RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2024-12-14 07:31:47 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
36,847 Commits 27 Branches 0 Tags 473 MiB
e984e504b5
Commit Graph

36847 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Evan Cheng
e984e504b5 Fix a potential serious problem where kills belonging to the val# defined by a two-address instruction is also on the val# that defines the input. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47057 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 09:06:18 +00:00
Evan Cheng
e35a6d1dfb * Cannot safely commute an instruction there are other defs which can reach its uses. 
* Ignore copy instructions which have already been coalesced.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47056 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 08:41:08 +00:00
Eli Friedman
35327b3e57 Add a note pointing to PR1996. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47055 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 07:56:04 +00:00
Chris Lattner
86ca3cacc5 In SDISel, for targets that support FORMAL_ARGUMENTS nodes, lower this 
node as soon as we create it in SDISel.  Previously we would lower it in
legalize.  The problem with this is that it only exposes the argument
loads implied by FORMAL_ARGUMENTs after legalize, so that only dag combine 2
can hack on them.  This causes us to miss some optimizations because 
datatype expansion also happens here.

Exposing the loads early allows us to do optimizations on them.  For example
we now compile arg-cast.ll to:

_foo:
	movl	$2147483647, %eax
	andl	8(%esp), %eax
	ret

where we previously produced:

_foo:
	subl	$12, %esp
	movsd	16(%esp), %xmm0
	movsd	%xmm0, (%esp)
	movl	$2147483647, %eax
	andl	4(%esp), %eax
	addl	$12, %esp
	ret

It might also make sense to do this for ISD::CALL nodes, which have implicit
stores on many targets.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47054 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 07:39:09 +00:00
Chris Lattner
9f72d1a730 don't try to avoid inserting loads when lowering FORMAL_ARGUMENTS. 
DAGCombine is now quite good at zapifying them.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47053 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 07:35:30 +00:00
Chris Lattner
fec42eb6da teach dag combiner how to eliminate MERGE_VALUES nodes. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47052 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 07:25:05 +00:00
Nate Begeman
b5041b3085 readme updates 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47051 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 07:06:12 +00:00
Eli Friedman
261c704da1 Add test for PR1996. (This is my first time adding a test for a 
transform, so please review.)



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47050 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 06:55:57 +00:00
Nate Begeman
5e1dd5597e Add testcase for recent legalizer change 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47049 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 06:48:40 +00:00
Nate Begeman
0325d90348 Support legalizing insert_vector_elt on targets where the element 
type is not legal.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47048 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 06:43:04 +00:00
Evan Cheng
b2e72b0feb New tests. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47047 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 03:23:53 +00:00
Evan Cheng
70071434ae Initial support for copy elimination by commuting its definition MI. 
PR1877.
A3 = op A2 B0<kill>                                                                                                                                                                            
...                                                                                                                                                                                     
B1 = A3      <- this copy                                                                                                                                                                      
...                                                                                                                                                                                          
   = op A3   <- more uses                                                                                                                                                                      
                                                                                                                                                                                                 
==>                                                                                                                                                                                             
                                                                                                                                                                                            
B2 = op B0 A2<kill>                                                                                                                                                                            
...                                                                                                                                                                                          
B1 = B2      <- now an identify copy                                                                                                                                                           
...                                                                                                                                                                                          
   = op B2   <- more uses

This speeds up FreeBench/neural by 29%, Olden/bh by 12%, oopack_v1p8 by 53%.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47046 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 03:01:43 +00:00
Nate Begeman
ba8d51c1d7 Make register scavenging happy by not using a reg (CR0) that isn't defined 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47045 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 02:58:33 +00:00
Evan Cheng
d2b8d7bc51 - Added removeValNo() to remove all live ranges of a particular value#. 
- removeRange() can now update value# information.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47044 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 02:48:26 +00:00
Evan Cheng
a4d16a1f0d commuteInstr() can now commute non-ssa machine instrs. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47043 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 02:46:49 +00:00
Evan Cheng
1eb5cf9c7d Added debugging routine dumpUses. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47042 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 02:45:38 +00:00
Dan Gohman
fd29e0eb06 Convert SelectionDAG::ComputeMaskedBits to use APInt instead of uint64_t. 
Add an overload that supports the uint64_t interface for use by clients
that haven't been updated yet.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47039 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-13 00:35:47 +00:00
Dale Johannesen
3006c39e3a __DATA not __DATA__ is the right segment name on darwin. 
Spotted by Nick Kledzik.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47037 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 23:35:09 +00:00
Nate Begeman
bf1caa9854 Remove some dead code 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47036 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 22:54:40 +00:00
Nate Begeman
cdd1eeca2c SSE4.1 64b integer insert/extract pattern support 
Move formats into the formats file


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47035 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 22:51:28 +00:00
Dan Gohman
ffe3e2514a Change APInt::getBitsSet to accept a "half-open" range, where the 
hiBit parameter marks the index one past the last desired set bit.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47032 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 21:47:33 +00:00
Duncan Sands
caf4fbd5df Note that these instructions are for x86-32 linux 
(the only platform on which the Ada compiler even
builds).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47031 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 21:40:21 +00:00
Duncan Sands
15bfd89b99 Add more spacing. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47028 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 21:28:39 +00:00
Duncan Sands
d051f7a13d Add instructions for building Ada and Fortran. 
Adjust mentions of gcc4 to be 4.0/4.2 agnostic.
This file should probably be renamed tor
 GCCFEBuildInstrs.html...


