llvm/test/Transforms/ConstProp, branch release_26http://llvm.orghttps://git.amat.us/llvm/atom/test/Transforms/ConstProp?h=release_262009-09-04T19:24:58ZMerge 80762 from mainline.2009-09-04T19:24:58ZTanya Lattnertonic@nondot.org2009-09-04T19:24:58Zurn:sha1:15c0e3ee5d13e6686583fc6b9e504d0dcfb793c1
fix PR4848 an infinite loop when indexing down through a recursive gep
and we get the original pointer type. This doesn't mean that we're
at the first pointer being indexed. Correct the predicate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_26@81028 91177308-0d34-0410-b5e6-96231b3b80d8
Remove the vicmp and vfcmp instructions. Because we never had a release with2009-07-08T03:04:38ZNick Lewyckynicholas@mxc.ca2009-07-08T03:04:38Zurn:sha1:7f6aa2b162e5daaf7b9ccf05d749597d3d7cf460
these instructions, no autoupgrade or backwards compatibility support is
provided.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@74991 91177308-0d34-0410-b5e6-96231b3b80d8
Expand this test to handle more cases (remainder and shifts) of zero.2009-06-21T01:56:41ZNick Lewyckynicholas@mxc.ca2009-06-21T01:56:41Zurn:sha1:e47f59db448fd468bb32ade37a1fa7ee8a9bff20
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@73839 91177308-0d34-0410-b5e6-96231b3b80d8
implement PR4424: 0/x is always 0 for integer division.2009-06-21T01:15:55ZChris Lattnersabre@nondot.org2009-06-21T01:15:55Zurn:sha1:1db6866147b6b3028fdc77bc242de8f44edd71d9
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@73835 91177308-0d34-0410-b5e6-96231b3b80d8
Split the Add, Sub, and Mul instruction opcodes into separate2009-06-04T22:49:04ZDan Gohmangohman@apple.com2009-06-04T22:49:04Zurn:sha1:ae3a0be92e33bc716722aa600983fc1535acb122
integer and floating-point opcodes, introducing
FAdd, FSub, and FMul.
For now, the AsmParser, BitcodeReader, and IRBuilder all preserve
backwards compatability, and the Core LLVM APIs preserve backwards
compatibility for IR producers. Most front-ends won't need to change
immediately.
This implements the first step of the plan outlined here:
http://nondot.org/sabre/LLVMNotes/IntegerOverflow.txt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72897 91177308-0d34-0410-b5e6-96231b3b80d8
apply Eli's patch for PR2165 and provide a testcase.2008-10-16T05:26:51ZChris Lattnersabre@nondot.org2008-10-16T05:26:51Zurn:sha1:868510328fc0b25ca97f2fba466a77c2c4e29c45
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57625 91177308-0d34-0410-b5e6-96231b3b80d8
Fix a case where vector comparison constant folding would cause an2008-07-10T00:29:28ZChris Lattnersabre@nondot.org2008-07-10T00:29:28Zurn:sha1:fea85c46c0baae0d40584e3901607ed3db55d6ef
infinite recursion. part of PR2529
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53383 91177308-0d34-0410-b5e6-96231b3b80d8
elementwise comparison of vector constants was completely wrong. Fix2008-07-10T00:08:17ZChris Lattnersabre@nondot.org2008-07-10T00:08:17Zurn:sha1:330e7e4d250653243afaaeef4c01a7c12c314b39
it for PR2529
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53380 91177308-0d34-0410-b5e6-96231b3b80d8
'Optimize' test2008-07-08T18:33:33ZChris Lattnersabre@nondot.org2008-07-08T18:33:33Zurn:sha1:d038c275163169e8ac51cfa65a13f2965c20162c
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53242 91177308-0d34-0410-b5e6-96231b3b80d8
Fix three bugs:2008-07-08T05:46:34ZChris Lattnersabre@nondot.org2008-07-08T05:46:34Zurn:sha1:155a490258a3a58f77ead53daa2078c3898938b5
1) evaluate [v]fcmp true/false with undefs to true or false instead
of undef.
2) fix vector comparisons with undef to return a vector result instead
of i1
3) fix vector comparisons with evaluatable results to return vector
true/false instead of i1 true/false (PR2529)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53220 91177308-0d34-0410-b5e6-96231b3b80d8