Page MenuHomePhabricator

Petar.Avramovic (Petar Avramovic)
User

Projects

User does not belong to any projects.

User Details

User Since
Feb 21 2018, 6:39 AM (145 w, 1 d)

Recent Activity

Fri, Nov 20

Petar.Avramovic updated the diff for D90050: AMDGPU/GlobalISel: Add integer med3 combines.

Move combines after register bank select. Check for register bank of the result register and only combine for vgpr bank. There is mir test for sgpr bank min/max input for completeness. However sgpr bank min/max are transformed to compare + select in regbankselect and we should not expect them in post regbank combiner.

Fri, Nov 20, 3:46 AM · Restricted Project

Thu, Nov 19

Petar.Avramovic added a comment to D90050: AMDGPU/GlobalISel: Add integer med3 combines.

Should we move this combine after register bank select and skip min/max with sgpr bank?

Thu, Nov 19, 8:44 AM · Restricted Project
Petar.Avramovic updated the diff for D90050: AMDGPU/GlobalISel: Add integer med3 combines.

Address review comments.

Thu, Nov 19, 8:41 AM · Restricted Project
Petar.Avramovic updated the diff for D91336: AMDGPU/GlobalISel: Fix negative offset folding for buffer_load.
Thu, Nov 19, 6:26 AM · Restricted Project

Wed, Nov 18

Petar.Avramovic updated the diff for D90050: AMDGPU/GlobalISel: Add integer med3 combines.

Use MIPatternMatch to match source pattern. Use Optional instead of return 0. Separate from fp med3 function to avoid bitcasted G_FCONSTANT and explicitly match G_CONSTANT.

Wed, Nov 18, 12:03 PM · Restricted Project
Petar.Avramovic added inline comments to D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 11:27 AM · Restricted Project
Petar.Avramovic added inline comments to D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 9:26 AM · Restricted Project
Petar.Avramovic updated the diff for D91714: AMDGPU/GlobalISel: Check values of constants in isKnownNeverNaN.

Add braces.

Wed, Nov 18, 9:19 AM · Restricted Project
Petar.Avramovic added inline comments to D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 9:05 AM · Restricted Project
Petar.Avramovic added inline comments to D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 8:49 AM · Restricted Project
Petar.Avramovic added inline comments to D91714: AMDGPU/GlobalISel: Check values of constants in isKnownNeverNaN.
Wed, Nov 18, 8:38 AM · Restricted Project
Petar.Avramovic added inline comments to D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 8:34 AM · Restricted Project
Petar.Avramovic requested review of D91716: AMDGPU/GlobalISel: Calculate isKnownNeverNaN for fminnum and fmaxnum.
Wed, Nov 18, 8:29 AM · Restricted Project
Petar.Avramovic requested review of D91714: AMDGPU/GlobalISel: Check values of constants in isKnownNeverNaN.
Wed, Nov 18, 8:27 AM · Restricted Project

Thu, Nov 12

Petar.Avramovic added inline comments to D91336: AMDGPU/GlobalISel: Fix negative offset folding for buffer_load.
Thu, Nov 12, 9:48 AM · Restricted Project
Petar.Avramovic added inline comments to D91336: AMDGPU/GlobalISel: Fix negative offset folding for buffer_load.
Thu, Nov 12, 8:03 AM · Restricted Project
Petar.Avramovic updated the diff for D91336: AMDGPU/GlobalISel: Fix negative offset folding for buffer_load.

Pre-commit the test.

