usual macro-based approach. Requires GNU ld, and
# currently probably works only with Linux.
# -DTHREAD_LOCAL_ALLOC defines GC_local_malloc(), GC_local_malloc_atomic()
# and GC_local_gcj_malloc(). Needed for gc_gcj.h interface. These allocate
# in a way that usually does not involve acquisition of a global lock.
# Currently works only on platforms such as Linux which use pthread_support.c.
# Recommended for multiprocessors.
# -DUSE_COMPILER_TLS causes thread local allocation to use compiler-supported
# "__thread" thread-local variables. This is the default in HP/UX. It
# may help performance on recent Linux installations. (It failed for
# me on RedHat 8, but appears to work on RedHat 9.)
# -DPARALLEL_MARK allows the marker to run in multiple threads. Recommended
# for multiprocessors. Currently requires Linux on X86 or IA64, though
# support for other Posix platforms should be fairly easy to add,
# if the thread implementation is otherwise supported.
# -DNO_GETENV prevents the collector from looking at environment variables.
# These may otherwise alter its configuration, or turn off GC altogether.
# I don't know of a reason to disable this, except possibly if the
# resulting process runs as a privileged user?
# -DUSE_GLOBAL_ALLOC. Win32 only. Use GlobalAlloc instead of
# VirtualAlloc to allocate the heap. May be needed to work around
# a Windows NT/2000 issue. Incompatible with USE_MUNMAP.
# See README.win32 for details.
# -DMAKE_BACK_GRAPH. Enable GC_PRINT_BACK_HEIGHT environment variable.
# See README.environment for details. Experimental. Limited platform
# support. Implies DBG_HDRS_ALL. All allocation should be done using
# the debug interface.
# -DSTUBBORN_ALLOC allows allocation of "hard to change" objects, and thus
# makes incremental collection easier. Was enabled by default until 6.0.
# Rarely used, to my knowledge.
# -DHANDLE_FORK attempts to make GC_malloc() work in a child process fork()ed
# from a multithreaded parent. Currently only supported by pthread_support.c.
# (Similar code should work on Solaris or Irix, but it hasn't been tried.)
# -DTEST_WITH_SYSTEM_MALLOC causes gctest to allocate (and leak) large chunks
# of memory with the standard system malloc. This will cause the root
# set and collected heap to grow significantly if malloced memory is
# somehow getting traced by the collector. This has no impact on the
# generated library; it only affects the test.
# -DPOINTER_MASK=0x... causes candidate pointers to be ANDed with the
# given mask before being considered. If either this or the following
# macro is defined, it will be assumed that all pointers stored in
# the heap need to be processed this way. Stack and register pointers
# will be considered both with and without processing.
# These macros are normally needed only to support systems that use
# high-order pointer tags. EXPERIMENTAL.
# -DPOINTER_SHIFT=n causes the collector to left shift candidate pointers
# by the indicated amount before trying to interpret them. Applied
# after POINTER_MASK. EXPERIMENTAL. See also the preceding macro.
# -DDARWIN_DONT_PARSE_STACK Causes the Darwin port to discover thread
# stack bounds in the same way as other pthread ports, without trying to
# walk the frames onthe stack. This is recommended only as a fallback
# for applications that don't support proper stack unwinding.
#
CXXFLAGS= $(CFLAGS)
AR= ar
RANLIB= ranlib
OBJS= alloc.o reclaim.o allchblk.o misc.o mach_dep.o os_dep.o mark_rts.o headers.o mark.o obj_map.o blacklst.o finalize.o new_hblk.o dbg_mlc.o malloc.o stubborn.o checksums.o solaris_threads.o pthread_support.o pthread_stop_world.o darwin_stop_world.o typd_mlc.o ptr_chck.o mallocx.o solaris_pthreads.o gcj_mlc.o specific.o gc_dlopen.o backgraph.o win32_threads.o
CSRCS= reclaim.c allchblk.c misc.c alloc.c mach_dep.c os_dep.c mark_rts.c headers.c mark.c obj_map.c pcr_interface.c blacklst.c finalize.c new_hblk.c real_malloc.c dyn_load.c dbg_mlc.c malloc.c stubborn.c checksums.c solaris_threads.c pthread_support.c pthread_stop_world.c darwin_stop_world.c typd_mlc.c ptr_chck.c mallocx.c solaris_pthreads.c gcj_mlc.c specific.c gc_dlopen.c backgraph.c win32_threads.c
CORD_SRCS= cord/cordbscs.c cord/cordxtra.c cord/cordprnt.c cord/de.c cord/cordtest.c include/cord.h include/ec.h include/private/cord_pos.h cord/de_win.c cord/de_win.h cord/de_cmds.h cord/de_win.ICO cord/de_win.RC
CORD_OBJS= cord/cordbscs.o cord/cordxtra.o cord/cordprnt.o
SRCS= $(CSRCS) mips_sgi_mach_dep.s rs6000_mach_dep.s alpha_mach_dep.S \
sparc_mach_dep.S include/gc.h include/gc_typed.h \
include/private/gc_hdrs.h include/private/gc_priv.h \
include/private/gcconfig.h include/private/gc_pmark.h \
include/gc_inl.h include/gc_inline.h include/gc_mark.h \
threadlibs.c if_mach.c if_not_there.c gc_cpp.cc include/gc_cpp.h \
gcname.c include/weakpointer.h include/private/gc_locks.h \
gcc_support.c mips_ultrix_mach_dep.s include/gc_alloc.h \
include/new_gc_alloc.h include/gc_allocator.h \
include/javaxfc.h sparc_sunos4_mach_dep.s sparc_netbsd_mach_dep.s \
include/private/solaris_threads.h include/gc_backptr.h \
hpux_test_and_clear.s include/gc_gcj.h \
include/gc_local_alloc.h include/private/dbg_mlc.h \
include/private/specific.h powerpc_darwin_mach_dep.s \
include/leak_detector.h include/gc_amiga_redirects.h \
include/gc_pthread_redirects.h ia64_save_regs_in_stack.s \
include/gc_config_macros.h include/private/pthread_support.h \
include/private/pthread_stop_world.h include/private/darwin_semaphore.h \
include/private/darwin_stop_world.h $(CORD_SRCS)
DOC_FILES= README.QUICK doc/README.Mac doc/README.MacOSX doc/README.OS2 \
doc/README.amiga doc/README.cords doc/debugging.html \
doc/README.dj doc/README.hp doc/README.linux doc/README.rs6000 \
doc/README.sgi doc/README.solaris2 doc/README.uts \
doc/README.win32 doc/barrett_diagram doc/README \
doc/README.contributors doc/README.changes doc/gc.man \
doc/README.environment doc/tree.html doc/gcdescr.html \
doc/README.autoconf doc/README.macros doc/README.ews4800 \
doc/README.DGUX386 doc/README.arm.cross doc/leak.html \
doc/scale.html doc/gcinterface.html doc/README.darwin \
doc/simple_example.html
TESTS= tests/test.c tests/test_cpp.cc tests/trace_test.c \
tests/leak_test.c tests/thread_leak_test.c tests/middle.c
GNU_BUILD_FILES= configure.ac Makefile.am configure acinclude.m4 \
libtool.m4 install-sh configure.host Makefile.in \
aclocal.m4 config.sub config.guess \
include/Makefile.am include/Makefile.in \
doc/Makefile.am doc/Makefile.in \
ltmain.sh mkinstalldirs depcomp missing
OTHER_MAKEFILES= OS2_MAKEFILE NT_MAKEFILE NT_THREADS_MAKEFILE gc.mak \
BCC_MAKEFILE EMX_MAKEFILE WCC_MAKEFILE Makefile.dj \
PCR-Makefile SMakefile.amiga Makefile.DLLs \
digimars.mak Makefile.direct NT_STATIC_THREADS_MAKEFILE
# Makefile and Makefile.direct are copies of each other.
OTHER_FILES= Makefile setjmp_t.c callprocs pc_excludes \
MacProjects.sit.hqx MacOS.c \
Mac_files/datastart.c Mac_files/dataend.c \
Mac_files/MacOS_config.h Mac_files/MacOS_Test_config.h \
add_gc_prefix.c gc_cpp.cpp \
version.h AmigaOS.c \
$(TESTS) $(GNU_BUILD_FILES) $(OTHER_MAKEFILES)
CORD_INCLUDE_FILES= $(srcdir)/include/gc.h $(srcdir)/include/cord.h \
$(srcdir)/include/ec.h $(srcdir)/include/private/cord_pos.h
UTILS= if_mach if_not_there threadlibs
# Libraries needed for curses applications. Only needed for de.
CURSES= -lcurses -ltermlib
# The following is irrelevant on most systems. But a few
# versions of make otherwise fork the shell specified in
# the SHELL environment variable.
SHELL= /bin/sh
SPECIALCFLAGS = -I$(srcdir)/include
# Alternative flags to the C compiler for mach_dep.c.
# Mach_dep.c often doesn't like optimization, and it's
# not time-critical anyway.
# Set SPECIALCFLAGS to -q nodirect_code on Encore.
all: gc.a gctest
LEAKFLAGS=$(CFLAGS) -DFIND_LEAK
BSD-pkg-all: bsd-libgc.a bsd-libleak.a
bsd-libgc.a:
$(MAKE) CFLAGS="$(CFLAGS)" clean c++-t
mv gc.a bsd-libgc.a
bsd-libleak.a:
$(MAKE) -f Makefile.direct CFLAGS="$(LEAKFLAGS)" clean c++-nt
mv gc.a bsd-libleak.a
BSD-pkg-install: BSD-pkg-all
${CP} bsd-libgc.a libgc.a
${INSTALL_DATA} libgc.a ${PREFIX}/lib
${INSTALL_DATA} gc.h gc_cpp.h ${PREFIX}/include
${INSTALL_MAN} doc/gc.man ${PREFIX}/man/man3/gc.3
pcr: PCR-Makefile include/private/gc_private.h include/private/gc_hdrs.h \
include/private/gc_locks.h include/gc.h include/private/gcconfig.h \
mach_dep.o $(SRCS)
$(MAKE) -f PCR-Makefile depend
$(MAKE) -f PCR-Makefile
$(OBJS) tests/test.o dyn_load.o dyn_load_sunos53.o: \
$(srcdir)/include/private/gc_priv.h \
$(srcdir)/include/private/gc_hdrs.h $(srcdir)/include/private/gc_locks.h \
$(srcdir)/include/gc.h $(srcdir)/include/gc_pthread_redirects.h \
$(srcdir)/include/private/gcconfig.h $(srcdir)/include/gc_typed.h \
$(srcdir)/include/gc_config_macros.h Makefile
# The dependency on Makefile is needed. Changing
# options such as -DSILENT affects the size of GC_arrays,
# invalidating all .o files that rely on gc_priv.h
mark.o typd_mlc.o finalize.o ptr_chck.o: $(srcdir)/include/gc_mark.h $(srcdir)/include/private/gc_pmark.h
specific.o pthread_support.o: $(srcdir)/include/private/specific.h
solaris_threads.o solaris_pthreads.o: $(srcdir)/include/private/solaris_threads.h
dbg_mlc.o gcj_mlc.o: $(srcdir)/include/private/dbg_mlc.h
tests/test.o: tests $(srcdir)/tests/test.c
$(CC) $(CFLAGS) -c $(srcdir)/tests/test.c
mv test.o tests/test.o
tests:
mkdir tests
base_lib gc.a: $(OBJS) dyn_load.o $(UTILS)
echo > base_lib
rm -f dont_ar_1
./if_mach SPARC SUNOS5 touch dont_ar_1
./if_mach SPARC SUNOS5 $(AR) rus gc.a $(OBJS) dyn_load.o
./if_mach M68K AMIGA touch dont_ar_1
./if_mach M68K AMIGA $(AR) -vrus gc.a $(OBJS) dyn_load.o
./if_not_there dont_ar_1 $(AR) ru gc.a $(OBJS) dyn_load.o
./if_not_there dont_ar_1 $(RANLIB) gc.a || cat /dev/null
# ignore ranlib failure; that usually means it doesn't exist, and isn't needed
cords: $(CORD_OBJS) cord/cordtest $(UTILS)
rm -f dont_ar_3
./if_mach SPARC SUNOS5 touch dont_ar_3
./if_mach SPARC SUNOS5 $(AR) rus gc.a $(CORD_OBJS)
./if_mach M68K AMIGA touch dont_ar_3
./if_mach M68K AMIGA $(AR) -vrus gc.a $(CORD_OBJS)
./if_not_there dont_ar_3 $(AR) ru gc.a $(CORD_OBJS)
./if_not_there dont_ar_3 $(RANLIB) gc.a || cat /dev/null
gc_cpp.o: $(srcdir)/gc_cpp.cc $(srcdir)/include/gc_cpp.h $(srcdir)/include/gc.h Makefile
$(CXX) -c $(CXXFLAGS) $(srcdir)/gc_cpp.cc
test_cpp: $(srcdir)/tests/test_cpp.cc $(srcdir)/include/gc_cpp.h gc_cpp.o $(srcdir)/include/gc.h \
base_lib $(UTILS)
rm -f test_cpp
./if_mach HP_PA HPUX $(CXX) $(CXXFLAGS) -o test_cpp $(srcdir)/tests/test_cpp.cc gc_cpp.o gc.a -ldld `./threadlibs`
./if_not_there test_cpp $(CXX) $(CXXFLAGS) -o test_cpp $(srcdir)/tests/test_cpp.cc gc_cpp.o gc.a `./threadlibs`
c++-t: c++
./test_cpp 1
c++-nt: c++
@echo "Use ./test_cpp 1 to test the leak library"
c++: gc_cpp.o $(srcdir)/include/gc_cpp.h test_cpp
rm -f dont_ar_4
./if_mach SPARC SUNOS5 touch dont_ar_4
./if_mach SPARC SUNOS5 $(AR) rus gc.a gc_cpp.o
./if_mach M68K AMIGA touch dont_ar_4
./if_mach M68K AMIGA $(AR) -vrus gc.a gc_cpp.o
./if_not_there dont_ar_4 $(AR) ru gc.a gc_cpp.o
./if_not_there dont_ar_4 $(RANLIB) gc.a || cat /dev/null
./test_cpp 1
echo > c++
dyn_load_sunos53.o: dyn_load.c
$(CC) $(CFLAGS) -DSUNOS53_SHARED_LIB -c $(srcdir)/dyn_load.c -o $@
# SunOS5 shared library version of the collector
sunos5gc.so: $(OBJS) dyn_load_sunos53.o
$(CC) -G -o sunos5gc.so $(OBJS) dyn_load_sunos53.o -ldl
ln sunos5gc.so libgc.so
# Alpha/OSF shared library version of the collector
libalphagc.so: $(OBJS)
ld -shared -o libalphagc.so $(OBJS) dyn_load.o -lc
ln libalphagc.so libgc.so
# IRIX shared library version of the collector
libirixgc.so: $(OBJS) dyn_load.o
ld -shared $(ABI_FLAG) -o libirixgc.so $(OBJS) dyn_load.o -lc
ln libirixgc.so libgc.so
# Linux shared library version of the collector
liblinuxgc.so: $(OBJS) dyn_load.o
gcc -shared -o liblinuxgc.so $(OBJS) dyn_load.o
ln liblinuxgc.so libgc.so
# Alternative Linux rule. This is preferable, but is likely to break the
# Makefile for some non-linux platforms.
# LIBOBJS= $(patsubst %.o, %.lo, $(OBJS))
#
#.SUFFIXES: .lo $(SUFFIXES)
#
#.c.lo:
# $(CC) $(CFLAGS) $(CPPFLAGS) -fPIC -c $< -o $@
#
# liblinuxgc.so: $(LIBOBJS) dyn_load.lo
# gcc -shared -Wl,-soname=libgc.so.0 -o libgc.so.0 $(LIBOBJS) dyn_load.lo
# touch liblinuxgc.so
mach_dep.o: $(srcdir)/mach_dep.c $(srcdir)/mips_sgi_mach_dep.s \
$(srcdir)/mips_ultrix_mach_dep.s \
$(srcdir)/rs6000_mach_dep.s $(srcdir)/powerpc_darwin_mach_dep.s \
$(srcdir)/sparc_mach_dep.S $(srcdir)/sparc_sunos4_mach_dep.s \
$(srcdir)/ia64_save_regs_in_stack.s \
$(srcdir)/sparc_netbsd_mach_dep.s $(UTILS)
rm -f mach_dep.o
./if_mach MIPS IRIX5 $(CC) -c -o mach_dep.o $(srcdir)/mips_sgi_mach_dep.s
./if_mach MIPS RISCOS $(AS) -o mach_dep.o $(srcdir)/mips_ultrix_mach_dep.s
./if_mach MIPS ULTRIX $(AS) -o mach_dep.o $(srcdir)/mips_ultrix_mach_dep.s
./if_mach POWERPC DARWIN $(AS) -o mach_dep.o $(srcdir)/powerpc_darwin_mach_dep.s
./if_mach ALPHA LINUX $(CC) -c -o mach_dep.o $(srcdir)/alpha_mach_dep.S
./if_mach SPARC SUNOS5 $(CC) -c -o mach_dep.o $(srcdir)/sparc_mach_dep.S
./if_mach SPARC SUNOS4 $(AS) -o mach_dep.o $(srcdir)/sparc_sunos4_mach_dep.s
./if_mach SPARC OPENBSD $(AS) -o mach_dep.o $(srcdir)/sparc_sunos4_mach_dep.s
./if_mach SPARC NETBSD $(AS) -o mach_dep.o $(srcdir)/sparc_netbsd_mach_dep.s
./if_mach IA64 "" as $(AS_ABI_FLAG) -o ia64_save_regs_in_stack.o $(srcdir)/ia64_save_regs_in_stack.s
./if_mach IA64 "" $(CC) -c -o mach_dep1.o $(SPECIALCFLAGS) $(srcdir)/mach_dep.c
./if_mach IA64 "" ld -r -o mach_dep.o mach_dep1.o ia64_save_regs_in_stack.o
./if_not_there mach_dep.o $(CC) -c $(SPECIALCFLAGS) $(srcdir)/mach_dep.c
mark_rts.o: $(srcdir)/mark_rts.c $(UTILS)
rm -f mark_rts.o
-./if_mach ALPHA OSF1 $(CC) -c $(CFLAGS) -Wo,-notail $(srcdir)/mark_rts.c
./if_not_there mark_rts.o $(CC) -c $(CFLAGS) $(srcdir)/mark_rts.c
# Work-around for DEC optimizer tail recursion elimination bug.
# The ALPHA-specific line should be removed if gcc is used.
alloc.o: version.h
cord:
mkdir cord
cord/cordbscs.o: cord $(srcdir)/cord/cordbscs.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordbscs.c
mv cordbscs.o cord/cordbscs.o
# not all compilers understand -o filename
cord/cordxtra.o: cord $(srcdir)/cord/cordxtra.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordxtra.c
mv cordxtra.o cord/cordxtra.o
cord/cordprnt.o: cord $(srcdir)/cord/cordprnt.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordprnt.c
mv cordprnt.o cord/cordprnt.o
cord/cordtest: $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a $(UTILS)
rm -f cord/cordtest
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o cord/cordtest $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a -lucb
./if_mach HP_PA HPUX $(CC) $(CFLAGS) -o cord/cordtest $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a -ldld `./threadlibs`
./if_mach M68K AMIGA $(CC) $(CFLAGS) -UGC_AMIGA_MAKINGLIB -o cord/cordtest $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a `./threadlibs`
./if_not_there cord/cordtest $(CC) $(CFLAGS) -o cord/cordtest $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a `./threadlibs`
cord/de: $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(UTILS)
rm -f cord/de
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES) -lucb `./threadlibs`
./if_mach HP_PA HPUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES) -ldld `./threadlibs`
./if_mach RS6000 "" $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses
./if_mach POWERPC DARWIN $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a
./if_mach I386 LINUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses `./threadlibs`
./if_mach ALPHA LINUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses `./threadlibs`
./if_mach IA64 LINUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses `./threadlibs`
./if_mach M68K AMIGA $(CC) $(CFLAGS) -UGC_AMIGA_MAKINGLIB -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses
./if_not_there cord/de $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES) `./threadlibs`
if_mach: $(srcdir)/if_mach.c $(srcdir)/include/private/gcconfig.h
$(HOSTCC) $(HOSTCFLAGS) -o if_mach $(srcdir)/if_mach.c
threadlibs: $(srcdir)/threadlibs.c $(srcdir)/include/private/gcconfig.h Makefile
$(HOSTCC) $(HOSTCFLAGS) -o threadlibs $(srcdir)/threadlibs.c
if_not_there: $(srcdir)/if_not_there.c
$(HOSTCC) $(HOSTCFLAGS) -o if_not_there $(srcdir)/if_not_there.c
clean:
rm -f gc.a *.o *.exe tests/*.o gctest gctest_dyn_link test_cpp \
setjmp_test mon.out gmon.out a.out core if_not_there if_mach \
threadlibs $(CORD_OBJS) cord/cordtest cord/de
-rm -f *~
gctest: tests/test.o gc.a $(UTILS)
rm -f gctest
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o gctest tests/test.o gc.a -lucb
./if_mach HP_PA HPUX $(CC) $(CFLAGS) -o gctest tests/test.o gc.a -ldld `./threadlibs`
./if_mach M68K AMIGA $(CC) $(CFLAGS) -UGC_AMIGA_MAKINGLIB -o gctest tests/test.o gc.a `./threadlibs`
./if_not_there gctest $(CC) $(CFLAGS) -o gctest tests/test.o gc.a `./threadlibs`
# If an optimized setjmp_test generates a segmentation fault,
# odds are your compiler is broken. Gctest may still work.
# Try compiling setjmp_t.c unoptimized.
setjmp_test: $(srcdir)/setjmp_t.c $(srcdir)/include/gc.h $(UTILS)
$(CC) $(CFLAGS) -o setjmp_test $(srcdir)/setjmp_t.c
test: KandRtest cord/cordtest
cord/cordtest
# Those tests that work even with a K&R C compiler:
KandRtest: setjmp_test gctest
./setjmp_test
./gctest
add_gc_prefix: $(srcdir)/add_gc_prefix.c $(srcdir)/version.h
$(CC) -o add_gc_prefix $(srcdir)/add_gc_prefix.c
gcname: $(srcdir)/gcname.c $(srcdir)/version.h
$(CC) -o gcname $(srcdir)/gcname.c
gc.tar: $(SRCS) $(DOC_FILES) $(OTHER_FILES) add_gc_prefix gcname
cp Makefile Makefile.old
cp Makefile.direct Makefile
rm -f `./gcname`
ln -s . `./gcname`
./add_gc_prefix $(SRCS) $(DOC_FILES) $(OTHER_FILES) > /tmp/gc.tar-files
tar cvfh gc.tar `cat /tmp/gc.tar-files`
cp gc.tar `./gcname`.tar
gzip `./gcname`.tar
rm `./gcname`
pc_gc.tar: $(SRCS) $(OTHER_FILES)
tar cvfX pc_gc.tar pc_excludes $(SRCS) $(OTHER_FILES)
floppy: pc_gc.tar
-mmd a:/cord
-mmd a:/cord/private
-mmd a:/include
-mmd a:/include/private
mkdir /tmp/pc_gc
cat pc_gc.tar | (cd /tmp/pc_gc; tar xvf -)
-mcopy -tmn /tmp/pc_gc/* a:
-mcopy -tmn /tmp/pc_gc/cord/* a:/cord
-mcopy -mn /tmp/pc_gc/cord/de_win.ICO a:/cord
-mcopy -tmn /tmp/pc_gc/cord/private/* a:/cord/private
-mcopy -tmn /tmp/pc_gc/include/* a:/include
-mcopy -tmn /tmp/pc_gc/include/private/* a:/include/private
rm -r /tmp/pc_gc
gc.tar.Z: gc.tar
compress gc.tar
gc.tar.gz: gc.tar
gzip gc.tar
lint: $(CSRCS) tests/test.c
lint -DLINT $(CSRCS) tests/test.c | egrep -v "possible pointer alignment problem|abort|exit|sbrk|mprotect|syscall|change in ANSI|improper alignment"
# BTL: added to test shared library version of collector.
# Currently works only under SunOS5. Requires GC_INIT call from statically
# loaded client code.
ABSDIR = `pwd`
gctest_dyn_link: tests/test.o libgc.so
$(CC) -L$(ABSDIR) -R$(ABSDIR) -o gctest_dyn_link tests/test.o -lgc -ldl -lthread
gctest_irix_dyn_link: tests/test.o libirixgc.so
$(CC) -L$(ABSDIR) -o gctest_irix_dyn_link tests/test.o -lirixgc
# The following appear to be dead, especially since libgc_globals.h
# is apparently lost.
test_dll.o: tests/test.c libgc_globals.h
$(CC) $(CFLAGS) -DGC_USE_DLL -c tests/test.c -o test_dll.o
test_dll: test_dll.o libgc_dll.a libgc.dll
$(CC) test_dll.o -L$(ABSDIR) -lgc_dll -o test_dll
SYM_PREFIX-libgc=GC
# Uncomment the following line to build a GNU win32 DLL
# include Makefile.DLLs
reserved_namespace: $(SRCS)
for file in $(SRCS) tests/test.c tests/test_cpp.cc; do \
sed s/GC_/_GC_/g < $$file > tmp; \
cp tmp $$file; \
done
user_namespace: $(SRCS)
for file in $(SRCS) tests/test.c tests/test_cpp.cc; do \
sed s/_GC_/GC_/g < $$file > tmp; \
cp tmp $$file; \
done
-430
View File
@@ -1,430 +0,0 @@
# This Makefile is intended only for DJGPP use.
# It is mainly a copy of the main Makefile, but tends to get out of sync
# with it. A merge would probably be appropriate.
# Primary targets:
# gc.a - builds basic library
# libgc.a - builds library for use with g++ "-fgc-keyword" extension
# -fgc-keyword was never really available. Historical
# interest only.
# c++ - adds C++ interface to library
# cords - adds cords (heavyweight strings) to library
# test - prints porting information, then builds basic version of gc.a,
# and runs some tests of collector and cords. Does not add cords or
# c++ interface to gc.a
# cord/de$(EXE_SUFFIX) - builds dumb editor based on cords.
ABI_FLAG=
CC=gcc $(ABI_FLAG)
CXX=gxx $(ABI_FLAG)
AS=gcc -c -x assembler-with-cpp $(ABI_FLAG)
# The above doesn't work with gas, which doesn't run cpp.
# Define AS as `gcc -c -x assembler-with-cpp' instead.
# Under Irix 6, you will have to specify the ABI (-o32, -n32, or -64)
# if you use something other than the default ABI on your machine.
# special defines for DJGPP
CXXLD=gxx $(ABI_FLAG)
EXE_SUFFIX=.exe
srcdir= .
VPATH= $(srcdir)
CFLAGS= -gstabs+ -O2 -I$(srcdir)/include -DATOMIC_UNCOLLECTABLE -DNO_SIGNALS -DALL_INTERIOR_POINTERS -DNO_EXECUTE_PERMISSION -DSILENT
# Setjmp_test may yield overly optimistic results when compiled
# without optimization.
# -DSILENT disables statistics printing, and improves performance.
# -DFIND_LEAK causes GC_find_leak to be initially set.
# This causes the collector to assume that all inaccessible
# objects should have been explicitly deallocated, and reports exceptions.
# Finalization and the test program are not usable in this mode.
# -DALL_INTERIOR_POINTERS allows all pointers to the interior
# of objects to be recognized. (See gc_priv.h for consequences.)
# -DSMALL_CONFIG tries to tune the collector for small heap sizes,
# usually causing it to use less space in such situations.
# Incremental collection no longer works in this case.
# -DLARGE_CONFIG tunes the collector for unusually large heaps.
# Necessary for heaps larger than about 500 MB on most machines.
# Recommended for heaps larger than about 64 MB.
# -DDONT_ADD_BYTE_AT_END is meaningful only with
# -DALL_INTERIOR_POINTERS. Normally -DALL_INTERIOR_POINTERS
# causes all objects to be padded so that pointers just past the end of
# an object can be recognized. This can be expensive. (The padding
# is normally more than one byte due to alignment constraints.)
# -DDONT_ADD_BYTE_AT_END disables the padding.
# -DNO_SIGNALS does not disable signals during critical parts of
# the GC process. This is no less correct than many malloc
# implementations, and it sometimes has a significant performance
# impact. However, it is dangerous for many not-quite-ANSI C
# programs that call things like printf in asynchronous signal handlers.
# This is on by default. Turning it off has not been extensively tested with
# compilers that reorder stores. It should have been.
# -DNO_EXECUTE_PERMISSION may cause some or all of the heap to not
# have execute permission, i.e. it may be impossible to execute
# code from the heap. Currently this only affects the incremental
# collector on UNIX machines. It may greatly improve its performance,
# since this may avoid some expensive cache synchronization.
# -DGC_NO_OPERATOR_NEW_ARRAY declares that the C++ compiler does not support
# the new syntax "operator new[]" for allocating and deleting arrays.
# See gc_cpp.h for details. No effect on the C part of the collector.
# This is defined implicitly in a few environments. Must also be defined
# by clients that use gc_cpp.h.
# -DREDIRECT_MALLOC=X causes malloc, realloc, and free to be defined
# as aliases for X, GC_realloc, and GC_free, respectively.
# Calloc is redefined in terms of the new malloc. X should
# be either GC_malloc or GC_malloc_uncollectable.
# The former is occasionally useful for working around leaks in code
# you don't want to (or can't) look at. It may not work for
# existing code, but it often does. Neither works on all platforms,
# since some ports use malloc or calloc to obtain system memory.
# (Probably works for UNIX, and win32.)
# -DIGNORE_FREE turns calls to free into a noop. Only useful with
# -DREDIRECT_MALLOC.
# -DNO_DEBUGGING removes GC_dump and the debugging routines it calls.
# Reduces code size slightly at the expense of debuggability.
# -DJAVA_FINALIZATION makes it somewhat safer to finalize objects out of
# order by specifying a nonstandard finalization mark procedure (see
# finalize.c). Objects reachable from finalizable objects will be marked
# in a sepearte postpass, and hence their memory won't be reclaimed.
# Not recommended unless you are implementing a language that specifies
# these semantics. Since 5.0, determines only only the initial value
# of GC_java_finalization variable.
# -DFINALIZE_ON_DEMAND causes finalizers to be run only in response
# to explicit GC_invoke_finalizers() calls.
# In 5.0 this became runtime adjustable, and this only determines the
# initial value of GC_finalize_on_demand.
# -DATOMIC_UNCOLLECTABLE includes code for GC_malloc_atomic_uncollectable.
# This is useful if either the vendor malloc implementation is poor,
# or if REDIRECT_MALLOC is used.
# -DHBLKSIZE=ddd, where ddd is a power of 2 between 512 and 16384, explicitly
# sets the heap block size. Each heap block is devoted to a single size and
# kind of object. For the incremental collector it makes sense to match
# the most likely page size. Otherwise large values result in more
# fragmentation, but generally better performance for large heaps.
# -DPRINT_BLACK_LIST Whenever a black list entry is added, i.e. whenever
# the garbage collector detects a value that looks almost, but not quite,
# like a pointer, print both the address containing the value, and the
# value of the near-bogus-pointer. Can be used to identifiy regions of
# memory that are likely to contribute misidentified pointers.
# -DKEEP_BACK_PTRS Add code to save back pointers in debugging headers
# for objects allocated with the debugging allocator. If all objects
# through GC_MALLOC with GC_DEBUG defined, this allows the client
# to determine how particular or randomly chosen objects are reachable
# for debugging/profiling purposes. The gc_backptr.h interface is
# implemented only if this is defined.
# -DGC_ASSERTIONS Enable some internal GC assertion checking. Currently
# this facility is only used in a few places. It is intended primarily
# for debugging of the garbage collector itself, but could also
# -DDBG_HDRS_ALL Make sure that all objects have debug headers. Increases
# the reliability (from 99.9999% to 100%) of some of the debugging
# code (especially KEEP_BACK_PTRS). Makes -DSHORT_DBG_HDRS possible.
# Assumes that all client allocation is done through debugging
# allocators.
# -DSHORT_DBG_HDRS Assume that all objects have debug headers. Shorten
# the headers to minimize object size, at the expense of checking for
# writes past the end of an object. This is intended for environments
# in which most client code is written in a "safe" language, such as
# Scheme or Java. Assumes that all client allocation is done using
# the GC_debug_ functions (or through the macros that expand to these.
# (Also eliminates the field for the requested object size.)
# occasionally be useful for debugging of client code. Slows down the
# collector somewhat, but not drastically.
# -DCHECKSUMS reports on erroneously clear dirty bits, and unexpectedly
# altered stubborn objects, at substantial performance cost.
# Use only for debugging of the incremental collector.
# -DGC_GCJ_SUPPORT includes support for gcj (and possibly other systems
# that include a pointer to a type descriptor in each allocated object).
# Building this way requires an ANSI C compiler.
# -DUSE_I686_PREFETCH causes the collector to issue Pentium III style
# prefetch instructions. No effect except on X86 Linux platforms.
# Assumes a very recent gcc-compatible compiler and assembler.
# (Gas prefetcht0 support was added around May 1999.)
# Empirically the code appears to still run correctly on Pentium II
# processors, though with no performance benefit. May not run on other
# X86 processors? In some cases this improves performance by
# 15% or so.
# -DUSE_3DNOW_PREFETCH causes the collector to issue AMD 3DNow style
# prefetch instructions. Same restrictions as USE_I686_PREFETCH.
# UNTESTED!!
# -DGC_USE_LD_WRAP in combination with the gld flags listed in README.linux
# causes the collector some system and pthread calls in a more transparent
# fashion than the usual macro-based approach. Requires GNU ld, and
# currently probably works only with Linux.
CXXFLAGS= $(CFLAGS) -DGC_OPERATOR_NEW_ARRAY
AR= ar
RANLIB= ranlib
OBJS= alloc.o reclaim.o allchblk.o misc.o mach_dep.o os_dep.o mark_rts.o headers.o mark.o obj_map.o blacklst.o finalize.o new_hblk.o dbg_mlc.o malloc.o stubborn.o checksums.o solaris_threads.o typd_mlc.o ptr_chck.o mallocx.o solaris_pthreads.o gcj_mlc.o specific.o
CSRCS= reclaim.c allchblk.c misc.c alloc.c mach_dep.c os_dep.c mark_rts.c headers.c mark.c obj_map.c pcr_interface.c blacklst.c finalize.c new_hblk.c real_malloc.c dyn_load.c dbg_mlc.c malloc.c stubborn.c checksums.c solaris_threads.c typd_mlc.c ptr_chck.c mallocx.c solaris_pthreads.c gcj_mlc.c specific.c
CORD_SRCS= cord/cordbscs.c cord/cordxtra.c cord/cordprnt.c cord/de.c cord/cordtest.c include/cord.h include/ec.h include/private/cord_pos.h cord/de_win.c cord/de_win.h cord/de_cmds.h cord/de_win.ICO cord/de_win.RC cord/SCOPTIONS.amiga cord/SMakefile.amiga
CORD_OBJS= cord/cordbscs.o cord/cordxtra.o cord/cordprnt.o
SRCS= $(CSRCS) mips_sgi_mach_dep.S rs6000_mach_dep.s alpha_mach_dep.S \
sparc_mach_dep.S include/gc.h include/gc_typed.h \
include/private/gc_hdrs.h include/private/gc_priv.h \
include/private/gcconfig.h include/private/gc_mark.h \
include/gc_inl.h include/gc_inline.h gc.man \
threadlibs.c if_mach.c if_not_there.c gc_cpp.cc include/gc_cpp.h \
include/weakpointer.h include/private/gc_locks.h \
gcc_support.c mips_ultrix_mach_dep.s include/gc_alloc.h \
include/new_gc_alloc.h include/javaxfc.h sparc_sunos4_mach_dep.s \
include/private/solaris_threads.h include/gc_backptr.h \
hpux_test_and_clear.s include/gc_gcj.h \
include/gc_local_alloc.h include/private/dbg_mlc.h \
include/private/specific.h powerpc_darwin_mach_dep.s \
include/leak_detector.h $(CORD_SRCS)
OTHER_FILES= Makefile PCR-Makefile OS2_MAKEFILE NT_MAKEFILE BCC_MAKEFILE \
README tests/test.c test_cpp.cc setjmp_t.c SMakefile.amiga \
SCoptions.amiga README.amiga README.win32 cord/README \
README.rs6000 README.QUICK callprocs pc_excludes \
barrett_diagram README.OS2 README.Mac MacProjects.sit.hqx \
MacOS.c EMX_MAKEFILE README.debugging \
Mac_files/datastart.c Mac_files/dataend.c \
Mac_files/MacOS_config.h Mac_files/MacOS_Test_config.h \
add_gc_prefix.c README.solaris2 README.sgi README.hp README.uts \
win32_threads.c NT_THREADS_MAKEFILE gc.mak README.dj Makefile.dj \
README.alpha README.linux README.MacOSX version.h Makefile.DLLs \
WCC_MAKEFILE nursery.c include/gc_nursery.h include/gc_copy_descr.h
CORD_INCLUDE_FILES= $(srcdir)/include/gc.h $(srcdir)/include/cord.h \
$(srcdir)/include/ec.h $(srcdir)/include/private/cord_pos.h
UTILS= if_mach$(EXE_SUFFIX) if_not_there$(EXE_SUFFIX)
# Libraries needed for curses applications. Only needed for de.
CURSES= -lcurses -ltermlib
# The following is irrelevant on most systems. But a few
# versions of make otherwise fork the shell specified in
# the SHELL environment variable.
SHELL= /bin/sh
SPECIALCFLAGS = -I$(srcdir)/include
# Alternative flags to the C compiler for mach_dep.c.
# Mach_dep.c often doesn't like optimization, and it's
# not time-critical anyway.
# Set SPECIALCFLAGS to -q nodirect_code on Encore.
all: gc.a gctest$(EXE_SUFFIX)
$(OBJS) test.o dyn_load.o dyn_load_sunos53.o: \
$(srcdir)/include/private/gc_priv.h \
$(srcdir)/include/private/gc_hdrs.h $(srcdir)/include/private/gc_locks.h \
$(srcdir)/include/gc.h \
$(srcdir)/include/private/gcconfig.h $(srcdir)/include/gc_typed.h \
Makefile
# The dependency on Makefile is needed. Changing
# options such as -DSILENT affects the size of GC_arrays,
# invalidating all .o files that rely on gc_priv.h
mark.o typd_mlc.o finalize.o: $(srcdir)/include/gc_mark.h
base_lib gc.a: $(OBJS) dyn_load.o $(UTILS)
echo > base_lib
rm -f on_sparc_sunos5_1
./if_mach SPARC SUNOS5 touch on_sparc_sunos5_1
./if_mach SPARC SUNOS5 $(AR) rus gc.a $(OBJS) dyn_load.o
./if_not_there on_sparc_sunos5_1 $(AR) ru gc.a $(OBJS) dyn_load.o
-./if_not_there on_sparc_sunos5_1 $(RANLIB) gc.a
# ignore ranlib failure; that usually means it doesn't exist, and isn't needed
cords: $(CORD_OBJS) cord/cordtest$(EXE_SUFFIX) $(UTILS)
rm -f on_sparc_sunos5_3
./if_mach SPARC SUNOS5 touch on_sparc_sunos5_3
./if_mach SPARC SUNOS5 $(AR) rus gc.a $(CORD_OBJS)
./if_not_there on_sparc_sunos5_3 $(AR) ru gc.a $(CORD_OBJS)
-./if_not_there on_sparc_sunos5_3 $(RANLIB) gc.a
gc_cpp.o: $(srcdir)/gc_cpp.cc $(srcdir)/include/gc_cpp.h $(srcdir)/include/gc.h Makefile
$(CXX) -c $(CXXFLAGS) $(srcdir)/gc_cpp.cc
test_cpp$(EXE_SUFFIX): $(srcdir)/test_cpp.cc $(srcdir)/include/gc_cpp.h gc_cpp.o $(srcdir)/include/gc.h \
base_lib $(UTILS)
rm -f test_cpp test_cpp$(EXE_SUFFIX)
./if_mach HP_PA "" $(CXX) $(CXXFLAGS) -o test_cpp $(srcdir)/test_cpp.cc gc_cpp.o gc.a -ldld
./if_not_there test_cpp$(EXE_SUFFIX) $(CXXLD) $(CXXFLAGS) -o test_cpp$(EXE_SUFFIX) $(srcdir)/test_cpp.cc gc_cpp.o gc.a
rm -f test_cpp
c++: gc_cpp.o $(srcdir)/include/gc_cpp.h test_cpp$(EXE_SUFFIX)
rm -f on_sparc_sunos5_4
./if_mach SPARC SUNOS5 touch on_sparc_sunos5_4
./if_mach SPARC SUNOS5 $(AR) rus gc.a gc_cpp.o
./if_not_there on_sparc_sunos5_4 $(AR) ru gc.a gc_cpp.o
-./if_not_there on_sparc_sunos5_4 $(RANLIB) gc.a
./test_cpp$(EXE_SUFFIX) 1
echo > c++
dyn_load_sunos53.o: dyn_load.c
$(CC) $(CFLAGS) -DSUNOS53_SHARED_LIB -c $(srcdir)/dyn_load.c -o $@
# SunOS5 shared library version of the collector
sunos5gc.so: $(OBJS) dyn_load_sunos53.o
$(CC) -G -o sunos5gc.so $(OBJS) dyn_load_sunos53.o -ldl
ln sunos5gc.so libgc.so
# Alpha/OSF shared library version of the collector
libalphagc.so: $(OBJS)
ld -shared -o libalphagc.so $(OBJS) dyn_load.o -lc
ln libalphagc.so libgc.so
# IRIX shared library version of the collector
libirixgc.so: $(OBJS) dyn_load.o
ld -shared $(ABI_FLAG) -o libirixgc.so $(OBJS) dyn_load.o -lc
ln libirixgc.so libgc.so
# Linux shared library version of the collector
liblinuxgc.so: $(OBJS) dyn_load.o
gcc -shared -o liblinuxgc.so $(OBJS) dyn_load.o -lo
ln liblinuxgc.so libgc.so
mach_dep.o: $(srcdir)/mach_dep.c $(srcdir)/mips_sgi_mach_dep.S $(srcdir)/mips_ultrix_mach_dep.s \
$(srcdir)/rs6000_mach_dep.s $(srcdir)/powerpc_darwin_mach_dep.s $(UTILS)
rm -f mach_dep.o
./if_mach MIPS IRIX5 $(AS) -o mach_dep.o $(srcdir)/mips_sgi_mach_dep.S
./if_mach MIPS RISCOS $(AS) -o mach_dep.o $(srcdir)/mips_ultrix_mach_dep.s
./if_mach MIPS ULTRIX $(AS) -o mach_dep.o $(srcdir)/mips_ultrix_mach_dep.s
./if_mach RS6000 "" $(AS) -o mach_dep.o $(srcdir)/rs6000_mach_dep.s
./if_mach POWERPC MACOSX $(AS) -o mach_dep.o $(srcdir)/powerpc_darwin_mach_dep.s
./if_mach ALPHA "" $(AS) -o mach_dep.o $(srcdir)/alpha_mach_dep.S
./if_mach SPARC SUNOS5 $(AS) -o mach_dep.o $(srcdir)/sparc_mach_dep.S
./if_mach SPARC SUNOS4 $(AS) -o mach_dep.o $(srcdir)/sparc_sunos4_mach_dep.s
./if_not_there mach_dep.o $(CC) -c $(SPECIALCFLAGS) $(srcdir)/mach_dep.c
mark_rts.o: $(srcdir)/mark_rts.c if_mach if_not_there $(UTILS)
rm -f mark_rts.o
-./if_mach ALPHA OSF1 $(CC) -c $(CFLAGS) -Wo,-notail $(srcdir)/mark_rts.c
./if_not_there mark_rts.o $(CC) -c $(CFLAGS) $(srcdir)/mark_rts.c
# Work-around for DEC optimizer tail recursion elimination bug.
# The ALPHA-specific line should be removed if gcc is used.
alloc.o: version.h
cord/cordbscs.o: $(srcdir)/cord/cordbscs.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordbscs.c
mv cordbscs.o cord/cordbscs.o
# not all compilers understand -o filename
cord/cordxtra.o: $(srcdir)/cord/cordxtra.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordxtra.c
mv cordxtra.o cord/cordxtra.o
cord/cordprnt.o: $(srcdir)/cord/cordprnt.c $(CORD_INCLUDE_FILES)
$(CC) $(CFLAGS) -c -I$(srcdir) $(srcdir)/cord/cordprnt.c
mv cordprnt.o cord/cordprnt.o
cord/cordtest$(EXE_SUFFIX): $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a $(UTILS) /tmp
rm -f cord/cordtest$(EXE_SUFFIX)
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o cord/cordtest$(EXE_SUFFIX) $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a -lucb
./if_mach HP_PA "" $(CC) $(CFLAGS) -o cord/cordtest$(EXE_SUFFIX) $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a -ldld
./if_not_there cord/cordtest$(EXE_SUFFIX) $(CC) $(CFLAGS) -o cord/cordtest $(srcdir)/cord/cordtest.c $(CORD_OBJS) gc.a
rm -f cord/cordtest cordtest
-mv cordtest$(EXE_SUFFIX) cord/
/tmp: $(UTILS)
./if_not_there /tmp mkdir /tmp
cord/de$(EXE_SUFFIX): $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(UTILS)
rm -f cord/de cord/de$(EXE_SUFFIX)
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES) -lucb `./threadlibs`
./if_mach HP_PA "" $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES) -ldld
./if_mach RS6000 "" $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses
./if_mach I386 LINUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses `./threadlibs`
./if_mach ALPHA LINUX $(CC) $(CFLAGS) -o cord/de $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a -lcurses
./if_not_there cord/de$(EXE_SUFFIX) $(CC) $(CFLAGS) -o cord/de$(EXE_SUFFIX) $(srcdir)/cord/de.c cord/cordbscs.o cord/cordxtra.o gc.a $(CURSES)
if_mach$(EXE_SUFFIX): $(srcdir)/if_mach.c $(srcdir)/include/private/gcconfig.h
rm -f if_mach if_mach$(EXE_SUFFIX)
$(CC) $(CFLAGS) -o if_mach $(srcdir)/if_mach.c
threadlibs$(EXE_SUFFIX): $(srcdir)/threadlibs.c $(srcdir)include/private/gcconfig.h Makefile
rm -f threadlibs threadlibs$(EXE_SUFFIX)
$(CC) $(CFLAGS) -o threadlibs $(srcdir)/threadlibs.c
if_not_there$(EXE_SUFFIX): $(srcdir)/if_not_there.c
rm -f if_not_there if_not_there$(EXE_SUFFIX)
$(CC) $(CFLAGS) -o if_not_there $(srcdir)/if_not_there.c
# Clean removes *.o several times,
# because as the first one doesn't seem to get them all!
clean:
rm -f gc.a *.o
rm -f *.o
rm -f *.o
rm -f cord/*.o
rm -f gctest gctest_dyn_link test_cpp
rm -f setjmp_test mon.out gmon.out a.out core if_not_there if_mach
rm -f threadlibs $(CORD_OBJS) cordtest cord/cordtest de cord/de
rm -f gctest$(EXE_SUFFIX) gctest_dyn_link$(EXE_SUFFIX) test_cpp$(EXE_SUFFIX)
rm -f setjmp_test$(EXE_SUFFIX) if_not_there$(EXE_SUFFIX) if_mach$(EXE_SUFFIX)
rm -f threadlibs$(EXE_SUFFIX) cord/cordtest$(EXE_SUFFIX)
-rm -f *~
gctest$(EXE_SUFFIX): tests/test.o gc.a if_mach$(EXE_SUFFIX) if_not_there$(EXE_SUFFIX)
rm -f gctest gctest$(EXE_SUFFIX)
./if_mach SPARC DRSNX $(CC) $(CFLAGS) -o gctest tests/test.o gc.a -lucb
./if_mach HP_PA "" $(CC) $(CFLAGS) -o gctest tests/test.o gc.a -ldld
./if_not_there gctest$(EXE_SUFFIX) $(CC) $(CFLAGS) -o gctest$(EXE_SUFFIX) tests/test.o gc.a
rm -f gctest
# If an optimized setjmp_test generates a segmentation fault,
# odds are your compiler is broken. Gctest may still work.
# Try compiling setjmp_t.c unoptimized.
setjmp_test$(EXE_SUFFIX): $(srcdir)/setjmp_t.c $(srcdir)/include/gc.h \
if_mach$(EXE_SUFFIX) if_not_there$(EXE_SUFFIX)
rm -f setjmp_test$(EXE_SUFFIX)
$(CC) $(CFLAGS) -o setjmp_test $(srcdir)/setjmp_t.c
rm -f setjmp_test
test: KandRtest cord/cordtest$(EXE_SUFFIX)
./cord/cordtest$(EXE_SUFFIX)
# Those tests that work even with a K&R C compiler:
KandRtest: setjmp_test$(EXE_SUFFIX) gctest$(EXE_SUFFIX)
./setjmp_test$(EXE_SUFFIX)
./gctest$(EXE_SUFFIX)
add_gc_prefix$(EXE_SUFFIX): add_gc_prefix.c
$(CC) -o add_gc_prefix$(EXE_SUFFIX) $(srcdir)/add_gc_prefix.c
rm -f add_gc_prefix
gc.tar: $(SRCS) $(OTHER_FILES) add_gc_prefix
./add_gc_prefix$(EXE_SUFFIX) $(SRCS) $(OTHER_FILES) > /tmp/gc.tar-files
(cd $(srcdir)/.. ; tar cvfh - `cat /tmp/gc.tar-files`) > gc.tar
pc_gc.tar: $(SRCS) $(OTHER_FILES)
tar cvfX pc_gc.tar pc_excludes $(SRCS) $(OTHER_FILES)
gc.tar.Z: gc.tar
compress gc.tar
gc.tar.gz: gc.tar
gzip gc.tar
lint: $(CSRCS) tests/test.c
lint -DLINT $(CSRCS) tests/test.c | egrep -v "possible pointer alignment problem|abort|exit|sbrk|mprotect|syscall"
