Benjamin Kramer [Tue, 17 Apr 2012 06:33:57 +0000 (06:33 +0000)]
Revert "SCEV: When expanding a GEP the final addition to the base pointer has NUW but not NSW."
This isn't right either, reverting for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154910
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Tue, 17 Apr 2012 06:20:42 +0000 (06:20 +0000)]
Test cases that assume layout should use -disable-code-place.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154908
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 17 Apr 2012 05:54:54 +0000 (05:54 +0000)]
Don't decode vperm2i128 or vperm2f128 into a shuffle if bit 3 or 7 of the immediate is set.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154907
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Tue, 17 Apr 2012 05:11:51 +0000 (05:11 +0000)]
Download and build the compiler-rt project.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154905
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Tue, 17 Apr 2012 04:15:51 +0000 (04:15 +0000)]
SlotIndexes used to store the index list in a crufty custom linked-list. I can't
for the life of me remember why I wrote it this way, but I can't see any good
reason for it now. This patch replaces the custom linked list with an ilist.
This change should preserve the existing numberings exactly, so no generated code
should change (if it does, file a bug!).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154904
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Tue, 17 Apr 2012 01:22:58 +0000 (01:22 +0000)]
Typo in an unused field.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154895
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Tue, 17 Apr 2012 01:19:37 +0000 (01:19 +0000)]
Update woefully stale credits entry. Everything listed there had been
re-done by wiser souls.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154894
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Tue, 17 Apr 2012 01:13:53 +0000 (01:13 +0000)]
Add a stub about the inline cost rewrite.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154892
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Tue, 17 Apr 2012 01:10:35 +0000 (01:10 +0000)]
Add a stub for block placement pass. This needs fleshing out w/ details
about what folks should expect w.r.t. the new algorithm.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154891
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Tue, 17 Apr 2012 01:01:55 +0000 (01:01 +0000)]
Don't tag libcxx and libcxxabi since we don't release them just yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154889
91177308-0d34-0410-b5e6-
96231b3b80d8
Kevin Enderby [Tue, 17 Apr 2012 00:49:27 +0000 (00:49 +0000)]
Fix ARM disassembly of VLD2 (single 2-element structure to all lanes)
instructions with writebacks. And add test a case for all opcodes handed by
DecodeVLD2DupInstruction() in ARMDisassembler.cpp .
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154884
91177308-0d34-0410-b5e6-
96231b3b80d8
Preston Gurd [Tue, 17 Apr 2012 00:21:35 +0000 (00:21 +0000)]
temporarily XFAIL this test until post RA
live-ins is properly enabled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154882
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Tue, 17 Apr 2012 00:01:04 +0000 (00:01 +0000)]
Tidy up. 80 columns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154881
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Mon, 16 Apr 2012 23:54:31 +0000 (23:54 +0000)]
Typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154879
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Mon, 16 Apr 2012 23:54:23 +0000 (23:54 +0000)]
Make comment here more clear.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154878
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 23:11:39 +0000 (23:11 +0000)]
Disable the atom scheduling test after r154874 broke it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154877
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 23:00:25 +0000 (23:00 +0000)]
ARM two-operand forms for vhadd and vhsub instructions.
rdar://
11252521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154875
91177308-0d34-0410-b5e6-
96231b3b80d8
Preston Gurd [Mon, 16 Apr 2012 22:52:28 +0000 (22:52 +0000)]
Temporarily turn off anti-dependency checking
during Post RA scheduling in X86,
until the X86 target is changed to properly set up
post RA liveness.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154874
91177308-0d34-0410-b5e6-
96231b3b80d8
Preston Gurd [Mon, 16 Apr 2012 22:26:48 +0000 (22:26 +0000)]
Add files which were not included by commit 154868.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154872
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 22:20:48 +0000 (22:20 +0000)]
Relax this test a touch to cope with different assembly variants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154870
91177308-0d34-0410-b5e6-
96231b3b80d8
Preston Gurd [Mon, 16 Apr 2012 22:12:58 +0000 (22:12 +0000)]
Implement GDB integration for source level debugging of code JITed using
the MCJIT execution engine.
The GDB JIT debugging integration support works by registering a loaded
object image with a pre-defined function that GDB will monitor if GDB
is attached. GDB integration support is implemented for ELF only at this
time. This integration requires GDB version 7.0 or newer.
