Page MenuHomePhabricator

craig.topper (Craig Topper)
User

Projects

User does not belong to any projects.

User Details

User Since
Jul 30 2013, 7:58 PM (320 w, 5 d)

Recent Activity

Yesterday

craig.topper created D67902: [X86] Use TargetConstant for condition code on X86ISD::SETCC/CMOV/BRCOND nodes..
Sun, Sep 22, 11:55 PM · Restricted Project
craig.topper added a comment to D67898: [SDAG] Peek through bitcasts before commuting setcc operands.

I've removed the canonicalization in LowerIntVSETCC_AVX512 that was causing this in r372543. And I've moved it to X86ISelDAGToDAG in r372544. At the time LowerIntVSETCC_AVX512 we were using isel patterns for vptest and this was an attempt reduce the size of the table. We're now using custom isel in X86ISelDAGToDAG so the canonicalization is much cheaper to handle there.

Sun, Sep 22, 10:58 PM · Restricted Project
craig.topper committed rG03b5a13ee32e: [X86] Canonicalize all zeroes vector to RHS in X86DAGToDAGISel::tryVPTESTM. (authored by craig.topper).
[X86] Canonicalize all zeroes vector to RHS in X86DAGToDAGISel::tryVPTESTM.
Sun, Sep 22, 10:36 PM
craig.topper committed rG5e26064c4061: [X86] Remove SETEQ/SETNE canonicalization code from LowerIntVSETCC_AVX512 to… (authored by craig.topper).
[X86] Remove SETEQ/SETNE canonicalization code from LowerIntVSETCC_AVX512 to…
Sun, Sep 22, 10:36 PM
craig.topper committed rG1f058538e0e4: [X86] Add 32-bit command line to avx512f-vec-test-testn.ll (authored by craig.topper).
[X86] Add 32-bit command line to avx512f-vec-test-testn.ll
Sun, Sep 22, 10:36 PM
craig.topper committed rGda4a4707d2ed: [X86] Convert to Constant arguments to MMX shift by i32 intrinsics to… (authored by craig.topper).
[X86] Convert to Constant arguments to MMX shift by i32 intrinsics to…
Sun, Sep 22, 6:21 PM
craig.topper committed rG5efc928ab278: [X86] Remove stale FIXME. (authored by craig.topper).
[X86] Remove stale FIXME.
Sun, Sep 22, 6:21 PM
craig.topper committed rGa533e87792ca: [X86][SelectionDAGBuilder] Move the hack for handling MMX shift by i32… (authored by craig.topper).
[X86][SelectionDAGBuilder] Move the hack for handling MMX shift by i32…
Sun, Sep 22, 6:05 PM
craig.topper committed rGe4c17651243e: [X86] Require last argument to LWPINS/LWPVAL builtins to be an ICE. Add ImmArg… (authored by craig.topper).
[X86] Require last argument to LWPINS/LWPVAL builtins to be an ICE. Add ImmArg…
Sun, Sep 22, 4:49 PM
craig.topper accepted D67874: Prefer AVX512 memcpy when applicable.

LGTM

Sun, Sep 22, 2:55 PM · Restricted Project
craig.topper accepted D67875: [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to BZHI (PR43381).

LGTM with those comment fixes.

Sun, Sep 22, 2:54 PM · Restricted Project
craig.topper committed rGc4802de31b1a: [X86] Fix some VCVTPS2PH isel patterns where 'i32' was used instead of 'timm' (authored by craig.topper).
[X86] Fix some VCVTPS2PH isel patterns where 'i32' was used instead of 'timm'
Sun, Sep 22, 1:10 PM
craig.topper added a comment to D67875: [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to BZHI (PR43381).

What if we just do the larger than 32-bit mask? Its not clear that making BZHI just to fold a load is an improvement. You have to materialize an immediate instead so the total uops increased.