# BTL: added to test shared library version of collector.
# Currently works only under SunOS5. Requires GC_INIT call from statically
# loaded client code.
ABSDIR = `pwd`
gctest_dyn_link: test.o libgc.so
$(CC) -L$(ABSDIR) -R$(ABSDIR) -o gctest_dyn_link test.o -lgc -ldl -lthread
test_dll.o: tests/test.c libgc_globals.h
$(CC) $(CFLAGS) -DGC_USE_DLL -c tests/test.c -o test_dll.o
test_dll: test_dll.o libgc_dll.a libgc.dll
$(CC) test_dll.o -L$(ABSDIR) -lgc_dll -o test_dll
SYM_PREFIX-libgc=GC
# Uncomment the following line to build a GNU win32 DLL
# include Makefile.DLLs
-60
View File
@@ -1,60 +0,0 @@
# Makefile for Windows NT. Assumes Microsoft compiler, and a single thread.
# DLLs are included in the root set under NT, but not under win32S.
# Use "nmake nodebug=1 all" for optimized versions of library, gctest and editor.
MY_CPU=X86
CPU=$(MY_CPU)
!include <ntwin32.mak>
OBJS= alloc.obj reclaim.obj allchblk.obj misc.obj mach_dep.obj os_dep.obj mark_rts.obj headers.obj mark.obj obj_map.obj blacklst.obj finalize.obj new_hblk.obj dbg_mlc.obj malloc.obj stubborn.obj dyn_load.obj typd_mlc.obj ptr_chck.obj gc_cpp.obj mallocx.obj
all: gctest.exe cord\de.exe test_cpp.exe
.c.obj:
$(cc) $(cdebug) $(cflags) $(cvars) -Iinclude -DSILENT -DALL_INTERIOR_POINTERS -D__STDC__ -DGC_NOT_DLL -DGC_BUILD $*.c /Fo$*.obj
.cpp.obj:
$(cc) $(cdebug) $(cflags) $(cvars) -Iinclude -DSILENT -DALL_INTERIOR_POINTERS -DGC_NOT_DLL -DGC_BUILD $*.CPP /Fo$*.obj
$(OBJS) tests\test.obj: include\private\gc_priv.h include\private\gc_hdrs.h include\gc.h include\private\gcconfig.h include\private\gc_locks.h include\private\gc_pmark.h include\gc_mark.h
gc.lib: $(OBJS)
lib /MACHINE:i386 /out:gc.lib $(OBJS)
# The original NT SDK used lib32 instead of lib
gctest.exe: tests\test.obj gc.lib
# The following works for win32 debugging. For win32s debugging use debugtype:coff
# and add mapsympe line.
# This produces a "GUI" applications that opens no windows and writes to the log file
# "gc.log". This is done to make the result runnable under win32s.
$(link) -debug:full -debugtype:cv $(guiflags) -stack:131072 -out:$*.exe tests\test.obj $(guilibs) gc.lib
# mapsympe -n -o gctest.sym gctest.exe
cord\de_win.rbj: cord\de_win.res
cvtres /MACHINE:$(MY_CPU) /OUT:cord\de_win.rbj cord\de_win.res
cord\de.obj cord\de_win.obj: include\cord.h include\private\cord_pos.h cord\de_win.h cord\de_cmds.h
cord\de_win.res: cord\de_win.rc cord\de_win.h cord\de_cmds.h
$(rc) $(rcvars) -r -fo cord\de_win.res $(cvars) cord\de_win.rc
# Cord/de is a real win32 gui application.
cord\de.exe: cord\cordbscs.obj cord\cordxtra.obj cord\de.obj cord\de_win.obj cord\de_win.rbj gc.lib
$(link) -debug:full -debugtype:cv $(guiflags) -stack:16384 -out:cord\de.exe cord\cordbscs.obj cord\cordxtra.obj cord\de.obj cord\de_win.obj cord\de_win.rbj gc.lib $(guilibs)
gc_cpp.obj: include\gc_cpp.h include\gc.h
gc_cpp.cpp: gc_cpp.cc
copy gc_cpp.cc gc_cpp.cpp
test_cpp.cpp: tests\test_cpp.cc
copy tests\test_cpp.cc test_cpp.cpp
# This generates the C++ test executable. The executable expects
# a single numeric argument, which is the number of iterations.
# The output appears in the file "gc.log".
test_cpp.exe: test_cpp.obj include\gc_cpp.h include\gc.h gc.lib
$(link) -debug:full -debugtype:cv $(guiflags) -stack:16384 -out:test_cpp.exe test_cpp.obj gc.lib $(guilibs)
-60
View File
@@ -1,60 +0,0 @@
# Makefile for Windows NT. Assumes Microsoft compiler.
# DLLs are included in the root set under NT, but not under win32S.
# Use "nmake nodebug=1 all" for optimized versions of library, gctest and editor.
MY_CPU=X86
CPU=$(MY_CPU)
!include <ntwin32.mak>
OBJS= alloc.obj reclaim.obj allchblk.obj misc.obj mach_dep.obj os_dep.obj mark_rts.obj headers.obj mark.obj obj_map.obj blacklst.obj finalize.obj new_hblk.obj dbg_mlc.obj malloc.obj stubborn.obj dyn_load.obj typd_mlc.obj ptr_chck.obj gc_cpp.obj mallocx.obj win32_threads.obj
all: gctest.exe cord\de.exe test_cpp.exe
.c.obj:
$(cc) $(cdebug) $(cflags) $(cvars) -Iinclude -DSILENT -DALL_INTERIOR_POINTERS -D__STDC__ -DGC_NOT_DLL -DGC_WIN32_THREADS $*.c /Fo$*.obj
.cpp.obj:
$(cc) $(cdebug) $(cflags) $(cvars) -Iinclude -DSILENT -DALL_INTERIOR_POINTERS -DGC_NOT_DLL $*.CPP -DGC_WIN32_THREADS /Fo$*.obj
$(OBJS) tests\test.obj: include\private\gc_priv.h include\private\gc_hdrs.h include\gc.h include\private\gcconfig.h include\private\gc_locks.h include\private\gc_pmark.h include\gc_mark.h
gc.lib: $(OBJS)
lib /MACHINE:i386 /out:gc.lib $(OBJS)
# The original NT SDK used lib32 instead of lib
gctest.exe: tests\test.obj gc.lib
# The following works for win32 debugging. For win32s debugging use debugtype:coff
# and add mapsympe line.
# This produces a "GUI" applications that opens no windows and writes to the log file
# "gc.log". This is done to make the result runnable under win32s.
$(link) -debug:full -debugtype:cv $(guiflags) -stack:131072 -out:$*.exe tests\test.obj $(guilibs) gc.lib
# mapsympe -n -o gctest.sym gctest.exe
cord\de_win.rbj: cord\de_win.res
cvtres /MACHINE:$(MY_CPU) /OUT:cord\de_win.rbj cord\de_win.res
cord\de.obj cord\de_win.obj: include\cord.h include\private\cord_pos.h cord\de_win.h cord\de_cmds.h
cord\de_win.res: cord\de_win.rc cord\de_win.h cord\de_cmds.h
$(rc) $(rcvars) -r -fo cord\de_win.res $(cvars) cord\de_win.rc
# Cord/de is a real win32 gui application.
cord\de.exe: cord\cordbscs.obj cord\cordxtra.obj cord\de.obj cord\de_win.obj cord\de_win.rbj gc.lib
$(link) -debug:full -debugtype:cv $(guiflags) -stack:16384 -out:cord\de.exe cord\cordbscs.obj cord\cordxtra.obj cord\de.obj cord\de_win.obj cord\de_win.rbj gc.lib $(guilibs)
gc_cpp.obj: include\gc_cpp.h include\gc.h
gc_cpp.cpp: gc_cpp.cc
copy gc_cpp.cc gc_cpp.cpp
test_cpp.cpp: tests\test_cpp.cc
copy tests\test_cpp.cc test_cpp.cpp
# This generates the C++ test executable. The executable expects
# a single numeric argument, which is the number of iterations.
# The output appears in the file "gc.log".
test_cpp.exe: test_cpp.obj include\gc_cpp.h include\gc.h gc.lib
$(link) -debug:full -debugtype:cv $(guiflags) -stack:16384 -out:test_cpp.exe test_cpp.obj gc.lib $(guilibs)
File diff suppressed because it is too large. Load diff
-45
View File
@@ -1,45 +0,0 @@
# Makefile for OS/2. Assumes IBM's compiler, static linking, and a single thread.
# Adding dynamic linking support seems easy, but takes a little bit of work.
# Adding thread support may be nontrivial, since we haven't yet figured out how to
# look at another thread's registers.
# Significantly revised for GC version 4.4 by Mark Boulter (Jan 1994).
OBJS= alloc.obj reclaim.obj allchblk.obj misc.obj mach_dep.obj os_dep.obj mark_rts.obj headers.obj mark.obj obj_map.obj blacklst.obj finalize.obj new_hblk.obj dbg_mlc.obj malloc.obj stubborn.obj typd_mlc.obj ptr_chck.obj mallocx.obj
CORDOBJS= cord\cordbscs.obj cord\cordxtra.obj cord\cordprnt.obj
CC= icc
CFLAGS= /O /Q /DSILENT /DSMALL_CONFIG /DALL_INTERIOR_POINTERS
# Use /Ti instead of /O for debugging
# Setjmp_test may yield overly optimistic results when compiled
# without optimization.
all: $(OBJS) gctest.exe cord\cordtest.exe
$(OBJS) test.obj: include\private\gc_priv.h include\private\gc_hdrs.h include\gc.h include\private\gcconfig.h
## ERASE THE LIB FIRST - if it is already there then this command will fail
## (make sure its there or erase will fail!)
gc.lib: $(OBJS)
echo . > gc.lib
erase gc.lib
LIB gc.lib $(OBJS), gc.lst
mach_dep.obj: mach_dep.c
$(CC) $(CFLAGS) /C mach_dep.c
gctest.exe: test.obj gc.lib
$(CC) $(CFLAGS) /B"/STACK:524288" /Fegctest test.obj gc.lib
cord\cordbscs.obj: cord\cordbscs.c include\cord.h include\private\cord_pos.h
$(CC) $(CFLAGS) /C /Focord\cordbscs cord\cordbscs.c
cord\cordxtra.obj: cord\cordxtra.c include\cord.h include\private\cord_pos.h include\ec.h
$(CC) $(CFLAGS) /C /Focord\cordxtra cord\cordxtra.c
cord\cordprnt.obj: cord\cordprnt.c include\cord.h include\private\cord_pos.h include\ec.h
$(CC) $(CFLAGS) /C /Focord\cordprnt cord\cordprnt.c
cord\cordtest.exe: cord\cordtest.c include\cord.h include\private\cord_pos.h include\ec.h $(CORDOBJS) gc.lib
$(CC) $(CFLAGS) /B"/STACK:65536" /Fecord\cordtest cord\cordtest.c gc.lib $(CORDOBJS)
-68
View File
@@ -1,68 +0,0 @@
#
# Default target
#
default: gc.o
include ../config/common.mk
#
# compilation flags, etc.
#
CPPFLAGS = $(INCLUDE) $(CONFIG_CPPFLAGS) \
-DPCR_NO_RENAME -DPCR_NO_HOSTDEP_ERR
#CFLAGS = -DPCR -DSILENT $(CONFIG_CFLAGS)
CFLAGS = -DPCR $(CONFIG_CFLAGS)
SPECIALCFLAGS = # For code involving asm's
ASPPFLAGS = $(INCLUDE) $(CONFIG_ASPPFLAGS) \
-DPCR_NO_RENAME -DPCR_NO_HOSTDEP_ERR -DASM
ASFLAGS = $(CONFIG_ASFLAGS)
LDRFLAGS = $(CONFIG_LDRFLAGS)
LDFLAGS = $(CONFIG_LDFLAGS)
#
#
#
#
# BEGIN PACKAGE-SPECIFIC PART
#
#
#
#
# Fix to point to local pcr installation directory.
PCRDIR= ..
COBJ= alloc.o reclaim.o allchblk.o misc.o os_dep.o mark_rts.o headers.o mark.o obj_map.o pcr_interface.o blacklst.o finalize.o new_hblk.o real_malloc.o dyn_load.o dbg_mlc.o malloc.o stubborn.o checksums.o solaris_threads.o typd_mlc.o ptr_chck.o mallocx.o
CSRC= reclaim.c allchblk.c misc.c alloc.c mach_dep.c os_dep.c mark_rts.c headers.c mark.c obj_map.c pcr_interface.c blacklst.c finalize.c new_hblk.c real_malloc.c dyn_load.c dbg_mlc.c malloc.c stubborn.c checksums.c solaris_threads.c typd_mlc.c ptr_chck.c mallocx.c
SHELL= /bin/sh
default: gc.o
gc.o: $(COBJ) mach_dep.o
$(LDR) $(CONFIG_LDRFLAGS) -o gc.o $(COBJ) mach_dep.o
mach_dep.o: mach_dep.c mips_mach_dep.s rs6000_mach_dep.s if_mach if_not_there
rm -f mach_dep.o
./if_mach MIPS "" as -o mach_dep.o mips_mach_dep.s
./if_mach RS6000 "" as -o mach_dep.o rs6000_mach_dep.s
./if_mach ALPHA "" as -o mach_dep.o alpha_mach_dep.s
./if_mach SPARC SUNOS5 as -o mach_dep.o sparc_mach_dep.s
./if_not_there mach_dep.o $(CC) -c $(SPECIALCFLAGS) mach_dep.c
if_mach: if_mach.c gcconfig.h
$(CC) $(CFLAGS) -o if_mach if_mach.c
if_not_there: if_not_there.c
$(CC) $(CFLAGS) -o if_not_there if_not_there.c
-67
View File
@@ -1,67 +0,0 @@
This is a modified version of Boehm GC 6.1 for Mono.
* There are two main changes to the upstream version:
- Makefile changes:
libgc has a lot of configurable options which are AC_DEFINE()d in its
configure.ac. To make it easier to build and bundle it with Mono, I
replaced most of the orignal configure.ac and the makefiles with custom
versions which just define what we actually need for Mono.
This means that you can just run configure in this directory and it'll
do the right thing. Later on, we'll just include this package in Mono
and use AC_CONFIG_SUBDIRS().
- Threading changes
The original libgc has several *_threads.c files for each possible threading
implementation.
For Mono, we're using a vtable
typedef struct
{
void (* initialize) (void);
void (* lock) (void);
void (* unlock) (void);
void (* stop_world) (void);
void (* push_thread_structures) (void);
void (* push_all_stacks) (void);
void (* start_world) (void);
} GCThreadFunctions;
extern GCThreadFunctions *gc_thread_vtable;
and a mono_threads.c file.
- Deleted files
Some files from the original distribution have been deleted in this version.
These are files which weren't actually linked into the library so they were not
needed. When importing a new upstream version, you can either keep them removed
or just replace them with their new upstream versions.
- include/private/gc_locks.h
This file has been replaced with a custom version.
When importing a new upstream version, keep this custom version, ie. don't import the
new upstream gc_locks.h.
* Importing a new upstream version
This is really simple. Just import the new version to the vendor branch (LIBGC) in CVS
and then merge it into the main trunk.
To get a diff to the original version:
cvs diff -u -r LIBGC
When importing new upstream versions, don't import the new configure.ac or any of the
Makefile.am's; they've been replaced by custom versions. Just import all the new source
files and it should be fine.
April 4th, 2003
Martin Baulig <[email protected]>
-85
View File
@@ -1,85 +0,0 @@
Copyright 1988, 1989 Hans-J. Boehm, Alan J. Demers
Copyright (c) 1991-1995 by Xerox Corporation. All rights reserved.
Copyright (c) 1996-1999 by Silicon Graphics. All rights reserved.
Copyright (c) 1999-2001 by Hewlett-Packard. All rights reserved.
THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
Permission is hereby granted to use or copy this program
for any purpose, provided the above notices are retained on all copies.
Permission to modify the code and to distribute modified code is granted,
provided the above notices are retained, and a notice that the code was
modified is included with the above copyright notice.
A few files have other copyright holders. A few of the files needed
to use the GNU-style build procedure come with a modified GPL license
that appears not to significantly restrict use of the collector, though
use of those files for a purpose other than building the collector may
require the resulting code to be covered by the GPL.
For more details and the names of other contributors, see the
doc/README* files and include/gc.h. This file describes typical use of
the collector on a machine that is already supported.
For the version number, see doc/README or version.h.
INSTALLATION:
Under UN*X, Linux:
Alternative 1 (the old way): type "make test" in this directory.
Link against gc.a.
Alternative 2 (the new way): type
"./configure --prefix=<dir>; make; make check; make install".
Link against <dir>/lib/libgc.a or <dir>/lib/libgc.so.
See README.autoconf for details
Under OS/2 or Windows 95, 98, Me, NT, or 2000:
copy the appropriate makefile to MAKEFILE, read it, and type "nmake test".
(Under Windows, this assumes you have Microsoft command-line tools
installed, and have DOS configured with enough environment space to run them.)
Read the machine specific README in the doc directory if one exists.
The only way to develop code with the collector for Windows 3.1 is
to develop under Windows NT or 95+, and then to use win32S.
If you need thread support, you will need to either follow the special
platform-dependent instructions (win32), or add a suitable define
option as described in Makefile.
If you wish to use the cord (structured string) library, type
"make cords". (This requires an ANSI C compiler. You may need
to redefine CC in the Makefile. The CORD_printf implementation in
cordprnt.c is known to be less than perfectly portable. The rest
of the package should still work.)
If you wish to use the collector from C++, type
"make c++". These add further files to gc.a and to the include
subdirectory. See cord/cord.h and include/gc_cpp.h.
TYPICAL USE:
Include "gc.h" from the include subdirectory. Link against the
appropriate library ("gc.a" under UN*X). Replace calls to malloc
by calls to GC_MALLOC, and calls to realloc by calls to GC_REALLOC.
If the object is known to never contain pointers, use GC_MALLOC_ATOMIC
instead of GC_MALLOC.
Define GC_DEBUG before including gc.h for additional checking.
More documentation on the collector interface can be found at
http://www.hpl.hp.com/personal/Hans_Boehm/gc/gcinterface.html,
in doc/README, and in include/gc.h .
WARNINGS:
Do not store the only pointer to an object in memory allocated
with system malloc, since the collector usually does not scan
memory allocated in this way.
Use with threads may be supported on your system, but requires the
collector to be built with thread support. See Makefile. The collector
does not guarantee to scan thread-local storage (e.g. of the kind
accessed with pthread_getspecific()). The collector does scan
thread stacks though, so generally the best solution is to ensure that
any pointers stored in thread-local storage are also stored on the
thread's stack for the duration of their lifetime.
-178
View File
@@ -1,178 +0,0 @@
# Rewritten smakefile for amiga / sas/c. -Kjetil M.
# Dont use the cord-package if you define parm=both or parm=reg.
#----------------TOOLS--------------------------------
CC=sc
LINKER=slink
LIBER=oml
#----------------CPU OPTIONS--------------------------
CPU=68060
#----------------FPU OPTIONS--------------------------
MATH=8
MATHLIB=LIB:scm881.lib
#----------------COMPILER OPTIONS---------------------
IGNORE= IGNORE=85 IGNORE=154 IGNORE=161 IGNORE=100
OPTIMIZE=optimize optimizetime optglobal optimizerdepth=100 optimizerpeephole optloop OPTSCHED optimizerinlocal optimizerrecurdepth=100
# optimizerinline optimizercomplexity=100
OPT= $(OPTIMIZE) CPU=$(CPU) math=$(MATH) NOSTACKCHECK VERBOSE \
MAPHUNK NOVERSION NOICONS nodebug \
DEFINE SILENT \
parm=reg \
DEFINE __USE_SYSBASE
SOPT= $(OPT) $(IGNORE) \
DEFINE AMIGA_SKIP_SEG \
DEFINE ATOMIC_UNCOLLECTABLE \
DEFINE GC_AMIGA_FASTALLOC \
DEFINE GC_AMIGA_RETRY \
DEFINE GC_AMIGA_PRINTSTATS \
DEFINE GC_AMIGA_GC
#DEFINE ALL_INTERIOR_POINTERS \
SCOPT= $(SOPT) define GC_AMIGA_MAKINGLIB
CSCOPT= $(OPT) DEFINE AMIGA IGNORE=100 IGNORE=161
#------------------LINKING----------------------------
all: gctest setjmp_t cord/cordtest
clean:
delete *.lib gctest setjmp_t *.o *.lnk cord/*.o cord/*.lib cord/*.lnk cord/cordtest
smake
test: setjmp_t gctest cord/cordtest
setjmp_t
gctest
cord/cordtest
gctest: gc$(CPU).lib GCAmigaOS$(CPU).lib test.o
$(LINKER) LIB:c.o test.o TO gctest LIB gc$(CPU).lib LIB:sc.lib $(MATHLIB)
setjmp_t: setjmp_t.o gc.h
$(LINKER) LIB:c.o setjmp_t.o to setjmp_t lib LIB:sc.lib
cord/cordtest: cord/cordtest.o cord/cord$(CPU).lib gc$(CPU).lib
slink LIB:c.o cord/cordtest.o LIB $(MATHLIB) gc$(CPU).lib cord/cord$(CPU).lib LIB:sc.lib TO cord/cordtest
#------------------LIBBING----------------------------
OBJS= alloc.o reclaim.o allchblk.o misc.o mach_dep.o os_dep.o mark_rts.o headers.o mark.o obj_map.o blacklst.o finalize.o new_hblk.o real_malloc.o dyn_load.o dbg_mlc.o malloc.o stubborn.o checksums.o typd_mlc.o ptr_chck.o mallocx.o
gc$(CPU).lib: $(OBJS)
$(LIBER) gc$(CPU).lib r $(OBJS)
COBJS = cord/cordbscs.o cord/cordprnt.o cord/cordxtra.o
cord/cord$(CPU).lib: $(COBJS)
oml cord/cord$(CPU).lib r $(COBJS)
#------------------COMPILING--------------------------
INC= gc_private.h gc_hdrs.h gc.h gcconfig.h
alloc.o : alloc.c $(INC)
$(CC) alloc.c $(SCOPT) ignore=7
reclaim.o : reclaim.c $(INC)
$(CC) reclaim.c $(SCOPT)
allchblk.o : allchblk.c $(INC)
$(CC) allchblk.c $(SCOPT)
misc.o : misc.c $(INC)
$(CC) misc.c $(SCOPT)
os_dep.o : os_dep.c $(INC) AmigaOS.c
$(CC) os_dep.c $(SCOPT)
mark_rts.o : mark_rts.c $(INC)
$(CC) mark_rts.c $(SCOPT)
headers.o : headers.c $(INC)
$(CC) headers.c $(SCOPT)
mark.o : mark.c $(INC)
$(CC) mark.c $(SCOPT)
obj_map.o : obj_map.c $(INC)
$(CC) obj_map.c $(SCOPT)
blacklst.o : blacklst.c $(INC)
$(CC) blacklst.c $(SCOPT)
finalize.o : finalize.c $(INC)
$(CC) finalize.c $(SCOPT) noopt #Could sas/c still have problems with this one? Gctest sometimes fails to finalize all.
new_hblk.o : new_hblk.c $(INC)
$(CC) new_hblk.c $(SCOPT)
real_malloc.o : real_malloc.c $(INC)
$(CC) real_malloc.c $(SCOPT)
dyn_load.o : dyn_load.c $(INC)
$(CC) dyn_load.c $(SCOPT)
dbg_mlc.o : dbg_mlc.c $(INC)
$(CC) dbg_mlc.c $(SCOPT)
malloc.o : malloc.c $(INC)
$(CC) malloc.c $(SCOPT)
mallocx.o : mallocx.c $(INC)
$(CC) mallocx.c $(SCOPT)
stubborn.o : stubborn.c $(INC)
$(CC) stubborn.c $(SCOPT)
checksums.o : checksums.c $(INC)
$(CC) checksums.c $(SCOPT)
typd_mlc.o: typd_mlc.c $(INC)
$(CC) typd_mlc.c $(SCOPT)
mach_dep.o : mach_dep.c $(INC)
$(CC) mach_dep.c $(SCOPT)
ptr_chck.o: ptr_chck.c $(INC)
$(CC) ptr_chck.c $(SCOPT)
test.o : test.c $(INC)
$(CC) test.c $(SOPT)
setjmp_t: setjmp_t.c gc.h
$(CC) setjmp_t.c $(SOPT)
# cords:
cord/cordbscs.o: cord/cordbscs.c
sc cord/cordbscs.c $(CSCOPT)
cord/cordprnt.o: cord/cordprnt.c
sc cord/cordprnt.c $(CSCOPT)
cord/cordxtra.o: cord/cordxtra.c
sc cord/cordxtra.c $(CSCOPT)
cord/cordtest.o: cord/cordtest.c
sc cord/cordtest.c $(CSCOPT)
-196
View File
@@ -1,196 +0,0 @@
# Makefile for Watcom C/C++ 10.5, 10.6, 11.0 on NT, OS2 and DOS4GW.
# May work with Watcom 10.0.
# Uncoment one of the lines below for cross compilation.
SYSTEM=MSWIN32
#SYSTEM=DOS4GW
#SYSTEM=OS2
# The collector can be built either as dynamic or as static library.
# Select the library type you need.
#MAKE_AS_DLL=1
MAKE_AS_LIB=1
# Select calling conventions.
# Possible choices are r and s.
CALLING=s
# Select target CPU.
# Possible choices are 3, 4, 5, and 6.
# The last choice available only since version 11.0.
CPU=5
# Set optimization options.
# Watcom before 11.0 does not support option "-oh".
OPTIM=-oneatx -s
#OPTIM=-ohneatx -s
DEFS=-DALL_INTERIOR_POINTERS -DSILENT -DNO_SIGNALS #-DSMALL_CONFIG #-DGC_DEBUG
#####
!ifndef SYSTEM
!ifdef __MSDOS__
SYSTEM=DOS4GW
!else ifdef __NT__
SYSTEM=MSWIN32
!else ifdef __OS2__
SYSTEM=OS2
!else
SYSTEM=Unknown
!endif
!endif
!define $(SYSTEM)
!ifdef DOS4GW
SYSFLAG=-DDOS4GW -bt=dos
!else ifdef MSWIN32
SYSFLAG=-DMSWIN32 -bt=nt
!else ifdef OS2
SYSFLAG=-DOS2 -bt=os2
!else
!error undefined or unsupported target platform: $(SYSTEM)
!endif
!ifdef MAKE_AS_DLL
DLLFLAG=-bd -DGC_DLL
TEST_DLLFLAG=-DGC_DLL
!else ifdef MAKE_AS_LIB
DLLFLAG=
TEST_DLLFLAG=
!else
!error Either MAKE_AS_LIB or MAKE_AS_DLL should be defined
!endif
CC=wcc386
CXX=wpp386
# -DUSE_GENERIC is required !
CFLAGS=-$(CPU)$(CALLING) $(OPTIM) -zp4 -zc $(SYSFLAG) $(DLLFLAG) -DGC_BUILD -DUSE_GENERIC $(DEFS)
CXXFLAGS= $(CFLAGS)
TEST_CFLAGS=-$(CPU)$(CALLING) $(OPTIM) -zp4 -zc $(SYSFLAG) $(TEST_DLLFLAG) $(DEFS)
TEST_CXXFLAGS= $(TEST_CFLAGS)
OBJS= alloc.obj reclaim.obj allchblk.obj misc.obj &
mach_dep.obj os_dep.obj mark_rts.obj headers.obj mark.obj &
obj_map.obj blacklst.obj finalize.obj new_hblk.obj &
dbg_mlc.obj malloc.obj stubborn.obj dyn_load.obj &
typd_mlc.obj ptr_chck.obj mallocx.obj
all: gc.lib gctest.exe test_cpp.exe
!ifdef MAKE_AS_DLL
gc.lib: gc.dll gc_cpp.obj
*wlib -b -c -n -p=512 $@ +gc.dll +gc_cpp.obj
gc.dll: $(OBJS) .AUTODEPEND
@%create $*.lnk
!ifdef DOS4GW
@%append $*.lnk sys os2v2_dll
!else ifdef MSWIN32
@%append $*.lnk sys nt_dll
!else ifdef OS2
@%append $*.lnk sys os2v2_dll
!endif
@%append $*.lnk name $*
@for %i in ($(OBJS)) do @%append $*.lnk file '%i'
!ifeq CALLING s
@%append $*.lnk export GC_is_marked
@%append $*.lnk export GC_incr_words_allocd
@%append $*.lnk export GC_incr_mem_freed
@%append $*.lnk export GC_generic_malloc_words_small
!else
@%append $*.lnk export GC_is_marked_
@%append $*.lnk export GC_incr_words_allocd_
@%append $*.lnk export GC_incr_mem_freed_
@%append $*.lnk export GC_generic_malloc_words_small_
!endif
*wlink @$*.lnk
!else
gc.lib: $(OBJS) gc_cpp.obj
@%create $*.lb1
@for %i in ($(OBJS)) do @%append $*.lb1 +'%i'
@%append $*.lb1 +'gc_cpp.obj'
*wlib -b -c -n -p=512 $@ @$*.lb1
!endif
gctest.exe: test.obj gc.lib
%create $*.lnk
!ifdef DOS4GW
@%append $*.lnk sys dos4g
!else ifdef MSWIN32
@%append $*.lnk sys nt
!else ifdef OS2
@%append $*.lnk sys os2v2
!endif
@%append $*.lnk op case
@%append $*.lnk op stack=256K
@%append $*.lnk name $*
@%append $*.lnk file test.obj
@%append $*.lnk library gc.lib
!ifdef MAKE_AS_DLL
!ifeq CALLING s
@%append $*.lnk import GC_is_marked gc
!else
@%append $*.lnk import GC_is_marked_ gc
!endif
!endif
*wlink @$*.lnk
test_cpp.exe: test_cpp.obj gc.lib
%create $*.lnk
!ifdef DOS4GW
@%append $*.lnk sys dos4g
!else ifdef MSWIN32
@%append $*.lnk sys nt
!else ifdef OS2
@%append $*.lnk sys os2v2
!endif
@%append $*.lnk op case
@%append $*.lnk op stack=256K
@%append $*.lnk name $*
@%append $*.lnk file test_cpp.obj
@%append $*.lnk library gc.lib
!ifdef MAKE_AS_DLL
!ifeq CALLING s
@%append $*.lnk import GC_incr_words_allocd gc
@%append $*.lnk import GC_incr_mem_freed gc
@%append $*.lnk import GC_generic_malloc_words_small gc
!else
@%append $*.lnk import GC_incr_words_allocd_ gc
@%append $*.lnk import GC_incr_mem_freed_ gc
@%append $*.lnk import GC_generic_malloc_words_small_ gc
!endif
!endif
*wlink @$*.lnk
gc_cpp.obj: gc_cpp.cc .AUTODEPEND
$(CXX) $(TEST_CXXFLAGS) -iinclude $*.cc
test.obj: tests\test.c .AUTODEPEND
$(CC) $(TEST_CFLAGS) $*.c
test_cpp.obj: tests\test_cpp.cc .AUTODEPEND
$(CXX) $(TEST_CXXFLAGS) -iinclude $*.cc
.c.obj: .AUTODEPEND
$(CC) $(CFLAGS) $*.c
.cc.obj: .AUTODEPEND
$(CXX) $(CXXFLAGS) $*.cc
clean : .SYMBOLIC
@if exist *.obj del *.obj
@if exist *.map del *.map
@if exist *.lnk del *.lnk
@if exist *.lb1 del *.lb1
@if exist *.sym del *.sym
@if exist *.err del *.err
@if exist *.tmp del *.tmp
@if exist *.lst del *.lst
@if exist *.exe del *.exe
@if exist *.log del *.log
@if exist *.lib del *.lib
@if exist *.dll del *.dll
-182
View File
@@ -1,182 +0,0 @@
dnl dolt, a replacement for libtool
dnl Copyright © 2007-2008 Josh Triplett <[email protected]>
dnl Copying and distribution of this file, with or without modification,
dnl are permitted in any medium without royalty provided the copyright
dnl notice and this notice are preserved.
dnl
dnl To use dolt, invoke the DOLT macro immediately after the libtool macros.
dnl Optionally, copy this file into acinclude.m4, to avoid the need to have it
dnl installed when running autoconf on your project.
AC_DEFUN([DOLT], [
AC_REQUIRE([AC_CANONICAL_HOST])
# dolt, a replacement for libtool
# Josh Triplett <[email protected]>
AC_PATH_PROG(DOLT_BASH, bash)
AC_MSG_CHECKING([if dolt supports this host])
dolt_supported=yes
if test x$DOLT_BASH = x; then
dolt_supported=no
fi
if test x$GCC != xyes; then
dolt_supported=no
fi
case $host in
i?86-*-linux*|i?86-apple-darwin*|x86_64-*-linux*|powerpc-*-linux*|powerpc64-*-linux* \
|amd64-*-freebsd*|i?86-*-freebsd*|ia64-*-freebsd*|arm*-*-linux*|sparc*-*-linux*|mips*-*-linux*|x86_64-apple-darwin*|aarch64*)
pic_options='-fPIC'
;;
?86-pc-cygwin*|i?86-pc-cygwin*|i?86-pc-mingw32*)
pic_options='-DDLL_EXPORT'
;;
i?86-apple-darwin*|arm-apple-darwin*)
pic_options='-fno-common'
;;
*)
dolt_supported=no
;;
esac
if test x$dolt_supported = xno ; then
AC_MSG_RESULT([no, falling back to libtool])
LTCOMPILE='$(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(COMPILE)'
LTCXXCOMPILE='$(LIBTOOL) --tag=CXX $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CXXCOMPILE)'
else
AC_MSG_RESULT([yes, replacing libtool])
dnl Start writing out doltcompile.
cat <<__DOLTCOMPILE__EOF__ >doltcompile
#!$DOLT_BASH
__DOLTCOMPILE__EOF__
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
args=("$[]@")
for ((arg=0; arg<${#args@<:@@@:>@}; arg++)) ; do
if test x"${args@<:@$arg@:>@}" = x-o ; then
objarg=$((arg+1))
break
fi
done
if test x$objarg = x ; then
echo 'Error: no -o on compiler command line' 1>&2
exit 1
fi
lo="${args@<:@$objarg@:>@}"
obj="${lo%.lo}"
if test x"$lo" = x"$obj" ; then
echo "Error: libtool object file name \"$lo\" does not end in .lo" 1>&2
exit 1
fi
objbase="${obj##*/}"
__DOLTCOMPILE__EOF__
dnl Write out shared compilation code.
if test x$enable_shared = xyes; then
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
libobjdir="${obj%$objbase}.libs"
if test ! -d "$libobjdir" ; then
mkdir_out="$(mkdir "$libobjdir" 2>&1)"
mkdir_ret=$?
if test "$mkdir_ret" -ne 0 && test ! -d "$libobjdir" ; then
echo "$mkdir_out" 1>&2
exit $mkdir_ret
fi
fi
pic_object="$libobjdir/$objbase.o"
args@<:@$objarg@:>@="$pic_object"
__DOLTCOMPILE__EOF__
cat <<__DOLTCOMPILE__EOF__ >>doltcompile
"\${args@<:@@@:>@}" $pic_options -DPIC || exit \$?
__DOLTCOMPILE__EOF__
fi
dnl Write out static compilation code.
dnl Avoid duplicate compiler output if also building shared objects.
if test x$enable_static = xyes; then
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
non_pic_object="$obj.o"
args@<:@$objarg@:>@="$non_pic_object"
__DOLTCOMPILE__EOF__
if test x$enable_shared = xyes; then
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
"${args@<:@@@:>@}" >/dev/null 2>&1 || exit $?
__DOLTCOMPILE__EOF__
else
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
"${args@<:@@@:>@}" || exit $?
__DOLTCOMPILE__EOF__
fi
fi
dnl Write out the code to write the .lo file.
dnl The second line of the .lo file must match "^# Generated by .*libtool"
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
{
echo "# $lo - a libtool object file"
echo "# Generated by doltcompile, not libtool"
__DOLTCOMPILE__EOF__
if test x$enable_shared = xyes; then
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
echo "pic_object='.libs/${objbase}.o'"
__DOLTCOMPILE__EOF__
else
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
echo pic_object=none
__DOLTCOMPILE__EOF__
fi
if test x$enable_static = xyes; then
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
echo "non_pic_object='${objbase}.o'"
__DOLTCOMPILE__EOF__
else
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
echo non_pic_object=none
__DOLTCOMPILE__EOF__
fi
cat <<'__DOLTCOMPILE__EOF__' >>doltcompile
} > "$lo"
__DOLTCOMPILE__EOF__
dnl Done writing out doltcompile; substitute it for libtool compilation.
chmod +x doltcompile
LTCOMPILE='$(top_builddir)/doltcompile $(COMPILE)'
LTCXXCOMPILE='$(top_builddir)/doltcompile $(CXXCOMPILE)'
dnl automake ignores LTCOMPILE and LTCXXCOMPILE when it has separate CFLAGS for
dnl a target, so write out a libtool wrapper to handle that case.
dnl Note that doltlibtool does not handle inferred tags or option arguments
dnl without '=', because automake does not use them.
cat <<__DOLTLIBTOOL__EOF__ > doltlibtool
#!$DOLT_BASH
__DOLTLIBTOOL__EOF__
cat <<'__DOLTLIBTOOL__EOF__' >>doltlibtool
top_builddir_slash="${0%%doltlibtool}"
: ${top_builddir_slash:=./}
args=()
modeok=false
tagok=false
for arg in "$[]@"; do
case "$arg" in
--silent) ;;
--mode=compile) modeok=true ;;
--tag=CC|--tag=CXX) tagok=true ;;
--quiet) ;;
*) args@<:@${#args[@]}@:>@="$arg" ;;
esac
done
if $modeok && $tagok ; then
. ${top_builddir_slash}doltcompile "${args@<:@@@:>@}"
else
exec ${top_builddir_slash}libtool "$[]@"
fi
__DOLTLIBTOOL__EOF__
dnl Done writing out doltlibtool; substitute it for libtool.
chmod +x doltlibtool
LIBTOOL='$(top_builddir)/doltlibtool'
fi
AC_SUBST(LTCOMPILE)
AC_SUBST(LTCXXCOMPILE)
# end dolt
])
-20
View File
@@ -1,20 +0,0 @@
# include <stdio.h>
# include "version.h"
int main(argc, argv, envp)
int argc;
char ** argv;
char ** envp;
{
int i;
for (i = 1; i < argc; i++) {
if (GC_ALPHA_VERSION == GC_NOT_ALPHA) {
printf("gc%d.%d/%s ", GC_VERSION_MAJOR, GC_VERSION_MINOR, argv[i]);
} else {
printf("gc%d.%dalpha%d/%s ", GC_VERSION_MAJOR,
GC_VERSION_MINOR, GC_ALPHA_VERSION, argv[i]);
}
}
return(0);
}
-688
View File
@@ -1,688 +0,0 @@
/*
* Copyright (c) 1991-1995 by Xerox Corporation. All rights reserved.
* Copyright (c) 1996-1999 by Silicon Graphics. All rights reserved.
* Copyright (c) 1999-2003 by Hewlett-Packard Company. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/*
* Support code for Irix (>=6.2) Pthreads and for AIX pthreads.
* This relies on properties
* not guaranteed by the Pthread standard. It may or may not be portable
* to other implementations.
*
* Note that there is a lot of code duplication between this file and
* (pthread_support.c, pthread_stop_world.c). They should be merged.
* Pthread_support.c should be directly usable.
*
* Please avoid adding new ports here; use the generic pthread support
* as a base instead.
*/
# if defined(GC_IRIX_THREADS) || defined(GC_AIX_THREADS)
# include "private/gc_priv.h"
# include <pthread.h>
# include <assert.h>
# include <semaphore.h>
# include <time.h>
# include <errno.h>
# include <unistd.h>
# include <sys/mman.h>
# include <sys/time.h>
#undef pthread_create
#undef pthread_sigmask
#undef pthread_join
#if defined(GC_IRIX_THREADS) && !defined(MUTEX_RECURSIVE_NP)
#define MUTEX_RECURSIVE_NP PTHREAD_MUTEX_RECURSIVE
#endif
void GC_thr_init();
#if 0
void GC_print_sig_mask()
{
sigset_t blocked;
int i;
if (pthread_sigmask(SIG_BLOCK, NULL, &blocked) != 0)
ABORT("pthread_sigmask");
GC_printf0("Blocked: ");
for (i = 1; i <= MAXSIG; i++) {
if (sigismember(&blocked, i)) { GC_printf1("%ld ",(long) i); }
}
GC_printf0("\n");
}
#endif
/* We use the allocation lock to protect thread-related data structures. */
/* The set of all known threads. We intercept thread creation and */
/* joins. We never actually create detached threads. We allocate all */
/* new thread stacks ourselves. These allow us to maintain this */
/* data structure. */
/* Protected by GC_thr_lock. */
/* Some of this should be declared volatile, but that's incosnsistent */
/* with some library routine declarations. */
typedef struct GC_Thread_Rep {
struct GC_Thread_Rep * next; /* More recently allocated threads */
/* with a given pthread id come */
/* first. (All but the first are */
/* guaranteed to be dead, but we may */
/* not yet have registered the join.) */
pthread_t id;
word stop;
# define NOT_STOPPED 0
# define PLEASE_STOP 1
# define STOPPED 2
word flags;
# define FINISHED 1 /* Thread has exited. */
# define DETACHED 2 /* Thread is intended to be detached. */
ptr_t stack_cold; /* cold end of the stack */
ptr_t stack_hot; /* Valid only when stopped. */
/* But must be within stack region at */
/* all times. */
void * status; /* Used only to avoid premature */
/* reclamation of any data it might */
/* reference. */
} * GC_thread;
GC_thread GC_lookup_thread(pthread_t id);
/*
* The only way to suspend threads given the pthread interface is to send
* signals. Unfortunately, this means we have to reserve
* a signal, and intercept client calls to change the signal mask.
*/
#if 0 /* DOB: 6.1 */
# if defined(GC_AIX_THREADS)
# define SIG_SUSPEND SIGUSR1
# else
# define SIG_SUSPEND (SIGRTMIN + 6)
# endif
#endif
pthread_mutex_t GC_suspend_lock = PTHREAD_MUTEX_INITIALIZER;
/* Number of threads stopped so far */
pthread_cond_t GC_suspend_ack_cv = PTHREAD_COND_INITIALIZER;
pthread_cond_t GC_continue_cv = PTHREAD_COND_INITIALIZER;
void GC_suspend_handler(int sig)
{
int dummy;
GC_thread me;
sigset_t all_sigs;
sigset_t old_sigs;
int i;
if (sig != SIG_SUSPEND) ABORT("Bad signal in suspend_handler");
me = GC_lookup_thread(pthread_self());
/* The lookup here is safe, since I'm doing this on behalf */
/* of a thread which holds the allocation lock in order */
/* to stop the world. Thus concurrent modification of the */
/* data structure is impossible. */
if (PLEASE_STOP != me -> stop) {
/* Misdirected signal. */
pthread_mutex_unlock(&GC_suspend_lock);
return;
}
pthread_mutex_lock(&GC_suspend_lock);
me -> stack_hot = (ptr_t)(&dummy);
me -> stop = STOPPED;
pthread_cond_signal(&GC_suspend_ack_cv);
pthread_cond_wait(&GC_continue_cv, &GC_suspend_lock);
pthread_mutex_unlock(&GC_suspend_lock);
/* GC_printf1("Continuing 0x%x\n", pthread_self()); */
}
GC_bool GC_thr_initialized = FALSE;
# define THREAD_TABLE_SZ 128 /* Must be power of 2 */
volatile GC_thread GC_threads[THREAD_TABLE_SZ];
void GC_push_thread_structures GC_PROTO((void))
{
GC_push_all((ptr_t)(GC_threads), (ptr_t)(GC_threads)+sizeof(GC_threads));
}
/* Add a thread to GC_threads. We assume it wasn't already there. */
/* Caller holds allocation lock. */
GC_thread GC_new_thread(pthread_t id)
{
int hv = ((word)id) % THREAD_TABLE_SZ;
GC_thread result;
static struct GC_Thread_Rep first_thread;
static GC_bool first_thread_used = FALSE;
GC_ASSERT(I_HOLD_LOCK());
if (!first_thread_used) {
result = &first_thread;
first_thread_used = TRUE;
/* Dont acquire allocation lock, since we may already hold it. */
} else {
result = (struct GC_Thread_Rep *)
GC_generic_malloc_inner(sizeof(struct GC_Thread_Rep), NORMAL);
}
if (result == 0) return(0);
result -> id = id;
result -> next = GC_threads[hv];
GC_threads[hv] = result;
/* result -> flags = 0; */
/* result -> stop = 0; */
return(result);
}
/* Delete a thread from GC_threads. We assume it is there. */
/* (The code intentionally traps if it wasn't.) */
/* Caller holds allocation lock. */
/* We explicitly pass in the GC_thread we're looking for, since */
/* if a thread has been joined, but we have not yet */
/* been notified, then there may be more than one thread */
/* in the table with the same pthread id. */
/* This is OK, but we need a way to delete a specific one. */
void GC_delete_gc_thread(pthread_t id, GC_thread gc_id)
{
int hv = ((word)id) % THREAD_TABLE_SZ;
register GC_thread p = GC_threads[hv];
register GC_thread prev = 0;
GC_ASSERT(I_HOLD_LOCK());
while (p != gc_id) {
prev = p;
p = p -> next;
}
if (prev == 0) {
GC_threads[hv] = p -> next;
} else {
prev -> next = p -> next;
}
}
/* Return a GC_thread corresponding to a given thread_t. */
/* Returns 0 if it's not there. */
/* Caller holds allocation lock or otherwise inhibits */
/* updates. */
/* If there is more than one thread with the given id we */
/* return the most recent one. */
GC_thread GC_lookup_thread(pthread_t id)
{
int hv = ((word)id) % THREAD_TABLE_SZ;
register GC_thread p = GC_threads[hv];
/* I either hold the lock, or i'm being called from the stop-the-world
* handler. */
#if defined(GC_AIX_THREADS)
GC_ASSERT(I_HOLD_LOCK()); /* no stop-the-world handler needed on AIX */
#endif
while (p != 0 && !pthread_equal(p -> id, id)) p = p -> next;
return(p);
}
#if defined(GC_AIX_THREADS)
void GC_stop_world()
{
pthread_t my_thread = pthread_self();
register int i;
register GC_thread p;
register int result;
struct timespec timeout;
GC_ASSERT(I_HOLD_LOCK());
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
if (p -> id != my_thread) {
pthread_suspend_np(p->id);
}
}
}
/* GC_printf1("World stopped 0x%x\n", pthread_self()); */
}
void GC_start_world()
{
GC_thread p;
unsigned i;
pthread_t my_thread = pthread_self();
/* GC_printf0("World starting\n"); */
GC_ASSERT(I_HOLD_LOCK());
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
if (p -> id != my_thread) {
pthread_continue_np(p->id);
}
}
}
}
#else /* GC_AIX_THREADS */
/* Caller holds allocation lock. */
void GC_stop_world()
{
pthread_t my_thread = pthread_self();
register int i;
register GC_thread p;
register int result;
struct timespec timeout;
GC_ASSERT(I_HOLD_LOCK());
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
if (p -> id != my_thread) {
if (p -> flags & FINISHED) {
p -> stop = STOPPED;
continue;
}
p -> stop = PLEASE_STOP;
result = pthread_kill(p -> id, SIG_SUSPEND);
/* GC_printf1("Sent signal to 0x%x\n", p -> id); */
switch(result) {
case ESRCH:
/* Not really there anymore. Possible? */
p -> stop = STOPPED;
break;
case 0:
break;
default:
ABORT("pthread_kill failed");
}
}
}
}
pthread_mutex_lock(&GC_suspend_lock);
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
while (p -> id != my_thread && p -> stop != STOPPED) {
clock_gettime(CLOCK_REALTIME, &timeout);
timeout.tv_nsec += 50000000; /* 50 msecs */
if (timeout.tv_nsec >= 1000000000) {
timeout.tv_nsec -= 1000000000;
++timeout.tv_sec;
}
result = pthread_cond_timedwait(&GC_suspend_ack_cv,
&GC_suspend_lock,
&timeout);
if (result == ETIMEDOUT) {
/* Signal was lost or misdirected. Try again. */
/* Duplicate signals should be benign. */
result = pthread_kill(p -> id, SIG_SUSPEND);
}
}
}
}
pthread_mutex_unlock(&GC_suspend_lock);
/* GC_printf1("World stopped 0x%x\n", pthread_self()); */
}
/* Caller holds allocation lock. */
void GC_start_world()
{
GC_thread p;
unsigned i;
/* GC_printf0("World starting\n"); */
GC_ASSERT(I_HOLD_LOCK());
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
p -> stop = NOT_STOPPED;
}
}
pthread_mutex_lock(&GC_suspend_lock);
/* All other threads are at pthread_cond_wait in signal handler. */
/* Otherwise we couldn't have acquired the lock. */
pthread_mutex_unlock(&GC_suspend_lock);
pthread_cond_broadcast(&GC_continue_cv);
}
#endif /* GC_AIX_THREADS */
/* We hold allocation lock. Should do exactly the right thing if the */
/* world is stopped. Should not fail if it isn't. */
void GC_push_all_stacks()
{
register int i;
register GC_thread p;
register ptr_t hot, cold;
pthread_t me = pthread_self();
/* GC_init() should have been called before GC_push_all_stacks is
* invoked, and GC_init calls GC_thr_init(), which sets
* GC_thr_initialized. */
GC_ASSERT(GC_thr_initialized);
/* GC_printf1("Pushing stacks from thread 0x%x\n", me); */
GC_ASSERT(I_HOLD_LOCK());
for (i = 0; i < THREAD_TABLE_SZ; i++) {
for (p = GC_threads[i]; p != 0; p = p -> next) {
if (p -> flags & FINISHED) continue;
cold = p->stack_cold;
if (!cold) cold=GC_stackbottom; /* 0 indicates 'original stack' */
if (pthread_equal(p -> id, me)) {
hot = GC_approx_sp();
} else {
# ifdef GC_AIX_THREADS
/* AIX doesn't use signals to suspend, so we need to get an */
/* accurate hot stack pointer. */
/* See http://publib16.boulder.ibm.com/pseries/en_US/libs/basetrf1/pthread_getthrds_np.htm */
pthread_t id = p -> id;
struct __pthrdsinfo pinfo;
int regbuf[64];
int val = sizeof(regbuf);
int retval = pthread_getthrds_np(&id, PTHRDSINFO_QUERY_ALL, &pinfo,
sizeof(pinfo), regbuf, &val);
if (retval != 0) {
printf("ERROR: pthread_getthrds_np() failed in GC\n");
abort();
}
/* according to the AIX ABI,
"the lowest possible valid stack address is 288 bytes (144 + 144)
less than the current value of the stack pointer. Functions may
use this stack space as volatile storage which is not preserved
across function calls."