Thu, Nov 12, 5:58 AM · Restricted Project
Petar.Avramovic requested review of D91336: AMDGPU/GlobalISel: Fix negative offset folding for buffer_load.
Thu, Nov 12, 4:50 AM · Restricted Project

Nov 3 2020

Petar.Avramovic added inline comments to D90052: AMDGPU/GlobalISel: Add clamp combine for IEEE=false.
Nov 3 2020, 5:52 AM · Restricted Project
Petar.Avramovic committed rG0031418dce3e: AMDGPU/GlobalISel: Use same builder/observer in post-legalizer-combiner (authored by Petar.Avramovic).
AMDGPU/GlobalISel: Use same builder/observer in post-legalizer-combiner
Nov 3 2020, 12:26 AM
Petar.Avramovic closed D90623: AMDGPU/GlobalISel: Use same builder/observer in post-legalizer-combiner.
Nov 3 2020, 12:26 AM · Restricted Project

Nov 2 2020

Petar.Avramovic updated the diff for D90623: AMDGPU/GlobalISel: Use same builder/observer in post-legalizer-combiner.

Use setInstrAndDebugLoc.

Nov 2 2020, 10:42 AM · Restricted Project
Petar.Avramovic requested review of D90623: AMDGPU/GlobalISel: Use same builder/observer in post-legalizer-combiner.
Nov 2 2020, 9:23 AM · Restricted Project

Oct 23 2020

Petar.Avramovic added inline comments to D90051: AMDGPU/GlobalISel: Add floating point med3 combine for IEEE=false.
Oct 23 2020, 11:34 AM · Restricted Project
Petar.Avramovic added inline comments to D90052: AMDGPU/GlobalISel: Add clamp combine for IEEE=false.
Oct 23 2020, 11:30 AM · Restricted Project
Petar.Avramovic requested review of D90052: AMDGPU/GlobalISel: Add clamp combine for IEEE=false.
Oct 23 2020, 8:43 AM · Restricted Project
Petar.Avramovic requested review of D90051: AMDGPU/GlobalISel: Add floating point med3 combine for IEEE=false.
Oct 23 2020, 8:42 AM · Restricted Project
Petar.Avramovic requested review of D90050: AMDGPU/GlobalISel: Add integer med3 combines.
Oct 23 2020, 8:40 AM · Restricted Project

Oct 20 2020

Petar.Avramovic added inline comments to D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.
Oct 20 2020, 4:14 AM · Restricted Project

Oct 16 2020

Petar.Avramovic updated the diff for D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.

Respect description of ISD::FMAXNUM and ISD::FMAXNUM_IEEE (their behavior is not affected by IEEE mode).
Legalizer will deal with use of FMAXNUM when IEEE=true (makes sure that potential SNaN input gets quieted) and we can try to fold FMAXNUM_IEEE later.
Most of the test changes come from clamp/fmed on top of fcanonicalize with IEEE=true.

Oct 16 2020, 6:49 AM · Restricted Project

Oct 13 2020

Petar.Avramovic added inline comments to D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.
Oct 13 2020, 8:04 AM · Restricted Project
Petar.Avramovic updated the diff for D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.

Fix typos.

Oct 13 2020, 8:04 AM · Restricted Project

Oct 1 2020

Petar.Avramovic updated the diff for D88573: [SelectionDAG] Add check for BUILD_VECTOR in isKnownNeverNaN.

Patch also affects constants with vector type. Effectively, during legalization, fcanonicalize will not be inserted when input was vector constant.

define amdgpu_kernel void @maxnum_v2f16_imm_b(<2 x half> addrspace(1)* %r, <2 x half> addrspace(1)* %a) {
entry:
  %a.val = load <2 x half>, <2 x half> addrspace(1)* %a
  %add = fadd <2 x half> %a.val, <half 9.0, half 8.0>
  %r.val = call <2 x half> @llvm.maxnum.v2f16(<2 x half> %add, <2 x half> <half 4.0, half 3.0>) ;<half 4.0, half 3.0> will not get fcanonicalized during legalization after this patch
  store <2 x half> %r.val, <2 x half> addrspace(1)* %r
  ret void
}

This won't be visible in end result since there is fcanonicalize + build_vector combine in SITargetLowering::performFCanonicalizeCombine. Test changes were for non-constant build_vectors.
Add comment to indicate vector constants are handled bellow.