Patch by Andy Kaylor!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154868
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 22:03:00 +0000 (22:03 +0000)]
Fix updateTerminator to be resiliant to degenerate terminators where
both fallthrough and a conditional branch target the same successor.
Gracefully delete the conditional branch and introduce any unconditional
branch needed to reach the actual successor. This fixes memory
corruption in 2009-06-15-RegScavengerAssert.ll and possibly other tests.
Also, while I'm here fix a latent bug I spotted by inspection. I never
applied the same fundamental fix to this fallthrough successor finding
logic that I did to the logic used when there are no conditional
branches. As a consequence it would have selected landing pads had they
be aligned in just the right way here. I don't have a test case as
I spotted this by inspection, and the previous time I found this
required have of TableGen's source code to produce it. =/ I hate backend
bugs. ;]
Thanks to Jim Grosbach for helping me reason through this and reviewing
the fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154867
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 21:18:49 +0000 (21:18 +0000)]
MC assembly parser handling for trailing comma in macro instantiation.
A trailing comma means no argument at all (i.e., as if the comma were not
present), not an empty argument to the invokee.
rdar://
11252521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154863
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 21:18:46 +0000 (21:18 +0000)]
ARM handle :lower16: and :upper16: after a '#' prefix.
rdar://
11252521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154862
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Mon, 16 Apr 2012 20:56:42 +0000 (20:56 +0000)]
FileCheckize these tests.
Add an extra test to ldr_post with an immediate increment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154859
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 16 Apr 2012 20:51:39 +0000 (20:51 +0000)]
Forgot to remove fast math metadata from the release notes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154858
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Mon, 16 Apr 2012 20:49:06 +0000 (20:49 +0000)]
Disable code placement for this test.
It makes it less sensitive to small changes in heuristics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154857
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 16 Apr 2012 19:39:33 +0000 (19:39 +0000)]
Remove support for the special 'fast' value for fpmath accuracy for the moment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154850
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Smith [Mon, 16 Apr 2012 18:43:53 +0000 (18:43 +0000)]
Fix incorrect atomics codegen introduced in r154705, and extend test to catch it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154845
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Mon, 16 Apr 2012 18:20:26 +0000 (18:20 +0000)]
This patch fixes 3 problems:
1. CHECKNEXT was used instead of CHECK-NEXT which caused the line to be
ignored which in turn hid the next 2 problems:
2. ('sh_offset', 0x{{{[0-9,a-f]+}}) had one too many leading curly braces and
failed to do it's job of accepting all hex digits and:
3. The check for the hex values for the code instructions didn't account for
blank separators.
Patch by Jack Carter.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154842
91177308-0d34-0410-b5e6-
96231b3b80d8
David Blaikie [Mon, 16 Apr 2012 18:10:13 +0000 (18:10 +0000)]
Remove unused variable
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154841
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 18:03:16 +0000 (18:03 +0000)]
ARM assembly two-operand forms for VRSHL.
rdar://
11252521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154840
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 18:03:14 +0000 (18:03 +0000)]
Tidy up. Test formatting.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154839
91177308-0d34-0410-b5e6-
96231b3b80d8
Akira Hatanaka [Mon, 16 Apr 2012 18:00:19 +0000 (18:00 +0000)]
Do not add offset in applyFixup. This has already been accounted for in Value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154838
91177308-0d34-0410-b5e6-
96231b3b80d8
Tobias Grosser [Mon, 16 Apr 2012 17:18:49 +0000 (17:18 +0000)]
Describe Polly as a 'Polyhedral Optimizer'
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154834
91177308-0d34-0410-b5e6-
96231b3b80d8
Tobias Grosser [Mon, 16 Apr 2012 17:17:00 +0000 (17:17 +0000)]
Add Polly news to the release notes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154833
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 17:14:11 +0000 (17:14 +0000)]
ARM two-operand aliases for VRHADD instructions.
rdar://
11252521
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154832
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 16 Apr 2012 17:14:07 +0000 (17:14 +0000)]
Tidy up. Testcase formatting.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154831
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Mon, 16 Apr 2012 17:06:49 +0000 (17:06 +0000)]
Link to the autovectorization EuroLLVM slides in the release notes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154830
91177308-0d34-0410-b5e6-
96231b3b80d8
Sirish Pande [Mon, 16 Apr 2012 17:05:06 +0000 (17:05 +0000)]
Hexagon V5 (Floating Point) Support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154829
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Mon, 16 Apr 2012 16:38:48 +0000 (16:38 +0000)]
Move to X86 directory because this fails on non-X86 platforms.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154825
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 16 Apr 2012 16:28:59 +0000 (16:28 +0000)]
Make it possible to indicate relaxed floating point requirements at the IR level
through the use of 'fpmath' metadata. Currently this only provides a 'fpaccuracy'
value, which may be a number in ULPs or the keyword 'fast', however the intent is
that this will be extended with additional information about NaN's, infinities
etc later. No optimizations have been hooked up to this so far.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154822
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 13:49:17 +0000 (13:49 +0000)]
Flip the new block-placement pass to be on by default.