ftp://ftp.penguinppc64.org/pub/people/amodra/PPC-elf64abi.txt.gz
*/
hot = (ptr_t)(unsigned long)pinfo.__pi_ustk-288;
cold = (ptr_t)pinfo.__pi_stackend; /* more precise */
/* push the registers too, because they won't be on stack */
GC_push_all_eager((ptr_t)&pinfo.__pi_context,
(ptr_t)((&pinfo.__pi_context)+1));
GC_push_all_eager((ptr_t)regbuf, ((ptr_t)regbuf)+val);
# else
hot = p -> stack_hot;
# endif
}
# ifdef STACK_GROWS_UP
GC_push_all_stack(cold, hot);
# else
/* printf("thread 0x%x: hot=0x%08x cold=0x%08x\n", p -> id, hot, cold); */
GC_push_all_stack(hot, cold);
# endif
}
}
}
/* We hold the allocation lock. */
void GC_thr_init()
{
GC_thread t;
struct sigaction act;
if (GC_thr_initialized) return;
GC_ASSERT(I_HOLD_LOCK());
GC_thr_initialized = TRUE;
#ifndef GC_AIX_THREADS
(void) sigaction(SIG_SUSPEND, 0, &act);
if (act.sa_handler != SIG_DFL)
ABORT("Previously installed SIG_SUSPEND handler");
/* Install handler. */
act.sa_handler = GC_suspend_handler;
act.sa_flags = SA_RESTART;
(void) sigemptyset(&act.sa_mask);
if (0 != sigaction(SIG_SUSPEND, &act, 0))
ABORT("Failed to install SIG_SUSPEND handler");
#endif
/* Add the initial thread, so we can stop it. */
t = GC_new_thread(pthread_self());
/* use '0' to indicate GC_stackbottom, since GC_init() has not
* completed by the time we are called (from GC_init_inner()) */
t -> stack_cold = 0; /* the original stack. */
t -> stack_hot = (ptr_t)(&t);
t -> flags = DETACHED;
}
int GC_pthread_sigmask(int how, const sigset_t *set, sigset_t *oset)
{
sigset_t fudged_set;
#ifdef GC_AIX_THREADS
return(pthread_sigmask(how, set, oset));
#endif
if (set != NULL && (how == SIG_BLOCK || how == SIG_SETMASK)) {
fudged_set = *set;
sigdelset(&fudged_set, SIG_SUSPEND);
set = &fudged_set;
}
return(pthread_sigmask(how, set, oset));
}
struct start_info {
void *(*start_routine)(void *);
void *arg;
word flags;
pthread_mutex_t registeredlock;
pthread_cond_t registered;
int volatile registereddone;
};
void GC_thread_exit_proc(void *arg)
{
GC_thread me;
LOCK();
me = GC_lookup_thread(pthread_self());
me -> flags |= FINISHED;
/* reclaim DETACHED thread right away; otherwise wait until join() */
if (me -> flags & DETACHED) {
GC_delete_gc_thread(pthread_self(), me);
}
UNLOCK();
}
int GC_pthread_join(pthread_t thread, void **retval)
{
int result;
GC_thread thread_gc_id;
LOCK();
thread_gc_id = GC_lookup_thread(thread);
/* This is guaranteed to be the intended one, since the thread id */
/* cant have been recycled by pthreads. */
UNLOCK();
GC_ASSERT(!(thread_gc_id->flags & DETACHED));
result = pthread_join(thread, retval);
/* Some versions of the Irix pthreads library can erroneously */
/* return EINTR when the call succeeds. */
if (EINTR == result) result = 0;
GC_ASSERT(thread_gc_id->flags & FINISHED);
LOCK();
/* Here the pthread thread id may have been recycled. */
GC_delete_gc_thread(thread, thread_gc_id);
UNLOCK();
return result;
}
void * GC_start_routine(void * arg)
{
int dummy;
struct start_info * si = arg;
void * result;
GC_thread me;
pthread_t my_pthread;
void *(*start)(void *);
void *start_arg;
my_pthread = pthread_self();
/* If a GC occurs before the thread is registered, that GC will */
/* ignore this thread. That's fine, since it will block trying to */
/* acquire the allocation lock, and won't yet hold interesting */
/* pointers. */
LOCK();
/* We register the thread here instead of in the parent, so that */
/* we don't need to hold the allocation lock during pthread_create. */
/* Holding the allocation lock there would make REDIRECT_MALLOC */
/* impossible. It probably still doesn't work, but we're a little */
/* closer ... */
/* This unfortunately means that we have to be careful the parent */
/* doesn't try to do a pthread_join before we're registered. */
me = GC_new_thread(my_pthread);
me -> flags = si -> flags;
me -> stack_cold = (ptr_t) &dummy; /* this now the 'start of stack' */
me -> stack_hot = me->stack_cold;/* this field should always be sensible */
UNLOCK();
start = si -> start_routine;
start_arg = si -> arg;
pthread_mutex_lock(&(si->registeredlock));
si->registereddone = 1;
pthread_cond_signal(&(si->registered));
pthread_mutex_unlock(&(si->registeredlock));
/* si went away as soon as we did this unlock */
pthread_cleanup_push(GC_thread_exit_proc, 0);
result = (*start)(start_arg);
me -> status = result;
pthread_cleanup_pop(1);
/* This involves acquiring the lock, ensuring that we can't exit */
/* while a collection that thinks we're alive is trying to stop */
/* us. */
return(result);
}
int
GC_pthread_create(pthread_t *new_thread,
const pthread_attr_t *attr,
void *(*start_routine)(void *), void *arg)
{
int result;
GC_thread t;
int detachstate;
word my_flags = 0;
struct start_info * si;
/* This is otherwise saved only in an area mmapped by the thread */
/* library, which isn't visible to the collector. */
LOCK();
/* GC_INTERNAL_MALLOC implicitly calls GC_init() if required */
si = (struct start_info *)GC_INTERNAL_MALLOC(sizeof(struct start_info),
NORMAL);
GC_ASSERT(GC_thr_initialized); /* initialized by GC_init() */
UNLOCK();
if (0 == si) return(ENOMEM);
pthread_mutex_init(&(si->registeredlock), NULL);
pthread_cond_init(&(si->registered),NULL);
pthread_mutex_lock(&(si->registeredlock));
si -> start_routine = start_routine;
si -> arg = arg;
pthread_attr_getdetachstate(attr, &detachstate);
if (PTHREAD_CREATE_DETACHED == detachstate) my_flags |= DETACHED;
si -> flags = my_flags;
result = pthread_create(new_thread, attr, GC_start_routine, si);
/* Wait until child has been added to the thread table. */
/* This also ensures that we hold onto si until the child is done */
/* with it. Thus it doesn't matter whether it is otherwise */
/* visible to the collector. */
if (0 == result) {
si->registereddone = 0;
while (!si->registereddone)
pthread_cond_wait(&(si->registered), &(si->registeredlock));
}
pthread_mutex_unlock(&(si->registeredlock));
pthread_cond_destroy(&(si->registered));
pthread_mutex_destroy(&(si->registeredlock));
LOCK();
GC_INTERNAL_FREE(si);
UNLOCK();
return(result);
}
/* For now we use the pthreads locking primitives on HP/UX */
VOLATILE GC_bool GC_collecting = 0; /* A hint that we're in the collector and */
/* holding the allocation lock for an */
/* extended period. */
/* Reasonably fast spin locks. Basically the same implementation */
/* as STL alloc.h. */
#define SLEEP_THRESHOLD 3
volatile unsigned int GC_allocate_lock = 0;
#define GC_TRY_LOCK() !GC_test_and_set(&GC_allocate_lock)
#define GC_LOCK_TAKEN GC_allocate_lock
void GC_lock()
{
# define low_spin_max 30 /* spin cycles if we suspect uniprocessor */
# define high_spin_max 1000 /* spin cycles for multiprocessor */
static unsigned spin_max = low_spin_max;
unsigned my_spin_max;
static unsigned last_spins = 0;
unsigned my_last_spins;
volatile unsigned junk;
# define PAUSE junk *= junk; junk *= junk; junk *= junk; junk *= junk
int i;
if (GC_TRY_LOCK()) {
return;
}
junk = 0;
my_spin_max = spin_max;
my_last_spins = last_spins;
for (i = 0; i < my_spin_max; i++) {
if (GC_collecting) goto yield;
if (i < my_last_spins/2 || GC_LOCK_TAKEN) {
PAUSE;
continue;
}
if (GC_TRY_LOCK()) {
/*
* got it!
* Spinning worked. Thus we're probably not being scheduled
* against the other process with which we were contending.
* Thus it makes sense to spin longer the next time.
*/
last_spins = i;
spin_max = high_spin_max;
return;
}
}
/* We are probably being scheduled against the other process. Sleep. */
spin_max = low_spin_max;
yield:
for (i = 0;; ++i) {
if (GC_TRY_LOCK()) {
return;
}
if (i < SLEEP_THRESHOLD) {
sched_yield();
} else {
struct timespec ts;
if (i > 26) i = 26;
/* Don't wait for more than about 60msecs, even */
/* under extreme contention. */
ts.tv_sec = 0;
ts.tv_nsec = 1 << i;
nanosleep(&ts, 0);
}
}
}
# else /* !GC_IRIX_THREADS && !GC_AIX_THREADS */
#ifndef LINT
int GC_no_Irix_threads;
#endif
# endif /* IRIX_THREADS */
-838
View File
@@ -1,838 +0,0 @@
/*
* Copyright 1988, 1989 Hans-J. Boehm, Alan J. Demers
* Copyright (c) 1991-1994 by Xerox Corporation. All rights reserved.
* Copyright (c) 1998-1999 by Silicon Graphics. All rights reserved.
* Copyright (c) 1999 by Hewlett-Packard Company. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* #define DEBUG */
#include <stdio.h>
#include "private/gc_priv.h"
GC_bool GC_use_entire_heap = 0;
/*
* Free heap blocks are kept on one of several free lists,
* depending on the size of the block. Each free list is doubly linked.
* Adjacent free blocks are coalesced.
*/
# define MAX_BLACK_LIST_ALLOC (2*HBLKSIZE)
/* largest block we will allocate starting on a black */
/* listed block. Must be >= HBLKSIZE. */
# define UNIQUE_THRESHOLD 32
/* Sizes up to this many HBLKs each have their own free list */
# define HUGE_THRESHOLD 256
/* Sizes of at least this many heap blocks are mapped to a */
/* single free list. */
# define FL_COMPRESSION 8
/* In between sizes map this many distinct sizes to a single */
/* bin. */
# define N_HBLK_FLS (HUGE_THRESHOLD - UNIQUE_THRESHOLD)/FL_COMPRESSION \
+ UNIQUE_THRESHOLD
struct hblk * GC_hblkfreelist[N_HBLK_FLS+1] = { 0 };
#ifndef USE_MUNMAP
word GC_free_bytes[N_HBLK_FLS+1] = { 0 };
/* Number of free bytes on each list. */
/* Is bytes + the number of free bytes on lists n .. N_HBLK_FLS */
/* > GC_max_large_allocd_bytes? */
# ifdef __GNUC__
__inline__
# endif
static GC_bool GC_enough_large_bytes_left(bytes,n)
word bytes;
int n;
{
int i;
for (i = N_HBLK_FLS; i >= n; --i) {
bytes += GC_free_bytes[i];
if (bytes > GC_max_large_allocd_bytes) return TRUE;
}
return FALSE;
}
# define INCR_FREE_BYTES(n, b) GC_free_bytes[n] += (b);
# define FREE_ASSERT(e) GC_ASSERT(e)
#else /* USE_MUNMAP */
# define INCR_FREE_BYTES(n, b)
# define FREE_ASSERT(e)
#endif /* USE_MUNMAP */
/* Map a number of blocks to the appropriate large block free list index. */
int GC_hblk_fl_from_blocks(blocks_needed)
word blocks_needed;
{
if (blocks_needed <= UNIQUE_THRESHOLD) return blocks_needed;
if (blocks_needed >= HUGE_THRESHOLD) return N_HBLK_FLS;
return (blocks_needed - UNIQUE_THRESHOLD)/FL_COMPRESSION
+ UNIQUE_THRESHOLD;
}
# define PHDR(hhdr) HDR(hhdr -> hb_prev)
# define NHDR(hhdr) HDR(hhdr -> hb_next)
# ifdef USE_MUNMAP
# define IS_MAPPED(hhdr) (((hhdr) -> hb_flags & WAS_UNMAPPED) == 0)
# else /* !USE_MMAP */
# define IS_MAPPED(hhdr) 1
# endif /* USE_MUNMAP */
# if !defined(NO_DEBUGGING)
void GC_print_hblkfreelist()
{
struct hblk * h;
word total_free = 0;
hdr * hhdr;
word sz;
int i;
for (i = 0; i <= N_HBLK_FLS; ++i) {
h = GC_hblkfreelist[i];
# ifdef USE_MUNMAP
if (0 != h) GC_printf1("Free list %ld:\n",
(unsigned long)i);
# else
if (0 != h) GC_printf2("Free list %ld (Total size %ld):\n",
(unsigned long)i,
(unsigned long)GC_free_bytes[i]);
# endif
while (h != 0) {
hhdr = HDR(h);
sz = hhdr -> hb_sz;
GC_printf2("\t0x%lx size %lu ", (unsigned long)h, (unsigned long)sz);
total_free += sz;
if (GC_is_black_listed(h, HBLKSIZE) != 0) {
GC_printf0("start black listed\n");
} else if (GC_is_black_listed(h, hhdr -> hb_sz) != 0) {
GC_printf0("partially black listed\n");
} else {
GC_printf0("not black listed\n");
}
h = hhdr -> hb_next;
}
}
# ifndef USE_MUNMAP
if (total_free != GC_large_free_bytes) {
GC_printf1("GC_large_free_bytes = %lu (INCONSISTENT!!)\n",
(unsigned long) GC_large_free_bytes);
}
# endif
GC_printf1("Total of %lu bytes on free list\n", (unsigned long)total_free);
}
/* Return the free list index on which the block described by the header */
/* appears, or -1 if it appears nowhere. */
int free_list_index_of(wanted)
hdr * wanted;
{
struct hblk * h;
hdr * hhdr;
int i;
for (i = 0; i <= N_HBLK_FLS; ++i) {
h = GC_hblkfreelist[i];
while (h != 0) {
hhdr = HDR(h);
if (hhdr == wanted) return i;
h = hhdr -> hb_next;
}
}
return -1;
}
void GC_dump_regions()
{
unsigned i;
ptr_t start, end;
ptr_t p;
size_t bytes;
hdr *hhdr;
for (i = 0; i < GC_n_heap_sects; ++i) {
start = GC_heap_sects[i].hs_start;
bytes = GC_heap_sects[i].hs_bytes;
end = start + bytes;
/* Merge in contiguous sections. */
while (i+1 < GC_n_heap_sects && GC_heap_sects[i+1].hs_start == end) {
++i;
end = GC_heap_sects[i].hs_start + GC_heap_sects[i].hs_bytes;
}
GC_printf2("***Section from 0x%lx to 0x%lx\n", start, end);
for (p = start; p < end;) {
hhdr = HDR(p);
GC_printf1("\t0x%lx ", (unsigned long)p);
if (IS_FORWARDING_ADDR_OR_NIL(hhdr)) {
GC_printf1("Missing header!!(%ld)\n", hhdr);
p += HBLKSIZE;
continue;
}
if (HBLK_IS_FREE(hhdr)) {
int correct_index = GC_hblk_fl_from_blocks(
divHBLKSZ(hhdr -> hb_sz));
int actual_index;
GC_printf1("\tfree block of size 0x%lx bytes",
(unsigned long)(hhdr -> hb_sz));
if (IS_MAPPED(hhdr)) {
GC_printf0("\n");
} else {
GC_printf0("(unmapped)\n");
}
actual_index = free_list_index_of(hhdr);
if (-1 == actual_index) {
GC_printf1("\t\tBlock not on free list %ld!!\n",
correct_index);
} else if (correct_index != actual_index) {
GC_printf2("\t\tBlock on list %ld, should be on %ld!!\n",
actual_index, correct_index);
}
p += hhdr -> hb_sz;
} else {
GC_printf1("\tused for blocks of size 0x%lx bytes\n",
(unsigned long)WORDS_TO_BYTES(hhdr -> hb_sz));
p += HBLKSIZE * OBJ_SZ_TO_BLOCKS(hhdr -> hb_sz);
}
}
}
}
# endif /* NO_DEBUGGING */
/* Initialize hdr for a block containing the indicated size and */
/* kind of objects. */
/* Return FALSE on failure. */
static GC_bool setup_header(hhdr, sz, kind, flags)
register hdr * hhdr;
word sz; /* object size in words */
int kind;
unsigned char flags;
{
register word descr;
/* Add description of valid object pointers */
if (!GC_add_map_entry(sz)) return(FALSE);
hhdr -> hb_map = GC_obj_map[sz > MAXOBJSZ? 0 : sz];
/* Set size, kind and mark proc fields */
hhdr -> hb_sz = sz;
hhdr -> hb_obj_kind = kind;
hhdr -> hb_flags = flags;
descr = GC_obj_kinds[kind].ok_descriptor;
if (GC_obj_kinds[kind].ok_relocate_descr) descr += WORDS_TO_BYTES(sz);
hhdr -> hb_descr = descr;
/* Clear mark bits */
GC_clear_hdr_marks(hhdr);
hhdr -> hb_last_reclaimed = (unsigned short)GC_gc_no;
return(TRUE);
}
#define FL_UNKNOWN -1
/*
* Remove hhdr from the appropriate free list.
* We assume it is on the nth free list, or on the size
* appropriate free list if n is FL_UNKNOWN.
*/
void GC_remove_from_fl(hhdr, n)
hdr * hhdr;
int n;
{
int index;
GC_ASSERT(((hhdr -> hb_sz) & (HBLKSIZE-1)) == 0);
# ifndef USE_MUNMAP
/* We always need index to mainatin free counts. */
if (FL_UNKNOWN == n) {
index = GC_hblk_fl_from_blocks(divHBLKSZ(hhdr -> hb_sz));
} else {
index = n;
}
# endif
if (hhdr -> hb_prev == 0) {
# ifdef USE_MUNMAP
if (FL_UNKNOWN == n) {
index = GC_hblk_fl_from_blocks(divHBLKSZ(hhdr -> hb_sz));
} else {
index = n;
}
# endif
GC_ASSERT(HDR(GC_hblkfreelist[index]) == hhdr);
GC_hblkfreelist[index] = hhdr -> hb_next;
} else {
hdr *phdr;
GET_HDR(hhdr -> hb_prev, phdr);
phdr -> hb_next = hhdr -> hb_next;
}
FREE_ASSERT(GC_free_bytes[index] >= hhdr -> hb_sz);
INCR_FREE_BYTES(index, - (signed_word)(hhdr -> hb_sz));
if (0 != hhdr -> hb_next) {
hdr * nhdr;
GC_ASSERT(!IS_FORWARDING_ADDR_OR_NIL(NHDR(hhdr)));
GET_HDR(hhdr -> hb_next, nhdr);
nhdr -> hb_prev = hhdr -> hb_prev;
}
}
/*
* Return a pointer to the free block ending just before h, if any.
*/
struct hblk * GC_free_block_ending_at(h)
struct hblk *h;
{
struct hblk * p = h - 1;
hdr * phdr;
GET_HDR(p, phdr);
while (0 != phdr && IS_FORWARDING_ADDR_OR_NIL(phdr)) {
p = FORWARDED_ADDR(p,phdr);
phdr = HDR(p);
}
if (0 != phdr) {
if(HBLK_IS_FREE(phdr)) {
return p;
} else {
return 0;
}
}
p = GC_prev_block(h - 1);
if (0 != p) {
phdr = HDR(p);
if (HBLK_IS_FREE(phdr) && (ptr_t)p + phdr -> hb_sz == (ptr_t)h) {
return p;
}
}
return 0;
}
/*
* Add hhdr to the appropriate free list.
* We maintain individual free lists sorted by address.
*/
void GC_add_to_fl(h, hhdr)
struct hblk *h;
hdr * hhdr;
{
int index = GC_hblk_fl_from_blocks(divHBLKSZ(hhdr -> hb_sz));
struct hblk *second = GC_hblkfreelist[index];
hdr * second_hdr;
# ifdef GC_ASSERTIONS
struct hblk *next = (struct hblk *)((word)h + hhdr -> hb_sz);
hdr * nexthdr = HDR(next);
struct hblk *prev = GC_free_block_ending_at(h);
hdr * prevhdr = HDR(prev);
GC_ASSERT(nexthdr == 0 || !HBLK_IS_FREE(nexthdr) || !IS_MAPPED(nexthdr));
GC_ASSERT(prev == 0 || !HBLK_IS_FREE(prevhdr) || !IS_MAPPED(prevhdr));
# endif
GC_ASSERT(((hhdr -> hb_sz) & (HBLKSIZE-1)) == 0);
GC_hblkfreelist[index] = h;
INCR_FREE_BYTES(index, hhdr -> hb_sz);
FREE_ASSERT(GC_free_bytes[index] <= GC_large_free_bytes)
hhdr -> hb_next = second;
hhdr -> hb_prev = 0;
if (0 != second) {
GET_HDR(second, second_hdr);
second_hdr -> hb_prev = h;
}
GC_invalidate_map(hhdr);
}
#ifdef USE_MUNMAP
/* Unmap blocks that haven't been recently touched. This is the only way */
/* way blocks are ever unmapped. */
void GC_unmap_old(void)
{
struct hblk * h;
hdr * hhdr;
word sz;
unsigned short last_rec, threshold;
int i;
# define UNMAP_THRESHOLD 6
for (i = 0; i <= N_HBLK_FLS; ++i) {
for (h = GC_hblkfreelist[i]; 0 != h; h = hhdr -> hb_next) {
hhdr = HDR(h);
if (!IS_MAPPED(hhdr)) continue;
threshold = (unsigned short)(GC_gc_no - UNMAP_THRESHOLD);
last_rec = hhdr -> hb_last_reclaimed;
if ((last_rec > GC_gc_no || last_rec < threshold)
&& threshold < GC_gc_no /* not recently wrapped */) {
sz = hhdr -> hb_sz;
GC_unmap((ptr_t)h, sz);
hhdr -> hb_flags |= WAS_UNMAPPED;
}
}
}
}
/* Merge all unmapped blocks that are adjacent to other free */
/* blocks. This may involve remapping, since all blocks are either */
/* fully mapped or fully unmapped. */
void GC_merge_unmapped(void)
{
struct hblk * h, *next;
hdr * hhdr, *nexthdr;
word size, nextsize;
int i;
for (i = 0; i <= N_HBLK_FLS; ++i) {
h = GC_hblkfreelist[i];
while (h != 0) {
GET_HDR(h, hhdr);
size = hhdr->hb_sz;
next = (struct hblk *)((word)h + size);
GET_HDR(next, nexthdr);
/* Coalesce with successor, if possible */
if (0 != nexthdr && HBLK_IS_FREE(nexthdr)) {
nextsize = nexthdr -> hb_sz;
if (IS_MAPPED(hhdr)) {
GC_ASSERT(!IS_MAPPED(nexthdr));
/* make both consistent, so that we can merge */
if (size > nextsize) {
GC_remap((ptr_t)next, nextsize);
} else {
GC_unmap((ptr_t)h, size);
hhdr -> hb_flags |= WAS_UNMAPPED;
}
} else if (IS_MAPPED(nexthdr)) {
GC_ASSERT(!IS_MAPPED(hhdr));
if (size > nextsize) {
GC_unmap((ptr_t)next, nextsize);
} else {
GC_remap((ptr_t)h, size);
hhdr -> hb_flags &= ~WAS_UNMAPPED;
hhdr -> hb_last_reclaimed = nexthdr -> hb_last_reclaimed;
}
} else {
/* Unmap any gap in the middle */
GC_unmap_gap((ptr_t)h, size, (ptr_t)next, nexthdr -> hb_sz);
}
/* If they are both unmapped, we merge, but leave unmapped. */
GC_remove_from_fl(hhdr, i);
GC_remove_from_fl(nexthdr, FL_UNKNOWN);
hhdr -> hb_sz += nexthdr -> hb_sz;
GC_remove_header(next);
GC_add_to_fl(h, hhdr);
/* Start over at beginning of list */
h = GC_hblkfreelist[i];
} else /* not mergable with successor */ {
h = hhdr -> hb_next;
}
} /* while (h != 0) ... */
} /* for ... */
}
#endif /* USE_MUNMAP */
/*
* Return a pointer to a block starting at h of length bytes.
* Memory for the block is mapped.
* Remove the block from its free list, and return the remainder (if any)
* to its appropriate free list.
* May fail by returning 0.
* The header for the returned block must be set up by the caller.
* If the return value is not 0, then hhdr is the header for it.
*/
struct hblk * GC_get_first_part(h, hhdr, bytes, index)
struct hblk *h;
hdr * hhdr;
word bytes;
int index;
{
word total_size = hhdr -> hb_sz;
struct hblk * rest;
hdr * rest_hdr;
GC_ASSERT((total_size & (HBLKSIZE-1)) == 0);
GC_remove_from_fl(hhdr, index);
if (total_size == bytes) return h;
rest = (struct hblk *)((word)h + bytes);
rest_hdr = GC_install_header(rest);
if (0 == rest_hdr) {
/* This may be very bad news ... */
WARN("Header allocation failed: Dropping block.\n", 0);
return(0);
}
rest_hdr -> hb_sz = total_size - bytes;
rest_hdr -> hb_flags = 0;
# ifdef GC_ASSERTIONS
/* Mark h not free, to avoid assertion about adjacent free blocks. */
hhdr -> hb_map = 0;
# endif
GC_add_to_fl(rest, rest_hdr);
return h;
}
/*
* H is a free block. N points at an address inside it.
* A new header for n has already been set up. Fix up h's header
* to reflect the fact that it is being split, move it to the
* appropriate free list.
* N replaces h in the original free list.
*
* Nhdr is not completely filled in, since it is about to allocated.
* It may in fact end up on the wrong free list for its size.
* (Hence adding it to a free list is silly. But this path is hopefully
* rare enough that it doesn't matter. The code is cleaner this way.)
*/
void GC_split_block(h, hhdr, n, nhdr, index)
struct hblk *h;
hdr * hhdr;
struct hblk *n;
hdr * nhdr;
int index; /* Index of free list */
{
word total_size = hhdr -> hb_sz;
word h_size = (word)n - (word)h;
struct hblk *prev = hhdr -> hb_prev;
struct hblk *next = hhdr -> hb_next;
/* Replace h with n on its freelist */
nhdr -> hb_prev = prev;
nhdr -> hb_next = next;
nhdr -> hb_sz = total_size - h_size;
nhdr -> hb_flags = 0;
if (0 != prev) {
HDR(prev) -> hb_next = n;
} else {
GC_hblkfreelist[index] = n;
}
if (0 != next) {
HDR(next) -> hb_prev = n;
}
INCR_FREE_BYTES(index, -(signed_word)h_size);
FREE_ASSERT(GC_free_bytes[index] > 0);
# ifdef GC_ASSERTIONS
nhdr -> hb_map = 0; /* Don't fail test for consecutive */
/* free blocks in GC_add_to_fl. */
# endif
# ifdef USE_MUNMAP
hhdr -> hb_last_reclaimed = GC_gc_no;
# endif
hhdr -> hb_sz = h_size;
GC_add_to_fl(h, hhdr);
GC_invalidate_map(nhdr);
}
struct hblk * GC_allochblk_nth();
/*
* Allocate (and return pointer to) a heap block
* for objects of size sz words, searching the nth free list.
*
* NOTE: We set obj_map field in header correctly.
* Caller is responsible for building an object freelist in block.
*
* Unlike older versions of the collectors, the client is responsible
* for clearing the block, if necessary.
*/
struct hblk *
GC_allochblk(sz, kind, flags)
word sz;
int kind;
unsigned flags; /* IGNORE_OFF_PAGE or 0 */
{
word blocks = OBJ_SZ_TO_BLOCKS(sz);
int start_list = GC_hblk_fl_from_blocks(blocks);
int i;
for (i = start_list; i <= N_HBLK_FLS; ++i) {
struct hblk * result = GC_allochblk_nth(sz, kind, flags, i);
if (0 != result) {
return result;
}
}
return 0;
}
/*
* The same, but with search restricted to nth free list.
*/
struct hblk *
GC_allochblk_nth(sz, kind, flags, n)
word sz;
int kind;
unsigned char flags; /* IGNORE_OFF_PAGE or 0 */
int n;
{
register struct hblk *hbp;
register hdr * hhdr; /* Header corr. to hbp */
register struct hblk *thishbp;
register hdr * thishdr; /* Header corr. to hbp */
signed_word size_needed; /* number of bytes in requested objects */
signed_word size_avail; /* bytes available in this block */
size_needed = HBLKSIZE * OBJ_SZ_TO_BLOCKS(sz);
/* search for a big enough block in free list */
hbp = GC_hblkfreelist[n];
for(; 0 != hbp; hbp = hhdr -> hb_next) {
GET_HDR(hbp, hhdr);
size_avail = hhdr->hb_sz;
if (size_avail < size_needed) continue;
if (size_avail != size_needed
&& !GC_use_entire_heap
&& !GC_dont_gc
&& USED_HEAP_SIZE >= GC_requested_heapsize
&& !TRUE_INCREMENTAL && GC_should_collect()) {
# ifdef USE_MUNMAP
continue;
# else
/* If we have enough large blocks left to cover any */
/* previous request for large blocks, we go ahead */
/* and split. Assuming a steady state, that should */
/* be safe. It means that we can use the full */
/* heap if we allocate only small objects. */
if (!GC_enough_large_bytes_left(GC_large_allocd_bytes, n)) {
continue;
}
/* If we are deallocating lots of memory from */
/* finalizers, fail and collect sooner rather */
/* than later. */
if (WORDS_TO_BYTES(GC_finalizer_mem_freed)
> (GC_heapsize >> 4)) {
continue;
}
# endif /* !USE_MUNMAP */
}
/* If the next heap block is obviously better, go on. */
/* This prevents us from disassembling a single large block */
/* to get tiny blocks. */
{
signed_word next_size;
thishbp = hhdr -> hb_next;
if (thishbp != 0) {
GET_HDR(thishbp, thishdr);
next_size = (signed_word)(thishdr -> hb_sz);
if (next_size < size_avail
&& next_size >= size_needed
&& !GC_is_black_listed(thishbp, (word)size_needed)) {
continue;
}
}
}
if ( !IS_UNCOLLECTABLE(kind) &&
(kind != PTRFREE || size_needed > MAX_BLACK_LIST_ALLOC)) {
struct hblk * lasthbp = hbp;
ptr_t search_end = (ptr_t)hbp + size_avail - size_needed;
signed_word orig_avail = size_avail;
signed_word eff_size_needed = ((flags & IGNORE_OFF_PAGE)?
HBLKSIZE
: size_needed);
while ((ptr_t)lasthbp <= search_end
&& (thishbp = GC_is_black_listed(lasthbp,
(word)eff_size_needed))
!= 0) {
lasthbp = thishbp;
}
size_avail -= (ptr_t)lasthbp - (ptr_t)hbp;
thishbp = lasthbp;
if (size_avail >= size_needed) {
if (thishbp != hbp &&
0 != (thishdr = GC_install_header(thishbp))) {
/* Make sure it's mapped before we mangle it. */
# ifdef USE_MUNMAP
if (!IS_MAPPED(hhdr)) {
GC_remap((ptr_t)hbp, hhdr -> hb_sz);
hhdr -> hb_flags &= ~WAS_UNMAPPED;
}
# endif
/* Split the block at thishbp */
GC_split_block(hbp, hhdr, thishbp, thishdr, n);
/* Advance to thishbp */
hbp = thishbp;
hhdr = thishdr;
/* We must now allocate thishbp, since it may */
/* be on the wrong free list. */
}
} else if (size_needed > (signed_word)BL_LIMIT
&& orig_avail - size_needed
> (signed_word)BL_LIMIT) {
/* Punt, since anything else risks unreasonable heap growth. */
if (++GC_large_alloc_warn_suppressed
>= GC_large_alloc_warn_interval) {
WARN("Repeated allocation of very large block "
"(appr. size %ld):\n"
"\tMay lead to memory leak and poor performance.\n",
size_needed);
GC_large_alloc_warn_suppressed = 0;
}
size_avail = orig_avail;
} else if (size_avail == 0 && size_needed == HBLKSIZE
&& IS_MAPPED(hhdr)) {
if (!GC_find_leak) {
static unsigned count = 0;
/* The block is completely blacklisted. We need */
/* to drop some such blocks, since otherwise we spend */
/* all our time traversing them if pointerfree */
/* blocks are unpopular. */
/* A dropped block will be reconsidered at next GC. */
if ((++count & 3) == 0) {
/* Allocate and drop the block in small chunks, to */
/* maximize the chance that we will recover some */
/* later. */
word total_size = hhdr -> hb_sz;
struct hblk * limit = hbp + divHBLKSZ(total_size);
struct hblk * h;
struct hblk * prev = hhdr -> hb_prev;
GC_words_wasted += BYTES_TO_WORDS(total_size);
GC_large_free_bytes -= total_size;
GC_remove_from_fl(hhdr, n);
for (h = hbp; h < limit; h++) {
if (h == hbp || 0 != (hhdr = GC_install_header(h))) {
(void) setup_header(
hhdr,
BYTES_TO_WORDS(HBLKSIZE),
PTRFREE, 0); /* Cant fail */
if (GC_debugging_started) {
BZERO(h, HBLKSIZE);
}
}
}
/* Restore hbp to point at free block */
hbp = prev;
if (0 == hbp) {
return GC_allochblk_nth(sz, kind, flags, n);
}
hhdr = HDR(hbp);
}
}
}
}
if( size_avail >= size_needed ) {
# ifdef USE_MUNMAP
if (!IS_MAPPED(hhdr)) {
GC_remap((ptr_t)hbp, hhdr -> hb_sz);
hhdr -> hb_flags &= ~WAS_UNMAPPED;
}
# endif
/* hbp may be on the wrong freelist; the parameter n */
/* is important. */
hbp = GC_get_first_part(hbp, hhdr, size_needed, n);
break;
}
}
if (0 == hbp) return 0;
/* Add it to map of valid blocks */
if (!GC_install_counts(hbp, (word)size_needed)) return(0);
/* This leaks memory under very rare conditions. */
/* Set up header */
if (!setup_header(hhdr, sz, kind, flags)) {
GC_remove_counts(hbp, (word)size_needed);
return(0); /* ditto */
}
/* Notify virtual dirty bit implementation that we are about to write. */
/* Ensure that pointerfree objects are not protected if it's avoidable. */
GC_remove_protection(hbp, divHBLKSZ(size_needed),
(hhdr -> hb_descr == 0) /* pointer-free */);
/* We just successfully allocated a block. Restart count of */
/* consecutive failures. */
{
extern unsigned GC_fail_count;
GC_fail_count = 0;
}
GC_large_free_bytes -= size_needed;
GC_ASSERT(IS_MAPPED(hhdr));
return( hbp );
}
struct hblk * GC_freehblk_ptr = 0; /* Search position hint for GC_freehblk */
/*
* Free a heap block.
*
* Coalesce the block with its neighbors if possible.
*
* All mark words are assumed to be cleared.
*/
void
GC_freehblk(hbp)
struct hblk *hbp;
{
struct hblk *next, *prev;
hdr *hhdr, *prevhdr, *nexthdr;
signed_word size;
GET_HDR(hbp, hhdr);
size = hhdr->hb_sz;
size = HBLKSIZE * OBJ_SZ_TO_BLOCKS(size);
GC_remove_counts(hbp, (word)size);
hhdr->hb_sz = size;
# ifdef USE_MUNMAP
hhdr -> hb_last_reclaimed = GC_gc_no;
# endif
/* Check for duplicate deallocation in the easy case */
if (HBLK_IS_FREE(hhdr)) {
GC_printf1("Duplicate large block deallocation of 0x%lx\n",
(unsigned long) hbp);
ABORT("Duplicate large block deallocation");
}
GC_ASSERT(IS_MAPPED(hhdr));
GC_invalidate_map(hhdr);
next = (struct hblk *)((word)hbp + size);
GET_HDR(next, nexthdr);
prev = GC_free_block_ending_at(hbp);
/* Coalesce with successor, if possible */
if(0 != nexthdr && HBLK_IS_FREE(nexthdr) && IS_MAPPED(nexthdr)) {
GC_remove_from_fl(nexthdr, FL_UNKNOWN);
hhdr -> hb_sz += nexthdr -> hb_sz;
GC_remove_header(next);
}
/* Coalesce with predecessor, if possible. */
if (0 != prev) {
prevhdr = HDR(prev);
if (IS_MAPPED(prevhdr)) {
GC_remove_from_fl(prevhdr, FL_UNKNOWN);
prevhdr -> hb_sz += hhdr -> hb_sz;
# ifdef USE_MUNMAP
prevhdr -> hb_last_reclaimed = GC_gc_no;
# endif
GC_remove_header(hbp);
hbp = prev;
hhdr = prevhdr;
}
}
/* FIXME: It is not clear we really always want to do these merges */
/* with -DUSE_MUNMAP, since it updates ages and hence prevents */
/* unmapping. */
GC_large_free_bytes += size;
GC_add_to_fl(hbp, hhdr);
}
-1145
View File
File diff suppressed because it is too large. Load diff
-86
View File
@@ -1,86 +0,0 @@
.arch ev6
.text
.align 4
.globl GC_push_regs
.ent GC_push_regs 2
GC_push_regs:
ldgp $gp, 0($27)
lda $sp, -16($sp)
stq $26, 0($sp)
.mask 0x04000000, 0
.frame $sp, 16, $26, 0
/* $0 integer result */
/* $1-$8 temp regs - not preserved cross calls */
/* $9-$15 call saved regs */
/* $16-$21 argument regs - not preserved cross calls */
/* $22-$28 temp regs - not preserved cross calls */
/* $29 global pointer - not preserved cross calls */
/* $30 stack pointer */
# define call_push(x) \
mov x, $16; \
jsr $26, GC_push_one; \
ldgp $gp, 0($26)
call_push($9)
call_push($10)
call_push($11)
call_push($12)
call_push($13)
call_push($14)
call_push($15)
/* $f0-$f1 floating point results */
/* $f2-$f9 call saved regs */
/* $f10-$f30 temp regs - not preserved cross calls */
/* Use the most efficient transfer method for this hardware. */
/* Bit 1 detects the FIX extension, which includes ftoit. */
amask 2, $0
bne $0, $use_stack
#undef call_push
#define call_push(x) \
ftoit x, $16; \
jsr $26, GC_push_one; \
ldgp $gp, 0($26)
call_push($f2)
call_push($f3)
call_push($f4)
call_push($f5)
call_push($f6)
call_push($f7)
call_push($f8)
call_push($f9)
ldq $26, 0($sp)
lda $sp, 16($sp)
ret $31, ($26), 1
.align 4
$use_stack:
#undef call_push
#define call_push(x) \
stt x, 8($sp); \
ldq $16, 8($sp); \
jsr $26, GC_push_one; \
ldgp $gp, 0($26)
call_push($f2)
call_push($f3)
call_push($f4)
call_push($f5)
call_push($f6)
call_push($f7)
call_push($f8)
call_push($f9)
ldq $26, 0($sp)
lda $sp, 16($sp)
ret $31, ($26), 1
.end GC_push_regs
-123
View File
@@ -1,123 +0,0 @@
#!/bin/sh
# Run this to generate all the initial makefiles, etc.
# Ripped off from GNOME macros version
DIE=0
srcdir=`dirname $0`
test -z "$srcdir" && srcdir=.
(autoconf --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`autoconf' installed to compile Mono."
echo "Download the appropriate package for your distribution,"
echo "or get the source tarball at ftp://ftp.gnu.org/pub/gnu/"
DIE=1
}
if [ -z "$LIBTOOLIZE" ]; then
LIBTOOLIZE=`which glibtoolize 2>/dev/null`
if [ ! -x "$LIBTOOLIZE" ]; then
LIBTOOLIZE=`which libtoolize`
fi
fi
(grep "^AM_PROG_LIBTOOL" $srcdir/configure.ac >/dev/null) && {
($LIBTOOLIZE --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`libtoolize' installed to compile Mono."
echo "Get ftp://ftp.gnu.org/gnu/libtool/libtool-1.2.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
}
grep "^AM_GNU_GETTEXT" $srcdir/configure.ac >/dev/null && {
grep "sed.*POTFILES" $srcdir/configure.ac >/dev/null || \
(gettext --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`gettext' installed to compile Mono."
echo "Get ftp://alpha.gnu.org/gnu/gettext-0.10.35.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
}
(automake --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`automake' installed to compile Mono."
echo "Get ftp://ftp.gnu.org/pub/gnu/automake-1.3.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
NO_AUTOMAKE=yes
}
# if no automake, don't bother testing for aclocal
test -n "$NO_AUTOMAKE" || (aclocal --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: Missing \`aclocal'. The version of \`automake'"
echo "installed doesn't appear recent enough."
echo "Get ftp://ftp.gnu.org/pub/gnu/automake-1.3.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
if test "$DIE" -eq 1; then
exit 1
fi
if test x$NOCONFIGURE = x && test -z "$*"; then
echo "**Warning**: I am going to run \`configure' with no arguments."
echo "If you wish to pass any to it, please specify them on the"
echo \`$0\'" command line."
echo
fi
am_opt="--add-missing --copy --gnu -Wno-obsolete"
case $CC in
xlc )
am_opt="$am_opt --include-deps";;
esac
if grep "^AC_PROG_LIBTOOL" configure.ac >/dev/null; then
if test -z "$NO_LIBTOOLIZE" ; then
echo "Running libtoolize..."
$LIBTOOLIZE --force --copy
fi
fi
ACLOCAL_FLAGS="$ACLOCAL_FLAGS -Wnone"
echo "Running aclocal $ACLOCAL_FLAGS ..."
aclocal $ACLOCAL_FLAGS || {
echo
echo "**Error**: aclocal failed. This may mean that you have not"
echo "installed all of the packages you need, or you may need to"
echo "set ACLOCAL_FLAGS to include \"-I \$prefix/share/aclocal\""
echo "for the prefix where you installed the packages whose"
echo "macros were not found"
exit 1
}
if grep "^AC_CONFIG_HEADERS" configure.ac >/dev/null; then
echo "Running autoheader..."
autoheader || { echo "**Error**: autoheader failed."; exit 1; }
fi
echo "Running automake $am_opt ..."
automake $am_opt ||
{ echo "**Error**: automake failed."; exit 1; }
echo "Running autoconf ..."
autoconf || { echo "**Error**: autoconf failed."; exit 1; }
conf_flags="--enable-maintainer-mode --enable-compile-warnings" #--enable-iso-c
if test x$NOCONFIGURE = x; then
echo Running $srcdir/configure $conf_flags "$@" ...
$srcdir/configure $conf_flags "$@" \
&& echo Now type \`make\' to compile $PKG_NAME || exit 1
else
echo Skipping configure process.
fi
-467
View File
@@ -1,467 +0,0 @@
/*
* Copyright (c) 2001 by Hewlett-Packard Company. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*
*/
/*
* This implements a full, though not well-tuned, representation of the
* backwards points-to graph. This is used to test for non-GC-robust
* data structures; the code is not used during normal garbage collection.
*
* One restriction is that we drop all back-edges from nodes with very
* high in-degree, and simply add them add them to a list of such
* nodes. They are then treated as permanent roots. Id this by itself
* doesn't introduce a space leak, then such nodes can't contribute to
* a growing space leak.
