llvm/lib/CodeGen/SelectionDAG, branch release_31http://llvm.orghttps://git.amat.us/llvm/atom/lib/CodeGen/SelectionDAG?h=release_312012-05-08T07:58:10ZNew Revision: 1557492012-05-08T07:58:10ZBill Wendlingisanbard@gmail.com2012-05-08T07:58:10Zurn:sha1:46bbc7c92be586e263a8cd8ae78f9d8d5b3178e4
URL: http://llvm.org/viewvc/llvm-project?rev=155749&view=rev
Log:
Reapply 155668: Fix the SD scheduler to avoid gluing the same node twice.
This time, also fix the caller of AddGlue to properly handle
incomplete chains. AddGlue had failure modes, but shamefully hid them
from its caller. It's luck ran out.
Fixes rdar://11314175: BuildSchedUnits assert.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_31@156376 91177308-0d34-0410-b5e6-96231b3b80d8
Merging r155954:2012-05-01T22:51:49ZBill Wendlingisanbard@gmail.com2012-05-01T22:51:49Zurn:sha1:e77bf55ebeab4abb6fa2504506e42dc8db97f232
------------------------------------------------------------------------
r155954 | void | 2012-05-01 15:50:45 -0700 (Tue, 01 May 2012) | 3 lines
Strip the pointer casts off of allocas so that the selection DAG can find them.
PR10799
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_31@155955 91177308-0d34-0410-b5e6-96231b3b80d8
Merging r155668:2012-04-26T23:04:56ZBill Wendlingisanbard@gmail.com2012-04-26T23:04:56Zurn:sha1:8d1440d0217ab3b09a407c42f7559ad641c7077f
------------------------------------------------------------------------
r155668 | atrick | 2012-04-26 14:48:25 -0700 (Thu, 26 Apr 2012) | 8 lines
Fix the SD scheduler to avoid gluing the same node twice.
DAGCombine strangeness may result in multiple loads from the same
offset. They both may try to glue themselves to another load. We could
insist that the redundant loads glue themselves to each other, but the
beter fix is to bail out from bad gluing at the time we detect it.
Fixes rdar://11314175: BuildSchedUnits assert.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_31@155672 91177308-0d34-0410-b5e6-96231b3b80d8
Fixes a problem in instruction selection with testing whether or not the 2012-04-17T22:23:10ZJoel Jonesjoel_k_jones@apple.com2012-04-17T22:23:10Zurn:sha1:d16ce17711e8e6231363fe1fd47800570b75d61d
transformation:
(X op C1) ^ C2 --> (X op C1) & ~C2 iff (C1&C2) == C2
should be done.
This change has been tested:
Using a debug+asserts build:
on the specific test case that brought this bug to light
make check-all
lnt nt
using this clang to build a release version of clang
Using the release+asserts clang-with-clang build:
on the specific test case that brought this bug to light
make check-all
lnt nt
Checking in because Evan wants it checked in. Test case forthcoming after
scrubbing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154955 91177308-0d34-0410-b5e6-96231b3b80d8
Remove dead SD nodes after the combining pass. Fixes PR12201.2012-04-16T03:33:22ZHal Finkelhfinkel@anl.gov2012-04-16T03:33:22Zurn:sha1:31490baf38680012000890dcb11ac4914ec94911
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154786 91177308-0d34-0410-b5e6-96231b3b80d8
When emulating vselect using OR/AND/XOR make sure to bitcast the result back to the original type.2012-04-15T15:08:09ZNadav Rotemnadav.rotem@intel.com2012-04-15T15:08:09Zurn:sha1:3ab32ea49ec13182f1397dc89c37551692f67140
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154764 91177308-0d34-0410-b5e6-96231b3b80d8
Reapply 154397. Original message:2012-04-11T08:26:11ZNadav Rotemnadav.rotem@intel.com2012-04-11T08:26:11Zurn:sha1:2e506198c8cc7eec8f2f4fb4719a07daef17f931
Fix a dagcombine optimization which assumes that the vsetcc result type is always
of the same size as the compared values. This is ture for SSE/AVX/NEON but not
for all targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154490 91177308-0d34-0410-b5e6-96231b3b80d8
Fix an overly indented line. Remove an 'else' after an 'if' that returns.2012-04-11T04:55:51ZCraig Toppercraig.topper@gmail.com2012-04-11T04:55:51Zurn:sha1:7eb46d834ccca887ab8fbd1d96080f9a4632bb48
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154479 91177308-0d34-0410-b5e6-96231b3b80d8
Inline implVisitAluOverflow by introducing a nested switch to convert the intrinsic to an nodetype.2012-04-11T04:34:11ZCraig Toppercraig.topper@gmail.com2012-04-11T04:34:11Zurn:sha1:c42e640dc8a970c7d16934a16551c4c1a3b3acae
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154478 91177308-0d34-0410-b5e6-96231b3b80d8
Optimize code a bit by calling push_back only once in some loops. Reduces compiled code size a bit.2012-04-11T03:06:35ZCraig Toppercraig.topper@gmail.com2012-04-11T03:06:35Zurn:sha1:23de31b13bfde8dfe3d5c21322f5e08608725521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154473 91177308-0d34-0410-b5e6-96231b3b80d8