llvm/lib/Transforms/InstCombine, branch release_27http://llvm.orghttps://git.amat.us/llvm/atom/lib/Transforms/InstCombine?h=release_272010-03-06T10:59:25ZLet the fallthrough handle whether or not we've changed anything2010-03-06T10:59:25ZEric Christopherechristo@apple.com2010-03-06T10:59:25Zurn:sha1:7b323a34fa034389e9e439f6c02eeb73e87ab9db
before we try to optimize.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97876 91177308-0d34-0410-b5e6-96231b3b80d8
Migrate _chk call lowering from SimplifyLibCalls to InstCombine. Stub2010-03-06T10:50:38ZEric Christopherechristo@apple.com2010-03-06T10:50:38Zurn:sha1:27ceaa13f820e33c44e29adc2095dd3e1e69d09f
out the remainder of the calls that we should lower in some way and
move the tests to the new correct directory. Fix up tests that are now
optimized more than they were before by -instcombine.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97875 91177308-0d34-0410-b5e6-96231b3b80d8
Temporarily revert:2010-03-06T03:11:35ZEric Christopherechristo@apple.com2010-03-06T03:11:35Zurn:sha1:75f0d6953a8492ec21f3b82489090a1152f1f516
Log:
Transform @llvm.objectsize to integer if the argument is a result of malloc of known size.
Modified:
llvm/trunk/lib/Transforms/InstCombine/InstCombineCalls.cpp
llvm/trunk/test/Transforms/InstCombine/objsize.ll
It appears to be causing swb and nightly test failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97866 91177308-0d34-0410-b5e6-96231b3b80d8
Transform @llvm.objectsize to integer if the argument is a result of malloc of known size.2010-03-06T01:01:42ZEvan Chengevan.cheng@apple.com2010-03-06T01:01:42Zurn:sha1:fc8ccfedbba651e324d827de6693aad1491314c7
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97860 91177308-0d34-0410-b5e6-96231b3b80d8
Instcombine should turn llvm.objectsize of a alloca with static size to an integer.2010-03-05T20:47:23ZEvan Chengevan.cheng@apple.com2010-03-05T20:47:23Zurn:sha1:a862326bc72be147d9f668b84b9187258e690dc4
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97827 91177308-0d34-0410-b5e6-96231b3b80d8
fix PR6512, a case where instcombine would incorrectly merge loads2010-03-05T18:53:28ZChris Lattnersabre@nondot.org2010-03-05T18:53:28Zurn:sha1:4c5fb1a311a7a5e3ceb802f5b26329d30f2713b1
from different addr spaces.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97813 91177308-0d34-0410-b5e6-96231b3b80d8
Fix PR6503. This turned into a much more interesting and nasty bug. Various 2010-03-05T08:46:26ZChris Lattnersabre@nondot.org2010-03-05T08:46:26Zurn:sha1:f34f48c57870cf4eb4d48e23b11e29b95d1950c3
parts of the cmp|cmp and cmp&cmp folding logic wasn't prepared for vectors
(unrelated to the bug but noticed while in the code) and the code was
*definitely* not safe to use by the (cast icmp)|(cast icmp) handling logic
that I added in r95855. Fix all this up by changing the various routines
to more consistently use IRBuilder and not pass in the I which had the wrong
type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97801 91177308-0d34-0410-b5e6-96231b3b80d8
simplify some functions and make them work with vector2010-03-05T07:47:57ZChris Lattnersabre@nondot.org2010-03-05T07:47:57Zurn:sha1:d6f645ac09a28b5b45f390b2597444514be8a70e
compares, noticed by inspection.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97795 91177308-0d34-0410-b5e6-96231b3b80d8
Add missing break for Intrinsic::objectsize case. It was falling through to the following Intrinsic::bswap code. I have no idea why it wasn't breaking stuff.2010-03-05T01:22:47ZEvan Chengevan.cheng@apple.com2010-03-05T01:22:47Zurn:sha1:f79d6246e6b6a83e31a1360e80f828707a51f98e
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97774 91177308-0d34-0410-b5e6-96231b3b80d8
This test case:2010-03-03T00:35:56ZBill Wendlingisanbard@gmail.com2010-03-03T00:35:56Zurn:sha1:7d9f2b93a356aa89186522bd61c5c565718ff555
long test(long x) { return (x & 123124) | 3; }
Currently compiles to:
_test:
orl $3, %edi
movq %rdi, %rax
andq $123127, %rax
ret
This is because instruction and DAG combiners canonicalize
(or (and x, C), D) -> (and (or, D), (C | D))
However, this is only profitable if (C & D) != 0. It gets in the way of the
3-addressification because the input bits are known to be zero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97616 91177308-0d34-0410-b5e6-96231b3b80d8