*/
#ifdef MAKE_BACK_GRAPH
#define MAX_IN 10 /* Maximum in-degree we handle directly */
#include "private/dbg_mlc.h"
#include <unistd.h>
#if !defined(DBG_HDRS_ALL) || (ALIGNMENT != CPP_WORDSZ/8) || !defined(UNIX_LIKE)
# error Configuration doesnt support MAKE_BACK_GRAPH
#endif
/* We store single back pointers directly in the object's oh_bg_ptr field. */
/* If there is more than one ptr to an object, we store q | FLAG_MANY, */
/* where q is a pointer to a back_edges object. */
/* Every once in a while we use a back_edges object even for a single */
/* pointer, since we need the other fields in the back_edges structure to */
/* be present in some fraction of the objects. Otherwise we get serious */
/* performance issues. */
#define FLAG_MANY 2
typedef struct back_edges_struct {
word n_edges; /* Number of edges, including those in continuation */
/* structures. */
unsigned short flags;
# define RETAIN 1 /* Directly points to a reachable object; */
/* retain for next GC. */
unsigned short height_gc_no;
/* If height > 0, then the GC_gc_no value when it */
/* was computed. If it was computed this cycle, then */
/* it is current. If it was computed during the */
/* last cycle, then it represents the old height, */
/* which is only saved for live objects referenced by */
/* dead ones. This may grow due to refs from newly */
/* dead objects. */
signed_word height;
/* Longest path through unreachable nodes to this node */
/* that we found using depth first search. */
# define HEIGHT_UNKNOWN ((signed_word)(-2))
# define HEIGHT_IN_PROGRESS ((signed_word)(-1))
ptr_t edges[MAX_IN];
struct back_edges_struct *cont;
/* Pointer to continuation structure; we use only the */
/* edges field in the continuation. */
/* also used as free list link. */
} back_edges;
/* Allocate a new back edge structure. Should be more sophisticated */
/* if this were production code. */
#define MAX_BACK_EDGE_STRUCTS 100000
static back_edges *back_edge_space = 0;
int GC_n_back_edge_structs = 0; /* Serves as pointer to never used */
/* back_edges space. */
static back_edges *avail_back_edges = 0;
/* Pointer to free list of deallocated */
/* back_edges structures. */
static back_edges * new_back_edges(void)
{
if (0 == back_edge_space) {
back_edge_space = (back_edges *)
GET_MEM(MAX_BACK_EDGE_STRUCTS*sizeof(back_edges));
}
if (0 != avail_back_edges) {
back_edges * result = avail_back_edges;
avail_back_edges = result -> cont;
result -> cont = 0;
return result;
}
if (GC_n_back_edge_structs >= MAX_BACK_EDGE_STRUCTS - 1) {
ABORT("needed too much space for back edges: adjust "
"MAX_BACK_EDGE_STRUCTS");
}
return back_edge_space + (GC_n_back_edge_structs++);
}
/* Deallocate p and its associated continuation structures. */
static void deallocate_back_edges(back_edges *p)
{
back_edges *last = p;
while (0 != last -> cont) last = last -> cont;
last -> cont = avail_back_edges;
avail_back_edges = p;
}
/* Table of objects that are currently on the depth-first search */
/* stack. Only objects with in-degree one are in this table. */
/* Other objects are identified using HEIGHT_IN_PROGRESS. */
/* FIXME: This data structure NEEDS IMPROVEMENT. */
#define INITIAL_IN_PROGRESS 10000
static ptr_t * in_progress_space = 0;
static size_t in_progress_size = 0;
static size_t n_in_progress = 0;
static void push_in_progress(ptr_t p)
{
if (n_in_progress >= in_progress_size)
if (in_progress_size == 0) {
in_progress_size = INITIAL_IN_PROGRESS;
in_progress_space = (ptr_t *)GET_MEM(in_progress_size * sizeof(ptr_t));
} else {
ptr_t * new_in_progress_space;
in_progress_size *= 2;
new_in_progress_space = (ptr_t *)
GET_MEM(in_progress_size * sizeof(ptr_t));
BCOPY(in_progress_space, new_in_progress_space,
n_in_progress * sizeof(ptr_t));
in_progress_space = new_in_progress_space;
/* FIXME: This just drops the old space. */
}
if (in_progress_space == 0)
ABORT("MAKE_BACK_GRAPH: Out of in-progress space: "
"Huge linear data structure?");
in_progress_space[n_in_progress++] = p;
}
static GC_bool is_in_progress(ptr_t p)
{
int i;
for (i = 0; i < n_in_progress; ++i) {
if (in_progress_space[i] == p) return TRUE;
}
return FALSE;
}
static void pop_in_progress(ptr_t p)
{
--n_in_progress;
GC_ASSERT(in_progress_space[n_in_progress] == p);
}
#define GET_OH_BG_PTR(p) \
(ptr_t)REVEAL_POINTER(((oh *)(p)) -> oh_bg_ptr)
#define SET_OH_BG_PTR(p,q) (((oh *)(p)) -> oh_bg_ptr) = HIDE_POINTER(q)
/* Execute s once for each predecessor q of p in the points-to graph. */
/* s should be a bracketed statement. We declare q. */
#define FOR_EACH_PRED(q, p, s) \
{ \
ptr_t q = GET_OH_BG_PTR(p); \
if (!((word)q & FLAG_MANY)) { \
if (q && !((word)q & 1)) s \
/* !((word)q & 1) checks for a misnterpreted freelist link */ \
} else { \
back_edges *orig_be_ = (back_edges *)((word)q & ~FLAG_MANY); \
back_edges *be_ = orig_be_; \
int total_, local_; \
int n_edges_ = be_ -> n_edges; \
for (total_ = 0, local_ = 0; total_ < n_edges_; ++local_, ++total_) { \
if (local_ == MAX_IN) { \
be_ = be_ -> cont; \
local_ = 0; \
} \
q = be_ -> edges[local_]; s \
} \
} \
}
/* Ensure that p has a back_edges structure associated with it. */
static void ensure_struct(ptr_t p)
{
ptr_t old_back_ptr = GET_OH_BG_PTR(p);
if (!((word)old_back_ptr & FLAG_MANY)) {
back_edges *be = new_back_edges();
be -> flags = 0;
if (0 == old_back_ptr) {
be -> n_edges = 0;
} else {
be -> n_edges = 1;
be -> edges[0] = old_back_ptr;
}
be -> height = HEIGHT_UNKNOWN;
be -> height_gc_no = GC_gc_no - 1;
GC_ASSERT(be >= back_edge_space);
SET_OH_BG_PTR(p, (word)be | FLAG_MANY);
}
}
/* Add the (forward) edge from p to q to the backward graph. Both p */
/* q are pointers to the object base, i.e. pointers to an oh. */
static void add_edge(ptr_t p, ptr_t q)
{
ptr_t old_back_ptr = GET_OH_BG_PTR(q);
back_edges * be, *be_cont;
word i;
static unsigned random_number = 13;
# define GOT_LUCKY_NUMBER (((++random_number) & 0x7f) == 0)
/* A not very random number we use to occasionally allocate a */
/* back_edges structure even for a single backward edge. This */
/* prevents us from repeatedly tracing back through very long */
/* chains, since we will have some place to store height and */
/* in_progress flags along the way. */
GC_ASSERT(p == GC_base(p) && q == GC_base(q));
if (!GC_HAS_DEBUG_INFO(q) || !GC_HAS_DEBUG_INFO(p)) {
/* This is really a misinterpreted free list link, since we saw */
/* a pointer to a free list. Dont overwrite it! */
return;
}
if (0 == old_back_ptr) {
SET_OH_BG_PTR(q, p);
if (GOT_LUCKY_NUMBER) ensure_struct(q);
return;
}
/* Check whether it was already in the list of predecessors. */
FOR_EACH_PRED(pred, q, { if (p == pred) return; });
ensure_struct(q);
old_back_ptr = GET_OH_BG_PTR(q);
be = (back_edges *)((word)old_back_ptr & ~FLAG_MANY);
for (i = be -> n_edges, be_cont = be; i > MAX_IN;
be_cont = be_cont -> cont, i -= MAX_IN) {}
if (i == MAX_IN) {
be_cont -> cont = new_back_edges();
be_cont = be_cont -> cont;
i = 0;
}
be_cont -> edges[i] = p;
be -> n_edges++;
if (be -> n_edges == 100) {
# if 0
if (GC_print_stats) {
GC_err_printf0("The following object has in-degree >= 100:\n");
GC_print_heap_obj(q);
}
# endif
}
}
typedef void (*per_object_func)(ptr_t p, word n_words, word gc_descr);
static void per_object_helper(struct hblk *h, word fn)
{
hdr * hhdr = HDR(h);
word sz = hhdr -> hb_sz;
word descr = hhdr -> hb_descr;
per_object_func f = (per_object_func)fn;
int i = 0;
do {
f((ptr_t)(h -> hb_body + i), sz, descr);
i += sz;
} while (i + sz <= BYTES_TO_WORDS(HBLKSIZE));
}
void GC_apply_to_each_object(per_object_func f)
{
GC_apply_to_all_blocks(per_object_helper, (word)f);
}
static void reset_back_edge(ptr_t p, word n_words, word gc_descr)
{
/* Skip any free list links, or dropped blocks */
if (GC_HAS_DEBUG_INFO(p)) {
ptr_t old_back_ptr = GET_OH_BG_PTR(p);
if ((word)old_back_ptr & FLAG_MANY) {
back_edges *be = (back_edges *)((word)old_back_ptr & ~FLAG_MANY);
if (!(be -> flags & RETAIN)) {
deallocate_back_edges(be);
SET_OH_BG_PTR(p, 0);
} else {
word *currentp;
GC_ASSERT(GC_is_marked(p));
/* Back edges may point to objects that will not be retained. */
/* Delete them for now, but remember the height. */
/* Some will be added back at next GC. */
be -> n_edges = 0;
if (0 != be -> cont) {
deallocate_back_edges(be -> cont);
be -> cont = 0;
}
GC_ASSERT(GC_is_marked(p));
/* We only retain things for one GC cycle at a time. */
be -> flags &= ~RETAIN;
}
} else /* Simple back pointer */ {
/* Clear to avoid dangling pointer. */
SET_OH_BG_PTR(p, 0);
}
}
}
static void add_back_edges(ptr_t p, word n_words, word gc_descr)
{
word *currentp = (word *)(p + sizeof(oh));
/* For now, fix up non-length descriptors conservatively. */
if((gc_descr & GC_DS_TAGS) != GC_DS_LENGTH) {
gc_descr = WORDS_TO_BYTES(n_words);
}
while (currentp < (word *)(p + gc_descr)) {
word current = *currentp++;
FIXUP_POINTER(current);
if (current >= (word)GC_least_plausible_heap_addr &&
current <= (word)GC_greatest_plausible_heap_addr) {
ptr_t target = GC_base((GC_PTR)current);
if (0 != target) {
add_edge(p, target);
}
}
}
}
/* Rebuild the representation of the backward reachability graph. */
/* Does not examine mark bits. Can be called before GC. */
void GC_build_back_graph(void)
{
GC_apply_to_each_object(add_back_edges);
}
/* Return an approximation to the length of the longest simple path */
/* through unreachable objects to p. We refer to this as the height */
/* of p. */
static word backwards_height(ptr_t p)
{
word result;
ptr_t back_ptr = GET_OH_BG_PTR(p);
back_edges *be;
if (0 == back_ptr) return 1;
if (!((word)back_ptr & FLAG_MANY)) {
if (is_in_progress(p)) return 0; /* DFS back edge, i.e. we followed */
/* an edge to an object already */
/* on our stack: ignore */
push_in_progress(p);
result = backwards_height(back_ptr)+1;
pop_in_progress(p);
return result;
}
be = (back_edges *)((word)back_ptr & ~FLAG_MANY);
if (be -> height >= 0 && be -> height_gc_no == GC_gc_no)
return be -> height;
/* Ignore back edges in DFS */
if (be -> height == HEIGHT_IN_PROGRESS) return 0;
result = (be -> height > 0? be -> height : 1);
be -> height = HEIGHT_IN_PROGRESS;
FOR_EACH_PRED(q, p, {
word this_height;
if (GC_is_marked(q) && !(FLAG_MANY & (word)GET_OH_BG_PTR(p))) {
if (GC_print_stats)
GC_printf2("Found bogus pointer from 0x%lx to 0x%lx\n", q, p);
/* Reachable object "points to" unreachable one. */
/* Could be caused by our lax treatment of GC descriptors. */
this_height = 1;
} else {
this_height = backwards_height(q);
}
if (this_height >= result) result = this_height + 1;
});
be -> height = result;
be -> height_gc_no = GC_gc_no;
return result;
}
word GC_max_height;
ptr_t GC_deepest_obj;
/* Compute the maximum height of every unreachable predecessor p of a */
/* reachable object. Arrange to save the heights of all such objects p */
/* so that they can be used in calculating the height of objects in the */
/* next GC. */
/* Set GC_max_height to be the maximum height we encounter, and */
/* GC_deepest_obj to be the corresponding object. */
static void update_max_height(ptr_t p, word n_words, word gc_descr)
{
if (GC_is_marked(p) && GC_HAS_DEBUG_INFO(p)) {
int i;
word p_height = 0;
ptr_t p_deepest_obj = 0;
ptr_t back_ptr;
back_edges *be = 0;
/* If we remembered a height last time, use it as a minimum. */
/* It may have increased due to newly unreachable chains pointing */
/* to p, but it can't have decreased. */
back_ptr = GET_OH_BG_PTR(p);
if (0 != back_ptr && ((word)back_ptr & FLAG_MANY)) {
be = (back_edges *)((word)back_ptr & ~FLAG_MANY);
if (be -> height != HEIGHT_UNKNOWN) p_height = be -> height;
}
FOR_EACH_PRED(q, p, {
if (!GC_is_marked(q) && GC_HAS_DEBUG_INFO(q)) {
word q_height;
q_height = backwards_height(q);
if (q_height > p_height) {
p_height = q_height;
p_deepest_obj = q;
}
}
});
if (p_height > 0) {
/* Remember the height for next time. */
if (be == 0) {
ensure_struct(p);
back_ptr = GET_OH_BG_PTR(p);
be = (back_edges *)((word)back_ptr & ~FLAG_MANY);
}
be -> flags |= RETAIN;
be -> height = p_height;
be -> height_gc_no = GC_gc_no;
}
if (p_height > GC_max_height) {
GC_max_height = p_height;
GC_deepest_obj = p_deepest_obj;
}
}
}
word GC_max_max_height = 0;
void GC_traverse_back_graph(void)
{
GC_max_height = 0;
GC_apply_to_each_object(update_max_height);
}
void GC_print_back_graph_stats(void)
{
GC_printf2("Maximum backwards height of reachable objects at GC %lu is %ld\n",
(unsigned long) GC_gc_no, GC_max_height);
if (GC_max_height > GC_max_max_height) {
GC_max_max_height = GC_max_height;
GC_printf0("The following unreachable object is last in a longest chain "
"of unreachable objects:\n");
GC_print_heap_obj(GC_deepest_obj);
}
if (GC_print_stats) {
GC_printf1("Needed max total of %ld back-edge structs\n",
GC_n_back_edge_structs);
}
GC_apply_to_each_object(reset_back_edge);
GC_deepest_obj = 0;
}
#endif /* MAKE_BACK_GRAPH */
-300
View File
@@ -1,300 +0,0 @@
/*
* Copyright 1988, 1989 Hans-J. Boehm, Alan J. Demers
* Copyright (c) 1991-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, August 9, 1995 6:09 pm PDT */
# include "private/gc_priv.h"
/*
* We maintain several hash tables of hblks that have had false hits.
* Each contains one bit per hash bucket; If any page in the bucket
* has had a false hit, we assume that all of them have.
* See the definition of page_hash_table in gc_private.h.
* False hits from the stack(s) are much more dangerous than false hits
* from elsewhere, since the former can pin a large object that spans the
* block, even though it does not start on the dangerous block.
*/
/*
* Externally callable routines are:
* GC_add_to_black_list_normal
* GC_add_to_black_list_stack
* GC_promote_black_lists
* GC_is_black_listed
*
* All require that the allocator lock is held.
*/
/* Pointers to individual tables. We replace one table by another by */
/* switching these pointers. */
word * GC_old_normal_bl;
/* Nonstack false references seen at last full */
/* collection. */
word * GC_incomplete_normal_bl;
/* Nonstack false references seen since last */
/* full collection. */
word * GC_old_stack_bl;
word * GC_incomplete_stack_bl;
word GC_total_stack_black_listed;
word GC_black_list_spacing = MINHINCR*HBLKSIZE; /* Initial rough guess */
void GC_clear_bl();
# if defined(__STDC__) || defined(__cplusplus)
void GC_default_print_heap_obj_proc(ptr_t p)
# else
void GC_default_print_heap_obj_proc(p)
ptr_t p;
# endif
{
ptr_t base = GC_base(p);
GC_err_printf2("start: 0x%lx, appr. length: %ld", base, GC_size(base));
}
void (*GC_print_heap_obj) GC_PROTO((ptr_t p)) =
GC_default_print_heap_obj_proc;
void GC_print_source_ptr(p)
ptr_t p;
{
ptr_t base = GC_base(p);
if (0 == base) {
if (0 == p) {
GC_err_printf0("in register");
} else {
GC_err_printf0("in root set");
}
} else {
GC_err_printf0("in object at ");
(*GC_print_heap_obj)(base);
}
}
void GC_bl_init()
{
if (!GC_all_interior_pointers) {
GC_old_normal_bl = (word *)
GC_scratch_alloc((word)(sizeof (page_hash_table)));
GC_incomplete_normal_bl = (word *)GC_scratch_alloc
((word)(sizeof(page_hash_table)));
if (GC_old_normal_bl == 0 || GC_incomplete_normal_bl == 0) {
GC_err_printf0("Insufficient memory for black list\n");
EXIT();
}
GC_clear_bl(GC_old_normal_bl);
GC_clear_bl(GC_incomplete_normal_bl);
}
GC_old_stack_bl = (word *)GC_scratch_alloc((word)(sizeof(page_hash_table)));
GC_incomplete_stack_bl = (word *)GC_scratch_alloc
((word)(sizeof(page_hash_table)));
if (GC_old_stack_bl == 0 || GC_incomplete_stack_bl == 0) {
GC_err_printf0("Insufficient memory for black list\n");
EXIT();
}
GC_clear_bl(GC_old_stack_bl);
GC_clear_bl(GC_incomplete_stack_bl);
}
void GC_clear_bl(doomed)
word *doomed;
{
BZERO(doomed, sizeof(page_hash_table));
}
void GC_copy_bl(old, new)
word *new, *old;
{
BCOPY(old, new, sizeof(page_hash_table));
}
static word total_stack_black_listed();
/* Signal the completion of a collection. Turn the incomplete black */
/* lists into new black lists, etc. */
void GC_promote_black_lists()
{
word * very_old_normal_bl = GC_old_normal_bl;
word * very_old_stack_bl = GC_old_stack_bl;
GC_old_normal_bl = GC_incomplete_normal_bl;
GC_old_stack_bl = GC_incomplete_stack_bl;
if (!GC_all_interior_pointers) {
GC_clear_bl(very_old_normal_bl);
}
GC_clear_bl(very_old_stack_bl);
GC_incomplete_normal_bl = very_old_normal_bl;
GC_incomplete_stack_bl = very_old_stack_bl;
GC_total_stack_black_listed = total_stack_black_listed();
# ifdef PRINTSTATS
GC_printf1("%ld bytes in heap blacklisted for interior pointers\n",
(unsigned long)GC_total_stack_black_listed);
# endif
if (GC_total_stack_black_listed != 0) {
GC_black_list_spacing =
HBLKSIZE*(GC_heapsize/GC_total_stack_black_listed);
}
if (GC_black_list_spacing < 3 * HBLKSIZE) {
GC_black_list_spacing = 3 * HBLKSIZE;
}
if (GC_black_list_spacing > MAXHINCR * HBLKSIZE) {
GC_black_list_spacing = MAXHINCR * HBLKSIZE;
/* Makes it easier to allocate really huge blocks, which otherwise */
/* may have problems with nonuniform blacklist distributions. */
/* This way we should always succeed immediately after growing the */
/* heap. */
}
}
void GC_unpromote_black_lists()
{
if (!GC_all_interior_pointers) {
GC_copy_bl(GC_old_normal_bl, GC_incomplete_normal_bl);
}
GC_copy_bl(GC_old_stack_bl, GC_incomplete_stack_bl);
}
/* P is not a valid pointer reference, but it falls inside */
/* the plausible heap bounds. */
/* Add it to the normal incomplete black list if appropriate. */
#ifdef PRINT_BLACK_LIST
void GC_add_to_black_list_normal(p, source)
ptr_t source;
#else
void GC_add_to_black_list_normal(p)
#endif
word p;
{
if (!(GC_modws_valid_offsets[p & (sizeof(word)-1)])) return;
{
register int index = PHT_HASH(p);
if (HDR(p) == 0 || get_pht_entry_from_index(GC_old_normal_bl, index)) {
# ifdef PRINT_BLACK_LIST
if (!get_pht_entry_from_index(GC_incomplete_normal_bl, index)) {
GC_err_printf2(
"Black listing (normal) 0x%lx referenced from 0x%lx ",
(unsigned long) p, (unsigned long) source);
GC_print_source_ptr(source);
GC_err_puts("\n");
}
# endif
set_pht_entry_from_index(GC_incomplete_normal_bl, index);
} /* else this is probably just an interior pointer to an allocated */
/* object, and isn't worth black listing. */
}
}
/* And the same for false pointers from the stack. */
#ifdef PRINT_BLACK_LIST
void GC_add_to_black_list_stack(p, source)
ptr_t source;
#else
void GC_add_to_black_list_stack(p)
#endif
word p;
{
register int index = PHT_HASH(p);
if (HDR(p) == 0 || get_pht_entry_from_index(GC_old_stack_bl, index)) {
# ifdef PRINT_BLACK_LIST
if (!get_pht_entry_from_index(GC_incomplete_stack_bl, index)) {
GC_err_printf2(
"Black listing (stack) 0x%lx referenced from 0x%lx ",
(unsigned long)p, (unsigned long)source);
GC_print_source_ptr(source);
GC_err_puts("\n");
}
# endif
set_pht_entry_from_index(GC_incomplete_stack_bl, index);
}
}
/*
* Is the block starting at h of size len bytes black listed? If so,
* return the address of the next plausible r such that (r, len) might not
* be black listed. (R may not actually be in the heap. We guarantee only
* that every smaller value of r after h is also black listed.)
* If (h,len) is not black listed, return 0.
* Knows about the structure of the black list hash tables.
*/
struct hblk * GC_is_black_listed(h, len)
struct hblk * h;
word len;
{
register int index = PHT_HASH((word)h);
register word i;
word nblocks = divHBLKSZ(len);
if (!GC_all_interior_pointers) {
if (get_pht_entry_from_index(GC_old_normal_bl, index)
|| get_pht_entry_from_index(GC_incomplete_normal_bl, index)) {
return(h+1);
}
}
for (i = 0; ; ) {
if (GC_old_stack_bl[divWORDSZ(index)] == 0
&& GC_incomplete_stack_bl[divWORDSZ(index)] == 0) {
/* An easy case */
i += WORDSZ - modWORDSZ(index);
} else {
if (get_pht_entry_from_index(GC_old_stack_bl, index)
|| get_pht_entry_from_index(GC_incomplete_stack_bl, index)) {
return(h+i+1);
}
i++;
}
if (i >= nblocks) break;
index = PHT_HASH((word)(h+i));
}
return(0);
}
/* Return the number of blacklisted blocks in a given range. */
/* Used only for statistical purposes. */
/* Looks only at the GC_incomplete_stack_bl. */
word GC_number_stack_black_listed(start, endp1)
struct hblk *start, *endp1;
{
register struct hblk * h;
word result = 0;
for (h = start; h < endp1; h++) {
register int index = PHT_HASH((word)h);
if (get_pht_entry_from_index(GC_old_stack_bl, index)) result++;
}
return(result);
}
/* Return the total number of (stack) black-listed bytes. */
static word total_stack_black_listed()
{
register unsigned i;
word total = 0;
for (i = 0; i < GC_n_heap_sects; i++) {
struct hblk * start = (struct hblk *) GC_heap_sects[i].hs_start;
word len = (word) GC_heap_sects[i].hs_bytes;
struct hblk * endp1 = start + len/HBLKSIZE;
total += GC_number_stack_black_listed(start, endp1);
}
return(total * HBLKSIZE);
}
-4
View File
@@ -1,4 +0,0 @@
#!/bin/sh
GC_DEBUG=1
export GC_DEBUG
$* 2>&1 | awk '{print "0x3e=c\""$0"\""};/^\t##PC##=/ {if ($2 != 0) {print $2"?i"}}' | adb $1 | sed "s/^ >/>/"
-199
View File
@@ -1,199 +0,0 @@
/*
* Copyright (c) 1992-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, March 29, 1995 12:51 pm PST */
# ifdef CHECKSUMS
# include "private/gc_priv.h"
/* This is debugging code intended to verify the results of dirty bit */
/* computations. Works only in a single threaded environment. */
/* We assume that stubborn objects are changed only when they are */
/* enabled for writing. (Certain kinds of writing are actually */
/* safe under other conditions.) */
# define NSUMS 10000
# define OFFSET 0x10000
typedef struct {
GC_bool new_valid;
word old_sum;
word new_sum;
struct hblk * block; /* Block to which this refers + OFFSET */
/* to hide it from collector. */
} page_entry;
page_entry GC_sums [NSUMS];
word GC_checksum(h)
struct hblk *h;
{
register word *p = (word *)h;
register word *lim = (word *)(h+1);
register word result = 0;
while (p < lim) {
result += *p++;
}
return(result | 0x80000000 /* doesn't look like pointer */);
}
# ifdef STUBBORN_ALLOC
/* Check whether a stubborn object from the given block appears on */
/* the appropriate free list. */
GC_bool GC_on_free_list(h)
struct hblk *h;
{
register hdr * hhdr = HDR(h);
register int sz = hhdr -> hb_sz;
ptr_t p;
if (sz > MAXOBJSZ) return(FALSE);
for (p = GC_sobjfreelist[sz]; p != 0; p = obj_link(p)) {
if (HBLKPTR(p) == h) return(TRUE);
}
return(FALSE);
}
# endif
int GC_n_dirty_errors;
int GC_n_changed_errors;
int GC_n_clean;
int GC_n_dirty;
void GC_update_check_page(h, index)
struct hblk *h;
int index;
{
page_entry *pe = GC_sums + index;
register hdr * hhdr = HDR(h);
struct hblk *b;
if (pe -> block != 0 && pe -> block != h + OFFSET) ABORT("goofed");
pe -> old_sum = pe -> new_sum;
pe -> new_sum = GC_checksum(h);
# if !defined(MSWIN32) && !defined(MSWINCE)
if (pe -> new_sum != 0x80000000 && !GC_page_was_ever_dirty(h)) {
GC_printf1("GC_page_was_ever_dirty(0x%lx) is wrong\n",
(unsigned long)h);
}
# endif
if (GC_page_was_dirty(h)) {
GC_n_dirty++;
} else {
GC_n_clean++;
}
b = h;
while (IS_FORWARDING_ADDR_OR_NIL(hhdr) && hhdr != 0) {
b -= (word)hhdr;
hhdr = HDR(b);
}
if (pe -> new_valid
&& hhdr != 0 && hhdr -> hb_descr != 0 /* may contain pointers */
&& pe -> old_sum != pe -> new_sum) {
if (!GC_page_was_dirty(h) || !GC_page_was_ever_dirty(h)) {
/* Set breakpoint here */GC_n_dirty_errors++;
}
# ifdef STUBBORN_ALLOC
if ( hhdr -> hb_map != GC_invalid_map
&& hhdr -> hb_obj_kind == STUBBORN
&& !GC_page_was_changed(h)
&& !GC_on_free_list(h)) {
/* if GC_on_free_list(h) then reclaim may have touched it */
/* without any allocations taking place. */
/* Set breakpoint here */GC_n_changed_errors++;
}
# endif
}
pe -> new_valid = TRUE;
pe -> block = h + OFFSET;
}
word GC_bytes_in_used_blocks;
void GC_add_block(h, dummy)
struct hblk *h;
word dummy;
{
register hdr * hhdr = HDR(h);
register bytes = WORDS_TO_BYTES(hhdr -> hb_sz);
bytes += HBLKSIZE-1;
bytes &= ~(HBLKSIZE-1);
GC_bytes_in_used_blocks += bytes;
}
void GC_check_blocks()
{
word bytes_in_free_blocks = GC_large_free_bytes;
GC_bytes_in_used_blocks = 0;
GC_apply_to_all_blocks(GC_add_block, (word)0);
GC_printf2("GC_bytes_in_used_blocks = %ld, bytes_in_free_blocks = %ld ",
GC_bytes_in_used_blocks, bytes_in_free_blocks);
GC_printf1("GC_heapsize = %ld\n", GC_heapsize);
if (GC_bytes_in_used_blocks + bytes_in_free_blocks != GC_heapsize) {
GC_printf0("LOST SOME BLOCKS!!\n");
}
}
/* Should be called immediately after GC_read_dirty and GC_read_changed. */
void GC_check_dirty()
{
register int index;
register unsigned i;
register struct hblk *h;
register ptr_t start;
GC_check_blocks();
GC_n_dirty_errors = 0;
GC_n_changed_errors = 0;
GC_n_clean = 0;
GC_n_dirty = 0;
index = 0;
for (i = 0; i < GC_n_heap_sects; i++) {
start = GC_heap_sects[i].hs_start;
for (h = (struct hblk *)start;
h < (struct hblk *)(start + GC_heap_sects[i].hs_bytes);
h++) {
GC_update_check_page(h, index);
index++;
if (index >= NSUMS) goto out;
}
}
out:
GC_printf2("Checked %lu clean and %lu dirty pages\n",
(unsigned long) GC_n_clean, (unsigned long) GC_n_dirty);
if (GC_n_dirty_errors > 0) {
GC_printf1("Found %lu dirty bit errors\n",
(unsigned long)GC_n_dirty_errors);
}
if (GC_n_changed_errors > 0) {
GC_printf1("Found %lu changed bit errors\n",
(unsigned long)GC_n_changed_errors);
GC_printf0("These may be benign (provoked by nonpointer changes)\n");
# ifdef THREADS
GC_printf0(
"Also expect 1 per thread currently allocating a stubborn obj.\n");
# endif
}
}
# else
extern int GC_quiet;
/* ANSI C doesn't allow translation units to be empty. */
/* So we guarantee this one is nonempty. */
# endif /* CHECKSUMS */
-569
View File
@@ -1,569 +0,0 @@
# Copyright (c) 1999-2001 by Red Hat, Inc. All rights reserved.
#
# THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
# OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
#
# Permission is hereby granted to use or copy this program
# for any purpose, provided the above notices are retained on all copies.
# Permission to modify the code and to distribute modified code is granted,
# provided the above notices are retained, and a notice that the code was
# modified is included with the above copyright notice.
#
# Original author: Tom Tromey
# Modified by: Grzegorz Jakacki <jakacki at acm dot org>
dnl Process this file with autoconf to produce configure.
AC_PREREQ([2.53])
AC_INIT([libgc-mono], [6.6], [[email protected]])
AM_INIT_AUTOMAKE([1.9 no-define foreign])
AC_CONFIG_SRCDIR(gcj_mlc.c)
AC_CONFIG_MACRO_DIR([m4])
AC_CANONICAL_HOST
AC_SUBST(PACKAGE)
AC_SUBST(GC_VERSION)
AC_PROG_CC
AC_PROG_CXX
m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])])
dnl automake 1.6 and later need the AM_PROG_AS macro.
ifdef([AM_PROG_AS],[AM_PROG_AS],[])
AC_CHECK_TOOL(AR, ar)
AC_CHECK_TOOL(RANLIB, ranlib, :) # :)
AC_PROG_INSTALL
AM_MAINTAINER_MODE
. [$]{srcdir}/configure.host
# We use a separate variable to pass down CPPFLAGS and CFLAGS from the main mono
# configure, because of autoconf brokeness
if test "x$CPPFLAGS_FOR_LIBGC" != "x"; then
CPPFLAGS=$CPPFLAGS_FOR_LIBGC
fi
if test "x$CFLAGS_FOR_LIBGC" != "x"; then
CFLAGS=$CFLAGS_FOR_LIBGC
fi
GC_CFLAGS=${gc_cflags}
AC_SUBST(GC_CFLAGS)
case $enable_embed_check in
no) ;;
*) AC_MSG_ERROR([This module is now part of 'mono' and cannot be built as a stand-alone module any longer.]) ;;
esac
THREADS=$with_libgc_threads
AC_ARG_ENABLE(win32-dllmain,
[ --enable-win32-dllmain Define the DllMain function in win32_threads.c even if the collector is not built as a dll],
)
AC_ARG_ENABLE(parallel-mark,
[ --enable-parallel-mark parallelize marking and free list construction],
[case "$THREADS" in
no | none | single)
AC_MSG_ERROR([Parallel mark requires --enable-threads=x spec])
;;
esac]
)
AC_ARG_ENABLE(cplusplus,
[ --enable-cplusplus install C++ support],
)
AM_CPPFLAGS=-I${srcdir}/include
THREADDLLIBS=
## Libraries needed to support dynamic loading and/or threads.
case "$THREADS" in
no | none | single)
THREADS=none
;;
posix | pthreads)
THREADS=posix
AC_CHECK_LIB(pthread, pthread_self, THREADDLLIBS="-lpthread",,)
case "$host" in
x86-*-linux* | ia64-*-linux* | i386-*-linux* | i486-*-linux* | i586-*-linux* | i686-*-linux* | x86_64-*-linux* | alpha*-*-linux* | s390*-*-linux* | sparc*-*-linux* | powerpc-*-linux*)
AC_DEFINE(GC_LINUX_THREADS)
AC_DEFINE(_REENTRANT)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
;;
*-*-linux* | *-*-nacl*)
AC_DEFINE(GC_LINUX_THREADS)
AC_DEFINE(_REENTRANT)
;;
*-*-aix*)
AC_DEFINE(GC_AIX_THREADS)
AC_DEFINE(_REENTRANT)
;;
*-*-haiku*)
AC_DEFINE(GC_HAIKU_THREADS)
AC_DEFINE(_REENTRANT)
;;
*-*-hpux*)
AC_MSG_WARN("Only HP/UX 11 threads are supported.")
AC_DEFINE(GC_HPUX_THREADS)
AC_DEFINE(_POSIX_C_SOURCE,199506L)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
THREADDLLIBS="-lpthread -lrt"
;;
*-*-kfreebsd*-gnu)
AC_DEFINE(GC_FREEBSD_THREADS)
INCLUDES="$INCLUDES -pthread"
THREADDLLIBS=-pthread
AC_DEFINE(_REENTRANT)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
AC_DEFINE(USE_COMPILER_TLS)
;;
*-*-freebsd*)
AC_DEFINE(GC_FREEBSD_THREADS)
if test "x$PTHREAD_CFLAGS" != "x"; then
AM_CPPFLAGS="$AM_CPPFLAGS $PTHREAD_CFLAGS"
fi
if test "x$PTHREAD_LIBS" = "x"; then
THREADDLLIBS=-pthread
else
THREADDLLIBS="$PTHREAD_LIBS"
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
;;
*-*-solaris*)
AC_DEFINE(GC_SOLARIS_THREADS)
AC_DEFINE(GC_SOLARIS_PTHREADS)
;;
*-*-irix*)
AC_DEFINE(GC_IRIX_THREADS)
;;
*-*-cygwin*)
AC_DEFINE(GC_WIN32_THREADS)
;;
*-*-darwin*)
AC_DEFINE(GC_DARWIN_THREADS)
AC_DEFINE(THREAD_LOCAL_ALLOC)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
;;
*-*-netbsd*)
AC_DEFINE(GC_NETBSD_THREADS)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
;;
*-*-openbsd*)
AC_DEFINE(GC_OPENBSD_THREADS)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
;;
*-*-osf*)
AC_DEFINE(GC_OSF1_THREADS)
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
AC_DEFINE(THREAD_LOCAL_ALLOC)
# May want to enable it in other cases, too.
# Measurements havent yet been done.
fi
AM_CPPFLAGS="$AM_CPPFLAGS -pthread"
THREADDLLIBS="-lpthread -lrt"
;;
*)
AC_MSG_ERROR("Pthreads not supported by the GC on this platform.")
;;
esac
;;
win32)
AC_DEFINE(GC_WIN32_THREADS)
dnl Wine getenv may not return NULL for missing entry
AC_DEFINE(NO_GETENV)
if test "${enable_win32_dllmain}" = yes; then
AC_DEFINE(GC_INSIDE_DLL)
fi
;;
dgux386)
THREADS=dgux386
AC_MSG_RESULT($THREADDLLIBS)
# Use pthread GCC switch
THREADDLLIBS=-pthread
if test "${enable_parallel_mark}" = yes; then
AC_DEFINE(PARALLEL_MARK)
fi
AC_DEFINE(THREAD_LOCAL_ALLOC)
AC_DEFINE(GC_DGUX386_THREADS)
AC_DEFINE(DGUX_THREADS)
# Enable _POSIX4A_DRAFT10_SOURCE with flag -pthread
AM_CPPFLAGS="-pthread $AM_CPPFLAGS"
;;
aix)
THREADS=posix
THREADDLLIBS=-lpthread
AC_DEFINE(GC_AIX_THREADS)
AC_DEFINE(_REENTRANT)
;;
decosf1 | irix | mach | os2 | solaris | dce | vxworks)
AC_MSG_ERROR(thread package $THREADS not yet supported)
;;
*)
AC_MSG_ERROR($THREADS is an unknown thread package)
;;
esac
AC_SUBST(THREADDLLIBS)
case "$host" in
powerpc-*-darwin*)
powerpc_darwin=true
;;
esac
AM_CONDITIONAL(POWERPC_DARWIN,test x$powerpc_darwin = xtrue)
# Check if the GCC builtin __sync_bool_compare_and_swap is available.
# It is preferred in gc_locks.h for PPC as GCC 4.4 has a problem with the inline assembly there.
AC_MSG_CHECKING(for __sync_bool_compare_and_swap)
AC_TRY_COMPILE([],[
volatile unsigned int foo = 0;
unsigned int r1 = __sync_bool_compare_and_swap(&foo, 0, 1);
return 0;
], [
AC_MSG_RESULT(yes)
AC_DEFINE(HAS___SYNC_BOOL_COMPARE_AND_SWAP)
], [
AC_MSG_RESULT(no)
])
AC_MSG_CHECKING(for xlc)
AC_TRY_COMPILE([],[
#ifndef __xlC__
# error
#endif
], [compiler_xlc=yes], [compiler_xlc=no])
AC_MSG_RESULT($compiler_xlc)
AM_CONDITIONAL(COMPILER_XLC,test $compiler_xlc = yes)
if test $compiler_xlc = yes -a "$powerpc_darwin" = true; then
# the darwin stack-frame-walking code is completely broken on xlc
AC_DEFINE(DARWIN_DONT_PARSE_STACK)
fi
# We never want libdl on darwin. It is a fake libdl that just ends up making
# dyld calls anyway
case "$host" in
*-*-darwin*) ;;
*)
AC_CHECK_LIB(dl, dlopen, THREADDLLIBS="$THREADDLLIBS -ldl")
;;
esac
AC_SUBST(EXTRA_TEST_LIBS)
target_all=libgc.la
AC_SUBST(target_all)
dnl If the target is an eCos system, use the appropriate eCos
dnl I/O routines.
dnl FIXME: this should not be a local option but a global target
dnl system; at present there is no eCos target.
TARGET_ECOS="no"
AC_ARG_WITH(ecos,
[ --with-ecos enable runtime eCos target support],
TARGET_ECOS="$with_ecos"
)
addobjs=
addlibs=
addincludes=
addtests=
CXXAM_CPPFLAGS=
case "$TARGET_ECOS" in
no)
;;
*)
AC_DEFINE(ECOS)
CXXINCLUDES="-I${TARGET_ECOS}/include"
addobjs="$addobjs ecos.lo"
;;
esac
if test "${enable_cplusplus}" = yes; then
addincludes="$addincludes include/gc_cpp.h include/gc_allocator.h"
addtests="$addtests test_cpp"
fi
AM_CONDITIONAL(CPLUSPLUS, test "${enable_cplusplus}" = yes)
AC_SUBST(CXX)
AC_SUBST(AM_CPPFLAGS)
AC_SUBST(CXXINCLUDES)
# Configuration of shared libraries
#
AC_MSG_CHECKING(whether to build shared libraries)
AC_ENABLE_SHARED
case "$host" in
alpha-*-openbsd*)
enable_shared=no
AC_MSG_RESULT(no)
;;
*)
AC_MSG_RESULT(yes)
;;
esac
# Configuration of machine-dependent code
#
AC_MSG_CHECKING(which machine-dependent code should be used)
machdep=
case "$host" in
alpha-*-openbsd*)
machdep="alpha_mach_dep.lo"
if test x"${ac_cv_lib_dl_dlopen}" != xyes ; then
AC_MSG_WARN(OpenBSD/Alpha without dlopen(). Shared library support is disabled)
fi
;;
alpha*-*-linux*)
machdep="alpha_mach_dep.lo"
;;
i?86-*-solaris2.[[89]] | i?86-*-solaris2.1?)
AC_DEFINE(SOLARIS25_PROC_VDB_BUG_FIXED)
;;
mipstx39-*-elf*)
machdep="mips_ultrix_mach_dep.lo"
AC_DEFINE(STACKBASE, __stackbase)
AC_DEFINE(DATASTART_IS_ETEXT)
;;
mips-dec-ultrix*)
machdep="mips_ultrix_mach-dep.lo"
;;
mips-nec-sysv*|mips-unknown-sysv*)
;;
mips*-*-linux*)
;;
mips-*-*)
machdep="mips_sgi_mach_dep.lo"
AC_DEFINE(NO_EXECUTE_PERMISSION)
;;
sparc-*-netbsd*)
machdep="sparc_netbsd_mach_dep.lo"
;;
sparc-sun-solaris2.3)
machdep="sparc_mach_dep.lo"
AC_DEFINE(SUNOS53_SHARED_LIB)
;;
sparc*-sun-solaris2.*)
machdep="sparc_mach_dep.lo"
;;
ia64-*-*)
machdep="mach_dep.lo ia64_save_regs_in_stack.lo"
target_ia64=true
;;
*-*-nacl*)
AC_DEFINE(NO_EXECUTE_PERMISSION)
;;
esac
if test x"$machdep" = x; then
AC_MSG_RESULT($machdep)
machdep="mach_dep.lo"
fi
addobjs="$addobjs $machdep"
#
# Adding object files directly to _LIBADD breaks -j builds, so we need to add the sources
# instead, but _SOURCES can't contain autoconf substitutions, so add them using automake
# conditionals.
#
#AC_SUBST(addobjs)
AC_SUBST(addincludes)
AC_SUBST(addlibs)
AC_SUBST(addtests)
AM_CONDITIONAL(TARGET_IA64,test x$target_ia64 = xtrue)
AC_PROG_LIBTOOL
# Use dolt (http://dolt.freedesktop.org/) instead of libtool for building.
DOLT
#
# Check for AViiON Machines running DGUX
#
ac_is_dgux=no
AC_CHECK_HEADER(sys/dg_sys_info.h,
[ac_is_dgux=yes;])
## :GOTCHA: we do not check anything but sys/dg_sys_info.h
if test $ac_is_dgux = yes; then
if test "$enable_full_debug" = "yes"; then
CFLAGS="-g -mstandard -DDGUX -D_DGUX_SOURCE -Di386 -mno-legend -O2"
CXXFLAGS="-g -mstandard -DDGUX -D_DGUX_SOURCE -Di386 -mno-legend -O2"
else
CFLAGS="-DDGUX -D_DGUX_SOURCE -Di386 -mno-legend -O2"
CXXFLAGS="-DDGUX -D_DGUX_SOURCE -Di386 -mno-legend -O2"
fi
AC_SUBST(CFLAGS)
AC_SUBST(CXXFLAGS)
fi
dnl We use these options to decide which functions to include.
AC_ARG_WITH(target-subdir,
[ --with-target-subdir=SUBDIR
configuring with a cross compiler])
AC_ARG_WITH(cross-host,
[ --with-cross-host=HOST configuring with a cross compiler])
# automake wants to see AC_EXEEXT. But we don't need it. And having
# it is actually a problem, because the compiler we're passed can't
# necessarily do a full link. So we fool automake here.
if false; then
# autoconf 2.50 runs AC_EXEEXT by default, and the macro expands
# to nothing, so nothing would remain between `then' and `fi' if it
# were not for the `:' below.
:
AC_EXEEXT
fi
AC_MSG_CHECKING([for threads package to use])
AC_MSG_RESULT($THREADS)
dnl As of 4.13a2, the collector will not properly work on Solaris when
dnl built with gcc and -O. So we remove -O in the appropriate case.
dnl Not needed anymore on Solaris.
AC_MSG_CHECKING(whether Solaris gcc optimization fix is necessary)
case "$host" in
*aix*)
if test "$GCC" = yes; then
AC_MSG_RESULT(yes)
new_CFLAGS=
for i in $CFLAGS; do
case "$i" in
-O*)
;;
*)
new_CFLAGS="$new_CFLAGS $i"
;;
esac
done
CFLAGS="$new_CFLAGS"
else
AC_MSG_RESULT(no)
fi
;;
*) AC_MSG_RESULT(no) ;;
esac
dnl We need to override the top-level CFLAGS. This is how we do it.
MY_CFLAGS="$CFLAGS"
AC_SUBST(MY_CFLAGS)
dnl Include defines that have become de facto standard.
dnl ALL_INTERIOR_POINTERS can be overridden in startup code.
AC_DEFINE(SILENT)
AC_DEFINE(NO_SIGNALS)
AC_DEFINE(NO_EXECUTE_PERMISSION)
dnl AC_DEFINE(ALL_INTERIOR_POINTERS)
dnl By default, make the library as general as possible.
AC_DEFINE(JAVA_FINALIZATION)
AC_DEFINE(GC_GCJ_SUPPORT)
AC_DEFINE(ATOMIC_UNCOLLECTABLE)
dnl This is something of a hack. When cross-compiling we turn off
dnl some functionality. We also enable the "small" configuration.
dnl These is only correct when targetting an embedded system. FIXME.
if test -n "${with_cross_host}"; then
AC_DEFINE(NO_SIGSET)
AC_DEFINE(NO_CLOCK)
AC_DEFINE(SMALL_CONFIG)
AC_DEFINE(NO_DEBUGGING)
fi
UNWINDLIBS=
AC_ARG_ENABLE(full-debug,
[ --enable-full-debug include full support for pointer backtracing etc.],
[ if test "$enable_full_debug" = "yes"; then
AC_MSG_WARN("Should define GC_DEBUG and use debug alloc. in clients.")
AC_DEFINE(KEEP_BACK_PTRS)
AC_DEFINE(DBG_HDRS_ALL)
case $host in
ia64-*-linux* )
AC_DEFINE(MAKE_BACK_GRAPH)
AC_DEFINE(SAVE_CALL_COUNT, 8)
AC_CHECK_LIB(unwind, backtrace, [
AC_DEFINE(GC_HAVE_BUILTIN_BACKTRACE)
UNWINDLIBS=-lunwind
AC_MSG_WARN("Client code may need to link against libunwind.")
])
;;
x86-*-linux* | i586-*-linux* | i686-*-linux* | x86_64-*-linux* )
AC_DEFINE(MAKE_BACK_GRAPH)
AC_MSG_WARN("Client must not use -fomit-frame-pointer.")
AC_DEFINE(SAVE_CALL_COUNT, 8)
;;
i[3456]86-*-dgux*)
AC_DEFINE(MAKE_BACK_GRAPH)
;;
esac ]
fi)
AC_SUBST(UNWINDLIBS)
AC_ARG_ENABLE(redirect-malloc,
[ --enable-redirect-malloc redirect malloc and friends to GC routines])
if test "${enable_redirect_malloc}" = yes; then
if test "${enable_full_debug}" = yes; then
AC_DEFINE(REDIRECT_MALLOC, GC_debug_malloc_replacement)
AC_DEFINE(REDIRECT_REALLOC, GC_debug_realloc_replacement)
AC_DEFINE(REDIRECT_FREE, GC_debug_free)
else
AC_DEFINE(REDIRECT_MALLOC, GC_malloc)
fi
fi
AC_DEFINE(_IN_LIBGC)
AC_ARG_ENABLE(gc-assertions,
[ --enable-gc-assertions collector-internal assertion checking])
if test "${enable_gc_assertions}" = yes; then
AC_DEFINE(GC_ASSERTIONS)
fi
AC_ARG_ENABLE(quiet-build, [ --enable-quiet-build Enable quiet libgc build (on by default)], enable_quiet_build=$enableval, enable_quiet_build=yes)
AM_CONDITIONAL(USE_LIBDIR, test -z "$with_cross_host")
AC_ARG_ENABLE(werror, [ --enable-werror Pass -Werror to the C compiler], werror_flag=$enableval, werror_flag=no)
if test x$werror_flag = xyes; then
WERROR_CFLAGS="-Werror"
fi
AC_SUBST([WERROR_CFLAGS])
if test "${multilib}" = "yes"; then
multilib_arg="--enable-multilib"
else
multilib_arg=
fi
AC_OUTPUT(Makefile
m4/Makefile
include/Makefile
include/private/Makefile
doc/Makefile,,
srcdir=${srcdir}
host=${host}
CONFIG_SHELL=${CONFIG_SHELL-/bin/sh}
CC="${CC}"
DEFS="$DEFS"
)
-59
View File
@@ -1,59 +0,0 @@
# configure.host
# This shell script handles all host based configuration for the garbage
# collector.
# It sets various shell variables based on the the host and the
# configuration options. You can modify this shell script without
# needing to rerun autoconf.
# This shell script should be invoked as
# . configure.host
# If it encounters an error, it will exit with a message.
# It uses the following shell variables:
# host The configuration host
# host_cpu The configuration host CPU
# target_optspace --enable-target-optspace ("yes", "no", "")
# GCC should be "yes" if using gcc
# It sets the following shell variables:
# gc_cflags Special CFLAGS to use when building
gc_cflags=""
if test :"$GCC": = :yes: ; then
:
else
case "$host" in
hppa*-*-hpux* )
if test :$GCC: != :"yes": ; then
gc_cflags="${gc_flags} +ESdbgasm"
fi
# :TODO: actaully we should check using Autoconf if
# the compiler supports this option.
;;
esac
fi
case "${target_optspace}:${host}" in
yes:*)
gc_cflags="${gc_cflags} -Os"
;;
:m32r-* | :d10v-* | :d30v-*)
gc_cflags="${gc_cflags} -Os"
;;
no:* | :*)
# Nothing.
;;