Sun, Sep 22, 1:07 PM · Restricted Project
craig.topper committed rG80fda375b225: [X86][TableGen] Allow timm to appear in output patterns. Use it to remove… (authored by craig.topper).
[X86][TableGen] Allow timm to appear in output patterns. Use it to remove…
Sun, Sep 22, 12:49 PM
craig.topper added inline comments to D67875: [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to BZHI (PR43381).
Sun, Sep 22, 12:20 PM · Restricted Project
craig.topper committed rGa1d86857fff2: [X86] Update commutable EVEX vcmp patterns to use timm instead of imm. (authored by craig.topper).
[X86] Update commutable EVEX vcmp patterns to use timm instead of imm.
Sun, Sep 22, 12:07 PM
craig.topper committed rGac8477126178: [X86] Add more tests for commuting evex vcmp instructions during isel to fold a… (authored by craig.topper).
[X86] Add more tests for commuting evex vcmp instructions during isel to fold a…
Sun, Sep 22, 12:07 PM
craig.topper updated subscribers of D67875: [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to BZHI (PR43381).
Sun, Sep 22, 11:51 AM · Restricted Project
craig.topper added inline comments to D67874: Prefer AVX512 memcpy when applicable.
Sun, Sep 22, 11:47 AM · Restricted Project
craig.topper added inline comments to D67875: [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to BZHI (PR43381).
Sun, Sep 22, 10:59 AM · Restricted Project
craig.topper added inline comments to D67874: Prefer AVX512 memcpy when applicable.
Sun, Sep 22, 10:44 AM · Restricted Project

Sat, Sep 21

craig.topper committed rG38014c553f0d: [X86] Add test memset and memcpy testcases for D67874. NFC (authored by craig.topper).
[X86] Add test memset and memcpy testcases for D67874. NFC
Sat, Sep 21, 11:52 PM
craig.topper added a comment to D67874: Prefer AVX512 memcpy when applicable.

I added some memset tests in r372494 and the RUN line changes here. Can you rebase?

Sat, Sep 21, 11:51 PM · Restricted Project
craig.topper added a comment to D67874: Prefer AVX512 memcpy when applicable.

Do we have a similar test for memset that should be updated?

Sat, Sep 21, 8:23 AM · Restricted Project

Fri, Sep 20

craig.topper committed rG04682939eb7e: [X86] Use sse_load_f32/f64 and timm in patterns for memory form of… (authored by craig.topper).
[X86] Use sse_load_f32/f64 and timm in patterns for memory form of…
Fri, Sep 20, 11:44 PM
craig.topper committed rG4fa12ac92ced: [X86] Add test case to show failure to fold load with getmantss due to isel… (authored by craig.topper).
[X86] Add test case to show failure to fold load with getmantss due to isel…
Fri, Sep 20, 11:44 PM
craig.topper added inline comments to D67850: Fix missed case of switching getConstant to getTargetConstant. Try 2..
Fri, Sep 20, 11:57 AM · Restricted Project
craig.topper committed rGc139d1e28109: [Mips] Remove immarg test for intrinsics that no longer have an immarg after… (authored by craig.topper).
[Mips] Remove immarg test for intrinsics that no longer have an immarg after…
Fri, Sep 20, 11:52 AM
craig.topper committed rG1b7b4b467f03: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold… (authored by craig.topper).
[SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold…
Fri, Sep 20, 9:51 AM
craig.topper added a comment to D67802: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold TargetConstant nodes to a Constant..

LGTM but could use tests

Fri, Sep 20, 9:38 AM · Restricted Project
craig.topper updated the diff for D67802: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold TargetConstant nodes to a Constant..

Remove ImmArg from Mips intrinsics

Fri, Sep 20, 9:01 AM · Restricted Project
craig.topper added inline comments to D67802: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold TargetConstant nodes to a Constant..
Fri, Sep 20, 9:01 AM · Restricted Project
craig.topper added reviewers for D67802: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold TargetConstant nodes to a Constant.: venkatra, atanasyan.
Fri, Sep 20, 12:30 AM · Restricted Project
craig.topper created D67802: [SelectionDAG][Mips][Sparc] Don't allow SimplifyDemandedBits to constant fold TargetConstant nodes to a Constant..
Fri, Sep 20, 12:27 AM · Restricted Project
craig.topper committed rG621c93ec1fd8: [X86] Convert tbm_bextri_u32/tbm_bextri_u64 intrinsics TargetConstant argument… (authored by craig.topper).
[X86] Convert tbm_bextri_u32/tbm_bextri_u64 intrinsics TargetConstant argument…
Fri, Sep 20, 12:04 AM

Thu, Sep 19

craig.topper committed rGa34f13f2bab3: [X86] Use timm in MMX pinsrw/pextrw isel patterns. Add missing test cases. (authored by craig.topper).
[X86] Use timm in MMX pinsrw/pextrw isel patterns. Add missing test cases.
Thu, Sep 19, 11:00 PM
craig.topper committed rG081cb7ef2370: [X86] Remove the special isBuildVectorOfConstantSDNodes handling from… (authored by craig.topper).
[X86] Remove the special isBuildVectorOfConstantSDNodes handling from…
Thu, Sep 19, 6:50 PM

