aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Revert r154800 which breaks windows builders.Argyrios Kyrtzidis2012-04-165-59/+0
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154802 91177308-0d34-0410-b5e6-96231b3b80d8
* Replace vpermd/vpermps intrinic patterns with custom lowering to target ↵Craig Topper2012-04-162-18/+14
| | | | | | specific nodes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154801 91177308-0d34-0410-b5e6-96231b3b80d8
* Add reverseColor to raw_ostream.Argyrios Kyrtzidis2012-04-165-0/+59
| | | | | | | 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
* Change type profile for vpermv back to using operand type for the mask ↵Craig Topper2012-04-162-5/+5
| | | | | | 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
* Flip the arguments when converting vpermd/vpermps intrinsics into ↵Craig Topper2012-04-161-3/+3
| | | | | | 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
* Add credit and release notes for r150307. By Kai Nacke.Bill Wendling2012-04-162-0/+5
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154796 91177308-0d34-0410-b5e6-96231b3b80d8
* Add a Fixme.Bill Wendling2012-04-161-0/+2
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154793 91177308-0d34-0410-b5e6-96231b3b80d8
* add configure flag --with-default-sysrootSebastian Pop2012-04-164-5/+27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154791 91177308-0d34-0410-b5e6-96231b3b80d8
* Say something about -vectorize in the release notes.Hal Finkel2012-04-161-0/+3
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154788 91177308-0d34-0410-b5e6-96231b3b80d8
* Simplify checking for pointer types in BBVectorize (this change was ↵Hal Finkel2012-04-161-5/+2
| | | | | | suggested by Duncan). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154787 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove dead SD nodes after the combining pass. Fixes PR12201.Hal Finkel2012-04-162-0/+47
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154786 91177308-0d34-0410-b5e6-96231b3b80d8
* Rewrite how machine block placement handles loop rotation.Chandler Carruth2012-04-162-69/+196
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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
* Merge vpermps/vpermd and vpermpd/vpermq SD nodes.Craig Topper2012-04-164-31/+20
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154782 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix SDTypeProfile for vpermps. The mask operand should be v8i32.Craig Topper2012-04-161-2/+4
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154781 91177308-0d34-0410-b5e6-96231b3b80d8
* Spacing fixes and 80 column fixes. Use 0 instead of 0x80 for undef indices ↵Craig Topper2012-04-151-9/+12
| | | | | | 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
* Remove AVX2 vpermq and vpermpd intrinsics. These can now be handled with ↵Craig Topper2012-04-153-39/+8
| | | | | | normal shuffle vectors. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154778 91177308-0d34-0410-b5e6-96231b3b80d8
* Make member variables of AsmToken private. Remove unnecessary forward ↵Craig Topper2012-04-153-5/+2
| | | | | | declarations. Remove an unnecessary include. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154775 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix class name.Jakub Staszak2012-04-151-1/+1
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154773 91177308-0d34-0410-b5e6-96231b3b80d8
* Do not convert between fp128 <-> ppc_fp128 since there is no legal cast ↵Nadav Rotem2012-04-151-9/+12
| | | | | | | | | | 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
* Fix filename and register numbers.Jakub Staszak2012-04-151-3/+3
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154771 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix PR12529. The Vxx family of instructions are only supported by AVX.Nadav Rotem2012-04-152-4/+14
| | | | | | | | Use non-vex instructions for SSE4. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154770 91177308-0d34-0410-b5e6-96231b3b80d8
* Add the MDBuilder helper class for conveniently creating metadata.Duncan Sands2012-04-153-0/+194
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154766 91177308-0d34-0410-b5e6-96231b3b80d8
* Wire up support for diagnostic ranges in the ARMAsmParser.Benjamin Kramer2012-04-151-3/+12
| | | | | | | | | | | | | | 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
* When emulating vselect using OR/AND/XOR make sure to bitcast the result back ↵Nadav Rotem2012-04-152-1/+10
| | | | | | to the original type. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154764 91177308-0d34-0410-b5e6-96231b3b80d8
* Added VPERM optimization for AVX2 shufflesElena Demikhovsky2012-04-156-4/+110
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154761 91177308-0d34-0410-b5e6-96231b3b80d8
* HexagonCopyToCombine.cpp: Silence two warnings, -Wunused-variable, with ↵NAKAMURA Takumi2012-04-151-0/+2
| | | | | | -Asserts. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154759 91177308-0d34-0410-b5e6-96231b3b80d8