esac
# Set any host dependent compiler flags.
# THIS TABLE IS SORTED. KEEP IT THAT WAY.
case "${host}" in
mips-tx39-*|mipstx39-unknown-*)
gc_cflags="${gc_cflags} -G 0"
;;
*)
;;
esac
-919
View File
@@ -1,919 +0,0 @@
/*
* Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*
* Author: Hans-J. Boehm ([email protected])
*/
/* Boehm, October 3, 1994 5:19 pm PDT */
# include "gc.h"
# include "cord.h"
# include <stdlib.h>
# include <stdio.h>
# include <string.h>
/* An implementation of the cord primitives. These are the only */
/* Functions that understand the representation. We perform only */
/* minimal checks on arguments to these functions. Out of bounds */
/* arguments to the iteration functions may result in client functions */
/* invoked on garbage data. In most cases, client functions should be */
/* programmed defensively enough that this does not result in memory */
/* smashes. */
typedef void (* oom_fn)(void);
oom_fn CORD_oom_fn = (oom_fn) 0;
# define OUT_OF_MEMORY { if (CORD_oom_fn != (oom_fn) 0) (*CORD_oom_fn)(); \
ABORT("Out of memory\n"); }
# define ABORT(msg) { fprintf(stderr, "%s\n", msg); abort(); }
typedef unsigned long word;
typedef union {
struct Concatenation {
char null;
char header;
char depth; /* concatenation nesting depth. */
unsigned char left_len;
/* Length of left child if it is sufficiently */
/* short; 0 otherwise. */
# define MAX_LEFT_LEN 255
word len;
CORD left; /* length(left) > 0 */
CORD right; /* length(right) > 0 */
} concatenation;
struct Function {
char null;
char header;
char depth; /* always 0 */
char left_len; /* always 0 */
word len;
CORD_fn fn;
void * client_data;
} function;
struct Generic {
char null;
char header;
char depth;
char left_len;
word len;
} generic;
char string[1];
} CordRep;
# define CONCAT_HDR 1
# define FN_HDR 4
# define SUBSTR_HDR 6
/* Substring nodes are a special case of function nodes. */
/* The client_data field is known to point to a substr_args */
/* structure, and the function is either CORD_apply_access_fn */
/* or CORD_index_access_fn. */
/* The following may be applied only to function and concatenation nodes: */
#define IS_CONCATENATION(s) (((CordRep *)s)->generic.header == CONCAT_HDR)
#define IS_FUNCTION(s) ((((CordRep *)s)->generic.header & FN_HDR) != 0)
#define IS_SUBSTR(s) (((CordRep *)s)->generic.header == SUBSTR_HDR)
#define LEN(s) (((CordRep *)s) -> generic.len)
#define DEPTH(s) (((CordRep *)s) -> generic.depth)
#define GEN_LEN(s) (CORD_IS_STRING(s) ? strlen(s) : LEN(s))
#define LEFT_LEN(c) ((c) -> left_len != 0? \
(c) -> left_len \
: (CORD_IS_STRING((c) -> left) ? \
(c) -> len - GEN_LEN((c) -> right) \
: LEN((c) -> left)))
#define SHORT_LIMIT (sizeof(CordRep) - 1)
/* Cords shorter than this are C strings */
/* Dump the internal representation of x to stdout, with initial */
/* indentation level n. */
void CORD_dump_inner(CORD x, unsigned n)
{
register size_t i;
for (i = 0; i < (size_t)n; i++) {
fputs(" ", stdout);
}
if (x == 0) {
fputs("NIL\n", stdout);
} else if (CORD_IS_STRING(x)) {
for (i = 0; i <= SHORT_LIMIT; i++) {
if (x[i] == '\0') break;
putchar(x[i]);
}
if (x[i] != '\0') fputs("...", stdout);
putchar('\n');
} else if (IS_CONCATENATION(x)) {
register struct Concatenation * conc =
&(((CordRep *)x) -> concatenation);
printf("Concatenation: %p (len: %d, depth: %d)\n",
x, (int)(conc -> len), (int)(conc -> depth));
CORD_dump_inner(conc -> left, n+1);
CORD_dump_inner(conc -> right, n+1);
} else /* function */{
register struct Function * func =
&(((CordRep *)x) -> function);
if (IS_SUBSTR(x)) printf("(Substring) ");
printf("Function: %p (len: %d): ", x, (int)(func -> len));
for (i = 0; i < 20 && i < func -> len; i++) {
putchar((*(func -> fn))(i, func -> client_data));
}
if (i < func -> len) fputs("...", stdout);
putchar('\n');
}
}
/* Dump the internal representation of x to stdout */
void CORD_dump(CORD x)
{
CORD_dump_inner(x, 0);
fflush(stdout);
}
CORD CORD_cat_char_star(CORD x, const char * y, size_t leny)
{
register size_t result_len;
register size_t lenx;
register int depth;
if (x == CORD_EMPTY) return(y);
if (leny == 0) return(x);
if (CORD_IS_STRING(x)) {
lenx = strlen(x);
result_len = lenx + leny;
if (result_len <= SHORT_LIMIT) {
register char * result = GC_MALLOC_ATOMIC(result_len+1);
if (result == 0) OUT_OF_MEMORY;
memcpy(result, x, lenx);
memcpy(result + lenx, y, leny);
result[result_len] = '\0';
return((CORD) result);
} else {
depth = 1;
}
} else {
register CORD right;
register CORD left;
register char * new_right;
register size_t right_len;
lenx = LEN(x);
if (leny <= SHORT_LIMIT/2
&& IS_CONCATENATION(x)
&& CORD_IS_STRING(right = ((CordRep *)x) -> concatenation.right)) {
/* Merge y into right part of x. */
if (!CORD_IS_STRING(left = ((CordRep *)x) -> concatenation.left)) {
right_len = lenx - LEN(left);
} else if (((CordRep *)x) -> concatenation.left_len != 0) {
right_len = lenx - ((CordRep *)x) -> concatenation.left_len;
} else {
right_len = strlen(right);
}
result_len = right_len + leny; /* length of new_right */
if (result_len <= SHORT_LIMIT) {
new_right = GC_MALLOC_ATOMIC(result_len + 1);
memcpy(new_right, right, right_len);
memcpy(new_right + right_len, y, leny);
new_right[result_len] = '\0';
y = new_right;
leny = result_len;
x = left;
lenx -= right_len;
/* Now fall through to concatenate the two pieces: */
}
if (CORD_IS_STRING(x)) {
depth = 1;
} else {
depth = DEPTH(x) + 1;
}
} else {
depth = DEPTH(x) + 1;
}
result_len = lenx + leny;
}
{
/* The general case; lenx, result_len is known: */
register struct Concatenation * result;
result = GC_NEW(struct Concatenation);
if (result == 0) OUT_OF_MEMORY;
result->header = CONCAT_HDR;
result->depth = depth;
if (lenx <= MAX_LEFT_LEN) result->left_len = lenx;
result->len = result_len;
result->left = x;
result->right = y;
if (depth >= MAX_DEPTH) {
return(CORD_balance((CORD)result));
} else {
return((CORD) result);
}
}
}
CORD CORD_cat(CORD x, CORD y)
{
register size_t result_len;
register int depth;
register size_t lenx;
if (x == CORD_EMPTY) return(y);
if (y == CORD_EMPTY) return(x);
if (CORD_IS_STRING(y)) {
return(CORD_cat_char_star(x, y, strlen(y)));
} else if (CORD_IS_STRING(x)) {
lenx = strlen(x);
depth = DEPTH(y) + 1;
} else {
register int depthy = DEPTH(y);
lenx = LEN(x);
depth = DEPTH(x) + 1;
if (depthy >= depth) depth = depthy + 1;
}
result_len = lenx + LEN(y);
{
register struct Concatenation * result;
result = GC_NEW(struct Concatenation);
if (result == 0) OUT_OF_MEMORY;
result->header = CONCAT_HDR;
result->depth = depth;
if (lenx <= MAX_LEFT_LEN) result->left_len = lenx;
result->len = result_len;
result->left = x;
result->right = y;
if (depth >= MAX_DEPTH) {
return(CORD_balance((CORD)result));
} else {
return((CORD) result);
}
}
}
CORD CORD_from_fn(CORD_fn fn, void * client_data, size_t len)
{
if (len <= 0) return(0);
if (len <= SHORT_LIMIT) {
register char * result;
register size_t i;
char buf[SHORT_LIMIT+1];
register char c;
for (i = 0; i < len; i++) {
c = (*fn)(i, client_data);
if (c == '\0') goto gen_case;
buf[i] = c;
}
buf[i] = '\0';
result = GC_MALLOC_ATOMIC(len+1);
if (result == 0) OUT_OF_MEMORY;
strcpy(result, buf);
result[len] = '\0';
return((CORD) result);
}
gen_case:
{
register struct Function * result;
result = GC_NEW(struct Function);
if (result == 0) OUT_OF_MEMORY;
result->header = FN_HDR;
/* depth is already 0 */
result->len = len;
result->fn = fn;
result->client_data = client_data;
return((CORD) result);
}
}
size_t CORD_len(CORD x)
{
if (x == 0) {
return(0);
} else {
return(GEN_LEN(x));
}
}
struct substr_args {
CordRep * sa_cord;
size_t sa_index;
};
char CORD_index_access_fn(size_t i, void * client_data)
{
register struct substr_args *descr = (struct substr_args *)client_data;
return(((char *)(descr->sa_cord))[i + descr->sa_index]);
}
char CORD_apply_access_fn(size_t i, void * client_data)
{
register struct substr_args *descr = (struct substr_args *)client_data;
register struct Function * fn_cord = &(descr->sa_cord->function);
return((*(fn_cord->fn))(i + descr->sa_index, fn_cord->client_data));
}
/* A version of CORD_substr that simply returns a function node, thus */
/* postponing its work. The fourth argument is a function that may */
/* be used for efficient access to the ith character. */
/* Assumes i >= 0 and i + n < length(x). */
CORD CORD_substr_closure(CORD x, size_t i, size_t n, CORD_fn f)
{
register struct substr_args * sa = GC_NEW(struct substr_args);
CORD result;
if (sa == 0) OUT_OF_MEMORY;
sa->sa_cord = (CordRep *)x;
sa->sa_index = i;
result = CORD_from_fn(f, (void *)sa, n);
((CordRep *)result) -> function.header = SUBSTR_HDR;
return (result);
}
# define SUBSTR_LIMIT (10 * SHORT_LIMIT)
/* Substrings of function nodes and flat strings shorter than */
/* this are flat strings. Othewise we use a functional */
/* representation, which is significantly slower to access. */
/* A version of CORD_substr that assumes i >= 0, n > 0, and i + n < length(x).*/
CORD CORD_substr_checked(CORD x, size_t i, size_t n)
{
if (CORD_IS_STRING(x)) {
if (n > SUBSTR_LIMIT) {
return(CORD_substr_closure(x, i, n, CORD_index_access_fn));
} else {
register char * result = GC_MALLOC_ATOMIC(n+1);
if (result == 0) OUT_OF_MEMORY;
strncpy(result, x+i, n);
result[n] = '\0';
return(result);
}
} else if (IS_CONCATENATION(x)) {
register struct Concatenation * conc
= &(((CordRep *)x) -> concatenation);
register size_t left_len;
register size_t right_len;
left_len = LEFT_LEN(conc);
right_len = conc -> len - left_len;
if (i >= left_len) {
if (n == right_len) return(conc -> right);
return(CORD_substr_checked(conc -> right, i - left_len, n));
} else if (i+n <= left_len) {
if (n == left_len) return(conc -> left);
return(CORD_substr_checked(conc -> left, i, n));
} else {
/* Need at least one character from each side. */
register CORD left_part;
register CORD right_part;
register size_t left_part_len = left_len - i;
if (i == 0) {
left_part = conc -> left;
} else {
left_part = CORD_substr_checked(conc -> left, i, left_part_len);
}
if (i + n == right_len + left_len) {
right_part = conc -> right;
} else {
right_part = CORD_substr_checked(conc -> right, 0,
n - left_part_len);
}
return(CORD_cat(left_part, right_part));
}
} else /* function */ {
if (n > SUBSTR_LIMIT) {
if (IS_SUBSTR(x)) {
/* Avoid nesting substring nodes. */
register struct Function * f = &(((CordRep *)x) -> function);
register struct substr_args *descr =
(struct substr_args *)(f -> client_data);
return(CORD_substr_closure((CORD)descr->sa_cord,
i + descr->sa_index,
n, f -> fn));
} else {
return(CORD_substr_closure(x, i, n, CORD_apply_access_fn));
}
} else {
char * result;
register struct Function * f = &(((CordRep *)x) -> function);
char buf[SUBSTR_LIMIT+1];
register char * p = buf;
register char c;
register int j;
register int lim = i + n;
for (j = i; j < lim; j++) {
c = (*(f -> fn))(j, f -> client_data);
if (c == '\0') {
return(CORD_substr_closure(x, i, n, CORD_apply_access_fn));
}
*p++ = c;
}
*p = '\0';
result = GC_MALLOC_ATOMIC(n+1);
if (result == 0) OUT_OF_MEMORY;
strcpy(result, buf);
return(result);
}
}
}
CORD CORD_substr(CORD x, size_t i, size_t n)
{
register size_t len = CORD_len(x);
if (i >= len || n <= 0) return(0);
/* n < 0 is impossible in a correct C implementation, but */
/* quite possible under SunOS 4.X. */
if (i + n > len) n = len - i;
# ifndef __STDC__
if (i < 0) ABORT("CORD_substr: second arg. negative");
/* Possible only if both client and C implementation are buggy. */
/* But empirically this happens frequently. */
# endif
return(CORD_substr_checked(x, i, n));
}
/* See cord.h for definition. We assume i is in range. */
int CORD_iter5(CORD x, size_t i, CORD_iter_fn f1,
CORD_batched_iter_fn f2, void * client_data)
{
if (x == 0) return(0);
if (CORD_IS_STRING(x)) {
register const char *p = x+i;
if (*p == '\0') ABORT("2nd arg to CORD_iter5 too big");
if (f2 != CORD_NO_FN) {
return((*f2)(p, client_data));
} else {
while (*p) {
if ((*f1)(*p, client_data)) return(1);
p++;
}
return(0);
}
} else if (IS_CONCATENATION(x)) {
register struct Concatenation * conc
= &(((CordRep *)x) -> concatenation);
if (i > 0) {
register size_t left_len = LEFT_LEN(conc);
if (i >= left_len) {
return(CORD_iter5(conc -> right, i - left_len, f1, f2,
client_data));
}
}
if (CORD_iter5(conc -> left, i, f1, f2, client_data)) {
return(1);
}
return(CORD_iter5(conc -> right, 0, f1, f2, client_data));
} else /* function */ {
register struct Function * f = &(((CordRep *)x) -> function);
register size_t j;
register size_t lim = f -> len;
for (j = i; j < lim; j++) {
if ((*f1)((*(f -> fn))(j, f -> client_data), client_data)) {
return(1);
}
}
return(0);
}
}
#undef CORD_iter
int CORD_iter(CORD x, CORD_iter_fn f1, void * client_data)
{
return(CORD_iter5(x, 0, f1, CORD_NO_FN, client_data));
}
int CORD_riter4(CORD x, size_t i, CORD_iter_fn f1, void * client_data)
{
if (x == 0) return(0);
if (CORD_IS_STRING(x)) {
register const char *p = x + i;
register char c;
for(;;) {
c = *p;
if (c == '\0') ABORT("2nd arg to CORD_riter4 too big");
if ((*f1)(c, client_data)) return(1);
if (p == x) break;
p--;
}
return(0);
} else if (IS_CONCATENATION(x)) {
register struct Concatenation * conc
= &(((CordRep *)x) -> concatenation);
register CORD left_part = conc -> left;
register size_t left_len;
left_len = LEFT_LEN(conc);
if (i >= left_len) {
if (CORD_riter4(conc -> right, i - left_len, f1, client_data)) {
return(1);
}
return(CORD_riter4(left_part, left_len - 1, f1, client_data));
} else {
return(CORD_riter4(left_part, i, f1, client_data));
}
} else /* function */ {
register struct Function * f = &(((CordRep *)x) -> function);
register size_t j;
for (j = i; ; j--) {
if ((*f1)((*(f -> fn))(j, f -> client_data), client_data)) {
return(1);
}
if (j == 0) return(0);
}
}
}
int CORD_riter(CORD x, CORD_iter_fn f1, void * client_data)
{
return(CORD_riter4(x, CORD_len(x) - 1, f1, client_data));
}
/*
* The following functions are concerned with balancing cords.
* Strategy:
* Scan the cord from left to right, keeping the cord scanned so far
* as a forest of balanced trees of exponentialy decreasing length.
* When a new subtree needs to be added to the forest, we concatenate all
* shorter ones to the new tree in the appropriate order, and then insert
* the result into the forest.
* Crucial invariants:
* 1. The concatenation of the forest (in decreasing order) with the
* unscanned part of the rope is equal to the rope being balanced.
* 2. All trees in the forest are balanced.
* 3. forest[i] has depth at most i.
*/
typedef struct {
CORD c;
size_t len; /* Actual length of c */
} ForestElement;
static size_t min_len [ MAX_DEPTH ];
static int min_len_init = 0;
int CORD_max_len;
typedef ForestElement Forest [ MAX_DEPTH ];
/* forest[i].len >= fib(i+1) */
/* The string is the concatenation */
/* of the forest in order of DECREASING */
/* indices. */
void CORD_init_min_len()
{
register int i;
register size_t last, previous, current;
min_len[0] = previous = 1;
min_len[1] = last = 2;
for (i = 2; i < MAX_DEPTH; i++) {
current = last + previous;
if (current < last) /* overflow */ current = last;
min_len[i] = current;
previous = last;
last = current;
}
CORD_max_len = last - 1;
min_len_init = 1;
}
void CORD_init_forest(ForestElement * forest, size_t max_len)
{
register int i;
for (i = 0; i < MAX_DEPTH; i++) {
forest[i].c = 0;
if (min_len[i] > max_len) return;
}
ABORT("Cord too long");
}
/* Add a leaf to the appropriate level in the forest, cleaning */
/* out lower levels as necessary. */
/* Also works if x is a balanced tree of concatenations; however */
/* in this case an extra concatenation node may be inserted above x; */
/* This node should not be counted in the statement of the invariants. */
void CORD_add_forest(ForestElement * forest, CORD x, size_t len)
{
register int i = 0;
register CORD sum = CORD_EMPTY;
register size_t sum_len = 0;
while (len > min_len[i + 1]) {
if (forest[i].c != 0) {
sum = CORD_cat(forest[i].c, sum);
sum_len += forest[i].len;
forest[i].c = 0;
}
i++;
}
/* Sum has depth at most 1 greter than what would be required */
/* for balance. */
sum = CORD_cat(sum, x);
sum_len += len;
/* If x was a leaf, then sum is now balanced. To see this */
/* consider the two cases in which forest[i-1] either is or is */
/* not empty. */
while (sum_len >= min_len[i]) {
if (forest[i].c != 0) {
sum = CORD_cat(forest[i].c, sum);
sum_len += forest[i].len;
/* This is again balanced, since sum was balanced, and has */
/* allowable depth that differs from i by at most 1. */
forest[i].c = 0;
}
i++;
}
i--;
forest[i].c = sum;
forest[i].len = sum_len;
}
CORD CORD_concat_forest(ForestElement * forest, size_t expected_len)
{
register int i = 0;
CORD sum = 0;
size_t sum_len = 0;
while (sum_len != expected_len) {
if (forest[i].c != 0) {
sum = CORD_cat(forest[i].c, sum);
sum_len += forest[i].len;
}
i++;
}
return(sum);
}
/* Insert the frontier of x into forest. Balanced subtrees are */
/* treated as leaves. This potentially adds one to the depth */
/* of the final tree. */
void CORD_balance_insert(CORD x, size_t len, ForestElement * forest)
{
register int depth;
if (CORD_IS_STRING(x)) {
CORD_add_forest(forest, x, len);
} else if (IS_CONCATENATION(x)
&& ((depth = DEPTH(x)) >= MAX_DEPTH
|| len < min_len[depth])) {
register struct Concatenation * conc
= &(((CordRep *)x) -> concatenation);
size_t left_len = LEFT_LEN(conc);
CORD_balance_insert(conc -> left, left_len, forest);
CORD_balance_insert(conc -> right, len - left_len, forest);
} else /* function or balanced */ {
CORD_add_forest(forest, x, len);
}
}
CORD CORD_balance(CORD x)
{
Forest forest;
register size_t len;
if (x == 0) return(0);
if (CORD_IS_STRING(x)) return(x);
if (!min_len_init) CORD_init_min_len();
len = LEN(x);
CORD_init_forest(forest, len);
CORD_balance_insert(x, len, forest);
return(CORD_concat_forest(forest, len));
}
/* Position primitives */
/* Private routines to deal with the hard cases only: */
/* P contains a prefix of the path to cur_pos. Extend it to a full */
/* path and set up leaf info. */
/* Return 0 if past the end of cord, 1 o.w. */
void CORD__extend_path(register CORD_pos p)
{
register struct CORD_pe * current_pe = &(p[0].path[p[0].path_len]);
register CORD top = current_pe -> pe_cord;
register size_t pos = p[0].cur_pos;
register size_t top_pos = current_pe -> pe_start_pos;
register size_t top_len = GEN_LEN(top);
/* Fill in the rest of the path. */
while(!CORD_IS_STRING(top) && IS_CONCATENATION(top)) {
register struct Concatenation * conc =
&(((CordRep *)top) -> concatenation);
register size_t left_len;
left_len = LEFT_LEN(conc);
current_pe++;
if (pos >= top_pos + left_len) {
current_pe -> pe_cord = top = conc -> right;
current_pe -> pe_start_pos = top_pos = top_pos + left_len;
top_len -= left_len;
} else {
current_pe -> pe_cord = top = conc -> left;
current_pe -> pe_start_pos = top_pos;
top_len = left_len;
}
p[0].path_len++;
}
/* Fill in leaf description for fast access. */
if (CORD_IS_STRING(top)) {
p[0].cur_leaf = top;
p[0].cur_start = top_pos;
p[0].cur_end = top_pos + top_len;
} else {
p[0].cur_end = 0;
}
if (pos >= top_pos + top_len) p[0].path_len = CORD_POS_INVALID;
}
char CORD__pos_fetch(register CORD_pos p)
{
/* Leaf is a function node */
struct CORD_pe * pe = &((p)[0].path[(p)[0].path_len]);
CORD leaf = pe -> pe_cord;
register struct Function * f = &(((CordRep *)leaf) -> function);
if (!IS_FUNCTION(leaf)) ABORT("CORD_pos_fetch: bad leaf");
return ((*(f -> fn))(p[0].cur_pos - pe -> pe_start_pos, f -> client_data));
}
void CORD__next(register CORD_pos p)
{
register size_t cur_pos = p[0].cur_pos + 1;
register struct CORD_pe * current_pe = &((p)[0].path[(p)[0].path_len]);
register CORD leaf = current_pe -> pe_cord;
/* Leaf is not a string or we're at end of leaf */
p[0].cur_pos = cur_pos;
if (!CORD_IS_STRING(leaf)) {
/* Function leaf */
register struct Function * f = &(((CordRep *)leaf) -> function);
register size_t start_pos = current_pe -> pe_start_pos;
register size_t end_pos = start_pos + f -> len;
if (cur_pos < end_pos) {
/* Fill cache and return. */
register size_t i;
register size_t limit = cur_pos + FUNCTION_BUF_SZ;
register CORD_fn fn = f -> fn;
register void * client_data = f -> client_data;
if (limit > end_pos) {
limit = end_pos;
}
for (i = cur_pos; i < limit; i++) {
p[0].function_buf[i - cur_pos] =
(*fn)(i - start_pos, client_data);
}
p[0].cur_start = cur_pos;
p[0].cur_leaf = p[0].function_buf;
p[0].cur_end = limit;
return;
}
}
/* End of leaf */
/* Pop the stack until we find two concatenation nodes with the */
/* same start position: this implies we were in left part. */
{
while (p[0].path_len > 0
&& current_pe[0].pe_start_pos != current_pe[-1].pe_start_pos) {
p[0].path_len--;
current_pe--;
}
if (p[0].path_len == 0) {
p[0].path_len = CORD_POS_INVALID;
return;
}
}
p[0].path_len--;
CORD__extend_path(p);
}
void CORD__prev(register CORD_pos p)
{
register struct CORD_pe * pe = &(p[0].path[p[0].path_len]);
if (p[0].cur_pos == 0) {
p[0].path_len = CORD_POS_INVALID;
return;
}
p[0].cur_pos--;
if (p[0].cur_pos >= pe -> pe_start_pos) return;
/* Beginning of leaf */
/* Pop the stack until we find two concatenation nodes with the */
/* different start position: this implies we were in right part. */
{
register struct CORD_pe * current_pe = &((p)[0].path[(p)[0].path_len]);
while (p[0].path_len > 0
&& current_pe[0].pe_start_pos == current_pe[-1].pe_start_pos) {
p[0].path_len--;
current_pe--;
}
}
p[0].path_len--;
CORD__extend_path(p);
}
#undef CORD_pos_fetch
#undef CORD_next
#undef CORD_prev
#undef CORD_pos_to_index
#undef CORD_pos_to_cord
#undef CORD_pos_valid
char CORD_pos_fetch(register CORD_pos p)
{
if (p[0].cur_start <= p[0].cur_pos && p[0].cur_pos < p[0].cur_end) {
return(p[0].cur_leaf[p[0].cur_pos - p[0].cur_start]);
} else {
return(CORD__pos_fetch(p));
}
}
void CORD_next(CORD_pos p)
{
if (p[0].cur_pos < p[0].cur_end - 1) {
p[0].cur_pos++;
} else {
CORD__next(p);
}
}
void CORD_prev(CORD_pos p)
{
if (p[0].cur_end != 0 && p[0].cur_pos > p[0].cur_start) {
p[0].cur_pos--;
} else {
CORD__prev(p);
}
}
size_t CORD_pos_to_index(CORD_pos p)
{
return(p[0].cur_pos);
}
CORD CORD_pos_to_cord(CORD_pos p)
{
return(p[0].path[0].pe_cord);
}
int CORD_pos_valid(CORD_pos p)
{
return(p[0].path_len != CORD_POS_INVALID);
}
void CORD_set_pos(CORD_pos p, CORD x, size_t i)
{
if (x == CORD_EMPTY) {
p[0].path_len = CORD_POS_INVALID;
return;
}
p[0].path[0].pe_cord = x;
p[0].path[0].pe_start_pos = 0;
p[0].path_len = 0;
p[0].cur_pos = i;
CORD__extend_path(p);
}
-396
View File
@@ -1,396 +0,0 @@
/*
* Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* An sprintf implementation that understands cords. This is probably */
/* not terribly portable. It assumes an ANSI stdarg.h. It further */
/* assumes that I can make copies of va_list variables, and read */
/* arguments repeatedly by applyting va_arg to the copies. This */
/* could be avoided at some performance cost. */
/* We also assume that unsigned and signed integers of various kinds */
/* have the same sizes, and can be cast back and forth. */
/* We assume that void * and char * have the same size. */
/* All this cruft is needed because we want to rely on the underlying */
/* sprintf implementation whenever possible. */
/* Boehm, September 21, 1995 6:00 pm PDT */
#include "cord.h"
#include "ec.h"
#include <stdio.h>
#include <stdarg.h>
#include <string.h>
#include "gc.h"
#define CONV_SPEC_LEN 50 /* Maximum length of a single */
/* conversion specification. */
#define CONV_RESULT_LEN 50 /* Maximum length of any */
/* conversion with default */
/* width and prec. */
static int ec_len(CORD_ec x)
{
return(CORD_len(x[0].ec_cord) + (x[0].ec_bufptr - x[0].ec_buf));
}
/* Possible nonumeric precision values. */
# define NONE -1
# define VARIABLE -2
/* Copy the conversion specification from CORD_pos into the buffer buf */
/* Return negative on error. */
/* Source initially points one past the leading %. */
/* It is left pointing at the conversion type. */
/* Assign field width and precision to *width and *prec. */
/* If width or prec is *, VARIABLE is assigned. */
/* Set *left to 1 if left adjustment flag is present. */
/* Set *long_arg to 1 if long flag ('l' or 'L') is present, or to */
/* -1 if 'h' is present. */
static int extract_conv_spec(CORD_pos source, char *buf,
int * width, int *prec, int *left, int * long_arg)
{
register int result = 0;
register int current_number = 0;
register int saw_period = 0;
register int saw_number = 0;
register int chars_so_far = 0;
register char current;
*width = NONE;
buf[chars_so_far++] = '%';
while(CORD_pos_valid(source)) {
if (chars_so_far >= CONV_SPEC_LEN) return(-1);
current = CORD_pos_fetch(source);
buf[chars_so_far++] = current;
switch(current) {
case '*':
saw_number = 1;
current_number = VARIABLE;
break;
case '0':
if (!saw_number) {
/* Zero fill flag; ignore */
break;
} /* otherwise fall through: */
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
saw_number = 1;
current_number *= 10;
current_number += current - '0';
break;
case '.':
saw_period = 1;
if(saw_number) {
*width = current_number;
saw_number = 0;
}
current_number = 0;
break;
case 'l':
case 'L':
*long_arg = 1;
current_number = 0;
break;
case 'h':
*long_arg = -1;
current_number = 0;
break;
case ' ':
case '+':
case '#':
current_number = 0;
break;
case '-':
*left = 1;
current_number = 0;
break;
case 'd':
case 'i':
case 'o':
case 'u':
case 'x':
case 'X':
case 'f':
case 'e':
case 'E':
case 'g':
case 'G':
case 'c':
case 'C':
case 's':
case 'S':
case 'p':
case 'n':
case 'r':
goto done;
default:
return(-1);
}
CORD_next(source);
}
return(-1);
done:
if (saw_number) {
if (saw_period) {
*prec = current_number;
} else {
*prec = NONE;
*width = current_number;
}
} else {
*prec = NONE;
}
buf[chars_so_far] = '\0';
return(result);
}
int CORD_vsprintf(CORD * out, CORD format, va_list args)
{
CORD_ec result;
register int count;
register char current;
CORD_pos pos;
char conv_spec[CONV_SPEC_LEN + 1];
CORD_ec_init(result);
for (CORD_set_pos(pos, format, 0); CORD_pos_valid(pos); CORD_next(pos)) {
current = CORD_pos_fetch(pos);
if (current == '%') {
CORD_next(pos);
if (!CORD_pos_valid(pos)) return(-1);
current = CORD_pos_fetch(pos);
if (current == '%') {
CORD_ec_append(result, current);
} else {
int width, prec;
int left_adj = 0;
int long_arg = 0;
CORD arg;
size_t len;
if (extract_conv_spec(pos, conv_spec,
&width, &prec,
&left_adj, &long_arg) < 0) {
return(-1);
}
current = CORD_pos_fetch(pos);
switch(current) {
case 'n':
/* Assign length to next arg */
if (long_arg == 0) {
int * pos_ptr;
pos_ptr = va_arg(args, int *);
*pos_ptr = ec_len(result);
} else if (long_arg > 0) {
long * pos_ptr;
pos_ptr = va_arg(args, long *);
*pos_ptr = ec_len(result);
} else {
short * pos_ptr;
pos_ptr = va_arg(args, short *);
*pos_ptr = ec_len(result);
}
goto done;
case 'r':
/* Append cord and any padding */
if (width == VARIABLE) width = va_arg(args, int);
if (prec == VARIABLE) prec = va_arg(args, int);
arg = va_arg(args, CORD);
len = CORD_len(arg);
if (prec != NONE && len > prec) {
if (prec < 0) return(-1);
arg = CORD_substr(arg, 0, prec);
len = prec;
}
if (width != NONE && len < width) {
char * blanks = GC_MALLOC_ATOMIC(width-len+1);
memset(blanks, ' ', width-len);
blanks[width-len] = '\0';
if (left_adj) {
arg = CORD_cat(arg, blanks);
} else {
arg = CORD_cat(blanks, arg);
}
}
CORD_ec_append_cord(result, arg);
goto done;
case 'c':
if (width == NONE && prec == NONE) {
register char c;
c = (char)va_arg(args, int);
CORD_ec_append(result, c);
goto done;
}
break;
case 's':
if (width == NONE && prec == NONE) {
char * str = va_arg(args, char *);
register char c;
while ((c = *str++)) {
CORD_ec_append(result, c);
}
goto done;
}
break;
default:
break;
}
/* Use standard sprintf to perform conversion */
{
register char * buf;
va_list vsprintf_args;
int max_size = 0;
int res;
# ifdef __va_copy
__va_copy(vsprintf_args, args);
# else
# if defined(__GNUC__) && !defined(__DJGPP__) /* and probably in other cases */
va_copy(vsprintf_args, args);
# else
vsprintf_args = args;
# endif
# endif
if (width == VARIABLE) width = va_arg(args, int);
if (prec == VARIABLE) prec = va_arg(args, int);
if (width != NONE) max_size = width;
if (prec != NONE && prec > max_size) max_size = prec;
max_size += CONV_RESULT_LEN;
if (max_size >= CORD_BUFSZ) {
buf = GC_MALLOC_ATOMIC(max_size + 1);
} else {
if (CORD_BUFSZ - (result[0].ec_bufptr-result[0].ec_buf)
< max_size) {
CORD_ec_flush_buf(result);
}
buf = result[0].ec_bufptr;
}
switch(current) {
case 'd':
case 'i':
case 'o':
case 'u':
case 'x':
case 'X':
case 'c':
if (long_arg <= 0) {
(void) va_arg(args, int);
} else if (long_arg > 0) {
(void) va_arg(args, long);
}
break;
case 's':
case 'p':
(void) va_arg(args, char *);
break;
case 'f':
case 'e':
case 'E':
case 'g':
case 'G':
(void) va_arg(args, double);
break;
default:
return(-1);
}
res = vsprintf(buf, conv_spec, vsprintf_args);
len = (size_t)res;
if ((char *)(GC_word)res == buf) {
/* old style vsprintf */
len = strlen(buf);
} else if (res < 0) {
return(-1);
}
if (buf != result[0].ec_bufptr) {
register char c;
while ((c = *buf++)) {
CORD_ec_append(result, c);
}
} else {
result[0].ec_bufptr = buf + len;
}
}
done:;
}
} else {
CORD_ec_append(result, current);
}
}
count = ec_len(result);
*out = CORD_balance(CORD_ec_to_cord(result));
return(count);
}
int CORD_sprintf(CORD * out, CORD format, ...)
{
va_list args;
int result;
va_start(args, format);
result = CORD_vsprintf(out, format, args);
va_end(args);
return(result);
}
int CORD_fprintf(FILE * f, CORD format, ...)
{
va_list args;
int result;
CORD out;
va_start(args, format);
result = CORD_vsprintf(&out, format, args);
va_end(args);
if (result > 0) CORD_put(out, f);
return(result);
}
int CORD_vfprintf(FILE * f, CORD format, va_list args)
{
int result;
CORD out;
result = CORD_vsprintf(&out, format, args);
if (result > 0) CORD_put(out, f);
return(result);
}
int CORD_printf(CORD format, ...)
{
va_list args;
int result;
CORD out;
va_start(args, format);
result = CORD_vsprintf(&out, format, args);
va_end(args);
if (result > 0) CORD_put(out, stdout);
return(result);
}
int CORD_vprintf(CORD format, va_list args)
{
int result;
CORD out;
result = CORD_vsprintf(&out, format, args);
if (result > 0) CORD_put(out, stdout);
return(result);
}
-235
View File
@@ -1,235 +0,0 @@
/*
* Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, August 24, 1994 11:58 am PDT */
# include "gc.h" /* For GC_INIT() only */
# include "cord.h"
# include <string.h>
# include <stdio.h>
# include <stdlib.h>
/* This is a very incomplete test of the cord package. It knows about */
/* a few internals of the package (e.g. when C strings are returned) */
/* that real clients shouldn't rely on. */
# define ABORT(string) \
{ int x = 0; fprintf(stderr, "FAILED: %s\n", string); x = 1 / x; abort(); }
int count;
int test_fn(char c, void * client_data)
{
if (client_data != (void *)13) ABORT("bad client data");
if (count < 64*1024+1) {
if ((count & 1) == 0) {
if (c != 'b') ABORT("bad char");
} else {
if (c != 'a') ABORT("bad char");
}
count++;
return(0);
} else {
if (c != 'c') ABORT("bad char");
count++;
return(1);
}
}
char id_cord_fn(size_t i, void * client_data)
{
return((char)i);
}
void test_basics()
{
CORD x = CORD_from_char_star("ab");
register int i;
char c;
CORD y;
CORD_pos p;
x = CORD_cat(x,x);
if (!CORD_IS_STRING(x)) ABORT("short cord should usually be a string");
if (strcmp(x, "abab") != 0) ABORT("bad CORD_cat result");
for (i = 1; i < 16; i++) {
x = CORD_cat(x,x);
}
x = CORD_cat(x,"c");
if (CORD_len(x) != 128*1024+1) ABORT("bad length");
count = 0;
if (CORD_iter5(x, 64*1024-1, test_fn, CORD_NO_FN, (void *)13) == 0) {
ABORT("CORD_iter5 failed");
}
if (count != 64*1024 + 2) ABORT("CORD_iter5 failed");
count = 0;
CORD_set_pos(p, x, 64*1024-1);
while(CORD_pos_valid(p)) {
(void) test_fn(CORD_pos_fetch(p), (void *)13);
CORD_next(p);
}
if (count != 64*1024 + 2) ABORT("Position based iteration failed");
y = CORD_substr(x, 1023, 5);
if (!CORD_IS_STRING(y)) ABORT("short cord should usually be a string");
if (strcmp(y, "babab") != 0) ABORT("bad CORD_substr result");
y = CORD_substr(x, 1024, 8);
if (!CORD_IS_STRING(y)) ABORT("short cord should usually be a string");
if (strcmp(y, "abababab") != 0) ABORT("bad CORD_substr result");
y = CORD_substr(x, 128*1024-1, 8);
if (!CORD_IS_STRING(y)) ABORT("short cord should usually be a string");
if (strcmp(y, "bc") != 0) ABORT("bad CORD_substr result");
x = CORD_balance(x);
if (CORD_len(x) != 128*1024+1) ABORT("bad length");
count = 0;
if (CORD_iter5(x, 64*1024-1, test_fn, CORD_NO_FN, (void *)13) == 0) {
ABORT("CORD_iter5 failed");
}
if (count != 64*1024 + 2) ABORT("CORD_iter5 failed");
y = CORD_substr(x, 1023, 5);
if (!CORD_IS_STRING(y)) ABORT("short cord should usually be a string");
if (strcmp(y, "babab") != 0) ABORT("bad CORD_substr result");
y = CORD_from_fn(id_cord_fn, 0, 13);
i = 0;
CORD_set_pos(p, y, i);
while(CORD_pos_valid(p)) {
c = CORD_pos_fetch(p);
if(c != i) ABORT("Traversal of function node failed");
CORD_next(p); i++;
}
if (i != 13) ABORT("Bad apparent length for function node");
}
void test_extras()
{
# if defined(__OS2__) || defined(__DJGPP__)
# define FNAME1 "tmp1"
# define FNAME2 "tmp2"
# elif defined(AMIGA)
# define FNAME1 "T:tmp1"
# define FNAME2 "T:tmp2"
# else
# define FNAME1 "/tmp/cord_test"
# define FNAME2 "/tmp/cord_test2"
# endif
register int i;
CORD y = "abcdefghijklmnopqrstuvwxyz0123456789";
CORD x = "{}";
CORD w, z;
FILE *f;
FILE *f1a, *f1b, *f2;
w = CORD_cat(CORD_cat(y,y),y);
z = CORD_catn(3,y,y,y);
if (CORD_cmp(w,z) != 0) ABORT("CORD_catn comparison wrong");
for (i = 1; i < 100; i++) {
x = CORD_cat(x, y);
}
z = CORD_balance(x);
if (CORD_cmp(x,z) != 0) ABORT("balanced string comparison wrong");
if (CORD_cmp(x,CORD_cat(z, CORD_nul(13))) >= 0) ABORT("comparison 2");
if (CORD_cmp(CORD_cat(x, CORD_nul(13)), z) <= 0) ABORT("comparison 3");
if (CORD_cmp(x,CORD_cat(z, "13")) >= 0) ABORT("comparison 4");
if ((f = fopen(FNAME1, "w")) == 0) ABORT("open failed");
if (CORD_put(z,f) == EOF) ABORT("CORD_put failed");
if (fclose(f) == EOF) ABORT("fclose failed");
w = CORD_from_file(f1a = fopen(FNAME1, "rb"));
if (CORD_len(w) != CORD_len(z)) ABORT("file length wrong");
if (CORD_cmp(w,z) != 0) ABORT("file comparison wrong");
if (CORD_cmp(CORD_substr(w, 50*36+2, 36), y) != 0)
ABORT("file substr wrong");
z = CORD_from_file_lazy(f1b = fopen(FNAME1, "rb"));
if (CORD_cmp(w,z) != 0) ABORT("File conversions differ");
if (CORD_chr(w, 0, '9') != 37) ABORT("CORD_chr failed 1");
if (CORD_chr(w, 3, 'a') != 38) ABORT("CORD_chr failed 2");
if (CORD_rchr(w, CORD_len(w) - 1, '}') != 1) ABORT("CORD_rchr failed");
x = y;
for (i = 1; i < 14; i++) {
x = CORD_cat(x,x);
}
if ((f = fopen(FNAME2, "w")) == 0) ABORT("2nd open failed");
# ifdef __DJGPP__
/* FIXME: DJGPP workaround. Why does this help? */
if (fflush(f) != 0) ABORT("fflush failed");
# endif
if (CORD_put(x,f) == EOF) ABORT("CORD_put failed");
if (fclose(f) == EOF) ABORT("fclose failed");
w = CORD_from_file(f2 = fopen(FNAME2, "rb"));
if (CORD_len(w) != CORD_len(x)) ABORT("file length wrong");
if (CORD_cmp(w,x) != 0) ABORT("file comparison wrong");
if (CORD_cmp(CORD_substr(w, 1000*36, 36), y) != 0)
ABORT("file substr wrong");
if (strcmp(CORD_to_char_star(CORD_substr(w, 1000*36, 36)), y) != 0)
ABORT("char * file substr wrong");
if (strcmp(CORD_substr(w, 1000*36, 2), "ab") != 0)
ABORT("short file substr wrong");
if (CORD_str(x,1,"9a") != 35) ABORT("CORD_str failed 1");
if (CORD_str(x,0,"9abcdefghijk") != 35) ABORT("CORD_str failed 2");
if (CORD_str(x,0,"9abcdefghijx") != CORD_NOT_FOUND)
ABORT("CORD_str failed 3");
if (CORD_str(x,0,"9>") != CORD_NOT_FOUND) ABORT("CORD_str failed 4");
if (remove(FNAME1) != 0) {
/* On some systems, e.g. OS2, this may fail if f1 is still open. */
if ((fclose(f1a) == EOF) & (fclose(f1b) == EOF))
ABORT("fclose(f1) failed");
if (remove(FNAME1) != 0) ABORT("remove 1 failed");
}
if (remove(FNAME2) != 0) {
if (fclose(f2) == EOF) ABORT("fclose(f2) failed");
if (remove(FNAME2) != 0) ABORT("remove 2 failed");
}
}
void test_printf()
{
CORD result;
char result2[200];
long l;
short s;
CORD x;
if (CORD_sprintf(&result, "%7.2f%ln", 3.14159F, &l) != 7)
ABORT("CORD_sprintf failed 1");
if (CORD_cmp(result, " 3.14") != 0)ABORT("CORD_sprintf goofed 1");
if (l != 7) ABORT("CORD_sprintf goofed 2");
if (CORD_sprintf(&result, "%-7.2s%hn%c%s", "abcd", &s, 'x', "yz") != 10)
ABORT("CORD_sprintf failed 2");
if (CORD_cmp(result, "ab xyz") != 0)ABORT("CORD_sprintf goofed 3");
if (s != 7) ABORT("CORD_sprintf goofed 4");
x = "abcdefghij";
x = CORD_cat(x,x);
x = CORD_cat(x,x);
x = CORD_cat(x,x);
if (CORD_sprintf(&result, "->%-120.78r!\n", x) != 124)
ABORT("CORD_sprintf failed 3");
(void) sprintf(result2, "->%-120.78s!\n", CORD_to_char_star(x));
if (CORD_cmp(result, result2) != 0)ABORT("CORD_sprintf goofed 5");
}
int main()
{
# ifdef THINK_C
printf("cordtest:\n");
# endif
GC_INIT();
test_basics();
test_extras();
test_printf();
CORD_fprintf(stderr, "SUCCEEDED\n");
return(0);
}
-621
View File
@@ -1,621 +0,0 @@
/*
* Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*
* Author: Hans-J. Boehm ([email protected])
*/
/*
* These are functions on cords that do not need to understand their
* implementation. They serve also serve as example client code for
* cord_basics.
*/
/* Boehm, December 8, 1995 1:53 pm PST */
# include <stdio.h>
# include <string.h>
# include <stdlib.h>
# include <stdarg.h>
# include "cord.h"
# include "ec.h"
# define I_HIDE_POINTERS /* So we get access to allocation lock. */
/* We use this for lazy file reading, */
/* so that we remain independent */
/* of the threads primitives. */
# include "gc.h"
/* For now we assume that pointer reads and writes are atomic, */
/* i.e. another thread always sees the state before or after */
/* a write. This might be false on a Motorola M68K with */
/* pointers that are not 32-bit aligned. But there probably */
/* aren't too many threads packages running on those. */
# define ATOMIC_WRITE(x,y) (x) = (y)
# define ATOMIC_READ(x) (*(x))
/* The standard says these are in stdio.h, but they aren't always: */
# ifndef SEEK_SET
# define SEEK_SET 0
# endif
# ifndef SEEK_END
# define SEEK_END 2
# endif
# define BUFSZ 2048 /* Size of stack allocated buffers when */
/* we want large buffers. */
typedef void (* oom_fn)(void);
# define OUT_OF_MEMORY { if (CORD_oom_fn != (oom_fn) 0) (*CORD_oom_fn)(); \
ABORT("Out of memory\n"); }
# define ABORT(msg) { fprintf(stderr, "%s\n", msg); abort(); }
CORD CORD_cat_char(CORD x, char c)
{
register char * string;
if (c == '\0') return(CORD_cat(x, CORD_nul(1)));
string = GC_MALLOC_ATOMIC(2);
if (string == 0) OUT_OF_MEMORY;
string[0] = c;
string[1] = '\0';
return(CORD_cat_char_star(x, string, 1));
}
CORD CORD_catn(int nargs, ...)
