From bc713b193fecd1566a333dc74dfbf5bc04cbc36e Mon Sep 17 00:00:00 2001 From: =?utf8?q?Thorsten=20Sch=C3=BCtt?= Date: Wed, 17 May 2023 17:36:55 +0200 Subject: [PATCH] [GlobalIsel][X86] fix legalization of G_CTLZ and G_CTPOP Note that the builders are protected by is64Bit(). More fine-grained availibility checks. Reviewed By: RKSimon Differential Revision: https://reviews.llvm.org/D150790 --- llvm/lib/Target/X86/X86LegalizerInfo.cpp | 32 ++++++++++++---------- .../test/CodeGen/X86/GlobalISel/legalize-ctpop.mir | 2 +- .../X86/GlobalISel/legalize-leading-zeros.mir | 2 +- 3 files changed, 19 insertions(+), 17 deletions(-) diff --git a/llvm/lib/Target/X86/X86LegalizerInfo.cpp b/llvm/lib/Target/X86/X86LegalizerInfo.cpp index b5b700c..e583d22 100644 --- a/llvm/lib/Target/X86/X86LegalizerInfo.cpp +++ b/llvm/lib/Target/X86/X86LegalizerInfo.cpp @@ -287,6 +287,23 @@ void X86LegalizerInfo::setLegalizerInfo64bit() { LegacyLegalizeActions::Legal); LegacyInfo.setAction({G_MERGE_VALUES, 1, s128}, LegacyLegalizeActions::Legal); LegacyInfo.setAction({G_UNMERGE_VALUES, s128}, LegacyLegalizeActions::Legal); + + if (Subtarget.hasPOPCNT()) { + // popcount + getActionDefinitionsBuilder(G_CTPOP) + .legalFor({{s16, s16}, {s32, s32}, {s64, s64}}) + .widenScalarToNextPow2(1, /*Min=*/16) + .clampScalar(1, s16, s64); + } + + if (Subtarget.hasLZCNT()) { + // count leading zeros (LZCNT) + getActionDefinitionsBuilder(G_CTLZ) + .legalFor({{s16, s16}, {s32, s32}, {s64, s64}}) + .widenScalarToNextPow2(1, /*Min=*/16) + .clampScalar(1, s16, s64); + } + } void X86LegalizerInfo::setLegalizerInfoSSE1() { @@ -389,21 +406,6 @@ void X86LegalizerInfo::setLegalizerInfoSSE42() { if (!Subtarget.hasSSE42()) return; - const LLT s16 = LLT::scalar(16); - const LLT s32 = LLT::scalar(32); - const LLT s64 = LLT::scalar(64); - - // popcount - getActionDefinitionsBuilder(G_CTPOP) - .legalFor({{s16, s16}, {s32, s32}, {s64, s64}}) - .widenScalarToNextPow2(1, /*Min=*/16) - .clampScalar(1, s16, s64); - - // count leading zeros (LZCNT) - getActionDefinitionsBuilder(G_CTLZ) - .legalFor({{s16, s16}, {s32, s32}, {s64, s64}}) - .widenScalarToNextPow2(1, /*Min=*/16) - .clampScalar(1, s16, s64); } void X86LegalizerInfo::setLegalizerInfoAVX() { diff --git a/llvm/test/CodeGen/X86/GlobalISel/legalize-ctpop.mir b/llvm/test/CodeGen/X86/GlobalISel/legalize-ctpop.mir index ab3af64..75ce295 100644 --- a/llvm/test/CodeGen/X86/GlobalISel/legalize-ctpop.mir +++ b/llvm/test/CodeGen/X86/GlobalISel/legalize-ctpop.mir @@ -1,5 +1,5 @@ # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 2 -# RUN: llc -mtriple=x86_64-linux-gnu -mattr=+sse4.2 -run-pass=legalizer %s -o - | FileCheck %s +# RUN: llc -mtriple=x86_64-linux-gnu -mattr=+popcnt -run-pass=legalizer %s -o - | FileCheck %s # test popcount for s16, s32, and s64 diff --git a/llvm/test/CodeGen/X86/GlobalISel/legalize-leading-zeros.mir b/llvm/test/CodeGen/X86/GlobalISel/legalize-leading-zeros.mir index 7e62848..cb4bae1 100644 --- a/llvm/test/CodeGen/X86/GlobalISel/legalize-leading-zeros.mir +++ b/llvm/test/CodeGen/X86/GlobalISel/legalize-leading-zeros.mir @@ -1,5 +1,5 @@ # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 2 -# RUN: llc -mtriple=x86_64-linux-gnu -mattr=+sse4.2 -run-pass=legalizer %s -o - | FileCheck %s +# RUN: llc -mtriple=x86_64-linux-gnu -mattr=+lzcnt -run-pass=legalizer %s -o - | FileCheck %s # test count leading zeros for s16, s32, and s64 -- 2.7.4