llvm/test/CodeGen/ARM, branch release_27http://llvm.orghttps://git.amat.us/llvm/atom/test/CodeGen/ARM?h=release_272010-03-06T19:39:36ZInitial bits of ARMv4-only support.2010-03-06T19:39:36ZAnton Korobeynikovasl@math.spbu.ru2010-03-06T19:39:36Zurn:sha1:ce7bf1c55f5238870bae2909cd368151f1d813d1
Patch by John Tytgat!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97886 91177308-0d34-0410-b5e6-96231b3b80d8
pr6478: The frame pointer spill frame index is only defined when there is a2010-03-04T21:42:36ZBob Wilsonbob.wilson@apple.com2010-03-04T21:42:36Zurn:sha1:436e6e7b5c85f12b7c2e41b7fd5c48e5d4d72912
frame pointer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97755 91177308-0d34-0410-b5e6-96231b3b80d8
pr6480: Don't try producing ld/st-multiple instructions when the address is2010-03-04T21:04:38ZBob Wilsonbob.wilson@apple.com2010-03-04T21:04:38Zurn:sha1:bbf39b0fd9c83f1d46ca5f858e66de66fb64ec98
an undef value. This is only going to come up for bugpoint-reduced tests --
correct programs will not access memory at undefined addresses -- so it's not
worth the effort of doing anything more aggressive.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97745 91177308-0d34-0410-b5e6-96231b3b80d8
Check for comparisons of +/- zero when optimizing less-than-or-equal and2010-02-24T22:15:53ZBob Wilsonbob.wilson@apple.com2010-02-24T22:15:53Zurn:sha1:e742bb53a3e68b88eac172782fd99cd98f89f4b3
greater-than-or-equal SELECT_CCs to NEON vmin/vmax instructions. This is
only allowed when UnsafeFPMath is set or when at least one of the operands
is known to be nonzero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97065 91177308-0d34-0410-b5e6-96231b3b80d8
Use NEON vmin/vmax instructions for floating-point selects.2010-02-18T06:05:53ZBob Wilsonbob.wilson@apple.com2010-02-18T06:05:53Zurn:sha1:9f6c4c141ffa9c8b13e90dce2f2285c4479ff403
Radar 7461718.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96572 91177308-0d34-0410-b5e6-96231b3b80d8
Fix pr6111: Avoid using the LR register for the target address of an indirect2010-02-16T17:24:15ZBob Wilsonbob.wilson@apple.com2010-02-16T17:24:15Zurn:sha1:1665b0a2246c83a2c123be105a1a167cf2b423fe
branch in ARM v4 code, since it gets clobbered by the return address before
it is used. Instead of adding a new register class containing all the GPRs
except LR, just use the existing tGPR class.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96360 91177308-0d34-0410-b5e6-96231b3b80d8
Reapply the new LoopStrengthReduction code, with compile time and2010-02-12T10:34:29ZDan Gohmangohman@apple.com2010-02-12T10:34:29Zurn:sha1:572645cf84060c0fc25cb91d38cb9079918b3a88
bug fixes, and with improved heuristics for analyzing foreign-loop
addrecs.
This change also flattens IVUsers, eliminating the stride-oriented
groupings, which makes it easier to work with.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@95975 91177308-0d34-0410-b5e6-96231b3b80d8
Delete dead PHI machine instructions. These can be created due to type2010-02-10T22:58:57ZBob Wilsonbob.wilson@apple.com2010-02-10T22:58:57Zurn:sha1:5e2b05a71e81b28712b77419f68657462494e103
legalization even when the IR-level optimizer has removed dead phis, such
as when the high half of an i64 value is unused on a 32-bit target.
I had to adjust a few test cases that had dead phis.
This is a partial fix for Radar 7627077.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@95816 91177308-0d34-0410-b5e6-96231b3b80d8
convert to filecheck.2010-02-08T23:47:34ZChris Lattnersabre@nondot.org2010-02-08T23:47:34Zurn:sha1:d863b766f57fb46d15ead90a48defa739b8393b5
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@95608 91177308-0d34-0410-b5e6-96231b3b80d8
Run codegen dce pass for all targets at all optimization levels. Previously it's2010-02-06T09:07:11ZEvan Chengevan.cheng@apple.com2010-02-06T09:07:11Zurn:sha1:00a99a35840451a291eb61a192a750908a4073ae
only run for x86 with fastisel. I've found it being very effective in
eliminating some obvious dead code as result of formal parameter lowering
especially when tail call optimization eliminated the need for some of the loads
from fixed frame objects. It also shrinks a number of the tests. A couple of
tests no longer make sense and are now eliminated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@95493 91177308-0d34-0410-b5e6-96231b3b80d8