{
register CORD result = CORD_EMPTY;
va_list args;
register int i;
va_start(args, nargs);
for (i = 0; i < nargs; i++) {
register CORD next = va_arg(args, CORD);
result = CORD_cat(result, next);
}
va_end(args);
return(result);
}
typedef struct {
size_t len;
size_t count;
char * buf;
} CORD_fill_data;
int CORD_fill_proc(char c, void * client_data)
{
register CORD_fill_data * d = (CORD_fill_data *)client_data;
register size_t count = d -> count;
(d -> buf)[count] = c;
d -> count = ++count;
if (count >= d -> len) {
return(1);
} else {
return(0);
}
}
int CORD_batched_fill_proc(const char * s, void * client_data)
{
register CORD_fill_data * d = (CORD_fill_data *)client_data;
register size_t count = d -> count;
register size_t max = d -> len;
register char * buf = d -> buf;
register const char * t = s;
while((buf[count] = *t++) != '\0') {
count++;
if (count >= max) {
d -> count = count;
return(1);
}
}
d -> count = count;
return(0);
}
/* Fill buf with len characters starting at i. */
/* Assumes len characters are available. */
void CORD_fill_buf(CORD x, size_t i, size_t len, char * buf)
{
CORD_fill_data fd;
fd.len = len;
fd.buf = buf;
fd.count = 0;
(void)CORD_iter5(x, i, CORD_fill_proc, CORD_batched_fill_proc, &fd);
}
int CORD_cmp(CORD x, CORD y)
{
CORD_pos xpos;
CORD_pos ypos;
register size_t avail, yavail;
if (y == CORD_EMPTY) return(x != CORD_EMPTY);
if (x == CORD_EMPTY) return(-1);
if (CORD_IS_STRING(y) && CORD_IS_STRING(x)) return(strcmp(x,y));
CORD_set_pos(xpos, x, 0);
CORD_set_pos(ypos, y, 0);
for(;;) {
if (!CORD_pos_valid(xpos)) {
if (CORD_pos_valid(ypos)) {
return(-1);
} else {
return(0);
}
}
if (!CORD_pos_valid(ypos)) {
return(1);
}
if ((avail = CORD_pos_chars_left(xpos)) <= 0
|| (yavail = CORD_pos_chars_left(ypos)) <= 0) {
register char xcurrent = CORD_pos_fetch(xpos);
register char ycurrent = CORD_pos_fetch(ypos);
if (xcurrent != ycurrent) return(xcurrent - ycurrent);
CORD_next(xpos);
CORD_next(ypos);
} else {
/* process as many characters as we can */
register int result;
if (avail > yavail) avail = yavail;
result = strncmp(CORD_pos_cur_char_addr(xpos),
CORD_pos_cur_char_addr(ypos), avail);
if (result != 0) return(result);
CORD_pos_advance(xpos, avail);
CORD_pos_advance(ypos, avail);
}
}
}
int CORD_ncmp(CORD x, size_t x_start, CORD y, size_t y_start, size_t len)
{
CORD_pos xpos;
CORD_pos ypos;
register size_t count;
register long avail, yavail;
CORD_set_pos(xpos, x, x_start);
CORD_set_pos(ypos, y, y_start);
for(count = 0; count < len;) {
if (!CORD_pos_valid(xpos)) {
if (CORD_pos_valid(ypos)) {
return(-1);
} else {
return(0);
}
}
if (!CORD_pos_valid(ypos)) {
return(1);
}
if ((avail = CORD_pos_chars_left(xpos)) <= 0
|| (yavail = CORD_pos_chars_left(ypos)) <= 0) {
register char xcurrent = CORD_pos_fetch(xpos);
register char ycurrent = CORD_pos_fetch(ypos);
if (xcurrent != ycurrent) return(xcurrent - ycurrent);
CORD_next(xpos);
CORD_next(ypos);
count++;
} else {
/* process as many characters as we can */
register int result;
if (avail > yavail) avail = yavail;
count += avail;
if (count > len) avail -= (count - len);
result = strncmp(CORD_pos_cur_char_addr(xpos),
CORD_pos_cur_char_addr(ypos), (size_t)avail);
if (result != 0) return(result);
CORD_pos_advance(xpos, (size_t)avail);
CORD_pos_advance(ypos, (size_t)avail);
}
}
return(0);
}
char * CORD_to_char_star(CORD x)
{
register size_t len = CORD_len(x);
char * result = GC_MALLOC_ATOMIC(len + 1);
if (result == 0) OUT_OF_MEMORY;
CORD_fill_buf(x, 0, len, result);
result[len] = '\0';
return(result);
}
CORD CORD_from_char_star(const char *s)
{
char * result;
size_t len = strlen(s);
if (0 == len) return(CORD_EMPTY);
result = GC_MALLOC_ATOMIC(len + 1);
if (result == 0) OUT_OF_MEMORY;
memcpy(result, s, len+1);
return(result);
}
const char * CORD_to_const_char_star(CORD x)
{
if (x == 0) return("");
if (CORD_IS_STRING(x)) return((const char *)x);
return(CORD_to_char_star(x));
}
char CORD_fetch(CORD x, size_t i)
{
CORD_pos xpos;
CORD_set_pos(xpos, x, i);
if (!CORD_pos_valid(xpos)) ABORT("bad index?");
return(CORD_pos_fetch(xpos));
}
int CORD_put_proc(char c, void * client_data)
{
register FILE * f = (FILE *)client_data;
return(putc(c, f) == EOF);
}
int CORD_batched_put_proc(const char * s, void * client_data)
{
register FILE * f = (FILE *)client_data;
return(fputs(s, f) == EOF);
}
int CORD_put(CORD x, FILE * f)
{
if (CORD_iter5(x, 0, CORD_put_proc, CORD_batched_put_proc, f)) {
return(EOF);
} else {
return(1);
}
}
typedef struct {
size_t pos; /* Current position in the cord */
char target; /* Character we're looking for */
} chr_data;
int CORD_chr_proc(char c, void * client_data)
{
register chr_data * d = (chr_data *)client_data;
if (c == d -> target) return(1);
(d -> pos) ++;
return(0);
}
int CORD_rchr_proc(char c, void * client_data)
{
register chr_data * d = (chr_data *)client_data;
if (c == d -> target) return(1);
(d -> pos) --;
return(0);
}
int CORD_batched_chr_proc(const char *s, void * client_data)
{
register chr_data * d = (chr_data *)client_data;
register char * occ = strchr(s, d -> target);
if (occ == 0) {
d -> pos += strlen(s);
return(0);
} else {
d -> pos += occ - s;
return(1);
}
}
size_t CORD_chr(CORD x, size_t i, int c)
{
chr_data d;
d.pos = i;
d.target = c;
if (CORD_iter5(x, i, CORD_chr_proc, CORD_batched_chr_proc, &d)) {
return(d.pos);
} else {
return(CORD_NOT_FOUND);
}
}
size_t CORD_rchr(CORD x, size_t i, int c)
{
chr_data d;
d.pos = i;
d.target = c;
if (CORD_riter4(x, i, CORD_rchr_proc, &d)) {
return(d.pos);
} else {
return(CORD_NOT_FOUND);
}
}
/* Find the first occurrence of s in x at position start or later. */
/* This uses an asymptotically poor algorithm, which should typically */
/* perform acceptably. We compare the first few characters directly, */
/* and call CORD_ncmp whenever there is a partial match. */
/* This has the advantage that we allocate very little, or not at all. */
/* It's very fast if there are few close misses. */
size_t CORD_str(CORD x, size_t start, CORD s)
{
CORD_pos xpos;
size_t xlen = CORD_len(x);
size_t slen;
register size_t start_len;
const char * s_start;
unsigned long s_buf = 0; /* The first few characters of s */
unsigned long x_buf = 0; /* Start of candidate substring. */
/* Initialized only to make compilers */
/* happy. */
unsigned long mask = 0;
register size_t i;
register size_t match_pos;
if (s == CORD_EMPTY) return(start);
if (CORD_IS_STRING(s)) {
s_start = s;
slen = strlen(s);
} else {
s_start = CORD_to_char_star(CORD_substr(s, 0, sizeof(unsigned long)));
slen = CORD_len(s);
}
if (xlen < start || xlen - start < slen) return(CORD_NOT_FOUND);
start_len = slen;
if (start_len > sizeof(unsigned long)) start_len = sizeof(unsigned long);
CORD_set_pos(xpos, x, start);
for (i = 0; i < start_len; i++) {
mask <<= 8;
mask |= 0xff;
s_buf <<= 8;
s_buf |= (unsigned char)s_start[i];
x_buf <<= 8;
x_buf |= (unsigned char)CORD_pos_fetch(xpos);
CORD_next(xpos);
}
for (match_pos = start; ; match_pos++) {
if ((x_buf & mask) == s_buf) {
if (slen == start_len ||
CORD_ncmp(x, match_pos + start_len,
s, start_len, slen - start_len) == 0) {
return(match_pos);
}
}
if ( match_pos == xlen - slen ) {
return(CORD_NOT_FOUND);
}
x_buf <<= 8;
x_buf |= (unsigned char)CORD_pos_fetch(xpos);
CORD_next(xpos);
}
}
void CORD_ec_flush_buf(CORD_ec x)
{
register size_t len = x[0].ec_bufptr - x[0].ec_buf;
char * s;
if (len == 0) return;
s = GC_MALLOC_ATOMIC(len+1);
memcpy(s, x[0].ec_buf, len);
s[len] = '\0';
x[0].ec_cord = CORD_cat_char_star(x[0].ec_cord, s, len);
x[0].ec_bufptr = x[0].ec_buf;
}
void CORD_ec_append_cord(CORD_ec x, CORD s)
{
CORD_ec_flush_buf(x);
x[0].ec_cord = CORD_cat(x[0].ec_cord, s);
}
/*ARGSUSED*/
char CORD_nul_func(size_t i, void * client_data)
{
return((char)(unsigned long)client_data);
}
CORD CORD_chars(char c, size_t i)
{
return(CORD_from_fn(CORD_nul_func, (void *)(unsigned long)c, i));
}
CORD CORD_from_file_eager(FILE * f)
{
register int c;
CORD_ec ecord;
CORD_ec_init(ecord);
for(;;) {
c = getc(f);
if (c == 0) {
/* Append the right number of NULs */
/* Note that any string of NULs is rpresented in 4 words, */
/* independent of its length. */
register size_t count = 1;
CORD_ec_flush_buf(ecord);
while ((c = getc(f)) == 0) count++;
ecord[0].ec_cord = CORD_cat(ecord[0].ec_cord, CORD_nul(count));
}
if (c == EOF) break;
CORD_ec_append(ecord, c);
}
(void) fclose(f);
return(CORD_balance(CORD_ec_to_cord(ecord)));
}
/* The state maintained for a lazily read file consists primarily */
/* of a large direct-mapped cache of previously read values. */
/* We could rely more on stdio buffering. That would have 2 */
/* disadvantages: */
/* 1) Empirically, not all fseek implementations preserve the */
/* buffer whenever they could. */
/* 2) It would fail if 2 different sections of a long cord */
/* were being read alternately. */
/* We do use the stdio buffer for read ahead. */
/* To guarantee thread safety in the presence of atomic pointer */
/* writes, cache lines are always replaced, and never modified in */
/* place. */
# define LOG_CACHE_SZ 14
# define CACHE_SZ (1 << LOG_CACHE_SZ)
# define LOG_LINE_SZ 9
# define LINE_SZ (1 << LOG_LINE_SZ)
typedef struct {
size_t tag;
char data[LINE_SZ];
/* data[i%LINE_SZ] = ith char in file if tag = i/LINE_SZ */
} cache_line;
typedef struct {
FILE * lf_file;
size_t lf_current; /* Current file pointer value */
cache_line * volatile lf_cache[CACHE_SZ/LINE_SZ];
} lf_state;
# define MOD_CACHE_SZ(n) ((n) & (CACHE_SZ - 1))
# define DIV_CACHE_SZ(n) ((n) >> LOG_CACHE_SZ)
# define MOD_LINE_SZ(n) ((n) & (LINE_SZ - 1))
# define DIV_LINE_SZ(n) ((n) >> LOG_LINE_SZ)
# define LINE_START(n) ((n) & ~(LINE_SZ - 1))
typedef struct {
lf_state * state;
size_t file_pos; /* Position of needed character. */
cache_line * new_cache;
} refill_data;
/* Executed with allocation lock. */
static char refill_cache(client_data)
refill_data * client_data;
{
register lf_state * state = client_data -> state;
register size_t file_pos = client_data -> file_pos;
FILE *f = state -> lf_file;
size_t line_start = LINE_START(file_pos);
size_t line_no = DIV_LINE_SZ(MOD_CACHE_SZ(file_pos));
cache_line * new_cache = client_data -> new_cache;
if (line_start != state -> lf_current
&& fseek(f, line_start, SEEK_SET) != 0) {
ABORT("fseek failed");
}
if (fread(new_cache -> data, sizeof(char), LINE_SZ, f)
<= file_pos - line_start) {
ABORT("fread failed");
}
new_cache -> tag = DIV_LINE_SZ(file_pos);
/* Store barrier goes here. */
ATOMIC_WRITE(state -> lf_cache[line_no], new_cache);
state -> lf_current = line_start + LINE_SZ;
return(new_cache->data[MOD_LINE_SZ(file_pos)]);
}
char CORD_lf_func(size_t i, void * client_data)
{
register lf_state * state = (lf_state *)client_data;
register cache_line * volatile * cl_addr =
&(state -> lf_cache[DIV_LINE_SZ(MOD_CACHE_SZ(i))]);
register cache_line * cl = (cache_line *)ATOMIC_READ(cl_addr);
if (cl == 0 || cl -> tag != DIV_LINE_SZ(i)) {
/* Cache miss */
refill_data rd;
rd.state = state;
rd.file_pos = i;
rd.new_cache = GC_NEW_ATOMIC(cache_line);
if (rd.new_cache == 0) OUT_OF_MEMORY;
return((char)(GC_word)
GC_call_with_alloc_lock((GC_fn_type) refill_cache, &rd));
}
return(cl -> data[MOD_LINE_SZ(i)]);
}
/*ARGSUSED*/
void CORD_lf_close_proc(void * obj, void * client_data)
{
if (fclose(((lf_state *)obj) -> lf_file) != 0) {
ABORT("CORD_lf_close_proc: fclose failed");
}
}
CORD CORD_from_file_lazy_inner(FILE * f, size_t len)
{
register lf_state * state = GC_NEW(lf_state);
register int i;
if (state == 0) OUT_OF_MEMORY;
if (len != 0) {
/* Dummy read to force buffer allocation. */
/* This greatly increases the probability */
/* of avoiding deadlock if buffer allocation */
/* is redirected to GC_malloc and the */
/* world is multithreaded. */
char buf[1];
(void) fread(buf, 1, 1, f);
rewind(f);
}
state -> lf_file = f;
for (i = 0; i < CACHE_SZ/LINE_SZ; i++) {
state -> lf_cache[i] = 0;
}
state -> lf_current = 0;
GC_REGISTER_FINALIZER(state, CORD_lf_close_proc, 0, 0, 0);
return(CORD_from_fn(CORD_lf_func, state, len));
}
CORD CORD_from_file_lazy(FILE * f)
{
register long len;
if (fseek(f, 0l, SEEK_END) != 0) {
ABORT("Bad fd argument - fseek failed");
}
if ((len = ftell(f)) < 0) {
ABORT("Bad fd argument - ftell failed");
}
rewind(f);
return(CORD_from_file_lazy_inner(f, (size_t)len));
}
# define LAZY_THRESHOLD (128*1024 + 1)
CORD CORD_from_file(FILE * f)
{
register long len;
if (fseek(f, 0l, SEEK_END) != 0) {
ABORT("Bad fd argument - fseek failed");
}
if ((len = ftell(f)) < 0) {
ABORT("Bad fd argument - ftell failed");
}
rewind(f);
if (len < LAZY_THRESHOLD) {
return(CORD_from_file_eager(f));
} else {
return(CORD_from_file_lazy_inner(f, (size_t)len));
}
}
-603
View File
@@ -1,603 +0,0 @@
/*
* Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*
* Author: Hans-J. Boehm ([email protected])
*/
/*
* A really simple-minded text editor based on cords.
* Things it does right:
* No size bounds.
* Inbounded undo.
* Shouldn't crash no matter what file you invoke it on (e.g. /vmunix)
* (Make sure /vmunix is not writable before you try this.)
* Scrolls horizontally.
* Things it does wrong:
* It doesn't handle tabs reasonably (use "expand" first).
* The command set is MUCH too small.
* The redisplay algorithm doesn't let curses do the scrolling.
* The rule for moving the window over the file is suboptimal.
*/
/* Boehm, February 6, 1995 12:27 pm PST */
/* Boehm, May 19, 1994 2:20 pm PDT */
#include <stdio.h>
#include "gc.h"
#include "cord.h"
#ifdef THINK_C
#define MACINTOSH
#include <ctype.h>
#endif
#if defined(__BORLANDC__) && !defined(WIN32)
/* If this is DOS or win16, we'll fail anyway. */
/* Might as well assume win32. */
# define WIN32
#endif
#if defined(WIN32)
# include <windows.h>
# include "de_win.h"
#elif defined(MACINTOSH)
# include <console.h>
/* curses emulation. */
# define initscr()
# define endwin()
# define nonl()
# define noecho() csetmode(C_NOECHO, stdout)
# define cbreak() csetmode(C_CBREAK, stdout)
# define refresh()
# define addch(c) putchar(c)
# define standout() cinverse(1, stdout)
# define standend() cinverse(0, stdout)
# define move(line,col) cgotoxy(col + 1, line + 1, stdout)
# define clrtoeol() ccleol(stdout)
# define de_error(s) { fprintf(stderr, s); getchar(); }
# define LINES 25
# define COLS 80
#else
# include <curses.h>
# define de_error(s) { fprintf(stderr, s); sleep(2); }
#endif
#include "de_cmds.h"
/* List of line number to position mappings, in descending order. */
/* There may be holes. */
typedef struct LineMapRep {
int line;
size_t pos;
struct LineMapRep * previous;
} * line_map;
/* List of file versions, one per edit operation */
typedef struct HistoryRep {
CORD file_contents;
struct HistoryRep * previous;
line_map map; /* Invalid for first record "now" */
} * history;
history now = 0;
CORD current; /* == now -> file_contents. */
size_t current_len; /* Current file length. */
line_map current_map = 0; /* Current line no. to pos. map */
size_t current_map_size = 0; /* Number of current_map entries. */
/* Not always accurate, but reset */
/* by prune_map. */
# define MAX_MAP_SIZE 3000
/* Current display position */
int dis_line = 0;
int dis_col = 0;
# define ALL -1
# define NONE - 2
int need_redisplay = 0; /* Line that needs to be redisplayed. */
/* Current cursor position. Always within file. */
int line = 0;
int col = 0;
size_t file_pos = 0; /* Character position corresponding to cursor. */
/* Invalidate line map for lines > i */
void invalidate_map(int i)
{
while(current_map -> line > i) {
current_map = current_map -> previous;
current_map_size--;
}
}
/* Reduce the number of map entries to save space for huge files. */
/* This also affects maps in histories. */
void prune_map()
{
line_map map = current_map;
int start_line = map -> line;
current_map_size = 0;
for(; map != 0; map = map -> previous) {
current_map_size++;
if (map -> line < start_line - LINES && map -> previous != 0) {
map -> previous = map -> previous -> previous;
}
}
}
/* Add mapping entry */
void add_map(int line, size_t pos)
{
line_map new_map = GC_NEW(struct LineMapRep);
if (current_map_size >= MAX_MAP_SIZE) prune_map();
new_map -> line = line;
new_map -> pos = pos;
new_map -> previous = current_map;
current_map = new_map;
current_map_size++;
}
/* Return position of column *c of ith line in */
/* current file. Adjust *c to be within the line.*/
/* A 0 pointer is taken as 0 column. */
/* Returns CORD_NOT_FOUND if i is too big. */
/* Assumes i > dis_line. */
size_t line_pos(int i, int *c)
{
int j;
size_t cur;
size_t next;
line_map map = current_map;
while (map -> line > i) map = map -> previous;
if (map -> line < i - 2) /* rebuild */ invalidate_map(i);
for (j = map -> line, cur = map -> pos; j < i;) {
cur = CORD_chr(current, cur, '\n');
if (cur == current_len-1) return(CORD_NOT_FOUND);
cur++;
if (++j > current_map -> line) add_map(j, cur);
}
if (c != 0) {
next = CORD_chr(current, cur, '\n');
if (next == CORD_NOT_FOUND) next = current_len - 1;
if (next < cur + *c) {
*c = next - cur;
}
cur += *c;
}
return(cur);
}
void add_hist(CORD s)
{
history new_file = GC_NEW(struct HistoryRep);
new_file -> file_contents = current = s;
current_len = CORD_len(s);
new_file -> previous = now;
if (now != 0) now -> map = current_map;
now = new_file;
}
void del_hist(void)
{
now = now -> previous;
current = now -> file_contents;
current_map = now -> map;
current_len = CORD_len(current);
}
/* Current screen_contents; a dynamically allocated array of CORDs */
CORD * screen = 0;
int screen_size = 0;
# ifndef WIN32
/* Replace a line in the curses stdscr. All control characters are */
/* displayed as upper case characters in standout mode. This isn't */
/* terribly appropriate for tabs. */
void replace_line(int i, CORD s)
{
register int c;
CORD_pos p;
size_t len = CORD_len(s);
if (screen == 0 || LINES > screen_size) {
screen_size = LINES;
screen = (CORD *)GC_MALLOC(screen_size * sizeof(CORD));
}
# if !defined(MACINTOSH)
/* A gross workaround for an apparent curses bug: */
if (i == LINES-1 && len == COLS) {
s = CORD_substr(s, 0, CORD_len(s) - 1);
}
# endif
if (CORD_cmp(screen[i], s) != 0) {
move(i, 0); clrtoeol(); move(i,0);
CORD_FOR (p, s) {
c = CORD_pos_fetch(p) & 0x7f;
if (iscntrl(c)) {
standout(); addch(c + 0x40); standend();
} else {
addch(c);
}
}
screen[i] = s;
}
}
#else
# define replace_line(i,s) invalidate_line(i)
#endif
/* Return up to COLS characters of the line of s starting at pos, */
/* returning only characters after the given column. */
CORD retrieve_line(CORD s, size_t pos, unsigned column)
{
CORD candidate = CORD_substr(s, pos, column + COLS);
/* avoids scanning very long lines */
int eol = CORD_chr(candidate, 0, '\n');
int len;
if (eol == CORD_NOT_FOUND) eol = CORD_len(candidate);
len = (int)eol - (int)column;
if (len < 0) len = 0;
return(CORD_substr(s, pos + column, len));
}
# ifdef WIN32
# define refresh();
CORD retrieve_screen_line(int i)
{
register size_t pos;
invalidate_map(dis_line + LINES); /* Prune search */
pos = line_pos(dis_line + i, 0);
if (pos == CORD_NOT_FOUND) return(CORD_EMPTY);
return(retrieve_line(current, pos, dis_col));
}
# endif
/* Display the visible section of the current file */
void redisplay(void)
{
register int i;
invalidate_map(dis_line + LINES); /* Prune search */
for (i = 0; i < LINES; i++) {
if (need_redisplay == ALL || need_redisplay == i) {
register size_t pos = line_pos(dis_line + i, 0);
if (pos == CORD_NOT_FOUND) break;
replace_line(i, retrieve_line(current, pos, dis_col));
if (need_redisplay == i) goto done;
}
}
for (; i < LINES; i++) replace_line(i, CORD_EMPTY);
done:
refresh();
need_redisplay = NONE;
}
int dis_granularity;
/* Update dis_line, dis_col, and dis_pos to make cursor visible. */
/* Assumes line, col, dis_line, dis_pos are in bounds. */
void normalize_display()
{
int old_line = dis_line;
int old_col = dis_col;
dis_granularity = 1;
if (LINES > 15 && COLS > 15) dis_granularity = 2;
while (dis_line > line) dis_line -= dis_granularity;
while (dis_col > col) dis_col -= dis_granularity;
while (line >= dis_line + LINES) dis_line += dis_granularity;
while (col >= dis_col + COLS) dis_col += dis_granularity;
if (old_line != dis_line || old_col != dis_col) {
need_redisplay = ALL;
}
}
# if defined(WIN32)
# elif defined(MACINTOSH)
# define move_cursor(x,y) cgotoxy(x + 1, y + 1, stdout)
# else
# define move_cursor(x,y) move(y,x)
# endif
/* Adjust display so that cursor is visible; move cursor into position */
/* Update screen if necessary. */
void fix_cursor(void)
{
normalize_display();
if (need_redisplay != NONE) redisplay();
move_cursor(col - dis_col, line - dis_line);
refresh();
# ifndef WIN32
fflush(stdout);
# endif
}
/* Make sure line, col, and dis_pos are somewhere inside file. */
/* Recompute file_pos. Assumes dis_pos is accurate or past eof */
void fix_pos()
{
int my_col = col;
if ((size_t)line > current_len) line = current_len;
file_pos = line_pos(line, &my_col);
if (file_pos == CORD_NOT_FOUND) {
for (line = current_map -> line, file_pos = current_map -> pos;
file_pos < current_len;
line++, file_pos = CORD_chr(current, file_pos, '\n') + 1);
line--;
file_pos = line_pos(line, &col);
} else {
col = my_col;
}
}
#if defined(WIN32)
# define beep() Beep(1000 /* Hz */, 300 /* msecs */)
#elif defined(MACINTOSH)
# define beep() SysBeep(1)
#else
/*
* beep() is part of some curses packages and not others.
* We try to match the type of the builtin one, if any.
*/
#ifdef __STDC__
int beep(void)
#else
int beep()
#endif
{
putc('\007', stderr);
return(0);
}
#endif
# define NO_PREFIX -1
# define BARE_PREFIX -2
int repeat_count = NO_PREFIX; /* Current command prefix. */
int locate_mode = 0; /* Currently between 2 ^Ls */
CORD locate_string = CORD_EMPTY; /* Current search string. */
char * arg_file_name;
#ifdef WIN32
/* Change the current position to whatever is currently displayed at */
/* the given SCREEN coordinates. */
void set_position(int c, int l)
{
line = l + dis_line;
col = c + dis_col;
fix_pos();
move_cursor(col - dis_col, line - dis_line);
}
#endif /* WIN32 */
/* Perform the command associated with character c. C may be an */
/* integer > 256 denoting a windows command, one of the above control */
/* characters, or another ASCII character to be used as either a */
/* character to be inserted, a repeat count, or a search string, */
/* depending on the current state. */
void do_command(int c)
{
int i;
int need_fix_pos;
FILE * out;
if ( c == '\r') c = '\n';
if (locate_mode) {
size_t new_pos;
if (c == LOCATE) {
locate_mode = 0;
locate_string = CORD_EMPTY;
return;
}
locate_string = CORD_cat_char(locate_string, (char)c);
new_pos = CORD_str(current, file_pos - CORD_len(locate_string) + 1,
locate_string);
if (new_pos != CORD_NOT_FOUND) {
need_redisplay = ALL;
new_pos += CORD_len(locate_string);
for (;;) {
file_pos = line_pos(line + 1, 0);
if (file_pos > new_pos) break;
line++;
}
col = new_pos - line_pos(line, 0);
file_pos = new_pos;
fix_cursor();
} else {
locate_string = CORD_substr(locate_string, 0,
CORD_len(locate_string) - 1);
beep();
}
return;
}
if (c == REPEAT) {
repeat_count = BARE_PREFIX; return;
} else if (c < 0x100 && isdigit(c)){
if (repeat_count == BARE_PREFIX) {
repeat_count = c - '0'; return;
} else if (repeat_count != NO_PREFIX) {
repeat_count = 10 * repeat_count + c - '0'; return;
}
}
if (repeat_count == NO_PREFIX) repeat_count = 1;
if (repeat_count == BARE_PREFIX && (c == UP || c == DOWN)) {
repeat_count = LINES - dis_granularity;
}
if (repeat_count == BARE_PREFIX) repeat_count = 8;
need_fix_pos = 0;
for (i = 0; i < repeat_count; i++) {
switch(c) {
case LOCATE:
locate_mode = 1;
break;
case TOP:
line = col = file_pos = 0;
break;
case UP:
if (line != 0) {
line--;
need_fix_pos = 1;
}
break;
case DOWN:
line++;
need_fix_pos = 1;
break;
case LEFT:
if (col != 0) {
col--; file_pos--;
}
break;
case RIGHT:
if (CORD_fetch(current, file_pos) == '\n') break;
col++; file_pos++;
break;
case UNDO:
del_hist();
need_redisplay = ALL; need_fix_pos = 1;
break;
case BS:
if (col == 0) {
beep();
break;
}
col--; file_pos--;
/* fall through: */
case DEL:
if (file_pos == current_len-1) break;
/* Can't delete trailing newline */
if (CORD_fetch(current, file_pos) == '\n') {
need_redisplay = ALL; need_fix_pos = 1;
} else {
need_redisplay = line - dis_line;
}
add_hist(CORD_cat(
CORD_substr(current, 0, file_pos),
CORD_substr(current, file_pos+1, current_len)));
invalidate_map(line);
break;
case WRITE:
{
CORD name = CORD_cat(CORD_from_char_star(arg_file_name),
".new");
if ((out = fopen(CORD_to_const_char_star(name), "wb")) == NULL
|| CORD_put(current, out) == EOF) {
de_error("Write failed\n");
need_redisplay = ALL;
} else {
fclose(out);
}
}
break;
default:
{
CORD left_part = CORD_substr(current, 0, file_pos);
CORD right_part = CORD_substr(current, file_pos, current_len);
add_hist(CORD_cat(CORD_cat_char(left_part, (char)c),
right_part));
invalidate_map(line);
if (c == '\n') {
col = 0; line++; file_pos++;
need_redisplay = ALL;
} else {
col++; file_pos++;
need_redisplay = line - dis_line;
}
break;
}
}
}
if (need_fix_pos) fix_pos();
fix_cursor();
repeat_count = NO_PREFIX;
}
/* OS independent initialization */
void generic_init(void)
{
FILE * f;
CORD initial;
if ((f = fopen(arg_file_name, "rb")) == NULL) {
initial = "\n";
} else {
initial = CORD_from_file(f);
if (initial == CORD_EMPTY
|| CORD_fetch(initial, CORD_len(initial)-1) != '\n') {
initial = CORD_cat(initial, "\n");
}
}
add_map(0,0);
add_hist(initial);
now -> map = current_map;
now -> previous = now; /* Can't back up further: beginning of the world */
need_redisplay = ALL;
fix_cursor();
}
#ifndef WIN32
main(argc, argv)
int argc;
char ** argv;
{
int c;
#if defined(MACINTOSH)
console_options.title = "\pDumb Editor";
cshow(stdout);
argc = ccommand(&argv);
#endif
GC_INIT();
if (argc != 2) goto usage;
arg_file_name = argv[1];
setvbuf(stdout, GC_MALLOC_ATOMIC(8192), _IOFBF, 8192);
initscr();
noecho(); nonl(); cbreak();
generic_init();
while ((c = getchar()) != QUIT) {
if (c == EOF) break;
do_command(c);
}
done:
move(LINES-1, 0);
clrtoeol();
refresh();
nl();
echo();
endwin();
exit(0);
usage:
fprintf(stderr, "Usage: %s file\n", argv[0]);
fprintf(stderr, "Cursor keys: ^B(left) ^F(right) ^P(up) ^N(down)\n");
fprintf(stderr, "Undo: ^U Write to <file>.new: ^W");
fprintf(stderr, "Quit:^D Repeat count: ^R[n]\n");
fprintf(stderr, "Top: ^T Locate (search, find): ^L text ^L\n");
exit(1);
}
#endif /* !WIN32 */
-33
View File
@@ -1,33 +0,0 @@
/*
* Copyright (c) 1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, May 19, 1994 2:24 pm PDT */
#ifndef DE_CMDS_H
# define DE_CMDS_H
# define UP 16 /* ^P */
# define DOWN 14 /* ^N */
# define LEFT 2 /* ^B */
# define RIGHT 6 /* ^F */
# define DEL 127 /* ^? */
# define BS 8 /* ^H */
# define UNDO 21 /* ^U */
# define WRITE 23 /* ^W */
# define QUIT 4 /* ^D */
# define REPEAT 18 /* ^R */
# define LOCATE 12 /* ^L */
# define TOP 20 /* ^T */
#endif
Binary file not shown.

Before

Width:  |  Height:  |  Size: 766 B

-78
View File
@@ -1,78 +0,0 @@
/*
* Copyright (c) 1991-1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to copy this garbage collector for any purpose,
* provided the above notices are retained on all copies.
*/
/* Boehm, May 13, 1994 9:50 am PDT */
#include "windows.h"
#include "de_cmds.h"
#include "de_win.h"
ABOUTBOX DIALOG 19, 21, 163, 47
STYLE DS_MODALFRAME | WS_POPUP | WS_CAPTION | WS_SYSMENU
CAPTION "About Demonstration Text Editor"
BEGIN
ICON "DE", -1, 8, 8, 13, 13, WS_CHILD | WS_VISIBLE
LTEXT "Demonstration Text Editor", -1, 44, 8, 118, 8, WS_CHILD | WS_VISIBLE | WS_GROUP
LTEXT "Version 4.1", -1, 44, 16, 60, 8, WS_CHILD | WS_VISIBLE | WS_GROUP
PUSHBUTTON "OK", IDOK, 118, 27, 24, 14, WS_CHILD | WS_VISIBLE | WS_TABSTOP
END
DE MENU
BEGIN
POPUP "&File"
BEGIN
MENUITEM "&Save\t^W", IDM_FILESAVE
MENUITEM "E&xit\t^D", IDM_FILEEXIT
END
POPUP "&Edit"
BEGIN
MENUITEM "Page &Down\t^R^N", IDM_EDITPDOWN
MENUITEM "Page &Up\t^R^P", IDM_EDITPUP
MENUITEM "U&ndo\t^U", IDM_EDITUNDO
MENUITEM "&Locate\t^L ... ^L", IDM_EDITLOCATE
MENUITEM "D&own\t^N", IDM_EDITDOWN
MENUITEM "U&p\t^P", IDM_EDITUP
MENUITEM "Le&ft\t^B", IDM_EDITLEFT
MENUITEM "&Right\t^F", IDM_EDITRIGHT
MENUITEM "Delete &Backward\tBS", IDM_EDITBS
MENUITEM "Delete F&orward\tDEL", IDM_EDITDEL
MENUITEM "&Top\t^T", IDM_EDITTOP
END
POPUP "&Help"
BEGIN
MENUITEM "&Contents", IDM_HELPCONTENTS
MENUITEM "&About...", IDM_HELPABOUT
END
MENUITEM "Page_&Down", IDM_EDITPDOWN
MENUITEM "Page_&Up", IDM_EDITPUP
END
DE ACCELERATORS
BEGIN
"^R", IDM_EDITREPEAT
"^N", IDM_EDITDOWN
"^P", IDM_EDITUP
"^L", IDM_EDITLOCATE
"^B", IDM_EDITLEFT
"^F", IDM_EDITRIGHT
"^T", IDM_EDITTOP
VK_DELETE, IDM_EDITDEL, VIRTKEY
VK_BACK, IDM_EDITBS, VIRTKEY
END
DE ICON cord\de_win.ICO
-366
View File
@@ -1,366 +0,0 @@
/*
* Copyright (c) 1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, February 6, 1995 12:29 pm PST */
/*
* The MS Windows specific part of de.
* This started as the generic Windows application template
* made available by Rob Haack ([email protected]), but
* significant parts didn't survive to the final version.
*
* This was written by a nonexpert windows programmer.
*/
#include "windows.h"
#include "gc.h"
#include "cord.h"
#include "de_cmds.h"
#include "de_win.h"
int LINES = 0;
int COLS = 0;
char szAppName[] = "DE";
char FullAppName[] = "Demonstration Editor";
HWND hwnd;
void de_error(char *s)
{
MessageBox( hwnd, (LPSTR) s,
(LPSTR) FullAppName,
MB_ICONINFORMATION | MB_OK );
InvalidateRect(hwnd, NULL, TRUE);
}
int APIENTRY WinMain (HINSTANCE hInstance, HINSTANCE hPrevInstance,
LPSTR command_line, int nCmdShow)
{
MSG msg;
WNDCLASS wndclass;
HANDLE hAccel;
if (!hPrevInstance)
{
wndclass.style = CS_HREDRAW | CS_VREDRAW;
wndclass.lpfnWndProc = WndProc;
wndclass.cbClsExtra = 0;
wndclass.cbWndExtra = DLGWINDOWEXTRA;
wndclass.hInstance = hInstance;
wndclass.hIcon = LoadIcon (hInstance, szAppName);
wndclass.hCursor = LoadCursor (NULL, IDC_ARROW);
wndclass.hbrBackground = GetStockObject(WHITE_BRUSH);
wndclass.lpszMenuName = "DE";
wndclass.lpszClassName = szAppName;
if (RegisterClass (&wndclass) == 0) {
char buf[50];
sprintf(buf, "RegisterClass: error code: 0x%X", GetLastError());
de_error(buf);
return(0);
}
}
/* Empirically, the command line does not include the command name ...
if (command_line != 0) {
while (isspace(*command_line)) command_line++;
while (*command_line != 0 && !isspace(*command_line)) command_line++;
while (isspace(*command_line)) command_line++;
} */
if (command_line == 0 || *command_line == 0) {
de_error("File name argument required");
return( 0 );
} else {
char *p = command_line;
while (*p != 0 && !isspace(*p)) p++;
arg_file_name = CORD_to_char_star(
CORD_substr(command_line, 0, p - command_line));
}
hwnd = CreateWindow (szAppName,
FullAppName,
WS_OVERLAPPEDWINDOW | WS_CAPTION, /* Window style */
CW_USEDEFAULT, 0, /* default pos. */
CW_USEDEFAULT, 0, /* default width, height */
NULL, /* No parent */
NULL, /* Window class menu */
hInstance, NULL);
if (hwnd == NULL) {
char buf[50];
sprintf(buf, "CreateWindow: error code: 0x%X", GetLastError());
de_error(buf);
return(0);
}
ShowWindow (hwnd, nCmdShow);
hAccel = LoadAccelerators( hInstance, szAppName );
while (GetMessage (&msg, NULL, 0, 0))
{
if( !TranslateAccelerator( hwnd, hAccel, &msg ) )
{
TranslateMessage (&msg);
DispatchMessage (&msg);
}
}
return msg.wParam;
}
/* Return the argument with all control characters replaced by blanks. */
char * plain_chars(char * text, size_t len)
{
char * result = GC_MALLOC_ATOMIC(len + 1);
register size_t i;
for (i = 0; i < len; i++) {
if (iscntrl(text[i])) {
result[i] = ' ';
} else {
result[i] = text[i];
}
}
result[len] = '\0';
return(result);
}
/* Return the argument with all non-control-characters replaced by */
/* blank, and all control characters c replaced by c + 32. */
char * control_chars(char * text, size_t len)
{
char * result = GC_MALLOC_ATOMIC(len + 1);
register size_t i;
for (i = 0; i < len; i++) {
if (iscntrl(text[i])) {
result[i] = text[i] + 0x40;
} else {
result[i] = ' ';
}
}
result[len] = '\0';
return(result);
}
int char_width;
int char_height;
void get_line_rect(int line, int win_width, RECT * rectp)
{
rectp -> top = line * char_height;
rectp -> bottom = rectp->top + char_height;
rectp -> left = 0;
rectp -> right = win_width;
}
int caret_visible = 0; /* Caret is currently visible. */
int screen_was_painted = 0;/* Screen has been painted at least once. */
void update_cursor(void);
LRESULT CALLBACK WndProc (HWND hwnd, UINT message,
WPARAM wParam, LPARAM lParam)
{
static FARPROC lpfnAboutBox;
static HANDLE hInstance;
HDC dc;
PAINTSTRUCT ps;
RECT client_area;
RECT this_line;
RECT dummy;
TEXTMETRIC tm;
register int i;
int id;
switch (message)
{
case WM_CREATE:
hInstance = ( (LPCREATESTRUCT) lParam)->hInstance;
lpfnAboutBox = MakeProcInstance( (FARPROC) AboutBox, hInstance );
dc = GetDC(hwnd);
SelectObject(dc, GetStockObject(SYSTEM_FIXED_FONT));
GetTextMetrics(dc, &tm);
ReleaseDC(hwnd, dc);
char_width = tm.tmAveCharWidth;
char_height = tm.tmHeight + tm.tmExternalLeading;
GetClientRect(hwnd, &client_area);
COLS = (client_area.right - client_area.left)/char_width;
LINES = (client_area.bottom - client_area.top)/char_height;
generic_init();
return(0);
case WM_CHAR:
if (wParam == QUIT) {
SendMessage( hwnd, WM_CLOSE, 0, 0L );
} else {
do_command(wParam);
}
return(0);
case WM_SETFOCUS:
CreateCaret(hwnd, NULL, char_width, char_height);
ShowCaret(hwnd);
caret_visible = 1;
update_cursor();
return(0);
case WM_KILLFOCUS:
HideCaret(hwnd);
DestroyCaret();
caret_visible = 0;
return(0);
case WM_LBUTTONUP:
{
unsigned xpos = LOWORD(lParam); /* From left */
unsigned ypos = HIWORD(lParam); /* from top */
set_position( xpos/char_width, ypos/char_height );
return(0);
}
case WM_COMMAND:
id = LOWORD(wParam);
if (id & EDIT_CMD_FLAG) {
if (id & REPEAT_FLAG) do_command(REPEAT);
do_command(CHAR_CMD(id));
return( 0 );
} else {
switch(id) {
case IDM_FILEEXIT:
SendMessage( hwnd, WM_CLOSE, 0, 0L );
return( 0 );
case IDM_HELPABOUT:
if( DialogBox( hInstance, "ABOUTBOX",
hwnd, lpfnAboutBox ) )
InvalidateRect( hwnd, NULL, TRUE );
return( 0 );
case IDM_HELPCONTENTS:
de_error(
"Cursor keys: ^B(left) ^F(right) ^P(up) ^N(down)\n"
"Undo: ^U Write: ^W Quit:^D Repeat count: ^R[n]\n"
"Top: ^T Locate (search, find): ^L text ^L\n");
return( 0 );
}
}
break;
case WM_CLOSE:
DestroyWindow( hwnd );
return 0;
case WM_DESTROY:
PostQuitMessage (0);
GC_win32_free_heap();
return 0;
case WM_PAINT:
dc = BeginPaint(hwnd, &ps);
GetClientRect(hwnd, &client_area);
COLS = (client_area.right - client_area.left)/char_width;
LINES = (client_area.bottom - client_area.top)/char_height;
SelectObject(dc, GetStockObject(SYSTEM_FIXED_FONT));
for (i = 0; i < LINES; i++) {
get_line_rect(i, client_area.right, &this_line);
if (IntersectRect(&dummy, &this_line, &ps.rcPaint)) {
CORD raw_line = retrieve_screen_line(i);
size_t len = CORD_len(raw_line);
char * text = CORD_to_char_star(raw_line);
/* May contain embedded NULLs */
char * plain = plain_chars(text, len);
char * blanks = CORD_to_char_star(CORD_chars(' ',
COLS - len));
char * control = control_chars(text, len);
# define RED RGB(255,0,0)
SetBkMode(dc, OPAQUE);
SetTextColor(dc, GetSysColor(COLOR_WINDOWTEXT));
TextOut(dc, this_line.left, this_line.top,
plain, len);
TextOut(dc, this_line.left + len * char_width, this_line.top,
blanks, COLS - len);
SetBkMode(dc, TRANSPARENT);
SetTextColor(dc, RED);
TextOut(dc, this_line.left, this_line.top,
control, strlen(control));
}
}
EndPaint(hwnd, &ps);
screen_was_painted = 1;
return 0;
}
return DefWindowProc (hwnd, message, wParam, lParam);
}
int last_col;
int last_line;
void move_cursor(int c, int l)
{
last_col = c;
last_line = l;
if (caret_visible) update_cursor();
}
void update_cursor(void)
{
SetCaretPos(last_col * char_width, last_line * char_height);
ShowCaret(hwnd);
}
void invalidate_line(int i)
{
RECT line;
if (!screen_was_painted) return;
/* Invalidating a rectangle before painting seems result in a */
/* major performance problem. */
get_line_rect(i, COLS*char_width, &line);
InvalidateRect(hwnd, &line, FALSE);
}
LRESULT CALLBACK AboutBox( HWND hDlg, UINT message,
WPARAM wParam, LPARAM lParam )
{
switch( message )
{
case WM_INITDIALOG:
SetFocus( GetDlgItem( hDlg, IDOK ) );
break;
case WM_COMMAND:
switch( wParam )
{
case IDOK:
EndDialog( hDlg, TRUE );
break;
}
break;
case WM_CLOSE:
EndDialog( hDlg, TRUE );
return TRUE;
}
return FALSE;
}
-103
View File
@@ -1,103 +0,0 @@
/*
* Copyright (c) 1994 by Xerox Corporation. All rights reserved.
*
* THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
* OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
*
* Permission is hereby granted to use or copy this program
* for any purpose, provided the above notices are retained on all copies.
* Permission to modify the code and to distribute modified code is granted,
* provided the above notices are retained, and a notice that the code was
* modified is included with the above copyright notice.
*/
/* Boehm, May 19, 1994 2:25 pm PDT */
/* cord.h, de_cmds.h, and windows.h should be included before this. */
# define OTHER_FLAG 0x100
# define EDIT_CMD_FLAG 0x200
# define REPEAT_FLAG 0x400
# define CHAR_CMD(i) ((i) & 0xff)
/* MENU: DE */
#define IDM_FILESAVE (EDIT_CMD_FLAG + WRITE)
#define IDM_FILEEXIT (OTHER_FLAG + 1)
#define IDM_HELPABOUT (OTHER_FLAG + 2)
#define IDM_HELPCONTENTS (OTHER_FLAG + 3)
#define IDM_EDITPDOWN (REPEAT_FLAG + EDIT_CMD_FLAG + DOWN)
#define IDM_EDITPUP (REPEAT_FLAG + EDIT_CMD_FLAG + UP)
#define IDM_EDITUNDO (EDIT_CMD_FLAG + UNDO)
#define IDM_EDITLOCATE (EDIT_CMD_FLAG + LOCATE)
#define IDM_EDITDOWN (EDIT_CMD_FLAG + DOWN)
#define IDM_EDITUP (EDIT_CMD_FLAG + UP)
#define IDM_EDITLEFT (EDIT_CMD_FLAG + LEFT)
#define IDM_EDITRIGHT (EDIT_CMD_FLAG + RIGHT)
#define IDM_EDITBS (EDIT_CMD_FLAG + BS)
#define IDM_EDITDEL (EDIT_CMD_FLAG + DEL)
#define IDM_EDITREPEAT (EDIT_CMD_FLAG + REPEAT)
#define IDM_EDITTOP (EDIT_CMD_FLAG + TOP)
/* Windows UI stuff */
LRESULT CALLBACK WndProc (HWND hwnd, UINT message,
UINT wParam, LONG lParam);
LRESULT CALLBACK AboutBox( HWND hDlg, UINT message,
UINT wParam, LONG lParam );
/* Screen dimensions. Maintained by de_win.c. */
extern int LINES;
extern int COLS;
/* File being edited. */
extern char * arg_file_name;
/* Current display position in file. Maintained by de.c */
extern int dis_line;
extern int dis_col;
/* Current cursor position in file. */
extern int line;
extern int col;
/*
* Calls from de_win.c to de.c
*/
CORD retrieve_screen_line(int i);
/* Get the contents of i'th screen line. */
/* Relies on COLS. */
void set_position(int x, int y);
/* Set column, row. Upper left of window = (0,0). */
void do_command(int);
/* Execute an editor command. */
/* Agument is a command character or one */
/* of the IDM_ commands. */
void generic_init(void);
/* OS independent initialization */
/*
* Calls from de.c to de_win.c
*/
void move_cursor(int column, int line);
/* Physically move the cursor on the display, */
/* so that it appears at */
/* (column, line). */
void invalidate_line(int line);
/* Invalidate line i on the screen. */
void de_error(char *s);
/* Display error message. */
-736
View File
@@ -1,736 +0,0 @@
#include "private/pthread_support.h"
# if defined(GC_DARWIN_THREADS)
#include <AvailabilityMacros.h>
#include "mono/utils/mono-compiler.h"
/* From "Inside Mac OS X - Mach-O Runtime Architecture" published by Apple
Page 49:
"The space beneath the stack pointer, where a new stack frame would normally
be allocated, is called the red zone. This area as shown in Figure 3-2 may
be used for any purpose as long as a new stack frame does not need to be
added to the stack."
Page 50: "If a leaf procedure's red zone usage would exceed 224 bytes, then
it must set up a stack frame just like routines that call other routines."