Wed, Sep 18

craig.topper committed rGc2d25ed1b36d: [X86] Prevent crash in LowerBUILD_VECTORvXi1 for v64i1 vectors on 32-bit… (authored by craig.topper).
[X86] Prevent crash in LowerBUILD_VECTORvXi1 for v64i1 vectors on 32-bit…
Wed, Sep 18, 11:57 PM
craig.topper committed rGd103bb654fe1: [X86] Change a SmallVector& argument to SmallVectorImpl&. NFC (authored by craig.topper).
[X86] Change a SmallVector& argument to SmallVectorImpl&. NFC
Wed, Sep 18, 11:26 PM
craig.topper committed rGeff4fd699982: [X86] Remove unused argument from a helper function. NFC (authored by craig.topper).
[X86] Remove unused argument from a helper function. NFC
Wed, Sep 18, 11:26 PM

Tue, Sep 17

craig.topper committed rG93e1f73b6b2d: [X86] Break non-power of 2 vXi1 vectors into scalars for argument passing with… (authored by craig.topper).
[X86] Break non-power of 2 vXi1 vectors into scalars for argument passing with…
Tue, Sep 17, 11:09 PM
craig.topper committed rG11082d53665d: [X86] Add test case for passing a v17i1 vector with avx512 (authored by craig.topper).
[X86] Add test case for passing a v17i1 vector with avx512
Tue, Sep 17, 11:09 PM
craig.topper committed rG4a07336a886d: [X86] Prevent assertion when calling a function that returns double with -mno… (authored by craig.topper).
[X86] Prevent assertion when calling a function that returns double with -mno…
Tue, Sep 17, 7:00 PM
craig.topper committed rGc198ffd8c34b: [X86] Use APInt::operator<<= and APInt::lshrInPlace. NFC (authored by craig.topper).
[X86] Use APInt::operator<<= and APInt::lshrInPlace. NFC
Tue, Sep 17, 11:21 AM
craig.topper committed rGb5ffbd0b1452: [SimplifyDemandedBits] Use APInt::intersects to instead of ANDing and comparing… (authored by craig.topper).
[SimplifyDemandedBits] Use APInt::intersects to instead of ANDing and comparing…
Tue, Sep 17, 11:21 AM
craig.topper committed rGf9a89b67885b: [X86] Simplify b2b KSHIFTL+KSHIFTR using demanded elts. (authored by craig.topper).
[X86] Simplify b2b KSHIFTL+KSHIFTR using demanded elts.
Tue, Sep 17, 11:03 AM
craig.topper committed rGf1ba94ade0b6: [X86] Call SimplifyDemandedVectorElts on KSHIFTL/KSHIFTR nodes during DAG… (authored by craig.topper).
[X86] Call SimplifyDemandedVectorElts on KSHIFTL/KSHIFTR nodes during DAG…
Tue, Sep 17, 11:03 AM
craig.topper committed rGb50894b9c369: [X86] Simplify some code in LowerBUILD_VECTORvXi1. NFCI (authored by craig.topper).
[X86] Simplify some code in LowerBUILD_VECTORvXi1. NFCI
Tue, Sep 17, 11:03 AM

Mon, Sep 16

craig.topper committed rG95aea7449493: [X86] Split oversized vXi1 vector arguments and return values into scalars on… (authored by craig.topper).
[X86] Split oversized vXi1 vector arguments and return values into scalars on…
Mon, Sep 16, 9:41 PM
craig.topper committed rG769dd59a27e5: [X86] Allow masked VBROADCAST instructions to be turned into BLENDM with a… (authored by craig.topper).
[X86] Allow masked VBROADCAST instructions to be turned into BLENDM with a…
Mon, Sep 16, 9:41 PM
craig.topper committed rG2cc57bedd51b: [X86] Add support for commuting EVEX VCMP instructons with any immediate value. (authored by craig.topper).
[X86] Add support for commuting EVEX VCMP instructons with any immediate value.
Mon, Sep 16, 9:41 PM
craig.topper committed rGd51576a3f06b: [X86] Add test case for missed opportunity to commute a VCMP instruction after… (authored by craig.topper).
[X86] Add test case for missed opportunity to commute a VCMP instruction after…
Mon, Sep 16, 9:41 PM
craig.topper committed rG359918dadf4d: [X86] Enable commuting of EVEX VCMP for all immediate values during isel. (authored by craig.topper).
[X86] Enable commuting of EVEX VCMP for all immediate values during isel.
Mon, Sep 16, 9:41 PM
craig.topper added a comment to D61675: [WIP] Update IRBuilder::CreateFNeg(...) to return a UnaryOperator.