Oct 1 2020, 4:11 AM · Restricted Project

Sep 30 2020

Petar.Avramovic updated the diff for D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.

Simplify logic in 'is safe to clamp' checks.

Sep 30 2020, 9:18 AM · Restricted Project
Petar.Avramovic requested review of D88574: AMDGPU/SelectionDAG Include fast-math-flags for fmed3 intrinsic.
Sep 30 2020, 7:41 AM · Restricted Project
Petar.Avramovic requested review of D88573: [SelectionDAG] Add check for BUILD_VECTOR in isKnownNeverNaN.
Sep 30 2020, 7:40 AM · Restricted Project
Petar.Avramovic requested review of D88572: AMDGPU/SelectionDAG Check for NaN, DX10Clamp and IEEE in fmed3 combine.
Sep 30 2020, 7:39 AM · Restricted Project

Sep 18 2020

Petar.Avramovic updated the diff for D86294: AMDGPU/GlobalISel: Add tablegen operator that looks through copies.

Allow patterns to explicitly ask to look through copies.

Sep 18 2020, 8:28 AM · Restricted Project

Sep 15 2020

Petar.Avramovic committed rG9b4fa8543431: GlobalISel/IRTranslator resetTargetOptions based on function attributes (authored by Petar.Avramovic).
GlobalISel/IRTranslator resetTargetOptions based on function attributes
Sep 15 2020, 1:26 AM
Petar.Avramovic closed D87511: GlobalISel/IRTranslator resetTargetOptions based on function attributes.
Sep 15 2020, 1:26 AM · Restricted Project

Sep 14 2020

Petar.Avramovic committed rG6e2a86ed5abf: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN (authored by Petar.Avramovic).
AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN
Sep 14 2020, 3:12 AM
Petar.Avramovic committed rG416346d1ca50: AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks' (authored by Petar.Avramovic).
AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks'
Sep 14 2020, 3:11 AM
Petar.Avramovic committed rG0c8f4cd65734: AMDGPU/GlobalISel Add test for non-leaf complex patterns (authored by Petar.Avramovic).
AMDGPU/GlobalISel Add test for non-leaf complex patterns
Sep 14 2020, 3:11 AM
Petar.Avramovic closed D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN.
Sep 14 2020, 3:11 AM · Restricted Project
Petar.Avramovic closed D87351: AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks'.
Sep 14 2020, 3:11 AM · Restricted Project
Petar.Avramovic committed rG09b8871f8d81: AMDGPU/GlobalISel/Emitter Support for predicate code that uses operands (authored by Petar.Avramovic).
AMDGPU/GlobalISel/Emitter Support for predicate code that uses operands
Sep 14 2020, 1:40 AM
Petar.Avramovic closed D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.
Sep 14 2020, 1:40 AM · Restricted Project

Sep 11 2020

Petar.Avramovic added a comment to D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN.

This will be fixed by in D87511.

Wouldn't it make more sense to change the order of the patches in the stack so that D87511 is applied first and this one is rebaed on top of it?

Sep 11 2020, 6:57 AM · Restricted Project
Petar.Avramovic updated the diff for D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN.

Move change in IRTranslator in another patch, since effects of that change can only be seen when function attributes are inspected during translation of a function that was not first in input file.
Move check for NoNaNsFPMath in isKnownNeverSNaN.
Reduce number of tests, only keep characteristic ones that will be updated when we add some combines and improvements to legalizer. Later switch to existing test in sdag that also checks for pattern commutes.

Sep 11 2020, 6:42 AM · Restricted Project
Petar.Avramovic requested review of D87511: GlobalISel/IRTranslator resetTargetOptions based on function attributes.
Sep 11 2020, 6:34 AM · Restricted Project
Petar.Avramovic updated the diff for D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.

Add assert for StoreIdx.