This is mostly to test the waters. I'd like to get results from FNT
build bots and other bots running on non-x86 platforms.
This feature has been pretty heavily tested over the last few months by
me, and it fixes several of the execution time regressions caused by the
inlining work by preventing inlining decisions from radically impacting
block layout.
I've seen very large improvements in yacr2 and ackermann benchmarks,
along with the expected noise across all of the benchmark suite whenever
code layout changes. I've analyzed all of the regressions and fixed
them, or found them to be impossible to fix. See my email to llvmdev for
more details.
I'd like for this to be in 3.1 as it complements the inliner changes,
but if any failures are showing up or anyone has concerns, it is just
a flag flip and so can be easily turned off.
I'm switching it on tonight to try and get at least one run through
various folks' performance suites in case SPEC or something else has
serious issues with it. I'll watch bots and revert if anything shows up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154816
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 13:49:09 +0000 (13:49 +0000)]
Remove an overly brittle test. This test will no longer be interesting
once we start changing the block layout, so just nuke it. If anyone has
ideas about how to craft a code layout agnostic form of the test please
let me know.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154815
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 16 Apr 2012 13:44:35 +0000 (13:44 +0000)]
Remove the methods for attaching metadata to instructions/retrieving metadata
from instructions. Chandler doesn't like them being here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154813
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 13:33:36 +0000 (13:33 +0000)]
Add a somewhat hacky heuristic to do something different from whole-loop
rotation. When there is a loop backedge which is an unconditional
branch, we will end up with a branch somewhere no matter what. Try
placing this backedge in a fallthrough position above the loop header as
that will definitely remove at least one branch from the loop iteration,
where whole loop rotation may not.
I haven't seen any benchmarks where this is important but loop-blocks.ll
tests for it, and so this will be covered when I flip the default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154812
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 16 Apr 2012 13:21:26 +0000 (13:21 +0000)]
Add convenience methods to MDBuilder for attaching metadata to instructions,
and retrieving it from instructions. I don't have a use for this but is seems
logical for it to exist. While there, remove some 'const' markings from methods
which are in fact 'const' in practice, but aren't logically 'const'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154811
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Mon, 16 Apr 2012 12:39:17 +0000 (12:39 +0000)]
Fix style violation in BBVectorize (pointed out by Bill Wendling)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154810
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Barton [Mon, 16 Apr 2012 11:32:10 +0000 (11:32 +0000)]
Add -disassemble support for -show-inst and -show-encode capability llvm-mc. Also refactor so all MC paraphernalia are created once for all uses as much as possible.
The test change is to account for the fact that the default disassembler behaviour has changed with regards to specifying the assembly syntax to use.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154809
91177308-0d34-0410-b5e6-
96231b3b80d8
Rafael Espindola [Mon, 16 Apr 2012 10:58:38 +0000 (10:58 +0000)]
Remove lto_codegen_set_whole_program_optimization. It is a work in progress,
so we don't want it to show up in the stable 3.1 interface.
While at it, add a comment about why LTOCodeGenerator manually creates the
internalize pass.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154807
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 09:31:23 +0000 (09:31 +0000)]
Tweak the loop rotation logic to check whether the loop is naturally
laid out in a form with a fallthrough into the header and a fallthrough
out of the bottom. In that case, leave the loop alone because any
rotation will introduce unnecessary branches. If either side looks like
it will require an explicit branch, then the rotation won't add any, do
it to ensure the branch occurs outside of the loop (if possible) and
maximize the benefit of the fallthrough in the bottom.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154806
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Mon, 16 Apr 2012 08:56:50 +0000 (08:56 +0000)]
Reapply 'Add reverseColor to raw_ostream'.
To be used in printing unprintable source in clang diagnostics.