What targets does clang enable FTZ/DAZ on? I don't think it does on X86.

Mon, Sep 16, 9:33 AM · Restricted Project

Fri, Sep 13

craig.topper committed rG8e0f104916f1: [X86] Use incDecVectorConstant to simplify the min/max code in LowerVSETCC. (authored by craig.topper).
[X86] Use incDecVectorConstant to simplify the min/max code in LowerVSETCC.
Fri, Sep 13, 8:02 AM

Thu, Sep 12

craig.topper committed rG4d1df2aa2380: [TargetRegisterInfo] Remove SVT argument from getCommonSubClass. (authored by craig.topper).
[TargetRegisterInfo] Remove SVT argument from getCommonSubClass.
Thu, Sep 12, 10:27 PM
craig.topper committed rG36e04d14e9fa: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC… (authored by craig.topper).
[PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC…
Thu, Sep 12, 3:08 PM
craig.topper updated the diff for D67513: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC register class..

Remove bad change

Thu, Sep 12, 2:37 PM · Restricted Project
craig.topper committed rGefe6724b9f55: [DAGCombiner][X86] Pass the CmpOpVT to reduceSelectOfFPConstantLoads so X86 can… (authored by craig.topper).
[DAGCombiner][X86] Pass the CmpOpVT to reduceSelectOfFPConstantLoads so X86 can…
Thu, Sep 12, 2:32 PM
craig.topper added inline comments to D67513: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC register class..
Thu, Sep 12, 2:32 PM · Restricted Project
craig.topper committed rG344c398e2a0b: [SelectionDAGBuilder] Simplify loop in visitSelect back to how it was before… (authored by craig.topper).
[SelectionDAGBuilder] Simplify loop in visitSelect back to how it was before…
Thu, Sep 12, 2:03 PM
craig.topper added a comment to D67513: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC register class..

Thanks for getting rid of SPE4RC, I really did not like it, but couldn't think of a better way. Glad you did.

It looks like you're changing more than just removing SPE4RC in this diff. Do the tests really change because of the removal?

Thu, Sep 12, 1:23 PM · Restricted Project
craig.topper added reviewers for D67513: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC register class.: jhibbits, nemanjai, kbarton, hfinkel.
Thu, Sep 12, 12:11 PM · Restricted Project
craig.topper created D67513: [PowerPC] Remove the SPE4RC register class and instead add f32 to the GPRC register class..
Thu, Sep 12, 12:10 PM · Restricted Project

Wed, Sep 11

craig.topper committed rGb8dd07527568: [LegalizeTypes] Remove code for softening a float type to itself. (authored by craig.topper).
[LegalizeTypes] Remove code for softening a float type to itself.
Wed, Sep 11, 10:57 PM
craig.topper committed rG635d383fad2b: [X86] Enable -mprefer-vector-width=256 by default for Skylake-avx512 and later… (authored by craig.topper).
[X86] Enable -mprefer-vector-width=256 by default for Skylake-avx512 and later…
Wed, Sep 11, 4:55 PM
craig.topper committed rG5278b0a04e86: [X86] Add test case for v16i64->v16i32 truncate on min-legal-vector-width=256. (authored by craig.topper).
[X86] Add test case for v16i64->v16i32 truncate on min-legal-vector-width=256.
Wed, Sep 11, 2:33 PM
craig.topper committed rG08474ca091fe: [X86] Move x86_64 fp128 conversion to libcalls from type legalization to DAG… (authored by craig.topper).
[X86] Move x86_64 fp128 conversion to libcalls from type legalization to DAG…
Wed, Sep 11, 2:30 PM
craig.topper added a comment to D67259: [X86] Enable -mprefer-vector-width=256 by default for Skylake-avx512 and later Intel CPUs..

So who wants to click Accept?

Wed, Sep 11, 12:48 PM · Restricted Project
craig.topper updated the diff for D67259: [X86] Enable -mprefer-vector-width=256 by default for Skylake-avx512 and later Intel CPUs..

Update clang release notes as well.