Sep 11 2020, 1:55 AM · Restricted Project

Sep 10 2020

Petar.Avramovic updated the diff for D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.

Factor string generation of operand names into function.

Sep 10 2020, 8:45 AM · Restricted Project
Petar.Avramovic added a comment to D87351: AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks'.

min3/max3 patterns are handled in combiner, inducing med3 patterns with constants. Patch imports med3 isel patterns: min(max(a, b), max(min(a, b), c)) but they need some combines/legalize fixes to be selected. From new imports only V_FRACT can be selected at the moment.

Sep 10 2020, 8:10 AM · Restricted Project
Petar.Avramovic added a comment to D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN.

With G_FCANONICALIZE out of the way, we can now select some of the med3 patterns that were imported in D87351.

Sep 10 2020, 8:02 AM · Restricted Project
Petar.Avramovic requested review of D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN.
Sep 10 2020, 7:56 AM · Restricted Project

Sep 9 2020

Petar.Avramovic updated the diff for D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.

Add more verbose comments, and switch to std::array as container for operands.

Sep 9 2020, 7:22 AM · Restricted Project
Petar.Avramovic updated the diff for D87351: AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks'.

Show how this emitter change affects isel, and update tablegen tests.
Add additional checks for complex suboperands for test added in D57980. Neither tablegen nor sdag have check for such pattern and it looks like an error to me (there is no way for tablegen to know that some c++ code that generates two operands will generate same operand x (first out operand) from different input operands - these are operand 1 and 2 of G_ADD instruction). Also, if I checked correctly, there were no patterns like that in llvm upstream. Skip such patterns anyway in GlobalIselEmitter.

Sep 9 2020, 4:06 AM · Restricted Project
Petar.Avramovic requested review of D87351: AMDGPU/GlobalISel/Emitter Recognize additional 'same operand checks'.
Sep 9 2020, 1:43 AM · Restricted Project

Sep 8 2020

Petar.Avramovic added inline comments to D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.
Sep 8 2020, 7:46 AM · Restricted Project
Petar.Avramovic updated the diff for D86294: AMDGPU/GlobalISel: Add tablegen operator that looks through copies.

Move support for predicate code that uses operands to a separate patch.

Sep 8 2020, 6:37 AM · Restricted Project
Petar.Avramovic requested review of D87285: AMDGPU/GlobalISelemitter Support for predicate code that uses operands.
Sep 8 2020, 6:33 AM · Restricted Project

Sep 4 2020

Petar.Avramovic updated the diff for D86294: AMDGPU/GlobalISel: Add tablegen operator that looks through copies.

Attempt to teach emitter to collect operands that predicate with 'let PredicateCodeUsesOperands = 1' requires.

Sep 4 2020, 7:51 AM · Restricted Project

Aug 20 2020

Petar.Avramovic requested review of D86294: AMDGPU/GlobalISel: Add tablegen operator that looks through copies.
Aug 20 2020, 8:12 AM · Restricted Project

Aug 17 2020

Petar.Avramovic accepted D86060: AMDGPU/GlobalISel: Start implementing computeKnownBitsForTargetInstr.
Aug 17 2020, 7:31 AM · Restricted Project
Petar.Avramovic accepted D86057: GlobalISel: Reduce G_SHL width if source is extension.
Aug 17 2020, 7:31 AM · Restricted Project

Aug 10 2020

Petar.Avramovic committed rG0d58d9e8fb93: AMDGPU/GlobalISel: Lower G_FREM (authored by Petar.Avramovic).
AMDGPU/GlobalISel: Lower G_FREM
Aug 10 2020, 1:18 AM
Petar.Avramovic closed D84324: AMDGPU/GlobalISel: Lower G_FREM.
Aug 10 2020, 1:18 AM · Restricted Project

Aug 7 2020

Petar.Avramovic updated the diff for D84324: AMDGPU/GlobalISel: Lower G_FREM.

Force promotion to 32-bit first when subtarget doesn't have 16-bit instruction.