Patch by Seth Cantrell, with a minor fix for mingw by me.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154805
91177308-0d34-0410-b5e6-
96231b3b80d8
Eli Bendersky [Mon, 16 Apr 2012 08:42:55 +0000 (08:42 +0000)]
Documentation fixes to LLVMBuild.html [PR 11563]
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154804
91177308-0d34-0410-b5e6-
96231b3b80d8
Argyrios Kyrtzidis [Mon, 16 Apr 2012 07:59:39 +0000 (07:59 +0000)]
Revert r154800 which breaks windows builders.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154802
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 16 Apr 2012 07:13:00 +0000 (07:13 +0000)]
Replace vpermd/vpermps intrinic patterns with custom lowering to target specific nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154801
91177308-0d34-0410-b5e6-
96231b3b80d8
Argyrios Kyrtzidis [Mon, 16 Apr 2012 07:07:38 +0000 (07:07 +0000)]
Add reverseColor to raw_ostream.
To be used in printing unprintable source in clang diagnostics.
Patch by Seth Cantrell!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154800
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 16 Apr 2012 06:43:40 +0000 (06:43 +0000)]
Change type profile for vpermv back to using operand type for the mask argument to match intrinsic behavior. Add a bitcast to the lowering code to convert mask from v8i32 to v8f32 for vpermps.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154798
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 16 Apr 2012 06:26:15 +0000 (06:26 +0000)]
Flip the arguments when converting vpermd/vpermps intrinsics into instructions. The intrinsic has the mask as the last operand, but the instruction has it as the second.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154797
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Mon, 16 Apr 2012 05:24:52 +0000 (05:24 +0000)]
Add credit and release notes for r150307. By Kai Nacke.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154796
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Mon, 16 Apr 2012 04:23:52 +0000 (04:23 +0000)]
Add a Fixme.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154793
91177308-0d34-0410-b5e6-
96231b3b80d8
Sebastian Pop [Mon, 16 Apr 2012 04:11:45 +0000 (04:11 +0000)]
add configure flag --with-default-sysroot
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154791
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Mon, 16 Apr 2012 03:49:43 +0000 (03:49 +0000)]
Say something about -vectorize in the release notes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154788
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Mon, 16 Apr 2012 03:49:42 +0000 (03:49 +0000)]
Simplify checking for pointer types in BBVectorize (this change was suggested by Duncan).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154787
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Mon, 16 Apr 2012 03:33:22 +0000 (03:33 +0000)]
Remove dead SD nodes after the combining pass. Fixes PR12201.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154786
91177308-0d34-0410-b5e6-
96231b3b80d8
Chandler Carruth [Mon, 16 Apr 2012 01:12:56 +0000 (01:12 +0000)]
Rewrite how machine block placement handles loop rotation.
This is a complex change that resulted from a great deal of
experimentation with several different benchmarks. The one which proved
the most useful is included as a test case, but I don't know that it
captures all of the relevant changes, as I didn't have specific
regression tests for each, they were more the result of reasoning about
what the old algorithm would possibly do wrong. I'm also failing at the
moment to craft more targeted regression tests for these changes, if
anyone has ideas, it would be welcome.
The first big thing broken with the old algorithm is the idea that we
can take a basic block which has a loop-exiting successor and a looping
successor and use the looping successor as the layout top in order to
get that particular block to be the bottom of the loop after layout.
This happens to work in many cases, but not in all.
The second big thing broken was that we didn't try to select the exit
which fell into the nearest enclosing loop (to which we exit at all). As
a consequence, even if the rotation worked perfectly, it would result in
one of two bad layouts. Either the bottom of the loop would get
fallthrough, skipping across a nearer enclosing loop and thereby making
it discontiguous, or it would be forced to take an explicit jump over
the nearest enclosing loop to earch its successor. The point of the
rotation is to get fallthrough, so we need it to fallthrough to the
nearest loop it can.
The fix to the first issue is to actually layout the loop from the loop
header, and then rotate the loop such that the correct exiting edge can
be a fallthrough edge. This is actually much easier than I anticipated
because we can handle all the hard parts of finding a viable rotation
before we do the layout. We just store that, and then rotate after
layout is finished. No inner loops get split across the post-rotation
backedge because we check for them when selecting the rotation.