Wed, Sep 11, 9:41 AM · Restricted Project

Tue, Sep 10

craig.topper added a comment to D67259: [X86] Enable -mprefer-vector-width=256 by default for Skylake-avx512 and later Intel CPUs..

Do we need to include some benchmark numbers?

Tue, Sep 10, 10:58 PM · Restricted Project
craig.topper added a comment to D67363: [BreakFalseDeps] ignore function with minsize attribute.

X86 looks good to me.

Tue, Sep 10, 10:39 PM · Restricted Project
craig.topper updated the diff for D67128: [X86] Move x86_64 fp128 conversion to libcalls from type legalization to DAG legalization.

Fix review comments

Tue, Sep 10, 10:21 PM · Restricted Project
craig.topper added inline comments to D67128: [X86] Move x86_64 fp128 conversion to libcalls from type legalization to DAG legalization.
Tue, Sep 10, 2:10 PM · Restricted Project
craig.topper added a comment to D67363: [BreakFalseDeps] ignore function with minsize attribute.

I believe this test case compiled with avx needs this pass.

Tue, Sep 10, 8:23 AM · Restricted Project
craig.topper added a comment to D67363: [BreakFalseDeps] ignore function with minsize attribute.

For VEX instructions don’t we just use the other input register to break the dependency without adding an instruction?

That sounds right, but it's not controlled by this pass. That's part of memory op folding?
If I'm seeing it correctly, that's already more aggressively optimizing for size than what we're doing here:

// Avoid partial and undef register update stalls unless optimizing for size.
if (!MF.getFunction().hasOptSize() &&
    (hasPartialRegUpdate(MI.getOpcode(), Subtarget, /*ForLoadFold*/true) ||
     shouldPreventUndefRegUpdateMemFold(MF, MI)))
  return nullptr;
Tue, Sep 10, 8:16 AM · Restricted Project
craig.topper added a comment to D67363: [BreakFalseDeps] ignore function with minsize attribute.

For VEX instructions don’t we just use the other input register to break the dependency without adding an instruction?

Tue, Sep 10, 7:18 AM · Restricted Project
craig.topper committed rGe8b432fa0e75: [LegalizeTypes] Teach SoftenFloatOp_SELECT_CC to handle operand 2 or 3 being… (authored by craig.topper).
[LegalizeTypes] Teach SoftenFloatOp_SELECT_CC to handle operand 2 or 3 being…
Tue, Sep 10, 1:01 AM

Mon, Sep 9

craig.topper committed rG0e533ca4bb82: [X86] Add broadcast load unfolding support for VCMPPS/PD. (authored by craig.topper).
[X86] Add broadcast load unfolding support for VCMPPS/PD.
Mon, Sep 9, 10:52 PM
craig.topper committed rG7c2fdf27795a: [X86] Add broadcast load unfold tests for VCMPPS/PD. (authored by craig.topper).
[X86] Add broadcast load unfold tests for VCMPPS/PD.
Mon, Sep 9, 10:52 PM
craig.topper accepted D66322: [X86] Updated target specific selection dag code to conservatively check for isAtomic in addition to isVolatile.

LGTM. Will you also be updating nonvolatile_load/nonvolatile_store in include/llvm/Target/TargetSelectionDAG.td. We use those in several places where we change the size of a load.

Mon, Sep 9, 4:01 PM · Restricted Project
craig.topper committed rG5ebd0a6e88a5: [SelectionDAG] Remove ISD::FP_ROUND_INREG (authored by craig.topper).
[SelectionDAG] Remove ISD::FP_ROUND_INREG
Mon, Sep 9, 10:54 AM
craig.topper committed rGce2cb0f09e7d: [X86] Allow _MM_FROUND_CUR_DIRECTION and _MM_FROUND_NO_EXC to be used together… (authored by craig.topper).
[X86] Allow _MM_FROUND_CUR_DIRECTION and _MM_FROUND_NO_EXC to be used together…
Mon, Sep 9, 10:48 AM
craig.topper updated the diff for D67312: [SelectionDAG] Remove ISD::FP_ROUND_INREG.