Aug 7 2020, 6:29 AM · Restricted Project
Petar.Avramovic updated the diff for D84324: AMDGPU/GlobalISel: Lower G_FREM.

Addressed review comments.

Aug 7 2020, 6:01 AM · Restricted Project

Aug 6 2020

Petar.Avramovic committed rGd893278bba01: [GlobalISel][InlineAsm] Fix matching input constraint to physreg (authored by Petar.Avramovic).
[GlobalISel][InlineAsm] Fix matching input constraint to physreg
Aug 6 2020, 5:36 AM
Petar.Avramovic closed D85122: [GlobalISel][InlineAsm] Fix matching input constraint to physreg.
Aug 6 2020, 5:36 AM · Restricted Project
Petar.Avramovic updated the diff for D84324: AMDGPU/GlobalISel: Lower G_FREM.

Switch to custom lowering and update to match changes in dag custom lowering for frem (and use same lowering for s16 also).

Aug 6 2020, 5:22 AM · Restricted Project
Petar.Avramovic updated subscribers of D85122: [GlobalISel][InlineAsm] Fix matching input constraint to physreg.

@hans, could you please commit this patch for 11.x release?

Aug 6 2020, 12:59 AM · Restricted Project

Aug 3 2020

Petar.Avramovic added a comment to D85122: [GlobalISel][InlineAsm] Fix matching input constraint to physreg.

might be worth to include this also for the 11.x release.

What is the procedure for that?

Aug 3 2020, 7:27 AM · Restricted Project
Petar.Avramovic updated the diff for D85122: [GlobalISel][InlineAsm] Fix matching input constraint to physreg.

Remove else branch.

Aug 3 2020, 7:26 AM · Restricted Project
Petar.Avramovic accepted D82878: AMDGPU/GlobalISel: Apply load bitcast to s.buffer.load intrinsic.
Aug 3 2020, 5:43 AM · Restricted Project
Petar.Avramovic accepted D84646: AMDGPU/GlobalISel: Remove old hacks for boolean selection.
Aug 3 2020, 5:40 AM · Restricted Project
Petar.Avramovic accepted D84376: AMDGPU/GlobalISel: Fix selecting broken copies for s32->s64 anyext.
Aug 3 2020, 5:09 AM · Restricted Project
Petar.Avramovic requested review of D85122: [GlobalISel][InlineAsm] Fix matching input constraint to physreg.
Aug 3 2020, 4:48 AM · Restricted Project
Petar.Avramovic accepted D73940: GlobalISel: Reimplement moreElementsVectorDst.
Aug 3 2020, 12:33 AM · Restricted Project

Jul 24 2020

Petar.Avramovic added a comment to D84533: AMDGPU/GlobalISel: Legalize s33 G_ADD, G_SUB and G_MUL.

Would it make more sense for narrow scalar to widen to the next multiple of 32 itself?

Probably yes, at least for add, sub and mul. It could calculate required scalar size and call widenScalar instead on reporting UnableToLegalize.
I am not so sure if this would be correct in general, maybe some opcodes prefer next-power-of-2 instead of multiple of NarrowSize?

Jul 24 2020, 7:50 AM · Restricted Project
Herald added a project to D84533: AMDGPU/GlobalISel: Legalize s33 G_ADD, G_SUB and G_MUL: Restricted Project.
Jul 24 2020, 7:22 AM · Restricted Project
Petar.Avramovic committed rG47bd41d099c0: AMDGPU/GlobalISel: Select set.inactive intrinsic (authored by Petar.Avramovic).
AMDGPU/GlobalISel: Select set.inactive intrinsic
Jul 24 2020, 1:14 AM
Petar.Avramovic closed D84407: AMDGPU/GlobalISel: Select set.inactive intrinsic.
Jul 24 2020, 1:14 AM · Restricted Project

Jul 23 2020

Herald added a project to D84407: AMDGPU/GlobalISel: Select set.inactive intrinsic: Restricted Project.
Jul 23 2020, 7:26 AM · Restricted Project

Jul 22 2020

Petar.Avramovic updated the diff for D84324: AMDGPU/GlobalISel: Lower G_FREM.