That fix exposed a latent problem with our exitting block selection --
we should allow the backedge to point into the middle of some inner-loop
chain as there is no real penalty to it, the whole point is that it
*won't* be a fallthrough edge. This may have blocked the rotation at all
in some cases, I have no idea and no test case as I've never seen it in
practice, it was just noticed by inspection.
Finally, all of these fixes, and studying the loops they produce,
highlighted another problem: in rotating loops like this, we sometimes
fail to align the destination of these backwards jumping edges. Fix this
by actually walking the backwards edges rather than relying on loopinfo.
This fixes regressions on heapsort if block placement is enabled as well
as lots of other cases where the previous logic would introduce an
abundance of unnecessary branches into the execution.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154783
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 16 Apr 2012 00:41:45 +0000 (00:41 +0000)]
Merge vpermps/vpermd and vpermpd/vpermq SD nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154782
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 16 Apr 2012 00:12:20 +0000 (00:12 +0000)]
Fix SDTypeProfile for vpermps. The mask operand should be v8i32.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154781
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 15 Apr 2012 23:48:57 +0000 (23:48 +0000)]
Spacing fixes and 80 column fixes. Use 0 instead of 0x80 for undef indices in vpermps/vpermd. Hardware only looks at lower 3-bits.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154780
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 15 Apr 2012 22:43:31 +0000 (22:43 +0000)]
Remove AVX2 vpermq and vpermpd intrinsics. These can now be handled with normal shuffle vectors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154778
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 15 Apr 2012 22:00:22 +0000 (22:00 +0000)]
Make member variables of AsmToken private. Remove unnecessary forward declarations. Remove an unnecessary include.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154775
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakub Staszak [Sun, 15 Apr 2012 20:22:36 +0000 (20:22 +0000)]
Fix class name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154773
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Sun, 15 Apr 2012 20:17:14 +0000 (20:17 +0000)]
Do not convert between fp128 <-> ppc_fp128 since there is no legal cast conversion between the two.
Patch by nobled <nobled@dreamwidth.org>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154772
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakub Staszak [Sun, 15 Apr 2012 20:13:47 +0000 (20:13 +0000)]
Fix filename and register numbers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154771
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Sun, 15 Apr 2012 19:36:44 +0000 (19:36 +0000)]
Fix PR12529. The Vxx family of instructions are only supported by AVX.
Use non-vex instructions for SSE4.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154770
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sun, 15 Apr 2012 18:03:49 +0000 (18:03 +0000)]
Add the MDBuilder helper class for conveniently creating metadata.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154766
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Sun, 15 Apr 2012 17:04:27 +0000 (17:04 +0000)]
Wire up support for diagnostic ranges in the ARMAsmParser.
As an example, attach range info to the "invalid instruction" message:
$ clang -arch arm -c asm.c
asm.c:2:11: error: invalid instruction
__asm__("foo r0");
^
<inline asm>:1:2: note: instantiated into assembly here
foo r0
^~~
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154765
91177308-0d34-0410-b5e6-
96231b3b80d8
Nadav Rotem [Sun, 15 Apr 2012 15:08:09 +0000 (15:08 +0000)]
When emulating vselect using OR/AND/XOR make sure to bitcast the result back to the original type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154764
91177308-0d34-0410-b5e6-
96231b3b80d8
Elena Demikhovsky [Sun, 15 Apr 2012 11:18:59 +0000 (11:18 +0000)]
Added VPERM optimization for AVX2 shuffles
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154761
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Sun, 15 Apr 2012 05:33:43 +0000 (05:33 +0000)]
HexagonCopyToCombine.cpp: Silence two warnings, -Wunused-variable, with -Asserts.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154759
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Sun, 15 Apr 2012 05:09:09 +0000 (05:09 +0000)]
Target/Hexagon: Tweak to fix msvc build.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154758
91177308-0d34-0410-b5e6-
96231b3b80d8
Anshuman Dasgupta [Sat, 14 Apr 2012 20:59:13 +0000 (20:59 +0000)]
Remove trailing whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154755
91177308-0d34-0410-b5e6-
96231b3b80d8
Anshuman Dasgupta [Sat, 14 Apr 2012 20:57:13 +0000 (20:57 +0000)]
Add VLIW packetizer to ReleaseNotes.html and CREDITS.TXT. Committing patch
by Sundeep Kushwaha.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154754
91177308-0d34-0410-b5e6-
96231b3b80d8
Brendon Cahoon [Sat, 14 Apr 2012 16:54:12 +0000 (16:54 +0000)]
Add the loop unrolling info to ReleaseNotes.html and CREDITS.TXT.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154752
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sat, 14 Apr 2012 15:43:22 +0000 (15:43 +0000)]
There is no need for setIsExact to be public. Make it private.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154750
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sat, 14 Apr 2012 12:36:06 +0000 (12:36 +0000)]
Rename "fpaccuracy" metadata to the more generic "fpmath". That's because I'm
thinking of generalizing it to be able to specify other freedoms beyond accuracy
(such as that NaN's don't have to be respected). I'd like the 3.1 release (the
first one with this metadata) to have the more generic name already rather than
having to auto-upgrade it in 3.2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154744
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Sat, 14 Apr 2012 09:04:57 +0000 (09:04 +0000)]
Make StringMap's copy ctor non-explicit.
