summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen
Commit message (Collapse)AuthorAgeFilesLines
* Re-apply r84295, with fixes to how the loop "top" and "bottom" blocks areDan Gohman2009-10-203-1/+280
| | | | | | | | | tracked. Instead of trying to manually keep track of these locations while doing complex modifications, just recompute them when they're needed. This fixes a bug in which the TopMBB and BotMBB were not correctly updated, leading to invalid transformations. llvm-svn: 84598
* If the physical register being spilled does not have an interval, spill its ↵Evan Cheng2009-10-201-0/+54
| | | | | | sub-registers instead. llvm-svn: 84586
* Enable allocation of R3 in Thumb1Jim Grosbach2009-10-191-1/+1
| | | | llvm-svn: 84563
* convert to filecheck syntax and make a lot more aggressive.Chris Lattner2009-10-191-8/+68
| | | | llvm-svn: 84517
* Revert r84295, this unbreaks llvm-gcc bootstrap on x86-64/linuxAnton Korobeynikov2009-10-192-208/+1
| | | | llvm-svn: 84516
* rename testChris Lattner2009-10-191-0/+0
| | | | llvm-svn: 84515
* Fix PR5247, "lock addq" pattern (and other atomics), it DOES modify EFLAGS.Torok Edwin2009-10-191-0/+69
| | | | | | | LLC was scheduling compares before the adds causing wrong branches to be taken in programs, resulting in misoptimized code wherever atomic adds where used. llvm-svn: 84485
* Teach lit that the .c files in 'test/CodeGen/CellSPU/useful-harnesses' ↵Daniel Dunbar2009-10-191-0/+1
| | | | | | aren't tests. llvm-svn: 84460
* Add support for matching shuffle patterns with palignr.Nate Begeman2009-10-193-15/+70
| | | | llvm-svn: 84459
* Turn on post-alloc scheduling for x86.Evan Cheng2009-10-1813-50/+38
| | | | llvm-svn: 84431
* Enhance CodePlacementOpt's unconditional intra-loop branch elimination logicDan Gohman2009-10-172-1/+208
| | | | | | | | | | | | to be more general and understand more varieties of loops. Teach CodePlacementOpt to reorganize the basic blocks of a loop so that they are contiguous. This also includes a fair amount of logic for preserving fall-through edges while doing so. This fixes a BranchFolding-ism where blocks which can't be made to use a fall-through edge and don't conveniently fit anywhere nearby get tossed out to the end of the function. llvm-svn: 84295
* Update tests to use FileCheckMon P Wang2009-10-1619-55/+70
| | | | llvm-svn: 84282
* Add test case for r84279Mon P Wang2009-10-161-0/+12
| | | | llvm-svn: 84280
* Move zext and sext casts fed by loads into the same block as theDan Gohman2009-10-162-2/+22
| | | | | | | load, to help SelectionDAG fold them into the loads, unless conditions are unfavorable. llvm-svn: 84271
* Enable post-alloc scheduling for all ARM variants except for Thumb1.Evan Cheng2009-10-161-5/+5
| | | | llvm-svn: 84249
* Branches must be the last instruction in a Thumb2 IT block. Approved by Evan ↵Sandeep Patel2009-10-151-0/+44
| | | | | | Cheng. llvm-svn: 84212
* Re-apply 84180 with the fixed test case.Sanjiv Gupta2009-10-151-3/+3
| | | | llvm-svn: 84195
* Move Blackfin intrinsics into the Target/Blackfin directory.Jakob Stoklund Olesen2009-10-151-0/+3
| | | | llvm-svn: 84194
* Make CodePlacementOpt align loops, rather than loop headers. TheDan Gohman2009-10-152-2/+13
| | | | | | | | | | header is just the entry block to the loop, and it needn't be at the top of the loop in the code layout. Remove the code that suppressed loop alignment for outer loops, so that outer loops are aligned. llvm-svn: 84158
* When LiveVariables is adding implicit-def to model "partial dead", add the ↵Evan Cheng2009-10-141-0/+15
| | | | | | earlyclobber marker if the superreg def has it. llvm-svn: 84153
* Fix this test to account for a movl $0 being emitted as an xor now,Dan Gohman2009-10-141-2/+3
| | | | | | and convert it to FileCheck. llvm-svn: 84065
* Revise ARM inline assembly memory operands to require the memory address toBob Wilson2009-10-131-1/+3
| | | | | | | be in a register. The previous use of ARM address mode 2 was completely arbitrary and inappropriate for Thumb. Radar 7137468. llvm-svn: 84022