Preserve flags, and use G_INTRINSIC_TRUNC.

Jul 22 2020, 8:57 AM · Restricted Project
Petar.Avramovic added inline comments to D84324: AMDGPU/GlobalISel: Lower G_FREM.
Jul 22 2020, 8:22 AM · Restricted Project
Herald added a project to D84324: AMDGPU/GlobalISel: Lower G_FREM: Restricted Project.
Jul 22 2020, 7:07 AM · Restricted Project
Petar.Avramovic committed rG44967fc60451: AMDGPU: Simplify f16 to i64 custom lowering (authored by Petar.Avramovic).
AMDGPU: Simplify f16 to i64 custom lowering
Jul 22 2020, 1:32 AM
Petar.Avramovic closed D84166: AMDGPU: Simplify f16 to i64 custom lowering.
Jul 22 2020, 1:32 AM · Restricted Project

Jul 21 2020

Petar.Avramovic updated the diff for D84166: AMDGPU: Simplify f16 to i64 custom lowering.

This update also simplifies f16 -> i64 on subtargets without has16BitInsts().
Such targets immediately promote f16 to f32 whenever they encounter f16 def using fp16_to_fp node. Recognize fp16_to_fp input to f32 to i64 and do the same thing as for f16 to i64 conversion. This gives similar(small difference for f16 vectors) results for subtargets with or without has16BitInsts() since f16->i32 gets selected like f16->f32->i32.
Update tests with more detailed checks.

Jul 21 2020, 7:59 AM · Restricted Project

Jul 20 2020

Petar.Avramovic committed rG6a1030aa0e4e: AMDGPU/GlobalISel: Legalize s16->s64 G_FPEXT (authored by Petar.Avramovic).
AMDGPU/GlobalISel: Legalize s16->s64 G_FPEXT
Jul 20 2020, 7:13 AM
Petar.Avramovic closed D84030: AMDGPU/GlobalISel: Legalize s16->s64 G_FPEXT.
Jul 20 2020, 7:13 AM · Restricted Project
Petar.Avramovic added inline comments to D84166: AMDGPU: Simplify f16 to i64 custom lowering.
Jul 20 2020, 6:28 AM · Restricted Project
Petar.Avramovic updated the diff for D84166: AMDGPU: Simplify f16 to i64 custom lowering.

Context was missing.

Jul 20 2020, 6:28 AM · Restricted Project
Herald added a project to D84166: AMDGPU: Simplify f16 to i64 custom lowering: Restricted Project.
Jul 20 2020, 5:48 AM · Restricted Project
Petar.Avramovic updated the diff for D84030: AMDGPU/GlobalISel: Legalize s16->s64 G_FPEXT.

Split G_FPEXT in legalizer. This could also be widenScalar for TypeIndex == 1.

Jul 20 2020, 3:59 AM · Restricted Project
Petar.Avramovic committed rGba938f6388ff: AMDGPU/GlobalISel: Legalize s16->s64 G_FPTOSI/G_FPTOUI (authored by Petar.Avramovic).
AMDGPU/GlobalISel: Legalize s16->s64 G_FPTOSI/G_FPTOUI
Jul 20 2020, 2:07 AM
Petar.Avramovic closed D84010: AMDGPU/GlobalISel: Legalize s16->s64 G_FPTOSI/G_FPTOUI.
Jul 20 2020, 2:06 AM · Restricted Project

Jul 17 2020

Petar.Avramovic updated the diff for D84010: AMDGPU/GlobalISel: Legalize s16->s64 G_FPTOSI/G_FPTOUI.
Jul 17 2020, 9:00 AM · Restricted Project