*/
#ifdef POWERPC
# if CPP_WORDSZ == 32
# define PPC_RED_ZONE_SIZE 224
# elif CPP_WORDSZ == 64
# define PPC_RED_ZONE_SIZE 320
# endif
#endif
typedef struct StackFrame {
unsigned long savedSP;
unsigned long savedCR;
unsigned long savedLR;
unsigned long reserved[2];
unsigned long savedRTOC;
} StackFrame;
unsigned long FindTopOfStack(unsigned int stack_start) {
StackFrame *frame;
if (stack_start == 0) {
# ifdef POWERPC
# if CPP_WORDSZ == 32
__asm__ volatile("lwz %0,0(r1)" : "=r" (frame));
# else
__asm__ volatile("ldz %0,0(r1)" : "=r" (frame));
# endif
# endif
} else {
frame = (StackFrame *)stack_start;
}
# ifdef DEBUG_THREADS
/* GC_printf1("FindTopOfStack start at sp = %p\n", frame); */
# endif
do {
if (frame->savedSP == 0) break;
/* if there are no more stack frames, stop */
frame = (StackFrame*)frame->savedSP;
/* we do these next two checks after going to the next frame
because the LR for the first stack frame in the loop
is not set up on purpose, so we shouldn't check it. */
if ((frame->savedLR & ~3) == 0) break; /* if the next LR is bogus, stop */
if ((~(frame->savedLR) & ~3) == 0) break; /* ditto */
} while (1);
# ifdef DEBUG_THREADS
/* GC_printf1("FindTopOfStack finish at sp = %p\n", frame); */
# endif
return (unsigned long)frame;
}
#ifdef DARWIN_DONT_PARSE_STACK
void GC_push_all_stacks() {
int i;
kern_return_t r;
GC_thread p;
pthread_t me;
ptr_t lo, hi, altstack_lo, altstack_hi;
#if defined(POWERPC)
ppc_thread_state_t state;
mach_msg_type_number_t thread_state_count = PPC_THREAD_STATE_COUNT;
#elif defined(I386)
i386_thread_state_t state;
mach_msg_type_number_t thread_state_count = i386_THREAD_STATE_COUNT;
#elif defined(ARM)
arm_thread_state_t state;
mach_msg_type_number_t thread_state_count = ARM_THREAD_STATE_COUNT;
#elif defined(X86_64)
x86_thread_state64_t state;
mach_msg_type_number_t thread_state_count = x86_THREAD_STATE64_COUNT;
#else
# error FIXME for non-x86 || ppc architectures
mach_msg_type_number_t thread_state_count = MACHINE_THREAD_STATE_COUNT;
#endif
me = pthread_self();
if (!GC_thr_initialized) GC_thr_init();
for(i=0;i<THREAD_TABLE_SZ;i++) {
for(p=GC_threads[i];p!=0;p=p->next) {
if(p -> flags & FINISHED) continue;
if(pthread_equal(p->id,me)) {
lo = GC_approx_sp();
} else {
/* Get the thread state (registers, etc) */
r = thread_get_state(
p->stop_info.mach_thread,
GC_MACH_THREAD_STATE_FLAVOR,
(natural_t*)&state,
&thread_state_count);
if(r != KERN_SUCCESS) continue;
#if defined(I386)
#if defined (TARGET_IOS) || (MAC_OS_X_VERSION_MIN_REQUIRED >= MAC_OS_X_VERSION_10_5)
lo = state.__esp;
GC_push_one(state.__eax);
GC_push_one(state.__ebx);
GC_push_one(state.__ecx);
GC_push_one(state.__edx);
GC_push_one(state.__edi);
GC_push_one(state.__esi);
GC_push_one(state.__ebp);
#else
lo = state.esp;
GC_push_one(state.eax);
GC_push_one(state.ebx);
GC_push_one(state.ecx);
GC_push_one(state.edx);
GC_push_one(state.edi);
GC_push_one(state.esi);
GC_push_one(state.ebp);
#endif
#elif defined(X86_64)
lo = state.__rsp;
GC_push_one(state.__rax);
GC_push_one(state.__rbx);
GC_push_one(state.__rcx);
GC_push_one(state.__rdx);
GC_push_one(state.__rdi);
GC_push_one(state.__rsi);
GC_push_one(state.__rbp);
GC_push_one(state.__r8);
GC_push_one(state.__r9);
GC_push_one(state.__r10);
GC_push_one(state.__r11);
GC_push_one(state.__r12);
GC_push_one(state.__r13);
GC_push_one(state.__r14);
GC_push_one(state.__r15);
#elif defined(POWERPC)
#if defined(_STRUCT_PPC_EXCEPTION_STATE) && defined(__DARWIN_UNIX03)
lo = (void*)(state.__r1 - PPC_RED_ZONE_SIZE);
GC_push_one(state.__r0);
GC_push_one(state.__r2);
GC_push_one(state.__r3);
GC_push_one(state.__r4);
GC_push_one(state.__r5);
GC_push_one(state.__r6);
GC_push_one(state.__r7);
GC_push_one(state.__r8);
GC_push_one(state.__r9);
GC_push_one(state.__r10);
GC_push_one(state.__r11);
GC_push_one(state.__r12);
GC_push_one(state.__r13);
GC_push_one(state.__r14);
GC_push_one(state.__r15);
GC_push_one(state.__r16);
GC_push_one(state.__r17);
GC_push_one(state.__r18);
GC_push_one(state.__r19);
GC_push_one(state.__r20);
GC_push_one(state.__r21);
GC_push_one(state.__r22);
GC_push_one(state.__r23);
GC_push_one(state.__r24);
GC_push_one(state.__r25);
GC_push_one(state.__r26);
GC_push_one(state.__r27);
GC_push_one(state.__r28);
GC_push_one(state.__r29);
GC_push_one(state.__r30);
GC_push_one(state.__r31);
#else
lo = (void*)(state.r1 - PPC_RED_ZONE_SIZE);
GC_push_one(state.r0);
GC_push_one(state.r2);
GC_push_one(state.r3);
GC_push_one(state.r4);
GC_push_one(state.r5);
GC_push_one(state.r6);
GC_push_one(state.r7);
GC_push_one(state.r8);
GC_push_one(state.r9);
GC_push_one(state.r10);
GC_push_one(state.r11);
GC_push_one(state.r12);
GC_push_one(state.r13);
GC_push_one(state.r14);
GC_push_one(state.r15);
GC_push_one(state.r16);
GC_push_one(state.r17);
GC_push_one(state.r18);
GC_push_one(state.r19);
GC_push_one(state.r20);
GC_push_one(state.r21);
GC_push_one(state.r22);
GC_push_one(state.r23);
GC_push_one(state.r24);
GC_push_one(state.r25);
GC_push_one(state.r26);
GC_push_one(state.r27);
GC_push_one(state.r28);
GC_push_one(state.r29);
GC_push_one(state.r30);
GC_push_one(state.r31);
#endif
#elif defined(ARM)
lo = (void*)state.__sp;
GC_push_one(state.__r[0]);
GC_push_one(state.__r[1]);
GC_push_one(state.__r[2]);
GC_push_one(state.__r[3]);
GC_push_one(state.__r[4]);
GC_push_one(state.__r[5]);
GC_push_one(state.__r[6]);
GC_push_one(state.__r[7]);
GC_push_one(state.__r[8]);
GC_push_one(state.__r[9]);
GC_push_one(state.__r[10]);
GC_push_one(state.__r[11]);
GC_push_one(state.__r[12]);
/* GC_push_one(state.__sp); */
GC_push_one(state.__lr);
GC_push_one(state.__pc);
GC_push_one(state.__cpsr);
#else
# error FIXME for non-x86 || ppc architectures
#endif
} /* p != me */
if(p->flags & MAIN_THREAD)
hi = GC_stackbottom;
else
hi = p->stack_end;
if (p->altstack && lo >= p->altstack && lo <= p->altstack + p->altstack_size) {
altstack_lo = lo;
altstack_hi = p->altstack + p->altstack_size;
lo = (char*)p->stack;
hi = (char*)p->stack + p->stack_size;
} else {
altstack_lo = NULL;
}
#if DEBUG_THREADS
GC_printf3("Darwin: Stack for thread 0x%lx = [%lx,%lx)\n",
(unsigned long) p -> id,
(unsigned long) lo,
(unsigned long) hi
);
#endif
if (lo)
GC_push_all_stack(lo,hi);
if (altstack_lo)
GC_push_all_stack(altstack_lo,altstack_hi);
} /* for(p=GC_threads[i]...) */
} /* for(i=0;i<THREAD_TABLE_SZ...) */
}
#else /* !DARWIN_DONT_PARSE_STACK; Use FindTopOfStack() */
void GC_push_all_stacks() {
int i;
task_t my_task;
kern_return_t r;
mach_port_t me;
ptr_t lo, hi;
thread_act_array_t act_list = 0;
mach_msg_type_number_t listcount = 0;
me = mach_thread_self();
if (!GC_thr_initialized) GC_thr_init();
my_task = current_task();
r = task_threads(my_task, &act_list, &listcount);
if(r != KERN_SUCCESS) ABORT("task_threads failed");
for(i = 0; i < listcount; i++) {
thread_act_t thread = act_list[i];
if (thread == me) {
lo = GC_approx_sp();
hi = (ptr_t)FindTopOfStack(0);
} else {
# if defined(POWERPC)
# if CPP_WORDSZ == 32
ppc_thread_state_t info;
# else
ppc_thread_state64_t info;
# endif
mach_msg_type_number_t outCount = THREAD_STATE_MAX;
r = thread_get_state(thread, GC_MACH_THREAD_STATE_FLAVOR,
(natural_t *)&info, &outCount);
if(r != KERN_SUCCESS) continue;
#if defined(_STRUCT_PPC_EXCEPTION_STATE)
lo = (void*)(info.__r1 - PPC_RED_ZONE_SIZE);
hi = (ptr_t)FindTopOfStack(info.__r1);
GC_push_one(info.__r0);
GC_push_one(info.__r2);
GC_push_one(info.__r3);
GC_push_one(info.__r4);
GC_push_one(info.__r5);
GC_push_one(info.__r6);
GC_push_one(info.__r7);
GC_push_one(info.__r8);
GC_push_one(info.__r9);
GC_push_one(info.__r10);
GC_push_one(info.__r11);
GC_push_one(info.__r12);
GC_push_one(info.__r13);
GC_push_one(info.__r14);
GC_push_one(info.__r15);
GC_push_one(info.__r16);
GC_push_one(info.__r17);
GC_push_one(info.__r18);
GC_push_one(info.__r19);
GC_push_one(info.__r20);
GC_push_one(info.__r21);
GC_push_one(info.__r22);
GC_push_one(info.__r23);
GC_push_one(info.__r24);
GC_push_one(info.__r25);
GC_push_one(info.__r26);
GC_push_one(info.__r27);
GC_push_one(info.__r28);
GC_push_one(info.__r29);
GC_push_one(info.__r30);
GC_push_one(info.__r31);
#else
lo = (void*)(info.r1 - PPC_RED_ZONE_SIZE);
hi = (ptr_t)FindTopOfStack(info.r1);
GC_push_one(info.r0);
GC_push_one(info.r2);
GC_push_one(info.r3);
GC_push_one(info.r4);
GC_push_one(info.r5);
GC_push_one(info.r6);
GC_push_one(info.r7);
GC_push_one(info.r8);
GC_push_one(info.r9);
GC_push_one(info.r10);
GC_push_one(info.r11);
GC_push_one(info.r12);
GC_push_one(info.r13);
GC_push_one(info.r14);
GC_push_one(info.r15);
GC_push_one(info.r16);
GC_push_one(info.r17);
GC_push_one(info.r18);
GC_push_one(info.r19);
GC_push_one(info.r20);
GC_push_one(info.r21);
GC_push_one(info.r22);
GC_push_one(info.r23);
GC_push_one(info.r24);
GC_push_one(info.r25);
GC_push_one(info.r26);
GC_push_one(info.r27);
GC_push_one(info.r28);
GC_push_one(info.r29);
GC_push_one(info.r30);
GC_push_one(info.r31);
#endif
# elif defined(I386) /* !POWERPC */
/* FIXME: Remove after testing: */
WARN("This is completely untested and likely will not work\n", 0);
i386_thread_state_t info;
mach_msg_type_number_t outCount = THREAD_STATE_MAX;
r = thread_get_state(thread, GC_MACH_THREAD_STATE_FLAVOR,
(natural_t *)&info, &outCount);
if(r != KERN_SUCCESS) continue;
#if defined (TARGET_IOS) || (MAC_OS_X_VERSION_MIN_REQUIRED >= MAC_OS_X_VERSION_10_5)
lo = (void*)info.__esp;
hi = (ptr_t)FindTopOfStack(info.__esp);
GC_push_one(info.__eax);
GC_push_one(info.__ebx);
GC_push_one(info.__ecx);
GC_push_one(info.__edx);
GC_push_one(info.__edi);
GC_push_one(info.__esi);
GC_push_one(info.__ebp);
/* GC_push_one(info.__esp); */
GC_push_one(info.__ss);
GC_push_one(info.__eip);
GC_push_one(info.__cs);
GC_push_one(info.__ds);
GC_push_one(info.__es);
GC_push_one(info.__fs);
GC_push_one(info.__gs);
#else
lo = (void*)info.esp;
hi = (ptr_t)FindTopOfStack(info.esp);
GC_push_one(info.eax);
GC_push_one(info.ebx);
GC_push_one(info.ecx);
GC_push_one(info.edx);
GC_push_one(info.edi);
GC_push_one(info.esi);
GC_push_one(info.ebp);
/* GC_push_one(info.esp); */
GC_push_one(info.ss);
GC_push_one(info.eip);
GC_push_one(info.cs);
GC_push_one(info.ds);
GC_push_one(info.es);
GC_push_one(info.fs);
GC_push_one(info.gs);
#endif
# elif defined(ARM) /* !I386 */
arm_thread_state_t info;
mach_msg_type_number_t outCount = THREAD_STATE_MAX;
r = thread_get_state(thread, GC_MACH_THREAD_STATE_FLAVOR,
(natural_t *)&info, &outCount);
if(r != KERN_SUCCESS) continue;
lo = (void*)info.__sp;
hi = (ptr_t)FindTopOfStack(info.__sp);
GC_push_one(info.__r[0]);
GC_push_one(info.__r[1]);
GC_push_one(info.__r[2]);
GC_push_one(info.__r[3]);
GC_push_one(info.__r[4]);
GC_push_one(info.__r[5]);
GC_push_one(info.__r[6]);
GC_push_one(info.__r[7]);
GC_push_one(info.__r[8]);
GC_push_one(info.__r[9]);
GC_push_one(info.__r[10]);
GC_push_one(info.__r[11]);
GC_push_one(info.__r[12]);
/* GC_push_one(info.__sp); */
GC_push_one(info.__lr);
GC_push_one(info.__pc);
GC_push_one(info.__cpsr);
# endif /* !ARM */
}
# if DEBUG_THREADS
GC_printf3("Darwin: Stack for thread 0x%lx = [%lx,%lx)\n",
(unsigned long) thread,
(unsigned long) lo,
(unsigned long) hi
);
# endif
GC_push_all_stack(lo, hi);
mach_port_deallocate(my_task, thread);
} /* for(p=GC_threads[i]...) */
vm_deallocate(my_task, (vm_address_t)act_list, sizeof(thread_t) * listcount);
mach_port_deallocate(my_task, me);
}
#endif /* !DARWIN_DONT_PARSE_STACK */
static mach_port_t GC_mach_handler_thread;
static int GC_use_mach_handler_thread = 0;
#define SUSPEND_THREADS_SIZE 2048
static struct GC_mach_thread GC_mach_threads[SUSPEND_THREADS_SIZE];
static int GC_mach_threads_count;
void GC_stop_init() {
int i;
for (i = 0; i < SUSPEND_THREADS_SIZE; i++) {
GC_mach_threads[i].thread = 0;
GC_mach_threads[i].already_suspended = 0;
}
GC_mach_threads_count = 0;
}
/* returns true if there's a thread in act_list that wasn't in old_list */
int GC_suspend_thread_list(thread_act_array_t act_list, int count,
thread_act_array_t old_list, int old_count) {
mach_port_t my_thread = mach_thread_self();
int i, j;
int changed = 0;
for(i = 0; i < count; i++) {
thread_act_t thread = act_list[i];
# if DEBUG_THREADS
GC_printf1("Attempting to suspend thread %p\n", thread);
# endif
/* find the current thread in the old list */
int found = 0;
for(j = 0; j < old_count; j++) {
thread_act_t old_thread = old_list[j];
if (old_thread == thread) {
found = 1;
break;
}
}
if (!found) {
/* add it to the GC_mach_threads list */
GC_mach_threads[GC_mach_threads_count].thread = thread;
/* default is not suspended */
GC_mach_threads[GC_mach_threads_count].already_suspended = 0;
changed = 1;
}
if (thread != my_thread &&
(!GC_use_mach_handler_thread
|| (GC_use_mach_handler_thread
&& GC_mach_handler_thread != thread))) {
struct thread_basic_info info;
mach_msg_type_number_t outCount = THREAD_INFO_MAX;
kern_return_t kern_result = thread_info(thread, THREAD_BASIC_INFO,
(thread_info_t)&info, &outCount);
if(kern_result != KERN_SUCCESS) {
/* the thread may have quit since the thread_threads () call
* we mark already_suspended so it's not dealt with anymore later
*/
if (!found) {
GC_mach_threads[GC_mach_threads_count].already_suspended = TRUE;
GC_mach_threads_count++;
}
continue;
}
# if DEBUG_THREADS
GC_printf2("Thread state for 0x%lx = %d\n", thread, info.run_state);
# endif
if (!found) {
GC_mach_threads[GC_mach_threads_count].already_suspended = info.suspend_count;
}
if (info.suspend_count) continue;
# if DEBUG_THREADS
GC_printf1("Suspending 0x%lx\n", thread);
# endif
/* Suspend the thread */
kern_result = thread_suspend(thread);
if(kern_result != KERN_SUCCESS) {
/* the thread may have quit since the thread_threads () call
* we mark already_suspended so it's not dealt with anymore later
*/
if (!found) {
GC_mach_threads[GC_mach_threads_count].already_suspended = TRUE;
GC_mach_threads_count++;
}
continue;
}
}
if (!found) GC_mach_threads_count++;
}
mach_port_deallocate(current_task(), my_thread);
return changed;
}
/* Caller holds allocation lock. */
void GC_stop_world()
{
int i, changes;
GC_thread p;
task_t my_task = current_task();
mach_port_t my_thread = mach_thread_self();
kern_return_t kern_result;
thread_act_array_t act_list, prev_list;
mach_msg_type_number_t listcount, prevcount;
if (GC_notify_event)
GC_notify_event (GC_EVENT_PRE_STOP_WORLD);
GC_process_togglerefs ();
# if DEBUG_THREADS
GC_printf1("Stopping the world from 0x%lx\n", mach_thread_self());
# endif
/* clear out the mach threads list table */
GC_stop_init();
/* Make sure all free list construction has stopped before we start. */
/* No new construction can start, since free list construction is */
/* required to acquire and release the GC lock before it starts, */
/* and we have the lock. */
# ifdef PARALLEL_MARK
GC_acquire_mark_lock();
GC_ASSERT(GC_fl_builder_count == 0);
/* We should have previously waited for it to become zero. */
# endif /* PARALLEL_MARK */
/* Loop stopping threads until you have gone over the whole list
twice without a new one appearing. thread_create() won't
return (and thus the thread stop) until the new thread
exists, so there is no window whereby you could stop a
thread, recognise it is stopped, but then have a new thread
it created before stopping show up later.
*/
changes = 1;
prev_list = NULL;
prevcount = 0;
do {
int result;
kern_result = task_threads(my_task, &act_list, &listcount);
if(kern_result == KERN_SUCCESS) {
result = GC_suspend_thread_list(act_list, listcount,
prev_list, prevcount);
changes = result;
if(prev_list != NULL) {
for(i = 0; i < prevcount; i++)
mach_port_deallocate(my_task, prev_list[i]);
vm_deallocate(my_task, (vm_address_t)prev_list, sizeof(thread_t) * prevcount);
}
prev_list = act_list;
prevcount = listcount;
}
} while (changes);
for(i = 0; i < listcount; i++)
mach_port_deallocate(my_task, act_list[i]);
vm_deallocate(my_task, (vm_address_t)act_list, sizeof(thread_t) * listcount);
# ifdef MPROTECT_VDB
if(GC_incremental) {
extern void GC_mprotect_stop();
GC_mprotect_stop();
}
# endif
# ifdef PARALLEL_MARK
GC_release_mark_lock();
# endif
#if DEBUG_THREADS
GC_printf1("World stopped from 0x%lx\n", my_thread);
#endif
mach_port_deallocate(my_task, my_thread);
if (GC_notify_event)
GC_notify_event (GC_EVENT_POST_STOP_WORLD);
}
/* Caller holds allocation lock, and has held it continuously since */
/* the world stopped. */
void GC_start_world()
{
task_t my_task = current_task();
mach_port_t my_thread = mach_thread_self();
int i, j;
GC_thread p;
kern_return_t kern_result;
thread_act_array_t act_list;
mach_msg_type_number_t listcount;
struct thread_basic_info info;
mach_msg_type_number_t outCount = THREAD_INFO_MAX;
if (GC_notify_event)
GC_notify_event (GC_EVENT_PRE_START_WORLD);
# if DEBUG_THREADS
GC_printf0("World starting\n");
# endif
# ifdef MPROTECT_VDB
if(GC_incremental) {
extern void GC_mprotect_resume();
GC_mprotect_resume();
}
# endif
kern_result = task_threads(my_task, &act_list, &listcount);
for(i = 0; i < listcount; i++) {
thread_act_t thread = act_list[i];
if (thread != my_thread &&
(!GC_use_mach_handler_thread ||
(GC_use_mach_handler_thread && GC_mach_handler_thread != thread))) {
for(j = 0; j < GC_mach_threads_count; j++) {
if (thread == GC_mach_threads[j].thread) {
if (GC_mach_threads[j].already_suspended) {
# if DEBUG_THREADS
GC_printf1("Not resuming already suspended thread %p\n", thread);
# endif
continue;
}
kern_result = thread_info(thread, THREAD_BASIC_INFO,
(thread_info_t)&info, &outCount);
if(kern_result != KERN_SUCCESS) continue;
# if DEBUG_THREADS
GC_printf2("Thread state for 0x%lx = %d\n", thread,
info.run_state);
GC_printf1("Resuming 0x%lx\n", thread);
# endif
/* Resume the thread */
kern_result = thread_resume(thread);
if(kern_result != KERN_SUCCESS) continue;
}
}
}
mach_port_deallocate(my_task, thread);
}
vm_deallocate(my_task, (vm_address_t)act_list, sizeof(thread_t) * listcount);
mach_port_deallocate(my_task, my_thread);
if (GC_notify_event)
GC_notify_event (GC_EVENT_POST_START_WORLD);
# if DEBUG_THREADS
GC_printf0("World started\n");
# endif
}
void GC_darwin_register_mach_handler_thread(mach_port_t thread) {
GC_mach_handler_thread = thread;
GC_use_mach_handler_thread = 1;
}
#endif
-1185
View File
File diff suppressed because it is too large. Load diff
-90
View File
@@ -1,90 +0,0 @@
# Makefile to build Hans Boehm garbage collector using the Digital Mars
# compiler from www.digitalmars.com
# Written by Walter Bright
DEFINES=-DNDEBUG -DSILENT -DGC_BUILD -D_WINDOWS -DGC_DLL -DALL_INTERIOR_POINTERS -D__STDC__ -DWIN32_THREADS
CFLAGS=-Iinclude $(DEFINES) -wx -g
LFLAGS=/ma/implib/co
CC=sc
.c.obj:
$(CC) -c $(CFLAGS) $*
.cpp.obj:
$(CC) -c $(CFLAGS) -Aa $*
OBJS= \
allchblk.obj\
alloc.obj\
blacklst.obj\
checksums.obj\
dbg_mlc.obj\
dyn_load.obj\
finalize.obj\
gc_cpp.obj\
headers.obj\
mach_dep.obj\
malloc.obj\
mallocx.obj\
mark.obj\
mark_rts.obj\
misc.obj\
new_hblk.obj\
obj_map.obj\
os_dep.obj\
ptr_chck.obj\
reclaim.obj\
stubborn.obj\
typd_mlc.obj\
win32_threads.obj
targets: gc.dll gc.lib gctest.exe
gc.dll: $(OBJS) gc.def digimars.mak
sc -ogc.dll $(OBJS) -L$(LFLAGS) gc.def kernel32.lib user32.lib
gc.def: digimars.mak
echo LIBRARY GC >gc.def
echo DESCRIPTION "Hans Boehm Garbage Collector" >>gc.def
echo EXETYPE NT >>gc.def
echo EXPORTS >>gc.def
echo GC_is_visible_print_proc >>gc.def
echo GC_is_valid_displacement_print_proc >>gc.def
clean:
del gc.def
del $(OBJS)
gctest.exe : gc.lib tests\test.obj
sc -ogctest.exe tests\test.obj gc.lib
tests\test.obj : tests\test.c
$(CC) -c -g -DNDEBUG -DSILENT -DGC_BUILD -D_WINDOWS -DGC_DLL \
-DALL_INTERIOR_POINTERS -DWIN32_THREADS \
-Iinclude tests\test.c -otests\test.obj
allchblk.obj: allchblk.c
alloc.obj: alloc.c
blacklst.obj: blacklst.c
checksums.obj: checksums.c
dbg_mlc.obj: dbg_mlc.c
dyn_load.obj: dyn_load.c
finalize.obj: finalize.c
gc_cpp.obj: gc_cpp.cpp
headers.obj: headers.c
mach_dep.obj: mach_dep.c
malloc.obj: malloc.c
mallocx.obj: mallocx.c
mark.obj: mark.c
mark_rts.obj: mark_rts.c
misc.obj: misc.c
new_hblk.obj: new_hblk.c
obj_map.obj: obj_map.c
os_dep.obj: os_dep.c
ptr_chck.obj: ptr_chck.c
reclaim.obj: reclaim.c
stubborn.obj: stubborn.c
typd_mlc.obj: typd_mlc.c
win32_threads.obj: win32_threads.c
-2
View File
@@ -1,2 +0,0 @@
/Makefile
/Makefile.in
-27
View File
@@ -1,27 +0,0 @@
#
#
# THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
# OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
#
# Permission is hereby granted to use or copy this program
# for any purpose, provided the above notices are retained on all copies.
# Permission to modify the code and to distribute modified code is granted,
# provided the above notices are retained, and a notice that the code was
# modified is included with the above copyright notice.
#
# Modified by: Grzegorz Jakacki <jakacki at acm dot org>
## Process this file with automake to produce Makefile.in.
# installed documentation
#
dist_pkgdata_DATA = barrett_diagram debugging.html gc.man \
gcdescr.html README README.amiga README.arm.cross \
README.autoconf README.changes README.contributors \
README.cords README.DGUX386 README.dj README.environment \
README.ews4800 README.hp README.linux README.Mac \
README.MacOSX README.macros README.OS2 README.rs6000 \
README.sgi README.solaris2 README.uts README.win32 \
tree.html leak.html gcinterface.html scale.html \
README.darwin simple_example.html
-622
View File
@@ -1,622 +0,0 @@
Copyright (c) 1988, 1989 Hans-J. Boehm, Alan J. Demers
Copyright (c) 1991-1996 by Xerox Corporation. All rights reserved.
Copyright (c) 1996-1999 by Silicon Graphics. All rights reserved.
Copyright (c) 1999-2004 Hewlett-Packard Development Company, L.P.
The file linux_threads.c is also
Copyright (c) 1998 by Fergus Henderson. All rights reserved.
The files Makefile.am, and configure.ac are
Copyright (c) 2001 by Red Hat Inc. All rights reserved.
Several files supporting GNU-style builds are copyrighted by the Free
Software Foundation, and carry a different license from that given
below.
THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
Permission is hereby granted to use or copy this program
for any purpose, provided the above notices are retained on all copies.
Permission to modify the code and to distribute modified code is granted,
provided the above notices are retained, and a notice that the code was
modified is included with the above copyright notice.
A few of the files needed to use the GNU-style build procedure come with
slightly different licenses, though they are all similar in spirit. A few
are GPL'ed, but with an exception that should cover all uses in the
collector. (If you are concerned about such things, I recommend you look
at the notice in config.guess or ltmain.sh.)
This is version 6.6 of a conservative garbage collector for C and C++.
You might find a more recent version of this at
http://www.hpl.hp.com/personal/Hans_Boehm/gc
OVERVIEW
This is intended to be a general purpose, garbage collecting storage
allocator. The algorithms used are described in:
Boehm, H., and M. Weiser, "Garbage Collection in an Uncooperative Environment",
Software Practice & Experience, September 1988, pp. 807-820.
Boehm, H., A. Demers, and S. Shenker, "Mostly Parallel Garbage Collection",
Proceedings of the ACM SIGPLAN '91 Conference on Programming Language Design
and Implementation, SIGPLAN Notices 26, 6 (June 1991), pp. 157-164.
Boehm, H., "Space Efficient Conservative Garbage Collection", Proceedings
of the ACM SIGPLAN '91 Conference on Programming Language Design and
Implementation, SIGPLAN Notices 28, 6 (June 1993), pp. 197-206.
Boehm H., "Reducing Garbage Collector Cache Misses", Proceedings of the
2000 International Symposium on Memory Management.
Possible interactions between the collector and optimizing compilers are
discussed in
Boehm, H., and D. Chase, "A Proposal for GC-safe C Compilation",
The Journal of C Language Translation 4, 2 (December 1992).
and
Boehm H., "Simple GC-safe Compilation", Proceedings
of the ACM SIGPLAN '96 Conference on Programming Language Design and
Implementation.
(Some of these are also available from
http://www.hpl.hp.com/personal/Hans_Boehm/papers/, among other places.)
Unlike the collector described in the second reference, this collector
operates either with the mutator stopped during the entire collection
(default) or incrementally during allocations. (The latter is supported
on only a few machines.) On the most common platforms, it can be built
with or without thread support. On a few platforms, it can take advantage
of a multiprocessor to speed up garbage collection.
Many of the ideas underlying the collector have previously been explored
by others. Notably, some of the run-time systems developed at Xerox PARC
in the early 1980s conservatively scanned thread stacks to locate possible
pointers (cf. Paul Rovner, "On Adding Garbage Collection and Runtime Types
to a Strongly-Typed Statically Checked, Concurrent Language" Xerox PARC
CSL 84-7). Doug McIlroy wrote a simpler fully conservative collector that
was part of version 8 UNIX (tm), but appears to not have received
widespread use.
Rudimentary tools for use of the collector as a leak detector are included
(see http://www.hpl.hp.com/personal/Hans_Boehm/gc/leak.html),
as is a fairly sophisticated string package "cord" that makes use of the
collector. (See doc/README.cords and H.-J. Boehm, R. Atkinson, and M. Plass,
"Ropes: An Alternative to Strings", Software Practice and Experience 25, 12
(December 1995), pp. 1315-1330. This is very similar to the "rope" package
in Xerox Cedar, or the "rope" package in the SGI STL or the g++ distribution.)
Further collector documantation can be found at
http://www.hpl.hp.com/personal/Hans_Boehm/gc
GENERAL DESCRIPTION
This is a garbage collecting storage allocator that is intended to be
used as a plug-in replacement for C's malloc.
Since the collector does not require pointers to be tagged, it does not
attempt to ensure that all inaccessible storage is reclaimed. However,
in our experience, it is typically more successful at reclaiming unused
memory than most C programs using explicit deallocation. Unlike manually
introduced leaks, the amount of unreclaimed memory typically stays
bounded.
In the following, an "object" is defined to be a region of memory allocated
by the routines described below.
Any objects not intended to be collected must be pointed to either
from other such accessible objects, or from the registers,
stack, data, or statically allocated bss segments. Pointers from
the stack or registers may point to anywhere inside an object.
The same is true for heap pointers if the collector is compiled with
ALL_INTERIOR_POINTERS defined, as is now the default.
Compiling without ALL_INTERIOR_POINTERS may reduce accidental retention
of garbage objects, by requiring pointers from the heap to to the beginning
of an object. But this no longer appears to be a significant
issue for most programs.
There are a number of routines which modify the pointer recognition
algorithm. GC_register_displacement allows certain interior pointers
to be recognized even if ALL_INTERIOR_POINTERS is nor defined.
GC_malloc_ignore_off_page allows some pointers into the middle of large objects
to be disregarded, greatly reducing the probablility of accidental
retention of large objects. For most purposes it seems best to compile
with ALL_INTERIOR_POINTERS and to use GC_malloc_ignore_off_page if
you get collector warnings from allocations of very large objects.
See README.debugging for details.
WARNING: pointers inside memory allocated by the standard "malloc" are not
seen by the garbage collector. Thus objects pointed to only from such a
region may be prematurely deallocated. It is thus suggested that the
standard "malloc" be used only for memory regions, such as I/O buffers, that
are guaranteed not to contain pointers to garbage collectable memory.
Pointers in C language automatic, static, or register variables,
are correctly recognized. (Note that GC_malloc_uncollectable has semantics
similar to standard malloc, but allocates objects that are traced by the
collector.)
WARNING: the collector does not always know how to find pointers in data
areas that are associated with dynamic libraries. This is easy to
remedy IF you know how to find those data areas on your operating
system (see GC_add_roots). Code for doing this under SunOS, IRIX 5.X and 6.X,
HP/UX, Alpha OSF/1, Linux, and win32 is included and used by default. (See
README.win32 for win32 details.) On other systems pointers from dynamic
library data areas may not be considered by the collector.
If you're writing a program that depends on the collector scanning
dynamic library data areas, it may be a good idea to include at least
one call to GC_is_visible() to ensure that those areas are visible
to the collector.
Note that the garbage collector does not need to be informed of shared
read-only data. However if the shared library mechanism can introduce
discontiguous data areas that may contain pointers, then the collector does
need to be informed.
Signal processing for most signals may be deferred during collection,
and during uninterruptible parts of the allocation process.
Like standard ANSI C mallocs, by default it is unsafe to invoke
malloc (and other GC routines) from a signal handler while another
malloc call may be in progress. Removing -DNO_SIGNALS from Makefile
attempts to remedy that. But that may not be reliable with a compiler that
substantially reorders memory operations inside GC_malloc.
The allocator/collector can also be configured for thread-safe operation.
(Full signal safety can also be achieved, but only at the cost of two system
calls per malloc, which is usually unacceptable.)
WARNING: the collector does not guarantee to scan thread-local storage
(e.g. of the kind accessed with pthread_getspecific()). The collector
does scan thread stacks, though, so generally the best solution is to
ensure that any pointers stored in thread-local storage are also
stored on the thread's stack for the duration of their lifetime.
(This is arguably a longstanding bug, but it hasn't been fixed yet.)
INSTALLATION AND PORTABILITY
As distributed, the macro SILENT is defined in Makefile.
In the event of problems, this can be removed to obtain a moderate
amount of descriptive output for each collection.
(The given statistics exhibit a few peculiarities.
Things don't appear to add up for a variety of reasons, most notably
fragmentation losses. These are probably much more significant for the
contrived program "test.c" than for your application.)
Note that typing "make test" will automatically build the collector
and then run setjmp_test and gctest. Setjmp_test will give you information
about configuring the collector, which is useful primarily if you have
a machine that's not already supported. Gctest is a somewhat superficial
test of collector functionality. Failure is indicated by a core dump or
a message to the effect that the collector is broken. Gctest takes about
35 seconds to run on a SPARCstation 2. It may use up to 8 MB of memory. (The
multi-threaded version will use more. 64-bit versions may use more.)
"Make test" will also, as its last step, attempt to build and test the
"cord" string library. This will fail without an ANSI C compiler, but
the garbage collector itself should still be usable.
The Makefile will generate a library gc.a which you should link against.
Typing "make cords" will add the cord library to gc.a.
Note that this requires an ANSI C compiler.
It is suggested that if you need to replace a piece of the collector
(e.g. GC_mark_rts.c) you simply list your version ahead of gc.a on the
ld command line, rather than replacing the one in gc.a. (This will
generate numerous warnings under some versions of AIX, but it still
works.)
All include files that need to be used by clients will be put in the
include subdirectory. (Normally this is just gc.h. "Make cords" adds
"cord.h" and "ec.h".)
The collector currently is designed to run essentially unmodified on
machines that use a flat 32-bit or 64-bit address space.
That includes the vast majority of Workstations and X86 (X >= 3) PCs.
(The list here was deleted because it was getting too long and constantly
out of date.)
It does NOT run under plain 16-bit DOS or Windows 3.X. There are however
various packages (e.g. win32s, djgpp) that allow flat 32-bit address
applications to run under those systemsif the have at least an 80386 processor,
and several of those are compatible with the collector.
In a few cases (Amiga, OS/2, Win32, MacOS) a separate makefile
or equivalent is supplied. Many of these have separate README.system
files.
Dynamic libraries are completely supported only under SunOS/Solaris,
(and even that support is not functional on the last Sun 3 release),
Linux, FreeBSD, NetBSD, IRIX 5&6, HP/UX, Win32 (not Win32S) and OSF/1
on DEC AXP machines plus perhaps a few others listed near the top
of dyn_load.c. On other machines we recommend that you do one of
the following:
1) Add dynamic library support (and send us the code).
2) Use static versions of the libraries.
3) Arrange for dynamic libraries to use the standard malloc.
This is still dangerous if the library stores a pointer to a
garbage collected object. But nearly all standard interfaces
prohibit this, because they deal correctly with pointers
to stack allocated objects. (Strtok is an exception. Don't
use it.)
In all cases we assume that pointer alignment is consistent with that
enforced by the standard C compilers. If you use a nonstandard compiler
you may have to adjust the alignment parameters defined in gc_priv.h.
Note that this may also be an issue with packed records/structs, if those
enforce less alignment for pointers.
A port to a machine that is not byte addressed, or does not use 32 bit
or 64 bit addresses will require a major effort. A port to plain MSDOS
or win16 is hard.
For machines not already mentioned, or for nonstandard compilers, the
following are likely to require change:
1. The parameters in gcconfig.h.
The parameters that will usually require adjustment are
STACKBOTTOM, ALIGNMENT and DATASTART. Setjmp_test
prints its guesses of the first two.
DATASTART should be an expression for computing the
address of the beginning of the data segment. This can often be
&etext. But some memory management units require that there be
some unmapped space between the text and the data segment. Thus
it may be more complicated. On UNIX systems, this is rarely
documented. But the adb "$m" command may be helpful. (Note
that DATASTART will usually be a function of &etext. Thus a
single experiment is usually insufficient.)
STACKBOTTOM is used to initialize GC_stackbottom, which
should be a sufficient approximation to the coldest stack address.
On some machines, it is difficult to obtain such a value that is
valid across a variety of MMUs, OS releases, etc. A number of
alternatives exist for using the collector in spite of this. See the
discussion in gcconfig.h immediately preceding the various
definitions of STACKBOTTOM.
2. mach_dep.c.
The most important routine here is one to mark from registers.
The distributed file includes a generic hack (based on setjmp) that
happens to work on many machines, and may work on yours. Try
compiling and running setjmp_t.c to see whether it has a chance of
working. (This is not correct C, so don't blame your compiler if it
doesn't work. Based on limited experience, register window machines
are likely to cause trouble. If your version of setjmp claims that
all accessible variables, including registers, have the value they
had at the time of the longjmp, it also will not work. Vanilla 4.2 BSD
on Vaxen makes such a claim. SunOS does not.)
If your compiler does not allow in-line assembly code, or if you prefer
not to use such a facility, mach_dep.c may be replaced by a .s file
(as we did for the MIPS machine and the PC/RT).
At this point enough architectures are supported by mach_dep.c
that you will rarely need to do more than adjust for assembler
syntax.
3. os_dep.c (and gc_priv.h).
Several kinds of operating system dependent routines reside here.
Many are optional. Several are invoked only through corresponding
macros in gc_priv.h, which may also be redefined as appropriate.
The routine GC_register_data_segments is crucial. It registers static
data areas that must be traversed by the collector. (User calls to
GC_add_roots may sometimes be used for similar effect.)
Routines to obtain memory from the OS also reside here.
Alternatively this can be done entirely by the macro GET_MEM
defined in gc_priv.h. Routines to disable and reenable signals
also reside here if they are need by the macros DISABLE_SIGNALS
and ENABLE_SIGNALS defined in gc_priv.h.
In a multithreaded environment, the macros LOCK and UNLOCK
in gc_priv.h will need to be suitably redefined.
The incremental collector requires page dirty information, which
is acquired through routines defined in os_dep.c. Unless directed
otherwise by gcconfig.h, these are implemented as stubs that simply
treat all pages as dirty. (This of course makes the incremental
collector much less useful.)
4. dyn_load.c
This provides a routine that allows the collector to scan data
segments associated with dynamic libraries. Often it is not
necessary to provide this routine unless user-written dynamic
libraries are used.
For a different version of UN*X or different machines using the
Motorola 68000, Vax, SPARC, 80386, NS 32000, PC/RT, or MIPS architecture,
it should frequently suffice to change definitions in gcconfig.h.
THE C INTERFACE TO THE ALLOCATOR
The following routines are intended to be directly called by the user.
Note that usually only GC_malloc is necessary. GC_clear_roots and GC_add_roots
calls may be required if the collector has to trace from nonstandard places
(e.g. from dynamic library data areas on a machine on which the
collector doesn't already understand them.) On some machines, it may
be desirable to set GC_stacktop to a good approximation of the stack base.
(This enhances code portability on HP PA machines, since there is no
good way for the collector to compute this value.) Client code may include
"gc.h", which defines all of the following, plus many others.
1) GC_malloc(nbytes)
- allocate an object of size nbytes. Unlike malloc, the object is
cleared before being returned to the user. Gc_malloc will
invoke the garbage collector when it determines this to be appropriate.
GC_malloc may return 0 if it is unable to acquire sufficient
space from the operating system. This is the most probable
consequence of running out of space. Other possible consequences
are that a function call will fail due to lack of stack space,
or that the collector will fail in other ways because it cannot
maintain its internal data structures, or that a crucial system
process will fail and take down the machine. Most of these
possibilities are independent of the malloc implementation.
2) GC_malloc_atomic(nbytes)
- allocate an object of size nbytes that is guaranteed not to contain any
pointers. The returned object is not guaranteed to be cleared.
(Can always be replaced by GC_malloc, but results in faster collection
times. The collector will probably run faster if large character
arrays, etc. are allocated with GC_malloc_atomic than if they are
statically allocated.)
3) GC_realloc(object, new_size)
- change the size of object to be new_size. Returns a pointer to the
new object, which may, or may not, be the same as the pointer to
the old object. The new object is taken to be atomic iff the old one
was. If the new object is composite and larger than the original object,
then the newly added bytes are cleared (we hope). This is very likely
to allocate a new object, unless MERGE_SIZES is defined in gc_priv.h.
Even then, it is likely to recycle the old object only if the object
is grown in small additive increments (which, we claim, is generally bad
coding practice.)
4) GC_free(object)
- explicitly deallocate an object returned by GC_malloc or
GC_malloc_atomic. Not necessary, but can be used to minimize
collections if performance is critical. Probably a performance
loss for very small objects (<= 8 bytes).
5) GC_expand_hp(bytes)
- Explicitly increase the heap size. (This is normally done automatically
if a garbage collection failed to GC_reclaim enough memory. Explicit
calls to GC_expand_hp may prevent unnecessarily frequent collections at
program startup.)
6) GC_malloc_ignore_off_page(bytes)
- identical to GC_malloc, but the client promises to keep a pointer to
the somewhere within the first 256 bytes of the object while it is
live. (This pointer should nortmally be declared volatile to prevent
interference from compiler optimizations.) This is the recommended
way to allocate anything that is likely to be larger than 100Kbytes
or so. (GC_malloc may result in failure to reclaim such objects.)
7) GC_set_warn_proc(proc)
- Can be used to redirect warnings from the collector. Such warnings
should be rare, and should not be ignored during code development.
8) GC_enable_incremental()
- Enables generational and incremental collection. Useful for large
heaps on machines that provide access to page dirty information.
Some dirty bit implementations may interfere with debugging
(by catching address faults) and place restrictions on heap arguments
to system calls (since write faults inside a system call may not be
handled well).
9) Several routines to allow for registration of finalization code.
User supplied finalization code may be invoked when an object becomes
unreachable. To call (*f)(obj, x) when obj becomes inaccessible, use
GC_register_finalizer(obj, f, x, 0, 0);
For more sophisticated uses, and for finalization ordering issues,
see gc.h.
The global variable GC_free_space_divisor may be adjusted up from its
default value of 4 to use less space and more collection time, or down for
the opposite effect. Setting it to 1 or 0 will effectively disable collections
and cause all allocations to simply grow the heap.
The variable GC_non_gc_bytes, which is normally 0, may be changed to reflect
the amount of memory allocated by the above routines that should not be
considered as a candidate for collection. Careless use may, of course, result
in excessive memory consumption.
Some additional tuning is possible through the parameters defined
near the top of gc_priv.h.
If only GC_malloc is intended to be used, it might be appropriate to define:
#define malloc(n) GC_malloc(n)
#define calloc(m,n) GC_malloc((m)*(n))
For small pieces of VERY allocation intensive code, gc_inl.h
includes some allocation macros that may be used in place of GC_malloc
and friends.
All externally visible names in the garbage collector start with "GC_".
To avoid name conflicts, client code should avoid this prefix, except when
accessing garbage collector routines or variables.
There are provisions for allocation with explicit type information.
This is rarely necessary. Details can be found in gc_typed.h.
THE C++ INTERFACE TO THE ALLOCATOR:
The Ellis-Hull C++ interface to the collector is included in
the collector distribution. If you intend to use this, type
"make c++" after the initial build of the collector is complete.
See gc_cpp.h for the definition of the interface. This interface
tries to approximate the Ellis-Detlefs C++ garbage collection
proposal without compiler changes.
Cautions:
1. Arrays allocated without new placement syntax are
allocated as uncollectable objects. They are traced by the
collector, but will not be reclaimed.
2. Failure to use "make c++" in combination with (1) will
result in arrays allocated using the default new operator.
This is likely to result in disaster without linker warnings.
3. If your compiler supports an overloaded new[] operator,
then gc_cpp.cc and gc_cpp.h should be suitably modified.
4. Many current C++ compilers have deficiencies that
break some of the functionality. See the comments in gc_cpp.h
for suggested workarounds.
USE AS LEAK DETECTOR:
The collector may be used to track down leaks in C programs that are
intended to run with malloc/free (e.g. code with extreme real-time or
portability constraints). To do so define FIND_LEAK in Makefile
This will cause the collector to invoke the report_leak
routine defined near the top of reclaim.c whenever an inaccessible
object is found that has not been explicitly freed. Such objects will
also be automatically reclaimed.
Productive use of this facility normally involves redefining report_leak
to do something more intelligent. This typically requires annotating
objects with additional information (e.g. creation time stack trace) that
identifies their origin. Such code is typically not very portable, and is
not included here, except on SPARC machines.
If all objects are allocated with GC_DEBUG_MALLOC (see next section),
then the default version of report_leak will report the source file
and line number at which the leaked object was allocated. This may
sometimes be sufficient. (On SPARC/SUNOS4 machines, it will also report
a cryptic stack trace. This can often be turned into a sympolic stack
trace by invoking program "foo" with "callprocs foo". Callprocs is
a short shell script that invokes adb to expand program counter values
to symbolic addresses. It was largely supplied by Scott Schwartz.)
Note that the debugging facilities described in the next section can
sometimes be slightly LESS effective in leak finding mode, since in
leak finding mode, GC_debug_free actually results in reuse of the object.
(Otherwise the object is simply marked invalid.) Also note that the test
program is not designed to run meaningfully in FIND_LEAK mode.
Use "make gc.a" to build the collector.
DEBUGGING FACILITIES:
The routines GC_debug_malloc, GC_debug_malloc_atomic, GC_debug_realloc,
and GC_debug_free provide an alternate interface to the collector, which
provides some help with memory overwrite errors, and the like.
Objects allocated in this way are annotated with additional
information. Some of this information is checked during garbage
collections, and detected inconsistencies are reported to stderr.
Simple cases of writing past the end of an allocated object should
be caught if the object is explicitly deallocated, or if the
collector is invoked while the object is live. The first deallocation
of an object will clear the debugging info associated with an
object, so accidentally repeated calls to GC_debug_free will report the
deallocation of an object without debugging information. Out of
memory errors will be reported to stderr, in addition to returning
NIL.
GC_debug_malloc checking during garbage collection is enabled
with the first call to GC_debug_malloc. This will result in some
slowdown during collections. If frequent heap checks are desired,
this can be achieved by explicitly invoking GC_gcollect, e.g. from
the debugger.
GC_debug_malloc allocated objects should not be passed to GC_realloc
or GC_free, and conversely. It is however acceptable to allocate only
some objects with GC_debug_malloc, and to use GC_malloc for other objects,
provided the two pools are kept distinct. In this case, there is a very
low probablility that GC_malloc allocated objects may be misidentified as
having been overwritten. This should happen with probability at most
one in 2**32. This probability is zero if GC_debug_malloc is never called.
GC_debug_malloc, GC_malloc_atomic, and GC_debug_realloc take two
additional trailing arguments, a string and an integer. These are not
interpreted by the allocator. They are stored in the object (the string is
not copied). If an error involving the object is detected, they are printed.
The macros GC_MALLOC, GC_MALLOC_ATOMIC, GC_REALLOC, GC_FREE, and
GC_REGISTER_FINALIZER are also provided. These require the same arguments
as the corresponding (nondebugging) routines. If gc.h is included
with GC_DEBUG defined, they call the debugging versions of these
functions, passing the current file name and line number as the two
extra arguments, where appropriate. If gc.h is included without GC_DEBUG
defined, then all these macros will instead be defined to their nondebugging
equivalents. (GC_REGISTER_FINALIZER is necessary, since pointers to
objects with debugging information are really pointers to a displacement
of 16 bytes form the object beginning, and some translation is necessary
when finalization routines are invoked. For details, about what's stored
in the header, see the definition of the type oh in debug_malloc.c)
INCREMENTAL/GENERATIONAL COLLECTION:
The collector normally interrupts client code for the duration of
a garbage collection mark phase. This may be unacceptable if interactive
response is needed for programs with large heaps. The collector
can also run in a "generational" mode, in which it usually attempts to
collect only objects allocated since the last garbage collection.
Furthermore, in this mode, garbage collections run mostly incrementally,
with a small amount of work performed in response to each of a large number of
GC_malloc requests.
This mode is enabled by a call to GC_enable_incremental().
Incremental and generational collection is effective in reducing
pause times only if the collector has some way to tell which objects
or pages have been recently modified. The collector uses two sources
of information:
1. Information provided by the VM system. This may be provided in
one of several forms. Under Solaris 2.X (and potentially under other
similar systems) information on dirty pages can be read from the
/proc file system. Under other systems (currently SunOS4.X) it is
possible to write-protect the heap, and catch the resulting faults.
On these systems we require that system calls writing to the heap
(other than read) be handled specially by client code.
See os_dep.c for details.
2. Information supplied by the programmer. We define "stubborn"
objects to be objects that are rarely changed. Such an object
can be allocated (and enabled for writing) with GC_malloc_stubborn.
Once it has been initialized, the collector should be informed with
a call to GC_end_stubborn_change. Subsequent writes that store
pointers into the object must be preceded by a call to
GC_change_stubborn.
This mechanism performs best for objects that are written only for
initialization, and such that only one stubborn object is writable
at once. It is typically not worth using for short-lived
objects. Stubborn objects are treated less efficiently than pointerfree
(atomic) objects.
A rough rule of thumb is that, in the absence of VM information, garbage
collection pauses are proportional to the amount of pointerful storage
plus the amount of modified "stubborn" storage that is reachable during
the collection.
Initial allocation of stubborn objects takes longer than allocation
of other objects, since other data structures need to be maintained.
We recommend against random use of stubborn objects in client
code, since bugs caused by inappropriate writes to stubborn objects
are likely to be very infrequently observed and hard to trace.
However, their use may be appropriate in a few carefully written
library routines that do not make the objects themselves available
for writing by client code.
BUGS:
Any memory that does not have a recognizable pointer to it will be
reclaimed. Exclusive-or'ing forward and backward links in a list
doesn't cut it.
Some C optimizers may lose the last undisguised pointer to a memory
object as a consequence of clever optimizations. This has almost
never been observed in practice. Send mail to [email protected]
for suggestions on how to fix your compiler.
This is not a real-time collector. In the standard configuration,
percentage of time required for collection should be constant across
heap sizes. But collection pauses will increase for larger heaps.
(On SPARCstation 2s collection times will be on the order of 300 msecs
per MB of accessible memory that needs to be scanned. Your mileage
may vary.) The incremental/generational collection facility helps,
but is portable only if "stubborn" allocation is used.
Please address bug reports to [email protected]. If you are
contemplating a major addition, you might also send mail to ask whether
it's already been done (or whether we tried and discarded it).
-215
View File
@@ -1,215 +0,0 @@
Garbage Collector (parallel iversion) for ix86 DG/UX Release R4.20MU07
*READ* the file README.QUICK.
You need the GCC-3.0.3 rev (DG/UX) compiler to build this tree.
This compiler has the new "dgux386" threads package implemented.
It also supports the switch "-pthread" needed to link correctly
the DG/UX's -lrte -lthread with -lgcc and the system's -lc.
Finally we support parralleli-mark for the SMP DG/UX machines.
To build the garbage collector do:
./configure --enable-parallel-mark
make
make gctest
Before you run "gctest" you need to set your LD_LIBRARY_PATH
correctly so that "gctest" can find the shared library libgc.
Alternatively you can do a configuration
./configure --enable-parallel-mark --disable-shared
to build only the static version of libgc.
To enable debugging messages please do:
1) Add the "--enable-full-debug" flag during configuration.
2) Edit the file linux-threads.c and uncommnect the line:
/* #define DEBUG_THREADS 1 */ to --->
#define DEBUG_THREADS 1
Then give "make" as usual.
In a machine with 4 CPUs (my own machine) the option parallel
mark (aka --enable-parallel-mark) makes a BIG difference.
Takis Psarogiannakopoulos
University of Cambridge
Centre for Mathematical Sciences
Department of Pure Mathematics
Wilberforce Road
Cambridge CB3 0WB ,UK , <[email protected]>
January 2002
Note (HB):
The integration of this patch is currently not complete.
The following patches against 6.1alpha3 where hard to move
to alpha4, and are not integrated. There may also be minor
problems with stylistic corrections made by me.
--- ltconfig.ORIG Mon Jan 28 20:22:18 2002
+++ ltconfig Mon Jan 28 20:44:00 2002
@@ -689,6 +689,11 @@
pic_flag=-Kconform_pic
fi
;;
+ dgux*)
+ pic_flag='-fPIC'
+ link_static='-Bstatic'
+ wl='-Wl,'
+ ;;
*)
pic_flag='-fPIC'
;;
@@ -718,6 +723,12 @@
# We can build DLLs from non-PIC.
;;
+ dgux*)
+ pic_flag='-KPIC'
+ link_static='-Bstatic'
+ wl='-Wl,'
+ ;;
+
osf3* | osf4* | osf5*)
# All OSF/1 code is PIC.
wl='-Wl,'
@@ -1154,6 +1165,22 @@
fi
;;
+ dgux*)
+ ld_shlibs=yes
+ # For both C/C++ ommit the deplibs. This is because we relying on the fact
+ # that compilation of execitables will put them in correct order
+ # in any case and sometimes are wrong when listed as deplibs (or missing some deplibs)
+ # However when GNU ld and --whole-archive needs to be used we have the problem
+ # that if the -fPIC *_s.a archive is linked through deplibs list we ommiting crucial
+ # .lo/.o files from the created shared lib. This I think is not the case here.
+ archive_cmds='$CC -shared -h $soname -o $lib $libobjs $linkopts'
+ thread_safe_flag_spec='-pthread'
+ wlarc=
+ hardcode_libdir_flag_spec='-L$libdir'
+ hardcode_shlibpath_var=no
+ ac_cv_archive_cmds_needs_lc=no
+ ;;
+
cygwin* | mingw*)
# hardcode_libdir_flag_spec is actually meaningless, as there is
# no search path for DLLs.
@@ -1497,7 +1524,7 @@
;;
dgux*)
- archive_cmds='$LD -G -h $soname -o $lib $libobjs $deplibs $linkopts'
+ archive_cmds='$CC -shared -h $soname -o $lib $libobjs $linkopts'
hardcode_libdir_flag_spec='-L$libdir'
hardcode_shlibpath_var=no
;;
@@ -2092,12 +2119,17 @@
;;
dgux*)
- version_type=linux
+ version_type=dgux
need_lib_prefix=no
need_version=no
- library_names_spec='${libname}${release}.so$versuffix ${libname}${release}.so$major $libname.so'
- soname_spec='${libname}${release}.so$major'
+ library_names_spec='$libname.so$versuffix'
+ soname_spec='$libname.so$versuffix'
shlibpath_var=LD_LIBRARY_PATH
+ thread_safe_flag_spec='-pthread'
+ wlarc=
+ hardcode_libdir_flag_spec='-L$libdir'
+ hardcode_shlibpath_var=no
+ ac_cv_archive_cmds_needs_lc=no
;;
sysv4*MP*)
--- ltmain.sh.ORIG Mon Jan 28 20:31:18 2002
+++ ltmain.sh Tue Jan 29 00:11:29 2002
@@ -1072,11 +1072,38 @@
esac
;;
+ -thread*)
+ # DG/UX GCC 2.95.x, 3.x.x rev (DG/UX) links -lthread
+ # with the switch -threads
+ if test "$arg" = "-threads"; then
+ case "$host" in
+ i[3456]86-*-dgux*)
+ deplibs="$deplibs $arg"
+ continue
+ ;;
+ esac
+ fi
+ ;;
+
+ -pthread*)
+ # DG/UX GCC 2.95.x, 3.x.x rev (DG/UX) links -lthread
+ # with the switch -pthread
+ if test "$arg" = "-pthread"; then
+ case "$host" in
+ i[3456]86-*-dgux*)
+ deplibs="$deplibs $arg"
+ continue
+ ;;
+ esac
+ fi
+ ;;
+
-l*)
if test "$arg" = "-lc"; then
case "$host" in
- *-*-cygwin* | *-*-mingw* | *-*-os2* | *-*-beos*)
+ *-*-cygwin* | *-*-mingw* | *-*-os2* | *-*-beos* | i[3456]86-*-dgux*)
# These systems don't actually have c library (as such)
+ # It is wrong in DG/UX to add -lc when creating shared/dynamic objs/libs
continue
;;
esac
@@ -1248,6 +1275,12 @@
temp_deplibs=
for deplib in $dependency_libs; do
case "$deplib" in
+ -thread*)
+ temp_deplibs="$temp_deplibs $deplib"
+ ;;
+ -pthread)
+ temp_deplibs="$temp_deplibs $deplib"
+ ;;
-R*) temp_xrpath=`$echo "X$deplib" | $Xsed -e 's/^-R//'`
case " $rpath $xrpath " in
*" $temp_xrpath "*) ;;
@@ -1709,6 +1742,13 @@
done
;;
+ dgux)
+ # Leave mostly blank for DG/UX
+ major=
+ versuffix=".$current.$revision";
+ verstring=
+ ;;
+
linux)
major=.`expr $current - $age`
versuffix="$major.$age.$revision"
@@ -1792,8 +1832,9 @@
dependency_libs="$deplibs"
case "$host" in
- *-*-cygwin* | *-*-mingw* | *-*-os2* | *-*-beos*)
+ *-*-cygwin* | *-*-mingw* | *-*-os2* | *-*-beos* | i[3456]86-*-dgux*)
# these systems don't actually have a c library (as such)!
+ # It is wrong in DG/UX to add -lc when creating shared/dynamic objs/libs
;;
*)
# Add libc to deplibs on all other systems.
-385
View File
@@ -1,385 +0,0 @@
Patrick Beard's Notes for building GC v4.12 with CodeWarrior Pro 2:
----------------------------------------------------------------------------
The current build environment for the collector is CodeWarrior Pro 2.