* Add ARMv6T2 SBFX/UBFX instructions. Approved by Anton Korobeynikov.Sandeep Patel2009-10-131-0/+37
| | | | llvm-svn: 84009
* Don't forget to mark RAX as live-out of the function when arranging forDan Gohman2009-10-121-1/+10
| | | | | | | | | it to hold the address of an sret return value, for x86-64 ABI purposes. Also, fix the test that was originally intended to test this to actually test it, using FileCheck. llvm-svn: 83853
* Eliminate some redundant llvm-as calls.Benjamin Kramer2009-10-1213-13/+13
| | | | llvm-svn: 83837
* Add missed mem-mem move patternsAnton Korobeynikov2009-10-112-0/+16
| | | | llvm-svn: 83812
* Add MSP430 mem-mem insts support. Patch by Brian Lucas with some my refinementsAnton Korobeynikov2009-10-112-0/+93
| | | | llvm-svn: 83811
* Add bunch of MSP430 'feature' tests. Patch by Brian Lucas with some my ↵Anton Korobeynikov2009-10-118-0/+343
| | | | | | refinements llvm-svn: 83809
* Implement 'm' memory operand properlyAnton Korobeynikov2009-10-111-11/+20
| | | | llvm-svn: 83785
* It seems that OR operation does not affect status reg at all.Anton Korobeynikov2009-10-101-0/+14
| | | | | | Remove impdef of SRW. This fixes PR4779 llvm-svn: 83739
* Update this test; the code is the same but it gets counted as oneDan Gohman2009-10-091-1/+1
| | | | | | fewer remat. llvm-svn: 83690
* Fix the x86 test-shrink optimization so that it doesn't shrink comparisonsDan Gohman2009-10-091-0/+23
| | | | | | | | when one of the bits being tested would end up being the sign bit in the narrower type, and a signed comparison is being performed, since this would change the result of the signed comparison. This fixes PR5132. llvm-svn: 83670
* Merge a bunch of NEON tests into larger files so they run faster.Bob Wilson2009-10-0987-3650/+3526
| | | | llvm-svn: 83667
* Convert some ARM tests with lots of greps to use FileCheck.Bob Wilson2009-10-095-58/+135
| | | | llvm-svn: 83651
* Fix a logic error that caused non-rematable loop invariants loads to be ↵Evan Cheng2009-10-091-0/+264
| | | | | | licm'ed out of loop. llvm-svn: 83622
* Commit one last NEON test to use FileCheck. That's all of them now!Bob Wilson2009-10-091-4/+13
| | | | llvm-svn: 83617
* Convert more NEON tests to use FileCheck.Bob Wilson2009-10-0911-115/+388
| | | | llvm-svn: 83616
* Reset kill markers after live interval is reconstructed.Evan Cheng2009-10-091-0/+34
| | | | llvm-svn: 83608
* Add codegen support for NEON vst4lane intrinsics with 128-bit vectors.Bob Wilson2009-10-091-0/+28
| | | | llvm-svn: 83600
* Add codegen support for NEON vst3lane intrinsics with 128-bit vectors.Bob Wilson2009-10-081-0/+28
| | | | llvm-svn: 83598
* Add codegen support for NEON vst2lane intrinsics with 128-bit vectors.Bob Wilson2009-10-081-0/+28
| | | | llvm-svn: 83596
* Convert more NEON tests to use FileCheck.Bob Wilson2009-10-089-46/+139
| | | | llvm-svn: 83595
* Add codegen support for NEON vld4lane intrinsics with 128-bit vectors.Bob Wilson2009-10-081-0/+53
| | | | | | Also fix some copy-and-paste errors in previous changes. llvm-svn: 83590
* Convert more NEON tests to use FileCheck.Bob Wilson2009-10-0813-100/+295
| | | | llvm-svn: 83587
* Add codegen support for NEON vld3lane intrinsics with 128-bit vectors.Bob Wilson2009-10-081-0/+47
| | | | llvm-svn: 83585
* Use lower16 / upper16 imm modifiers to asmprint 32-bit imms splitted via ↵Anton Korobeynikov2009-10-081-2/+2
| | | | | | movt/movw pair. llvm-svn: 83572
* Add codegen support for NEON vld2lane intrinsics with 128-bit vectors.Bob Wilson2009-10-081-0/+41
| | | | llvm-svn: 83568
* Add some peepholes for signed comparisons using ashr X, X, 32.Richard Osborne2009-10-081-0/+76
| | | | llvm-svn: 83549
* Convert more NEON tests to use FileCheck.Bob Wilson2009-10-088-46/+140
| | | | llvm-svn: 83528
* Add codegen support for NEON vst4 intrinsics with <1 x i64> vectors.Bob Wilson2009-10-081-0/+9
| | | | llvm-svn: 83526
OpenPOWER on IntegriCloud