llvm/test/CodeGen, branch release_28http://llvm.orghttps://git.amat.us/llvm/atom/test/CodeGen?h=release_282010-09-10T10:21:50ZApproved by Evan.2010-09-10T10:21:50ZBill Wendlingisanbard@gmail.com2010-09-10T10:21:50Zurn:sha1:3c941a5acb0c466307af6c585339ab588e290395
$ svn merge -c 113576 https://llvm.org/svn/llvm-project/llvm/trunk
--- Merging r113576 into '.':
U test/CodeGen/ARM/2007-01-19-InfiniteLoop.ll
U lib/Target/ARM/ARMLoadStoreOptimizer.cpp
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_28@113583 91177308-0d34-0410-b5e6-96231b3b80d8
Remove the rest of the nonexistent 64-bit AVX instructions.2010-09-03T21:23:00ZDale Johannesendalej@apple.com2010-09-03T21:23:00Zurn:sha1:caa9ba228d9d8326cd54457be6e9c2ae313db75c
Bruno, please review.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113014 91177308-0d34-0410-b5e6-96231b3b80d8
Re-apply r112883:2010-09-03T18:37:12ZJim Grosbachgrosbach@apple.com2010-09-03T18:37:12Zurn:sha1:65482b1bb873dd820f54a24a2f34bd65f2669e5c
"For ARM stack frames that utilize variable sized objects and have either
large local stack areas or require dynamic stack realignment, allocate a
base register via which to access the local frame. This allows efficient
access to frame indices not accessible via the FP (either due to being out
of range or due to dynamic realignment) or the SP (due to variable sized
object allocation). In particular, this greatly improves efficiency of access
to spill slots in Thumb functions which contain VLAs."
r112986 fixed a latent bug exposed by the above.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112989 91177308-0d34-0410-b5e6-96231b3b80d8
Revert "For ARM stack frames that utilize variable sized objects and have either", it is breaking oggenc with Clang for ARMv6.2010-09-03T15:26:42ZDaniel Dunbardaniel@zuster.org2010-09-03T15:26:42Zurn:sha1:6a8700301ca6f8f2f5f787c8d1f5206a7dfceed6
This reverts commit 8d6e29cfda270be483abf638850311670829ee65.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112962 91177308-0d34-0410-b5e6-96231b3b80d8
test/CodeGen/X86: Add explicit -mtriple=(i686|x86_64)-linux for Win32 host.2010-09-03T03:24:08ZNAKAMURA Takumigeek4civic@gmail.com2010-09-03T03:24:08Zurn:sha1:9dceb1305cb3366e20ce27f0370c19adf13c6b0c
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112947 91177308-0d34-0410-b5e6-96231b3b80d8
AVX doesn't support mm operations neither its instrinsics.2010-09-03T02:08:45ZBruno Cardoso Lopesbruno.cardoso@gmail.com2010-09-03T02:08:45Zurn:sha1:95f1e2d6b52d55773bca16a5d3b7b283344dc47d
The AVX versions of PALIGN and PABS* should only exist for
128-bit. Remove the unnecessary stuff.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112944 91177308-0d34-0410-b5e6-96231b3b80d8
Replace NEON vabdl, vaba, and vabal intrinsics with combinations of the2010-09-03T01:35:08ZBob Wilsonbob.wilson@apple.com2010-09-03T01:35:08Zurn:sha1:eb0c3d372906df9c61a31651a0ba278034447e94
vabd intrinsic and add and/or zext operations. In the case of vaba, this
also avoids the need for a DAG combine pattern to combine vabd with add.
Update tests. Auto-upgrade the old intrinsics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112941 91177308-0d34-0410-b5e6-96231b3b80d8
Properly emit __chkstk call instead of __alloca on non-mingw windows targets.2010-09-02T23:03:46ZAnton Korobeynikovasl@math.spbu.ru2010-09-02T23:03:46Zurn:sha1:ace53f2fbc65f1e26c084c733ffbdbe3516580a6
Patch by Cameron Esfahani!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112902 91177308-0d34-0410-b5e6-96231b3b80d8
For ARM stack frames that utilize variable sized objects and have either2010-09-02T22:29:01ZJim Grosbachgrosbach@apple.com2010-09-02T22:29:01Zurn:sha1:1755b3964f931bdd6fa9b4c0138f666ccfa12aca
large local stack areas or require dynamic stack realignment, allocate a
base register via which to access the local frame. This allows efficient
access to frame indices not accessible via the FP (either due to being out
of range or due to dynamic realignment) or the SP (due to variable sized
object allocation). In particular, this greatly improves efficiency of access
to spill slots in Thumb functions which contain VLAs.
rdar://7352504
rdar://8374540
rdar://8355680
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112883 91177308-0d34-0410-b5e6-96231b3b80d8
Don't narrow the load and store in a load+twiddle+store sequence unless2010-09-02T21:18:42ZDan Gohmangohman@apple.com2010-09-02T21:18:42Zurn:sha1:24bde5bce192119ee0fc4f94ef8757fd4031e5f6
there are clearly no stores between the load and the store. This fixes
this miscompile reported as PR7833.
This breaks the test/CodeGen/X86/narrow_op-2.ll optimization, which is
safe, but awkward to prove safe. Move it to X86's README.txt.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112861 91177308-0d34-0410-b5e6-96231b3b80d8