Projects for CodeWarrior Pro 2 (and for quite a few older versions)
are distributed in the file Mac_projects.sit.hqx. The project file
:Mac_projects:gc.prj builds static library versions of the collector.
:Mac_projects:gctest.prj builds the GC test suite.
Configuring the collector is still done by editing the files
:Mac_files:MacOS_config.h and :Mac_files:MacOS_Test_config.h.
Lars Farm's suggestions on building the collector:
----------------------------------------------------------------------------
Garbage Collection on MacOS - a manual 'MakeFile'
-------------------------------------------------
Project files and IDE's are great on the Macintosh, but they do have
problems when used as distribution media. This note tries to provide
porting instructions in pure TEXT form to avoid those problems. A manual
'makefile' if you like.
GC version: 4.12a2
Codewarrior: CWPro1
date: 18 July 1997
The notes may or may not apply to earlier or later versions of the
GC/CWPro. Actually, they do apply to earlier versions of both except that
until recently a project could only build one target so each target was a
separate project. The notes will most likely apply to future versions too.
Possibly with minor tweaks.
This is just to record my experiences. These notes do not mean I now
provide a supported port of the GC to MacOS. It works for me. If it works
for you, great. If it doesn't, sorry, try again...;-) Still, if you find
errors, please let me know.
mailto: [email protected]
address: Lars Farm
Krönvägen 33b
856 44 Sundsvall
Sweden
Porting to MacOS is a bit more complex than it first seems. Which MacOS?
68K/PowerPC? Which compiler? Each supports both 68K and PowerPC and offer a
large number of (unique to each environment) compiler settings. Each
combination of compiler/68K/PPC/settings require a unique combination of
standard libraries. And the IDE's does not select them for you. They don't
even check that the library is built with compatible setting and this is
the major source of problems when porting the GC (and otherwise too).
You will have to make choices when you configure the GC. I've made some
choices here, but there are other combinations of settings and #defines
that work too.
As for target settings the major obstacles may be:
- 68K Processor: check "4-byte Ints".
- PPC Processor: uncheck "Store Static Data in TOC".
What you need to do:
===================
1) Build the GC as a library
2) Test that the library works with 'test.c'.
3) Test that the C++ interface 'gc_cpp.cc/h' works with 'test_cpp.cc'.
1) The Libraries:
=================
I made one project with four targets (68K/PPC tempmem or appheap). One target
will suffice if you're able to decide which one you want. I wasn't...
Codewarrior allows a large number of compiler/linker settings. I used these:
Settings shared by all targets:
------------------------------
o Access Paths:
- User Paths: the GC folder
- System Paths: {Compiler}:Metrowerks Standard Library:
{Compiler}:MacOS Support:Headers:
{Compiler}:MacOS Support:MacHeaders:
o C/C++ language:
- inlining: normal
- direct to SOM: off
- enable/check: exceptions, RTTI, bool (and if you like pool strings)
PowerPC target settings
-----------------------
o Target Settings:
- name of target
- MacOS PPC Linker
o PPC Target
- name of library
o C/C++ language
- prefix file as described below
o PPC Processor
- Struct Alignment: PowerPC
- uncheck "Store Static Data in TOC" -- important!
I don't think the others matter, I use full optimization and its ok
o PPC Linker
- Factory Settings (SYM file with full paths, faster linking, dead-strip
static init, Main: __start)
68K target settings
-------------------
o Target Settings:
- name of target
- MacOS 68K Linker
o 68K Target
- name of library
- A5 relative data
o C/C++ language
- prefix file as described below
o 68K Processor
- Code model: smart
- Struct alignment: 68K
- FP: SANE
- enable 4-Byte Ints -- important!
I don't think the others matter. I selected...
- enable: 68020
- enable: global register allocation
o IR Optimizer
- enable: Optimize Space, Optimize Speed
I suppose the others would work too, but haven't tried...
o 68K Linker
- Factory Settings (New Style MacsBug,SYM file with full paths,
A6 Frames, fast link, Merge compiler glue into segment 1,
dead-strip static init)
Prefix Files to configure the GC sources
----------------------------------------
The Codewarrior equivalent of commandline compilers -DNAME=X is to use
prefix-files. A TEXT file that is automatically #included before the first byte
of every source file. I used these:
---- ( cut here ) ---- gc_prefix_tempmem.h -- 68K and PPC -----
#include "gc_prefix_common.h"
#undef USE_TEMPORARY_MEMORY
#define USE_TEMPORARY_MEMORY
---- ( cut here ) ---- gc_prefix_appmem.h -- 68K and PPC -----
#include "gc_prefix_common.h"
#undef USE_TEMPORARY_MEMORY
// #define USE_TEMPORARY_MEMORY
---- ( cut here ) ---- gc_prefix_common.h --------------------
// gc_prefix_common.h
// ------------------
// Codewarrior prefix file to configure the GC libraries
//
// prefix files are the Codewarrior equivalent of the
// command line option -Dname=x frequently seen in makefiles
#if !__MWERKS__
#error only tried this with Codewarrior
#endif
#if macintosh
#define MSL_USE_PRECOMPILED_HEADERS 0
#include <ansi_prefix.mac.h>
#ifndef __STDC__
#define __STDC__ 0
#endif
// See list of #defines to configure the library in: 'MakeFile'
// see also README
#define SILENT // no collection messages. In case
// of trouble you might want this off
#define ALL_INTERIOR_POINTERS // follows interior pointers.
//#define DONT_ADD_BYTE_AT_END // disables the padding if defined.
//#define SMALL_CONFIG // whether to use a smaller heap.
#define NO_SIGNALS // signals aren't real on the Macintosh.
#define ATOMIC_UNCOLLECTABLE // GC_malloc_atomic_uncollectable()
// define either or none as per personal preference
// used in malloc.c
#define REDIRECT_MALLOC GC_malloc
//#define REDIRECT_MALLOC GC_malloc_uncollectable
// if REDIRECT_MALLOC is #defined make sure that the GC library
// is listed before the ANSI/ISO libs in the Codewarrior
// 'Link order' panel
//#define IGNORE_FREE
// mac specific configs
//#define USE_TEMPORARY_MEMORY // use Macintosh temporary memory.
//#define SHARED_LIBRARY_BUILD // build for use in a shared library.
#else
// could build Win32 here too, or in the future
// Rhapsody PPC-mach, Rhapsody PPC-MacOS,
// Rhapsody Intel-mach, Rhapsody Intel-Win32,...
// ... ugh this will get messy ...
#endif
// make sure ints are at least 32-bit
// ( could be set to 16-bit by compiler settings (68K) )
struct gc_private_assert_intsize_{ char x[ sizeof(int)>=4 ? 1 : 0 ]; };
#if __powerc
#if __option(toc_data)
#error turn off "store static data in TOC" when using GC
// ... or find a way to add TOC to the root set...(?)
#endif
#endif
---- ( cut here ) ---- end of gc_prefix_common.h -----------------
Files to build the GC libraries:
--------------------------------
allchblk.c
alloc.c
blacklst.c
checksums.c
dbg_mlc.c
finalize.c
headers.c
mach_dep.c
MacOS.c -- contains MacOS code
malloc.c
mallocx.c
mark.c
mark_rts.c
misc.c
new_hblk.c
obj_map.c
os_dep.c -- contains MacOS code
ptr_chck.c
reclaim.c
stubborn.c
typd_mlc.c
gc++.cc -- this is 'gc_cpp.cc' with less 'inline' and
-- throw std::bad_alloc when out of memory
-- gc_cpp.cc works just fine too
2) Test that the library works with 'test.c'.
=============================================
The test app is just an ordinary ANSI-C console app. Make sure settings
match the library you're testing.
Files
-----
test.c
the GC library to test -- link order before ANSI libs
suitable Mac+ANSI libraries
prefix:
------
---- ( cut here ) ---- gc_prefix_testlib.h -- all libs -----
#define MSL_USE_PRECOMPILED_HEADERS 0
#include <ansi_prefix.mac.h>
#undef NDEBUG
#define ALL_INTERIOR_POINTERS /* for GC_priv.h */
---- ( cut here ) ----
3) Test that the C++ interface 'gc_cpp.cc/h' works with 'test_cpp.cc'.
The test app is just an ordinary ANSI-C console app. Make sure settings match
the library you're testing.
Files
-----
test_cpp.cc
the GC library to test -- link order before ANSI libs
suitable Mac+ANSI libraries
prefix:
------
same as for test.c
For convenience I used one test-project with several targets so that all
test apps are build at once. Two for each library to test: test.c and
gc_app.cc. When I was satisfied that the libraries were ok. I put the
libraries + gc.h + the c++ interface-file in a folder that I then put into
the MSL hierarchy so that I don't have to alter access-paths in projects
that use the GC.
After that, just add the proper GC library to your project and the GC is in
action! malloc will call GC_malloc and free GC_free, new/delete too. You
don't have to call free or delete. You may have to be a bit cautious about
delete if you're freeing other resources than RAM. See gc_cpp.h. You can
also keep coding as always with delete/free. That works too. If you want,
"include <gc.h> and tweak it's use a bit.
Symantec SPM
============
It has been a while since I tried the GC in SPM, but I think that the above
instructions should be sufficient to guide you through in SPM too. SPM
needs to know where the global data is. Use the files 'datastart.c' and
'dataend.c'. Put 'datastart.c' at the top of your project and 'dataend.c'
at the bottom of your project so that all data is surrounded. This is not
needed in Codewarrior because it provides intrinsic variables
__datastart__, __data_end__ that wraps all globals.
Source Changes (GC 4.12a2)
==========================
Very few. Just one tiny in the GC, not strictly needed.
- MacOS.c line 131 in routine GC_MacFreeTemporaryMemory()
change # if !defined(SHARED_LIBRARY_BUILD)
to # if !defined(SILENT) && !defined(SHARED_LIBRARY_BUILD)
To turn off a message when the application quits (actually, I faked
this change by #defining SHARED_LIBRARY_BUILD in a statically linked
library for more than a year without ill effects but perhaps this is
better).
- test_cpp.cc
made the first lines of main() look like this:
------------
int main( int argc, char* argv[] ) {
#endif
#if macintosh // MacOS
char* argv_[] = {"test_cpp","10"}; // doesn't
argv=argv_; // have a
argc = sizeof(argv_)/sizeof(argv_[0]); // commandline
#endif //
int i, iters, n;
# ifndef __GNUC__
alloc dummy_to_fool_the_compiler_into_doing_things_it_currently_cant_handle;
------------
- config.h [now gcconfig.h]
__MWERKS__ does not have to mean MACOS. You can use Codewarrior to
build a Win32 or BeOS library and soon a Rhapsody library. You may
have to change that #if...
It worked for me, hope it works for you.
Lars Farm
18 July 1997
----------------------------------------------------------------------------
Patrick Beard's instructions (may be dated):
v4.3 of the collector now runs under Symantec C++/THINK C v7.0.4, and
Metrowerks C/C++ v4.5 both 68K and PowerPC. Project files are provided
to build and test the collector under both development systems.
Configuration
-------------
To configure the collector, under both development systems, a prefix file
is used to set preprocessor directives. This file is called "MacOS_config.h".
Also to test the collector, "MacOS_Test_config.h" is provided.
Testing
-------
To test the collector (always a good idea), build one of the gctest projects,
gctest.¹ (Symantec C++/THINK C), mw/gctest.68K.¹, or mw/gctest.PPC.¹. The
test will ask you how many times to run; 1 should be sufficient.
Building
--------
For your convenience project files for the major Macintosh development
systems are provided.
For Symantec C++/THINK C, you must build the two projects gclib-1.¹ and
gclib-2.¹. It has to be split up because the collector has more than 32k
of static data and no library can have more than this in the Symantec
environment. (Future versions will probably fix this.)
For Metrowerks C/C++ 4.5 you build gc.68K.¹/gc.PPC.¹ and the result will
be a library called gc.68K.lib/gc.PPC.lib.
Using
-----
Under Symantec C++/THINK C, you can just add the gclib-1.¹ and gclib-2.¹
projects to your own project. Under Metrowerks, you add gc.68K.lib or
gc.PPC.lib and two additional files. You add the files called datastart.c
and dataend.c to your project, bracketing all files that use the collector.
See mw/gctest.¹ for an example.
Include the projects/libraries you built above into your own project,
#include "gc.h", and call GC_malloc. You don't have to call GC_free.
Patrick C. Beard
January 4, 1995
-1
View File
@@ -1 +0,0 @@
See README.darwin for the latest Darwin/MacOSX information.
-6
View File
@@ -1,6 +0,0 @@
The code assumes static linking, and a single thread. The editor de has
not been ported. The cord test program has. The supplied OS2_MAKEFILE
assumes the IBM C Set/2 environment, but the code shouldn't.
Since we haven't figured out hoe to do perform partial links or to build static
libraries, clients currently need to link against a long list of executables.
-322
View File
@@ -1,322 +0,0 @@
===========================================================================
Kjetil S. Matheussen's notes (28-11-2000)
===========================================================================
Compiles under SAS/C again. Should allso still compile under other
amiga compilers without big changes. I haven't checked if it still
works under gcc, because I don't have gcc for amiga. But I have
updated 'Makefile', and hope it compiles fine.
WHATS NEW:
1.
Made a pretty big effort in preventing GCs allocating-functions from returning
chip-mem.
The lower part of the new file AmigaOS.c does this in various ways, mainly by
wrapping GC_malloc, GC_malloc_atomic, GC_malloc_uncollectable,
GC_malloc_atomic_uncollectable, GC_malloc_stubborn, GC_malloc_ignore_off_page
and GC_malloc_atomic_ignore_off_page. GC_realloc is allso wrapped, but
doesn't do the same effort in preventing to return chip-mem.
Other allocating-functions (f.ex. GC_*_typed_) can probably be
used without any problems, but beware that the warn hook will not be called.
In case of problems, don't define GC_AMIGA_FASTALLOC.
Programs using more time actually using the memory allocated
(instead of just allocate and free rapidly) have
the most to earn on this, but even gctest now normally runs twice
as fast and uses less memory, on my poor 8MB machine.
The changes have only effect when there is no more
fast-mem left. But with the way GC works, it
could happen quite often. Beware that an atexit handler had to be added,
so using the abort() function will make a big memory-loss.
If you absolutely must call abort() instead of exit(), try calling
the GC_amiga_free_all_mem function before abort().
New amiga-spesific compilation flags:
GC_AMIGA_FASTALLOC - By NOT defining this option, GC will work like before,
it will not try to force fast-mem out of the OS, and
it will use normal calloc for allocation, and the rest
of the following flags will have no effect.
GC_AMIGA_ONLYFAST - Makes GC never to return chip-mem. GC_AMIGA_RETRY have
no effect if this flag is set.
GC_AMIGA_GC - If gc returns NULL, do a GC_gcollect, and try again. This
usually is a success with the standard GC configuration.
It is allso the most important flag to set to prevent
GC from returning chip-mem. Beware that it slows down a lot
when a program is rapidly allocating/deallocating when
theres either very little fast-memory left or verly little
chip-memory left. Its not a very common situation, but gctest
sometimes (very rare) use many minutes because of this.
GC_AMIGA_RETRY - If gc succeed allocating memory, but it is chip-mem,
try again and see if it is fast-mem. Most of the time,
it will actually return fast-mem for the second try.
I have set max number of retries to 9 or size/5000. You
can change this if you like. (see GC_amiga_rec_alloc())
GC_AMIGA_PRINTSTATS - Gather some statistics during the execution of a
program, and prints out the info when the atexit-handler
is called.
My reccomendation is to set all this flags, except GC_AMIGA_PRINTSTATS and
GC_AMIGA_ONLYFAST.
If your program demands high response-time, you should
not define GC_AMIGA_GC, and possible allso define GC_AMIGA_ONLYFAST.
GC_AMIGA_RETRY does not seem to slow down much.
Allso, when compiling up programs, and GC_AMIGA_FASTALLOC was not defined when
compilling gc, you can define GC_AMIGA_MAKINGLIB to avoid having these allocation-
functions wrapped. (see gc.h)
Note that GC_realloc must not be called before any of
the other above mentioned allocating-functions have been called. (shouldn't be
any programs doing so either, I hope).
Another note. The allocation-function is wrapped when defining
GC_AMIGA_FASTALLOC by letting the function go thru the new
GC_amiga_allocwrapper_do function-pointer (see gc.h). Means that
sending function-pointers, such as GC_malloc, GC_malloc_atomic, etc.,
for later to be called like f.ex this, (*GC_malloc_functionpointer)(size),
will not wrap the function. This is normally not a big problem, unless
all allocation function is called like this, which will cause the
atexit un-allocating function never to be called. Then you either
have to manually add the atexit handler, or call the allocation-
functions function-pointer functions like this;
(*GC_amiga_allocwrapper_do)(size,GC_malloc_functionpointer).
There are probably better ways this problem could be handled, unfortunately,
I didn't find any without rewriting or replacing a lot of the GC-code, which
I really didn't want to. (Making new GC_malloc_* functions, and just
define f.ex GC_malloc as GC_amiga_malloc should allso work).
New amiga-spesific function:
void GC_amiga_set_toany(void (*func)(void));
'func' is a function that will be called right before gc has to change
allocation-method from MEMF_FAST to MEMF_ANY. Ie. when it is likely
it will return chip-mem.
2. A few small compiler-spesific additions to make it compile with SAS/C again.
3. Updated and rewritten the smakefile, so that it works again and that
the "unnecesarry" 'SCOPTIONS' files could be removed. Allso included
the cord-smakefile stuff in the main smakefile, so that the cord smakefile
could be removed too. By writing smake -f Smakefile.smk, both gc.lib and
cord.lib will be made.
STILL MISSING:
Programs can not be started from workbench, at least not for SAS/C. (Martin
Tauchmanns note about that it now works with workbench is definitely wrong
when concerning SAS/C). I guess it works if you use the old "#if 0'ed"-code,
but I haven't tested it. I think the reason for MT to replace the
"#if 0'ed"-code was only because it was a bit to SAS/C-spesific. But I
don't know. An iconx-script solves this problem anyway.
BEWARE!
-To run gctest, set the stack to around 200000 bytes first.
-SAS/C-spesific: cord will crash if you compile gc.lib with
either parm=reg or parm=both. (missing legal prototypes for
function-pointers someplace is the reason I guess.).
tested with software: Radium, http://www.stud.ifi.uio.no/~ksvalast/radium/
tested with hardware: MC68060
[email protected]
===========================================================================
Martin Tauchmann's notes (1-Apr-99)
===========================================================================
Works now, also with the GNU-C compiler V2.7.2.1. <ftp://ftp.unina.it/pub/amiga/geekgadgets/amiga/m68k/snapshots/971125/amiga-bin/>
Modify the `Makefile`
CC=cc $(ABI_FLAG)
to
CC=gcc $(ABI_FLAG)
TECHNICAL NOTES
- `GC_get_stack_base()`, `GC_register_data_segments()` works now with every
C compiler; also Workbench.
- Removed AMIGA_SKIP_SEG, but the Code-Segment must not be scanned by GC.
PROBLEMS
- When the Linker, does`t merge all Code-Segments to an single one. LD of GCC
do it always.
- With ixemul.library V47.3, when an GC program launched from another program
(example: `Make` or `if_mach M68K AMIGA gctest`), `GC_register_data_segments()`
found the Segment-List of the caller program.
Can be fixed, if the run-time initialization code (for C programs, usually *crt0*)
support `__data` and `__bss`.
- PowerPC Amiga currently not supported.
- Dynamic libraries (dyn_load.c) not supported.
TESTED WITH SOFTWARE
`Optimized Oberon 2 C` (oo2c) <http://cognac.informatik.uni-kl.de/download/index.html>
TESTED WITH HARDWARE
MC68030
CONTACT
Please, contact me at <[email protected]>, when you change the
Amiga port. <http://martintauchmann.home.pages.de>
===========================================================================
Michel Schinz's notes
===========================================================================
WHO DID WHAT
The original Amiga port was made by Jesper Peterson. I (Michel Schinz)
modified it slightly to reflect the changes made in the new official
distributions, and to take advantage of the new SAS/C 6.x features. I also
created a makefile to compile the "cord" package (see the cord
subdirectory).
TECHNICAL NOTES
In addition to Jesper's notes, I have the following to say:
- Starting with version 4.3, gctest checks to see if the code segment is
added to the root set or not, and complains if it is. Previous versions
of this Amiga port added the code segment to the root set, so I tried to
fix that. The only problem is that, as far as I know, it is impossible to
know which segments are code segments and which are data segments (there
are indeed solutions to this problem, like scanning the program on disk
or patch the LoadSeg functions, but they are rather complicated). The
solution I have chosen (see os_dep.c) is to test whether the program
counter is in the segment we are about to add to the root set, and if it
is, to skip the segment. The problems are that this solution is rather
awkward and that it works only for one code segment. This means that if
your program has more than one code segment, all of them but one will be
added to the root set. This isn't a big problem in fact, since the
collector will continue to work correctly, but it may be slower.
Anyway, the code which decides whether to skip a segment or not can be
removed simply by not defining AMIGA_SKIP_SEG. But notice that if you do
so, gctest will complain (it will say that "GC_is_visible produced wrong
failure indication"). However, it may be useful if you happen to have
pointers stored in a code segment (you really shouldn't).
If anyone has a good solution to the problem of finding, when a program
is loaded in memory, whether a segment is a code or a data segment,
please let me know.
PROBLEMS
If you have any problem with this version, please contact me at
[email protected] (but do *not* send long files, since we pay for
every mail!).
===========================================================================
Jesper Peterson's notes
===========================================================================
ADDITIONAL NOTES FOR AMIGA PORT
These notes assume some familiarity with Amiga internals.
WHY I PORTED TO THE AMIGA
The sole reason why I made this port was as a first step in getting
the Sather(*) language on the Amiga. A port of this language will
be done as soon as the Sather 1.0 sources are made available to me.
Given this motivation, the garbage collection (GC) port is rather
minimal.
(*) For information on Sather read the comp.lang.sather newsgroup.
LIMITATIONS
This port assumes that the startup code linked with target programs
is that supplied with SAS/C versions 6.0 or later. This allows
assumptions to be made about where to find the stack base pointer
and data segments when programs are run from WorkBench, as opposed
to running from the CLI. The compiler dependent code is all in the
GC_get_stack_base() and GC_register_data_segments() functions, but
may spread as I add Amiga specific features.
Given that SAS/C was assumed, the port is set up to be built with
"smake" using the "SMakefile". Compiler options in "SCoptions" can
be set with "scopts" program. Both "smake" and "scopts" are part of
the SAS/C commercial development system.
In keeping with the porting philosophy outlined above, this port
will not behave well with Amiga specific code. Especially not inter-
process comms via messages, and setting up public structures like
Intuition objects or anything else in the system lists. For the
time being the use of this library is limited to single threaded
ANSI/POSIX compliant or near-complient code. (ie. Stick to stdio
for now). Given this limitation there is currently no mechanism for
allocating "CHIP" or "PUBLIC" memory under the garbage collector.
I'll add this after giving it considerable thought. The major
problem is the entire physical address space may have to me scanned,
since there is no telling who we may have passed memory to.
If you allocate your own stack in client code, you will have to
assign the pointer plus stack size to GC_stackbottom.
The initial stack size of the target program can be compiled in by
setting the __stack symbol (see SAS documentaion). It can be over-
ridden from the CLI by running the AmigaDOS "stack" program, or from
the WorkBench by setting the stack size in the tool types window.
SAS/C COMPILER OPTIONS (SCoptions)
You may wish to check the "CPU" code option is appropriate for your
intended target system.
Under no circumstances set the "StackExtend" code option in either
compiling the library or *ANY* client code.
All benign compiler warnings have been suppressed. These mainly
involve lack of prototypes in the code, and dead assignments
detected by the optimizer.
THE GOOD NEWS
The library as it stands is compatible with the GigaMem commercial
virtual memory software, and probably similar PD software.
The performance of "gctest" on an Amiga 2630 (68030 @ 25Mhz)
compares favourably with an HP9000 with similar architecture (a 325
with a 68030 I think).
-----------------------------------------------------------------------
The Amiga port has been brought to you by:
Jesper Peterson.
[email protected] (preferred, but 1 week turnaround)
[email protected] (that's orca<one>, 1 day turnaround)
At least one of these addresses should be around for a while, even
though I don't work for either of the companies involved.
-68
View File
@@ -1,68 +0,0 @@
From: Margaret Fleck
Here's the key details of what worked for me, in case anyone else needs them.
There may well be better ways to do some of this, but ....
-- Margaret
The badge4 has a StrongArm-1110 processor and a StrongArm-1111 coprocessor.
Assume that the garbage collector distribution is unpacked into /home/arm/gc6.0,
which is visible to both the ARM machine and a linux desktop (e.g. via NFS mounting).
Assume that you have a file /home/arm/config.site with contents something like the
example attached below. Notice that our local ARM toolchain lives in
/skiff/local.
Go to /home/arm/gc6.0 directory. Do
CONFIG_SITE=/home/arm/config.site ./configure --target=arm-linux
--prefix=/home/arm/gc6.0
On your desktop, do:
make
make install
The main garbage collector library should now be in ../gc6.0/lib/libgc.so.
To test the garbage collector, first do the following on your desktop
make gctest
./gctest
Then do the following on the ARM machine
cd .libs
./lt-gctest
Do not try to do "make test" (the usual way of running the test
program). This does not work and seems to erase some of the important
files.
The gctest program claims to have succeeded. Haven't run any further tests
with it, though I'll be doing so in the near future.
-------------------------------
# config.site for configure
# Modified from the one provided by Bradley D. LaRonde
# Edited by Andrej Cedilnik <[email protected]>
# Used some of solutions by Tilman Vogel <[email protected]>
# Ported for iPAQ Familiar by Oliver Kurth <[email protected]>
# Further modified by Margaret Fleck for the badge4
HOSTCC=gcc
# Names of the cross-compilers
CC=/skiff/local/bin/arm-linux-gcc
CXX=/skiff/local/bin/arm-linux-gcc
# The cross compiler specific options
CFLAGS="-O2 -fno-exceptions"
CXXFLAGS="-O2 -fno-exceptions"
CPPFLAGS="-O2 -fno-exceptions"
LDFLAGS=""
# Some other programs
AR=/skiff/local/bin/arm-linux-ar
RANLIB=/skiff/local/bin/arm-linux-ranlib
NM=/skiff/local/bin/arm-linux-nm
ac_cv_path_NM=/skiff/local/bin/arm-linux-nm
ac_cv_func_setpgrp_void=yes
x_includes=/skiff/local/arm-linux/include/X11
x_libraries=/skiff/local/arm-linux/lib/X11
-59
View File
@@ -1,59 +0,0 @@
As of GC6.0alpha8, we attempt to support GNU-style builds based on automake,
autoconf and libtool. This is based almost entirely on Tom Tromey's work
with gcj.
To build and install libraries use
configure; make; make install
The advantages of this process are:
1) It should eventually do a better job of automatically determining the
right compiler to use, etc. It probably already does in some cases.
2) It tries to automatically set a good set of default GC parameters for
the platform (e.g. thread support). It provides an easier way to configure
some of the others.
3) It integrates better with other projects using a GNU-style build process.
4) It builds both dynamic and static libraries.
The known disadvantages are:
1) The build scripts are much more complex and harder to debug (though largely
standard). I don't understand them all, and there's probably lots of redundant
stuff.
2) It probably doesn't work on all Un*x-like platforms yet. It probably will
never work on the rest.
3) The scripts are not yet complete. Some of the standard GNU targets don't
yet work. (Corrections/additions are very welcome.)
The distribution should contain all files needed to run "configure" and "make",
as well as the sources needed to regenerate the derived files. (If I missed
some, please let me know.)
Note that the distribution comes with a "Makefile" which will be overwritten
by "configure" with one that is not at all equiavelent to the original. The
distribution contains a copy of the original "Makefile" in "Makefile.direct".
Important options to configure:
--prefix=PREFIX install architecture-independent files in PREFIX
[/usr/local]
--exec-prefix=EPREFIX install architecture-dependent files in EPREFIX
[same as prefix]
--enable-threads=TYPE choose threading package
--enable-parallel-mark parallelize marking and free list construction
--enable-full-debug include full support for pointer backtracing etc.
Unless --prefix is set (or --exec-prefix or one of the more obscure options),
make install will install libgc.a and libgc.so in /usr/local/bin, which
would typically require the "make install" to be run as root.
Most commonly --enable-threads=posix or will be needed. --enable-parallel-mark
is recommended for multiprocessors if it is supported on the platform.
File diff suppressed because it is too large. Load diff
-57
View File
@@ -1,57 +0,0 @@
This is an attempt to acknowledge early contributions to the garbage
collector. Later contributions should instead be mentioned in
README.changes.
HISTORY -
Early versions of this collector were developed as a part of research
projects supported in part by the National Science Foundation
and the Defense Advance Research Projects Agency.
The garbage collector originated as part of the run-time system for
the Russell programming language implementation. The first version of the
garbage collector was written primarily by Al Demers. It was then refined
and mostly rewritten, primarily by Hans-J. Boehm, at Cornell U.,
the University of Washington, Rice University (where it was first used for
C and assembly code), Xerox PARC, SGI, and HP Labs. However, significant
contributions have also been made by many others.
Some other contributors:
More recent contributors are mentioned in the modification history in
README.changes. My apologies for any omissions.
The SPARC specific code was originally contributed by Mark Weiser.
The Encore Multimax modifications were supplied by
Kevin Kenny ([email protected]). The adaptation to the IBM PC/RT is largely
due to Vernon Lee, on machines made available to Rice by IBM.
Much of the HP specific code and a number of good suggestions for improving the
generic code are due to Walter Underwood.
Robert Brazile ([email protected]) originally supplied the ULTRIX code.
Al Dosser ([email protected]) and Regis Cridlig ([email protected])
subsequently provided updates and information on variation between ULTRIX
systems. Parag Patel ([email protected]) supplied the A/UX code.
Jesper Peterson([email protected]), Michel Schinz, and
Martin Tauchmann ([email protected]) supplied the Amiga port.
Thomas Funke ([email protected](?)) and
Brian D.Carlstrom ([email protected]) supplied the NeXT ports.
Douglas Steel ([email protected]) provided ICL DRS6000 code.
Bill Janssen ([email protected]) supplied the SunOS dynamic loader
specific code. Manuel Serrano ([email protected]) supplied linux and
Sony News specific code. Al Dosser provided Alpha/OSF/1 code. He and
Dave Detlefs([email protected]) also provided several generic bug fixes.
Alistair G. Crooks([email protected]) supplied the NetBSD and 386BSD ports.
Jeffrey Hsu ([email protected]) provided the FreeBSD port.
Brent Benson ([email protected]) ported the collector to
a Motorola 88K processor running CX/UX (Harris NightHawk).
Ari Huttunen ([email protected]) generalized the OS/2 port to
nonIBM development environments (a nontrivial task).
Patrick Beard ([email protected]) provided the initial MacOS port.
David Chase, then at Olivetti Research, suggested several improvements.
Scott Schwartz ([email protected]) supplied some of the
code to save and print call stacks for leak detection on a SPARC.
Jesse Hull and John Ellis supplied the C++ interface code.
Zhong Shao performed much of the experimentation that led to the
current typed allocation facility. (His dynamic type inference code hasn't
made it into the released version of the collector, yet.)
-53
View File
@@ -1,53 +0,0 @@
Copyright (c) 1993-1994 by Xerox Corporation. All rights reserved.
THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
Permission is hereby granted to use or copy this program
for any purpose, provided the above notices are retained on all copies.
Permission to modify the code and to distribute modified code is granted,
provided the above notices are retained, and a notice that the code was
modified is included with the above copyright notice.
Please send bug reports to Hans-J. Boehm ([email protected] or
[email protected]).
This is a string packages that uses a tree-based representation.
See cord.h for a description of the functions provided. Ec.h describes
"extensible cords", which are essentially output streams that write
to a cord. These allow for efficient construction of cords without
requiring a bound on the size of a cord.
More details on the data structure can be found in
Boehm, Atkinson, and Plass, "Ropes: An Alternative to Strings",
Software Practice and Experience 25, 12, December 1995, pp. 1315-1330.
A fundamentally similar "rope" data structure is also part of SGI's standard
template library implementation, and its descendents, which include the
GNU C++ library. That uses reference counting by default.
There is a short description of that data structure at
http://reality.sgi.com/boehm/ropeimpl.html . (The more official location
http://www.sgi.com/tech/stl/ropeimpl.html is missing a figure.)
All of these are descendents of the "ropes" in Xerox Cedar.
de.c is a very dumb text editor that illustrates the use of cords.
It maintains a list of file versions. Each version is simply a
cord representing the file contents. Nonetheless, standard
editing operations are efficient, even on very large files.
(Its 3 line "user manual" can be obtained by invoking it without
arguments. Note that ^R^N and ^R^P move the cursor by
almost a screen. It does not understand tabs, which will show
up as highlighred "I"s. Use the UNIX "expand" program first.)
To build the editor, type "make cord/de" in the gc directory.
This package assumes an ANSI C compiler such as gcc. It will
not compile with an old-style K&R compiler.
Note that CORD_printf iand friends use C functions with variable numbers
of arguments in non-standard-conforming ways. This code is known to
break on some platforms, notably PowerPC. It should be possible to
build the remainder of the library (everything but cordprnt.c) on
any platform that supports the collector.
-141
View File
@@ -1,141 +0,0 @@
6.5 update:
I disabled incremental GC on Darwin in this version, since I couldn't
get gctest to pass when the GC was built as a dynamic library. Building
with -DMPROTECT_VDB (and threads) on the command line should get you
back to the old state. - HB
./configure --enable-cplusplus results in a "make check" failure, probably
because the ::delete override ends up in a separate dl, and Darwin dynamic
loader semantics appear to be such that this is not really visible to the
main program, unlike on ELF systems. Someone who understands dynamic
loading needs to lookat this. For now, gc_cpp.o needs to be linked
statically, if needed. - HB
Darwin/MacOSX Support - December 16, 2003
=========================================
Important Usage Notes
=====================
GC_init() MUST be called before calling any other GC functions. This
is necessary to properly register segments in dynamic libraries. This
call is required even if you code does not use dynamic libraries as the
dyld code handles registering all data segments.
When your use of the garbage collector is confined to dylibs and you
cannot call GC_init() before your libraries' static initializers have
run and perhaps called GC_malloc(), create an initialization routine
for each library to call GC_init():
#include <gc/gc.h>
extern "C" void my_library_init() { GC_init(); }
Compile this code into a my_library_init.o, and link it into your
dylib. When you link the dylib, pass the -init argument with
_my_library_init (e.g. gcc -dynamiclib -o my_library.dylib a.o b.o c.o
my_library_init.o -init _my_library_init). This causes
my_library_init() to be called before any static initializers, and
will initialize the garbage collector properly.
Note: It doesn't hurt to call GC_init() more than once, so it's best,
if you have an application or set of libraries that all use the
garbage collector, to create an initialization routine for each of
them that calls GC_init(). Better safe than sorry.
The incremental collector is still a bit flaky on darwin. It seems to
work reliably with workarounds for a few possible bugs in place however
these workaround may not work correctly in all cases. There may also
be additional problems that I have not found.
Thread-local GC allocation will not work with threads that are not
created using the GC-provided override of pthread_create(). Threads
created without the GC-provided pthread_create() do not have the
necessary data structures in the GC to store this data.
Implementation Information
==========================
Darwin/MacOSX support is nearly complete. Thread support is reliable on
Darwin 6.x (MacOSX 10.2) and there have been reports of success on older
Darwin versions (MacOSX 10.1). Shared library support had also been
added and the gc can be run from a shared library. There is currently only
support for Darwin/PPC although adding x86 support should be trivial.
Thread support is implemented in terms of mach thread_suspend and
thread_resume calls. These provide a very clean interface to thread
suspension. This implementation doesn't rely on pthread_kill so the
code works on Darwin < 6.0 (MacOSX 10.1). All the code to stop and
start the world is located in darwin_stop_world.c.
Since not all uses of the GC enable clients to override pthread_create()
before threads have been created, the code for stopping the world has
been rewritten to look for threads using Mach kernel calls. Each
thread identified in this way is suspended and resumed as above. In
addition, since Mach kernel threads do not contain pointers to their
stacks, a stack-walking function has been written to find the stack
limits. Given an initial stack pointer (for the current thread, a
pointer to a stack-allocated local variable will do; for a non-active
thread, we grab the value of register 1 (on PowerPC)), it
will walk the PPC Mach-O-ABI compliant stack chain until it reaches the
top of the stack. This appears to work correctly for GCC-compiled C,
C++, Objective-C, and Objective-C++ code, as well as for Java
programs that use JNI. If you run code that does not follow the stack
layout or stack pointer conventions laid out in the PPC Mach-O ABI,
then this will likely crash the garbage collector.
The original incremental collector support unfortunatelly no longer works
on recent Darwin versions. It also relied on some undocumented kernel
structures. Mach, however, does have a very clean interface to exception
handing. The current implementation uses Mach's exception handling.
Much thanks goes to Andrew Stone, Dietmar Planitzer, Andrew Begel,
Jeff Sturm, and Jesse Rosenstock for all their work on the
Darwin/OS X port.
-Brian Alliet
[email protected]
Older Information (Most of this no longer applies to the current code)
======================================================================
While the GC should work on MacOS X Server, MacOS X and Darwin, I only tested
it on MacOS X Server.
I've added a PPC assembly version of GC_push_regs(), thus the setjmp() hack is
no longer necessary. Incremental collection is supported via mprotect/signal.
The current solution isn't really optimal because the signal handler must decode
the faulting PPC machine instruction in order to find the correct heap address.
Further, it must poke around in the register state which the kernel saved away
in some obscure register state structure before it calls the signal handler -
needless to say the layout of this structure is no where documented.
Threads and dynamic libraries are not yet supported (adding dynamic library
support via the low-level dyld API shouldn't be that hard).
The original MacOS X port was brought to you by Andrew Stone.
June, 1 2000
Dietmar Planitzer
[email protected]
Note from Andrew Begel:
One more fix to enable gc.a to link successfully into a shared library for
MacOS X. You have to add -fno-common to the CFLAGS in the Makefile. MacOSX
disallows common symbols in anything that eventually finds its way into a
shared library. (I don't completely understand why, but -fno-common seems to
work and doesn't mess up the garbage collector's functionality).
Feb 26, 2003
Jeff Sturm and Jesse Rosenstock provided a patch that adds thread support.
GC_MACOSX_THREADS should be defined in the build and in clients. Real
dynamic library support is still missing, i.e. dynamic library data segments
are still not scanned. Code that stores pointers to the garbage collected
heap in statically allocated variables should not reside in a dynamic
library. This still doesn't appear to be 100% reliable.
Mar 10, 2003
Brian Alliet contributed dynamic library support for MacOSX. It could also
use more testing.
-12
View File
@@ -1,12 +0,0 @@
[Original version supplied by Xiaokun Zhu <[email protected]>]
[This version came mostly from Gary Leavens. ]
Look first at Makefile.dj, and possibly change the definitions of
RM and MV if you don't have rm and mv installed.
Then use Makefile.dj to compile the garbage collector.
For example, you can do:
make -f Makefile.dj test
All the tests should work fine.
-151
View File
@@ -1,151 +0,0 @@
The garbage collector looks at a number of environment variables which are
then used to affect its operation. These are examined only on Un*x-like
platforms and win32.
GC_INITIAL_HEAP_SIZE=<bytes> - Initial heap size in bytes. May speed up
process start-up.
GC_MAXIMUM_HEAP_SIZE=<bytes> - Maximum collected heap size.
GC_LOOP_ON_ABORT - Causes the collector abort routine to enter a tight loop.
This may make it easier to debug, such a process, especially
for multithreaded platforms that don't produce usable core
files, or if a core file would be too large. On some
platforms, this also causes SIGSEGV to be caught and
result in an infinite loop in a handler, allowing
similar debugging techniques.
GC_PRINT_STATS - Turn on as much logging as is easily feasible without
adding signifcant runtime overhead. Doesn't work if
the collector is built with SMALL_CONFIG. Overridden
by setting GC_quiet. On by default if the collector
was built without -DSILENT.
GC_DUMP_REGULARLY - Generate a GC debugging dump GC_dump() on startup
and during every collection. Very verbose. Useful
if you have a bug to report, but please include only the
last complete dump.
GC_BACKTRACES=<n> - Generate n random backtraces (for heap profiling) after
each GC. Collector must have been built with
KEEP_BACK_PTRS. This won't generate useful output unless
most objects in the heap were allocated through debug
allocators. This is intended to be only a statistical
sample; individual traces may be erroneous due to
concurrent heap mutation.
GC_PRINT_ADDRESS_MAP - Linux only. Dump /proc/self/maps, i.e. various address
maps for the process, to stderr on every GC. Useful for
mapping root addresses to source for deciphering leak
reports.
GC_NPROCS=<n> - Linux w/threads only. Explicitly sets the number of processors
that the GC should expect to use. Note that setting this to 1
when multiple processors are available will preserve
correctness, but may lead to really horrible performance,
since the lock implementation will immediately yield without
first spinning.
GC_MARKERS=<n> - Linux w/threads and parallel marker only. Set the number
of marker threads. This is normaly set to the number of
processors. It is safer to adjust GC_MARKERS than GC_NPROCS,
since GC_MARKERS has no impact on the lock implementation.
GC_NO_BLACKLIST_WARNING - Prevents the collector from issuing
warnings about allocations of very large blocks.
Deprecated. Use GC_LARGE_ALLOC_WARN_INTERVAL instead.
GC_LARGE_ALLOC_WARN_INTERVAL=<n> - Print every nth warning about very large
block allocations, starting with the nth one. Small values
of n are generally benign, in that a bounded number of
such warnings generally indicate at most a bounded leak.
For best results it should be set at 1 during testing.
Default is 5. Very large numbers effectively disable the
warning.
GC_IGNORE_GCJ_INFO - Ignore the type descriptors implicitly supplied by
GC_gcj_malloc and friends. This is useful for debugging
descriptor generation problems, and possibly for
temporarily working around such problems. It forces a
fully conservative scan of all heap objects except
those known to be pointerfree, and may thus have other
adverse effects.
GC_PRINT_BACK_HEIGHT - Print max length of chain through unreachable objects
ending in a reachable one. If this number remains
bounded, then the program is "GC robust". This ensures
that a fixed number of misidentified pointers can only
result in a bounded space leak. This currently only
works if debugging allocation is used throughout.
It increases GC space and time requirements appreciably.
This feature is still somewhat experimental, and requires
that the collector have been built with MAKE_BACK_GRAPH
defined. For details, see Boehm, "Bounding Space Usage
of Conservative Garbage Collectors", POPL 2001, or
http://lib.hpl.hp.com/techpubs/2001/HPL-2001-251.html .
GC_RETRY_SIGNALS, GC_NO_RETRY_SIGNALS - Try to compensate for lost
thread suspend signals in linux_threads.c. On by
default for GC_OSF1_THREADS, off otherwise. Note
that this does not work around a possible loss of
thread restart signals. This seems to be necessary for
some versions of Tru64. Since we've previously seen
similar issues on some other operating systems, it
was turned into a runtime flag to enable last-minute
work-arounds.
GC_IGNORE_FB[=<n>] - (Win32 only.) Try to avoid treating a mapped
frame buffer as part of the root set. Certain (higher end?)
graphics cards seems to result in the graphics memory mapped
into the user address space as writable memory.
Unfortunately, there seems to be no systematic way to
identify such memory. Setting the environment variable to n
causes the collector to ignore mappings longer than n MB.
The default value of n is currently 15. (This should cover
a 16 MB graphics card, since the mapping appears to be slightly
shorter than all of graphics memory. It will fail if a dll
writes pointers to collectable objects into a data segment
whose length is >= 15MB. Empirically that's rare, but
certainly possible.) WARNING: Security sensitive applications
should probably disable this feature by setting
GC_disallow_ignore_fb, or by building with -DNO_GETENV,
since small values could force collection of reachable
objects, which is conceivably a (difficult to exploit)
security hole. GC_IGNORE_FB values less than 3 MB
are never honored, eliminating this risk for most,
but not all, applications. This feature is likely to disappear
if/when we find a less disgusting "solution".
IN VERSION 6.4 AND LATER, THIS SHOULD BE UNNECESSARY.
The following turn on runtime flags that are also program settable. Checked
only during initialization. We expect that they will usually be set through
other means, but this may help with debugging and testing:
GC_ENABLE_INCREMENTAL - Turn on incremental collection at startup. Note that,
depending on platform and collector configuration, this
may involve write protecting pieces of the heap to
track modifications. These pieces may include pointerfree
objects or not. Although this is intended to be
transparent, it may cause unintended system call failures.
Use with caution.
GC_PAUSE_TIME_TARGET - Set the desired garbage collector pause time in msecs.
This only has an effect if incremental collection is
enabled. If a collection requires appreciably more time
than this, the client will be restarted, and the collector
will need to do additional work to compensate. The
special value "999999" indicates that pause time is
unlimited, and the incremental collector will behave
completely like a simple generational collector. If
the collector is configured for parallel marking, and
run on a multiprocessor, incremental collection should
only be used with unlimited pause time.
GC_FIND_LEAK - Turns on GC_find_leak and thus leak detection. Forces a
collection at program termination to detect leaks that would
otherwise occur after the last GC.
GC_ALL_INTERIOR_POINTERS - Turns on GC_all_interior_pointers and thus interior
pointer recognition.
GC_DONT_GC - Turns off garbage collection. Use cautiously.
-81
View File
@@ -1,81 +0,0 @@
GC on EWS4800
-------------
1. About EWS4800
EWS4800 is 32bit/64bit workstation.
Vender: NEC Corporation
OS: UX/4800 R9.* - R13.* (SystemV R4.2)
CPU: R4000, R4400, R10000 (MIPS)
2. Compiler
32bit:
Use ANSI C compiler.
CC = /usr/abiccs/bin/cc
64bit:
Use 64bit ANSI C compiler.
CC = /usr/ccs64/bin/cc
AR = /usr/ccs64/bin/ar
3. ELF file format
*** Caution: The following infomation is empirical. ***
32bit:
ELF file has an unique format. (See a.out(4) and end(3C).)
&_start
: text segment
&etext
DATASTART
: data segment (initialized)
&edata
DATASTART2
: data segment (uninitialized)
&end
Here, DATASTART and DATASTART2 are macros of GC, and are defined as
the following equations. (See include/private/gcconfig.h.)
The algorithm for DATASTART is similar with the function
GC_SysVGetDataStart() in os_dep.c.
DATASTART = ((&etext + 0x3ffff) & ~0x3ffff) + (&etext & 0xffff)
Dynamically linked:
DATASTART2 = (&_gp + 0x8000 + 0x3ffff) & ~0x3ffff
Statically linked:
DATASTART2 = &edata
GC has to check addresses both between DATASTART and &edata, and
between DATASTART2 and &end. If a program accesses between &etext
and DATASTART, or between &edata and DATASTART2, the segmentation
error occurs and the program stops.
If a program is statically linked, there is not a gap between
&edata and DATASTART2. The global symbol &_DYNAMIC_LINKING is used
for the detection.
64bit:
ELF file has a simple format. (See end(3C).)
_ftext
: text segment
_etext
_fdata = DATASTART
: data segment (initialized)
_edata
_fbss
: data segment (uninitialized)
_end = DATAEND
--
Hironori SAKAMOTO <[email protected]>
When using the new "configure; make" build process, please
run configure with the --disable-shared option. "Make check" does not
yet pass with dynamic libraries. Ther reasons for that are not yet
understood. (HB, paraphrasing message from Hironori SAKAMOTO.)
-18
View File
@@ -1,18 +0,0 @@
Dynamic loading support requires that executables be linked with -ldld.
The alternative is to build the collector without defining DYNAMIC_LOADING
in gcconfig.h and ensuring that all garbage collectable objects are
accessible without considering statically allocated variables in dynamic
libraries.
The collector should compile with either plain cc or cc -Ae. Cc -Aa
fails to define _HPUX_SOURCE and thus will not configure the collector
correctly.
Incremental collection support was reccently added, and should now work.
In spite of past claims, pthread support under HP/UX 11 should now work.
Define GC_HPUX_THREADS for the build. Incremental collection still does not
work in combination with it.
The stack finding code can be confused by putenv calls before collector
initialization. Call GC_malloc or GC_init before any putenv calls.
Loaded 100 of 542 files, more files were not shown because too many files have changed in this diff. Show more