6 Commits

Author SHA1 Message Date
Steven Wu
d57c7490c5 bitcode support change for fast flags compatibility
Summary: The discussion and as per need, each vendor needs a way to keep the old fast flags and the new fast flags in the auto upgrade path of the IR upgrader.  This revision addresses that issue.

Patched by Michael Berg

Reviewers: qcolombet, hans, steven_wu

Reviewed By: qcolombet, steven_wu

Subscribers: dexonsmith, vsk, mehdi_amini, andrewrk, MatzeB, wristow, spatel

Differential Revision: https://reviews.llvm.org/D43253

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@325525 91177308-0d34-0410-b5e6-96231b3b80d8
2018-02-19 19:22:28 +00:00
Sanjay Patel
00e900afdb [IR] redefine 'UnsafeAlgebra' / 'reassoc' fast-math-flags and add 'trans' fast-math-flag
As discussed on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2016-November/107104.html
and again more recently:
http://lists.llvm.org/pipermail/llvm-dev/2017-October/118118.html

...this is a step in cleaning up our fast-math-flags implementation in IR to better match
the capabilities of both clang's user-visible flags and the backend's flags for SDNode.

As proposed in the above threads, we're replacing the 'UnsafeAlgebra' bit (which had the 
'umbrella' meaning that all flags are set) with a new bit that only applies to algebraic 
reassociation - 'AllowReassoc'.

We're also adding a bit to allow approximations for library functions called 'ApproxFunc' 
(this was initially proposed as 'libm' or similar).

...and we're out of bits. 7 bits ought to be enough for anyone, right? :) FWIW, I did 
look at getting this out of SubclassOptionalData via SubclassData (spacious 16-bits), 
but that's apparently already used for other purposes. Also, I don't think we can just 
add a field to FPMathOperator because Operator is not intended to be instantiated. 
We'll defer movement of FMF to another day.

We keep the 'fast' keyword. I thought about removing that, but seeing IR like this:
%f.fast = fadd reassoc nnan ninf nsz arcp contract afn float %op1, %op2
...made me think we want to keep the shortcut synonym.

Finally, this change is binary incompatible with existing IR as seen in the 
compatibility tests. This statement:
"Newer releases can ignore features from older releases, but they cannot miscompile 
them. For example, if nsw is ever replaced with something else, dropping it would be 
a valid way to upgrade the IR." 
( http://llvm.org/docs/DeveloperPolicy.html#ir-backwards-compatibility )
...provides the flexibility we want to make this change without requiring a new IR 
version. Ie, we're not loosening the FP strictness of existing IR. At worst, we will 
fail to optimize some previously 'fast' code because it's no longer recognized as 
'fast'. This should get fixed as we audit/squash all of the uses of 'isFast()'.

Note: an inter-dependent clang commit to use the new API name should closely follow 
commit.

Differential Revision: https://reviews.llvm.org/D39304



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@317488 91177308-0d34-0410-b5e6-96231b3b80d8
2017-11-06 16:27:15 +00:00
Martin Storsjo
6c132cb749 [AArch64] Extend CallingConv::X86_64_Win64 to AArch64 as well
Rename the enum value from X86_64_Win64 to plain Win64.

The symbol exposed in the textual IR is changed from 'x86_64_win64cc'
to 'win64cc', but the numeric value is kept, keeping support for
old bitcode.

Differential Revision: https://reviews.llvm.org/D34474

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@308208 91177308-0d34-0410-b5e6-96231b3b80d8
2017-07-17 20:05:19 +00:00
Konstantin Zhuravlyov
8f85685860 Enhance synchscope representation
OpenCL 2.0 introduces the notion of memory scopes in atomic operations to
  global and local memory. These scopes restrict how synchronization is
  achieved, which can result in improved performance.

  This change extends existing notion of synchronization scopes in LLVM to
  support arbitrary scopes expressed as target-specific strings, in addition to
  the already defined scopes (single thread, system).

  The LLVM IR and MIR syntax for expressing synchronization scopes has changed
  to use *syncscope("<scope>")*, where <scope> can be "singlethread" (this
  replaces *singlethread* keyword), or a target-specific name. As before, if
  the scope is not specified, it defaults to CrossThread/System scope.

  Implementation details:
    - Mapping from synchronization scope name/string to synchronization scope id
      is stored in LLVM context;
    - CrossThread/System and SingleThread scopes are pre-defined to efficiently
      check for known scopes without comparing strings;
    - Synchronization scope names are stored in SYNC_SCOPE_NAMES_BLOCK in
      the bitcode.

Differential Revision: https://reviews.llvm.org/D21723



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@307722 91177308-0d34-0410-b5e6-96231b3b80d8
2017-07-11 22:23:00 +00:00
Eric Christopher
d8caf85d6d Change the default attributes for llvm.prefetch to inaccessiblemem_or_argmemonly
so that we can perform some optimizations across it.

Fixes PR32365

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@298781 91177308-0d34-0410-b5e6-96231b3b80d8
2017-03-25 20:20:23 +00:00
Vedant Kumar
d1a05a83f8 [Bitcode] Add compatibility test for the 4.0 release
Fork off compatibility.ll for the 4.0 release. The *.bc file in this
commit was produced using a Release build of the release_40 branch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@298109 91177308-0d34-0410-b5e6-96231b3b80d8
2017-03-17 17:53:26 +00:00