* Target/Hexagon: Tweak to fix msvc build.NAKAMURA Takumi2012-04-151-2/+2
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154758 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove trailing whitespace.Anshuman Dasgupta2012-04-141-1/+1
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154755 91177308-0d34-0410-b5e6-96231b3b80d8
* Add VLIW packetizer to ReleaseNotes.html and CREDITS.TXT. Committing patchAnshuman Dasgupta2012-04-142-0/+7
| | | | | | | by Sundeep Kushwaha. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154754 91177308-0d34-0410-b5e6-96231b3b80d8
* Add the loop unrolling info to ReleaseNotes.html and CREDITS.TXT.Brendon Cahoon2012-04-142-0/+7
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154752 91177308-0d34-0410-b5e6-96231b3b80d8
* There is no need for setIsExact to be public. Make it private.Duncan Sands2012-04-141-2/+3
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154750 91177308-0d34-0410-b5e6-96231b3b80d8
* Rename "fpaccuracy" metadata to the more generic "fpmath". That's because I'mDuncan Sands2012-04-146-47/+47
| | | | | | | | | | 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
* Make StringMap's copy ctor non-explicit.Benjamin Kramer2012-04-141-1/+1
| | | | | | | 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
* Fix an error in BBVectorize important for vectorizing pointer types.Hal Finkel2012-04-142-0/+54
| | | | | | | | | | 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
* Enhance BBVectorize to more-properly handle pointer values and vectorize GEPs.Hal Finkel2012-04-143-2/+114
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154734 91177308-0d34-0410-b5e6-96231b3b80d8
* misched: Added CanHandleTerminators.Andrew Trick2012-04-132-3/+10
| | | | | | | | | 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
* Remove old code to strip out unwanted PPC slices for Apple llvmCore.Bob Wilson2012-04-131-13/+0
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154706 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix X86 codegen for 'atomicrmw nand' to generate *x = ~(*x & y), not *x = ↵Richard Smith2012-04-133-27/+48
| | | | | | ~*x & y. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154705 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove iostream from New Value Jump.Sirish Pande2012-04-131-1/+0
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154703 91177308-0d34-0410-b5e6-96231b3b80d8
* Add support to BBVectorize for vectorizing selects.Hal Finkel2012-04-133-0/+41
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154700 91177308-0d34-0410-b5e6-96231b3b80d8
* Add support for Hexagon Architectural feature, New Value Jump.Sirish Pande2012-04-137-10/+684
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154696 91177308-0d34-0410-b5e6-96231b3b80d8
* Pass to replace tranfer/copy instructions into combine instruction where ↵Sirish Pande2012-04-135-0/+484
| | | | | | possible. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154695 91177308-0d34-0410-b5e6-96231b3b80d8
* Reduce malloc traffic in DwarfAccelTableBenjamin Kramer2012-04-132-58/+28
| | | | | | | | | - 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
* Support for Hexagon backend.Tony Linthicum2012-04-131-0/+2
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154692 91177308-0d34-0410-b5e6-96231b3b80d8
* Support for Hexagon backend.Tony Linthicum2012-04-131-0/+4
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154691 91177308-0d34-0410-b5e6-96231b3b80d8
* On Darwin targets, only use vfma etc. if the source use fma() intrinsic ↵Evan Cheng2012-04-132-3/+6
| | | | | | explicitly. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154689 91177308-0d34-0410-b5e6-96231b3b80d8
* Add some comments, and fix a few places that missed setting Changed.Dan Gohman2012-04-131-2/+24
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154687 91177308-0d34-0410-b5e6-96231b3b80d8
* For ARM disassembly only print 32 unsigned bits for the address of branchKevin Enderby2012-04-131-2/+2
| | | | | | | | targets so if the branch target has the high bit set it does not get printed as: beq 0xffffffff8008c404 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154685 91177308-0d34-0410-b5e6-96231b3b80d8
* Consider ObjC runtime calls objc_storeWeak and others which make a copy ofDan Gohman2012-04-132-14/+160
| | | | | | | | their argument as "escape" points for objc_retainBlock optimization. This fixes rdar://11229925. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154682 91177308-0d34-0410-b5e6-96231b3b80d8
* By default, use Early-CSE instead of GVN for vectorization cleanup.Hal Finkel2012-04-131-2/+9
| | | | | | | | | | As has been suggested by Duncan and others, Early-CSE and GVN should do similar redundancy elimination, but Early-CSE is much less expensive. Most of my autovectorization benchmarks show a performance regresion, but all of these are < 0.1%, and so I think that it is still worth using the less expensive pass. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154673 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.1 at 2025-02-09 11:38:26 +0000