Add a sentence to the release notes

Mon, Sep 9, 10:03 AM · Restricted Project
craig.topper committed rGa88f58ff0e8b: [X86] Add broadcast load unfolding support for vpcmpeq/vpcmpgt/vpcmp/vpcmpu. (authored by craig.topper).
[X86] Add broadcast load unfolding support for vpcmpeq/vpcmpgt/vpcmp/vpcmpu.
Mon, Sep 9, 12:47 AM
craig.topper committed rG667f039c8ca7: [X86] Add broadcast load unfolding tests for vpcmpeq/vpcmpgt/vpcmp/vpcmpu. (authored by craig.topper).
[X86] Add broadcast load unfolding tests for vpcmpeq/vpcmpgt/vpcmp/vpcmpu.
Mon, Sep 9, 12:47 AM

Sun, Sep 8

craig.topper committed rG8c2ab1c4cbce: [X86] Add broadcast load unfold support for smin/umin/smax/umax. (authored by craig.topper).
[X86] Add broadcast load unfold support for smin/umin/smax/umax.
Sun, Sep 8, 11:34 PM
craig.topper committed rG68b2e1973fc5: [X86] Add broadcast load unfolding tests for smin/umin/smax/smin. (authored by craig.topper).
[X86] Add broadcast load unfolding tests for smin/umin/smax/smin.
Sun, Sep 8, 11:34 PM
craig.topper committed rGad7822329fd0: [X86] Add broadcast load unfolding support for VMAXPS/PD and VMINPS/PD. (authored by craig.topper).
[X86] Add broadcast load unfolding support for VMAXPS/PD and VMINPS/PD.
Sun, Sep 8, 9:25 PM
craig.topper committed rG8d42a796c2ad: [X86] Add broadcast load unfolding tests for vmaxps/pd and vminps/pd (authored by craig.topper).
[X86] Add broadcast load unfolding tests for vmaxps/pd and vminps/pd
Sun, Sep 8, 9:25 PM
craig.topper committed rG197901081bc5: [X86] Add fp128 test cases for ceil/floor/trunc/nearbyint/rint/round libcalls. (authored by craig.topper).
[X86] Add fp128 test cases for ceil/floor/trunc/nearbyint/rint/round libcalls.
Sun, Sep 8, 7:49 PM
craig.topper updated the diff for D67128: [X86] Move x86_64 fp128 conversion to libcalls from type legalization to DAG legalization.

Rebase

Sun, Sep 8, 7:11 PM · Restricted Project
craig.topper committed rGfb1e77505ae6: [X86] Add test cases for fptoui/fptosi/sitofp/uitofp between fp128 and i128. (authored by craig.topper).
[X86] Add test cases for fptoui/fptosi/sitofp/uitofp between fp128 and i128.
Sun, Sep 8, 6:37 PM
craig.topper committed rG72624b0e59d6: [X86] Use xorps to create fp128 +0.0 constants. (authored by craig.topper).
[X86] Use xorps to create fp128 +0.0 constants.
Sun, Sep 8, 6:36 PM
craig.topper committed rG861d34394939: [X86] Add avx and avx512f RUN lines to fp128-cast.ll (authored by craig.topper).
[X86] Add avx and avx512f RUN lines to fp128-cast.ll
Sun, Sep 8, 6:36 PM
craig.topper committed rG77dd86ee4aae: [X86] Add a hack to combineVSelectWithAllOnesOrZeros to turn selects with two… (authored by craig.topper).
[X86] Add a hack to combineVSelectWithAllOnesOrZeros to turn selects with two…
Sun, Sep 8, 1:57 PM
craig.topper committed rG9c119012566c: [X86] Remove call to getZeroVector from materializeVectorConstant. Add isel… (authored by craig.topper).
[X86] Remove call to getZeroVector from materializeVectorConstant. Add isel…
Sun, Sep 8, 1:57 PM
craig.topper committed rG97d41b891743: [X86] Use DAG.getConstant instead of getZeroVector in combinePMULDQ. (authored by craig.topper).
[X86] Use DAG.getConstant instead of getZeroVector in combinePMULDQ.
Sun, Sep 8, 12:25 PM
craig.topper committed rGdac34f52d3f9: [DAGCombiner][X86][ARM] Teach visitMULO to fold multiplies with 0 to 0 and no… (authored by craig.topper).
[DAGCombiner][X86][ARM] Teach visitMULO to fold multiplies with 0 to 0 and no…
Sun, Sep 8, 12:25 PM
craig.topper committed rG30837abd9623: [X86] Teach materializeVectorConstant to not call getZeroVector/getOnesVector… (authored by craig.topper).
[X86] Teach materializeVectorConstant to not call getZeroVector/getOnesVector…
Sun, Sep 8, 12:25 PM