- User Since
- Jun 17 2015, 7:07 AM (165 w, 5 d)
I am ok with committing this, but maybe we should have someone else's opinion as well?
Fri, Aug 17
- Report unroll-and-jam as not applied even if unroll is present as well.
- rename followup_cyclic/followup_noncyclic to followup_sequential/followup_coincident
- Move hasUnrollAndJamTransformation in LoopUnrollAndJamPass to different place
- Remove some unrelated whitespace changes made by clang-format
- Extract followup attribute names into constant
Thu, Aug 16
Tue, Aug 14
I am thinking about adding a LoopMetadataTacker (sort of a combination of LoopVectorizeHints and AssumptionTracker) analysis pass which would centralize the interpretation of that metadata and avoid the linear search through the metadata list when looking up a specific attribute.
Mon, Aug 13
To add some details: The original failure was external/skia/src/gpu/GrTestUtils.cpp of the aosp buildbot. It expects a library function __muloti4 which was lowered from the intrinsic llvm.smul.with.overflow.i128. The next smaller llvm.smul.with.overflow.i64 should be able to handle 64 bit values (not just 64 bits).
Fri, Aug 10
Thu, Aug 9
- Explicitly document followup of not applied transformations to be ignored
- Unroll/UnrollAndJam: Interpret enable/count/full as forced
- Unroll/UnrollAndJam: Add tests for disable_nonforced combined with enable/count/full
- Reduce size of unroll-pragmas_transform.ll
Should the total execution time of benchmarks be around second or less?
Maybe reduce the number of tested parameters to reduce the total runtime?
Could you close the review if it's not needed anymore?
Tue, Aug 7
Looks like an obvious change, but r338787 is by @MatzeB I'd wait for his ok.
@santosh Do you need someone to commit?
Thank you. LGTM.
Mon, Aug 6
Sat, Aug 4
Fri, Aug 3
For this patch the goal is to have the attributes in the AST in an order that is less surprising to consumers (including out-of-tree). If we change it now, new/revised code/diagnostics will be written to match this order.
Thu, Aug 2
Wed, Aug 1
- Remove TODOs about the attribute order
I am unsure how to proceed. Commit since already accepted? Wait for reconfirmation? Open new differential?
Rebase after de-listifying in r336945
Reopen after revert (and to be able to update the diff)
Tue, Jul 31
As mentioned in the last phone call, I think we should not use 'Stride' as an alternative to row-major indexing. The primary reason is that there are no unique coordinates for a single memory location which means we cannot accurately compute dependencies. Indeed, the delinearization stuff is all about ensuring that there is no unpredictable aliasing.
- Remove appendToArray helper function
- Replace some get("")->getAsXYZ by getXYZ("")
I find the summary confusing. Can we say that simple unrolling will be prioritized over llvm.loop.unroll_and_jam.count metadata and -unroll-and-jam-count= cmdline options?
Has this been recommitted yet?
Mon, Jul 30
Cannot commit this as it miscompiles a few test-suite tests (as does -polly-allow-nonaffine):
FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pabsw_239.test (64 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_packssdw_217.test (67 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_packusdw_296.test (73 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_paddsw_102.test (97 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_paddusw_104.test (104 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_paddw_92.test (110 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pavgw_127.test (119 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pavgw_128.test (120 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pcmpeqw_113.test (146 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pcmpgtw_114.test (160 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmaxsw_129.test (174 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmaxuw_283.test (186 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pminsw_130.test (194 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pminuw_284.test (206 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhuw_110.test (210 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhuw_134.test (212 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhuw_133.test (213 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhuw_135.test (214 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhw_106.test (227 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhw_107.test (228 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhw_109.test (230 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmulhw_108.test (231 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_pmullw_94.test (248 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_psubsw_103.test (265 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_psubusw_105.test (273 of 1340) FAIL: test-suite :: Bitcode/simd_ops/simd_ops_test_op_psubw_93.test (281 of 1340) NOEXE: test-suite :: External/SPEC/CFP2017rate/526.blender_r/526.blender_r.test (532 of 1340) FAIL: test-suite :: External/SPEC/CINT2006/401.bzip2/401.bzip2.test (554 of 1340) FAIL: test-suite :: External/SPEC/CINT2006/458.sjeng/458.sjeng.test (563 of 1340) FAIL: test-suite :: External/SPEC/CINT2017rate/502.gcc_r/502.gcc_r.test (569 of 1340) FAIL: test-suite :: External/SPEC/CINT2017speed/602.gcc_s/602.gcc_s.test (578 of 1340) FAIL: test-suite :: MultiSource/Benchmarks/TSVC/ControlLoops-flt/ControlLoops-flt.test (741 of 1340) FAIL: test-suite :: MultiSource/Benchmarks/TSVC/ControlLoops-dbl/ControlLoops-dbl.test (742 of 1340) FAIL: test-suite :: Polybench/Polybench-boost/linear-algebra/kernels/3mm/Pboost_3mm.test (1021 of 1340)
Sat, Jul 28
Did you check the execution time?
Fri, Jul 27
Wow, that's 10x fast with Polly!
What is the execution time of this?
Can be committed just before the first dependent patch being committed.
ping. Any other opinions on this?
Thu, Jul 26
Title: Added -> Add
Typo in title: requred
- Address @dmgreen's remarks