Without this gcc doesn't allow us to put a StringMap into a
std::map. Works with clang though.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154737
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Sat, 14 Apr 2012 07:32:50 +0000 (07:32 +0000)]
Fix an error in BBVectorize important for vectorizing pointer types.
When vectorizing pointer types it is important to realize that potential
pairs cannot be connected via the address pointer argument of a load or store.
This is because even after vectorization, the address is still a scalar because
the address of the higher half of the pair is implicit from the address of the
lower half (it need not be, and should not be, explicitly computed).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154735
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Sat, 14 Apr 2012 07:32:43 +0000 (07:32 +0000)]
Enhance BBVectorize to more-properly handle pointer values and vectorize GEPs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154734
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrew Trick [Fri, 13 Apr 2012 23:29:54 +0000 (23:29 +0000)]
misched: Added CanHandleTerminators.
This is a special flag for targets that really want their block
terminators in the DAG. The default scheduler cannot handle this
correctly, so it becomes the specialized scheduler's responsibility to
schedule terminators.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154712
91177308-0d34-0410-b5e6-
96231b3b80d8
Bob Wilson [Fri, 13 Apr 2012 22:58:53 +0000 (22:58 +0000)]
Remove old code to strip out unwanted PPC slices for Apple llvmCore.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154706
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Smith [Fri, 13 Apr 2012 22:47:00 +0000 (22:47 +0000)]
Fix X86 codegen for 'atomicrmw nand' to generate *x = ~(*x & y), not *x = ~*x & y.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154705
91177308-0d34-0410-b5e6-
96231b3b80d8
Sirish Pande [Fri, 13 Apr 2012 21:01:35 +0000 (21:01 +0000)]
Remove iostream from New Value Jump.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154703
91177308-0d34-0410-b5e6-
96231b3b80d8
Hal Finkel [Fri, 13 Apr 2012 20:45:45 +0000 (20:45 +0000)]
Add support to BBVectorize for vectorizing selects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154700
91177308-0d34-0410-b5e6-
96231b3b80d8
Sirish Pande [Fri, 13 Apr 2012 20:22:31 +0000 (20:22 +0000)]
Add support for Hexagon Architectural feature, New Value Jump.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154696
91177308-0d34-0410-b5e6-
96231b3b80d8
Sirish Pande [Fri, 13 Apr 2012 20:22:19 +0000 (20:22 +0000)]
Pass to replace tranfer/copy instructions into combine instruction where possible.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154695
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Fri, 13 Apr 2012 20:06:17 +0000 (20:06 +0000)]
Reduce malloc traffic in DwarfAccelTable
- Don't copy offsets into HashData, the underlying vector won't change once the table is finalized.
- Allocate HashData and HashDataContents in a BumpPtrAllocator.
- Allocate string map entries in the same allocator.
- Random cleanups.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154694
91177308-0d34-0410-b5e6-
96231b3b80d8
Tony Linthicum [Fri, 13 Apr 2012 19:09:44 +0000 (19:09 +0000)]
Support for Hexagon backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154692
91177308-0d34-0410-b5e6-
96231b3b80d8
Tony Linthicum [Fri, 13 Apr 2012 19:09:18 +0000 (19:09 +0000)]
Support for Hexagon backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154691
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 13 Apr 2012 18:59:28 +0000 (18:59 +0000)]
On Darwin targets, only use vfma etc. if the source use fma() intrinsic explicitly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154689
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 13 Apr 2012 18:57:48 +0000 (18:57 +0000)]
Add some comments, and fix a few places that missed setting Changed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154687
91177308-0d34-0410-b5e6-
96231b3b80d8