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47027 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 21:22:58 +00:00
Owen Anderson
30b4bd4d10 Re-apply the patch to improve the optimizations of memcpy's, with several 
bugs fixed.  This now passes PPC bootstrap.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47026 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 21:15:18 +00:00
Duncan Sands
014e04a5da Generalize getCopyFromParts and getCopyToParts to 
handle arbitrary precision integers and any number
of parts.  For example, on a 32 bit machine an i50
corresponds to two i32 parts.  getCopyToParts will
extend the i50 to an i64 then write half of the i64
to each part; getCopyFromParts will combine the two
i32 parts into an i64 then truncate the result to
i50.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47024 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 20:46:31 +00:00
Evan Cheng
394d28048c Revert r46916 PPCTargetAsmInfo.cpp. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47020 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 19:25:12 +00:00
Evan Cheng
9f143ce989 Only using x86-64 rip relative addressing in non-staic mode? 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47019 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 19:20:46 +00:00
Evan Cheng
073c5b721d Don't mask the isel bug. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47018 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 19:11:29 +00:00
Evan Cheng
aeb231cb59 This test assumes no SSE4.1. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47017 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 19:11:08 +00:00
Dan Gohman
14471be9ab Add a doxygen comment for SrcValueSDNode, to make its purpose 
clear and to clarify how it differs from MemOperandSDNode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47015 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 18:52:52 +00:00
Devang Patel
2be680481b Remove dead code. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47012 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 18:20:50 +00:00
Wojciech Matyjewicz
52f321f525 Now that ScalarEvolution::print writes to the correct stream, there is 
no need to redirect stderr into stdout.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47009 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 15:12:40 +00:00
Wojciech Matyjewicz
2a9054cfd6 Change negative grep into positive one in my yesterday's testcase. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47008 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 15:10:35 +00:00
Wojciech Matyjewicz
7b5b76835a Fix PR2002. Suppose n is the initial value for the induction 
variable (with step 1) and m is its final value. Then, the correct trip 
count is SMAX(m,n)-n. Previously, we used SMAX(0,m-n), but m-n may 
overflow and can't in general be interpreted as signed.

Patch by Nick Lewycky.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47007 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 15:09:36 +00:00
Eli Friedman
b6c36e4058 Fix for bug 1996: optimize out loads of undef. This code basically just 
checks for a malloc/alloca immediately followed by a load.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47006 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 12:08:14 +00:00
Evan Cheng
add251754c Update comment. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47002 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:59:55 +00:00
Evan Cheng
5ae45caf79 Unbreak various insert_vector_elt and extract_vector_elt tests in presence of SSE4. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47001 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:59:45 +00:00
Tanya Lattner
3c593c947c Test case for annotate builtin. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46999 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:46:33 +00:00
Evan Cheng
65b2e3df0d Fix some test cases. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46998 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:22:46 +00:00
Bill Wendling
0602df7403 More violation fixes. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46997 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:12:46 +00:00
Bill Wendling
58d96d6e3c Fix validation errors: di -> dt. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46996 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:09:05 +00:00
Bill Wendling
643eb5d7ba Reformatting. Adding "doc_code" divisions for code examples. Updated some 
of the examples to reflect the current .TD files.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46995 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 07:06:19 +00:00
Chris Lattner
2b659ef4bd rip out llvm 2.2 stuff in preparation for llvm 2.3 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46993 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 06:29:45 +00:00
Chris Lattner
18232a5fd1 heavily refactor this to have less duplication between BUILD_FOR_WEBSITE mode and not. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46991 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 06:07:47 +00:00
Chris Lattner
02b81fe5f5 remove unknown details 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46985 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 05:07:12 +00:00
Tanya Lattner
dd50f0f954 Add 2.2 release tag. 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46981 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-12 02:42:55 +00:00
Nate Begeman
5bb013c34a Stuff noticed while grepping code 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46979 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-11 23:47:56 +00:00
Dan Gohman
d3993d13be Correct the order of the arguments in the examples in the comments 
for APInt::getBitsSet. And fix an off-by-one bug in "wrapping" mode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46978 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-11 23:45:14 +00:00
Ted Kremenek
a618f82c9f The factories for ImutAVLTree/ImmutableSet/ImmutableMap now take an (optional) 
BumpPtrAllocator argument to their constructors.  This BumpPtrAllocator
will be used to allocate trees.  If no BumpPtrAllocator is provided, one
is created (as before).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46975 91177308-0d34-0410-b5e6-96231b3b80d8
 2008-02-11 23:11:12 +00:00