Skip to content

[CostModel][X86] merge fp comparison costs tests using -cost-kind=all #131874

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 19, 2025

Conversation

RKSimon
Copy link
Collaborator

@RKSimon RKSimon commented Mar 18, 2025

No description provided.

@llvmbot llvmbot added backend:X86 llvm:analysis Includes value tracking, cost tables and constant folding labels Mar 18, 2025
@llvmbot
Copy link
Member

llvmbot commented Mar 18, 2025

@llvm/pr-subscribers-backend-x86

@llvm/pr-subscribers-llvm-analysis

Author: Simon Pilgrim (RKSimon)

Changes

Patch is 442.92 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/131874.diff

4 Files Affected:

  • (removed) llvm/test/Analysis/CostModel/X86/fcmp-codesize.ll (-894)
  • (removed) llvm/test/Analysis/CostModel/X86/fcmp-latency.ll (-1102)
  • (removed) llvm/test/Analysis/CostModel/X86/fcmp-sizelatency.ll (-894)
  • (modified) llvm/test/Analysis/CostModel/X86/fcmp.ll (+893-893)
diff --git a/llvm/test/Analysis/CostModel/X86/fcmp-codesize.ll b/llvm/test/Analysis/CostModel/X86/fcmp-codesize.ll
deleted file mode 100644
index 1cb0f07b3f326..0000000000000
--- a/llvm/test/Analysis/CostModel/X86/fcmp-codesize.ll
+++ /dev/null
@@ -1,894 +0,0 @@
-; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+sse2 | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+sse3 | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+ssse3 | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+sse4.2 | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+avx | FileCheck %s --check-prefixes=AVX
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+avx2 | FileCheck %s --check-prefixes=AVX
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+avx512f | FileCheck %s --check-prefixes=AVX512
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX512
-;
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mcpu=slm | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mcpu=goldmont | FileCheck %s --check-prefixes=SSE
-; RUN: opt < %s -mtriple=x86_64-- -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=code-size -mcpu=btver2 | FileCheck %s --check-prefixes=AVX
-
-define i32 @cmp_float_oeq(i32 %arg) {
-; SSE-LABEL: 'cmp_float_oeq'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oeq float undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oeq <2 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oeq <4 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = fcmp oeq <8 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fcmp oeq <16 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oeq double undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oeq <2 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = fcmp oeq <4 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fcmp oeq <8 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F64 = fcmp oeq <16 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX-LABEL: 'cmp_float_oeq'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oeq float undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oeq <2 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oeq <4 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp oeq <8 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = fcmp oeq <16 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oeq double undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oeq <2 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp oeq <4 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = fcmp oeq <8 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F64 = fcmp oeq <16 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX512-LABEL: 'cmp_float_oeq'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oeq float undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oeq <2 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oeq <4 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp oeq <8 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fcmp oeq <16 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oeq double undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oeq <2 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp oeq <4 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fcmp oeq <8 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F64 = fcmp oeq <16 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-  %F32 = fcmp oeq float undef, undef
-  %V2F32 = fcmp oeq <2 x float> undef, undef
-  %V4F32 = fcmp oeq <4 x float> undef, undef
-  %V8F32 = fcmp oeq <8 x float> undef, undef
-  %V16F32 = fcmp oeq <16 x float> undef, undef
-
-  %F64 = fcmp oeq double undef, undef
-  %V2F64 = fcmp oeq <2 x double> undef, undef
-  %V4F64 = fcmp oeq <4 x double> undef, undef
-  %V8F64 = fcmp oeq <8 x double> undef, undef
-  %V16F64 = fcmp oeq <16 x double> undef, undef
-
-  ret i32 undef
-}
-
-define i32 @cmp_float_one(i32 %arg) {
-; SSE-LABEL: 'cmp_float_one'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp one float undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2F32 = fcmp one <2 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V4F32 = fcmp one <4 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8F32 = fcmp one <8 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V16F32 = fcmp one <16 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp one double undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2F64 = fcmp one <2 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4F64 = fcmp one <4 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8F64 = fcmp one <8 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16F64 = fcmp one <16 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX-LABEL: 'cmp_float_one'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp one float undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp one <2 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp one <4 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp one <8 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = fcmp one <16 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp one double undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp one <2 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp one <4 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = fcmp one <8 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F64 = fcmp one <16 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX512-LABEL: 'cmp_float_one'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp one float undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp one <2 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp one <4 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp one <8 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fcmp one <16 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp one double undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp one <2 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp one <4 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fcmp one <8 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F64 = fcmp one <16 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-  %F32 = fcmp one float undef, undef
-  %V2F32 = fcmp one <2 x float> undef, undef
-  %V4F32 = fcmp one <4 x float> undef, undef
-  %V8F32 = fcmp one <8 x float> undef, undef
-  %V16F32 = fcmp one <16 x float> undef, undef
-
-  %F64 = fcmp one double undef, undef
-  %V2F64 = fcmp one <2 x double> undef, undef
-  %V4F64 = fcmp one <4 x double> undef, undef
-  %V8F64 = fcmp one <8 x double> undef, undef
-  %V16F64 = fcmp one <16 x double> undef, undef
-
-  ret i32 undef
-}
-
-define i32 @cmp_float_ord(i32 %arg) {
-; SSE-LABEL: 'cmp_float_ord'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp ord float undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp ord <2 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp ord <4 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = fcmp ord <8 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fcmp ord <16 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp ord double undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp ord <2 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = fcmp ord <4 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fcmp ord <8 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F64 = fcmp ord <16 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX-LABEL: 'cmp_float_ord'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp ord float undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp ord <2 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp ord <4 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp ord <8 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = fcmp ord <16 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp ord double undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp ord <2 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp ord <4 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = fcmp ord <8 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F64 = fcmp ord <16 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX512-LABEL: 'cmp_float_ord'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp ord float undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp ord <2 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp ord <4 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp ord <8 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fcmp ord <16 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp ord double undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp ord <2 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp ord <4 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fcmp ord <8 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F64 = fcmp ord <16 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-  %F32 = fcmp ord float undef, undef
-  %V2F32 = fcmp ord <2 x float> undef, undef
-  %V4F32 = fcmp ord <4 x float> undef, undef
-  %V8F32 = fcmp ord <8 x float> undef, undef
-  %V16F32 = fcmp ord <16 x float> undef, undef
-
-  %F64 = fcmp ord double undef, undef
-  %V2F64 = fcmp ord <2 x double> undef, undef
-  %V4F64 = fcmp ord <4 x double> undef, undef
-  %V8F64 = fcmp ord <8 x double> undef, undef
-  %V16F64 = fcmp ord <16 x double> undef, undef
-
-  ret i32 undef
-}
-
-define i32 @cmp_float_oge(i32 %arg) {
-; SSE-LABEL: 'cmp_float_oge'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oge float undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oge <2 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oge <4 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = fcmp oge <8 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fcmp oge <16 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oge double undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oge <2 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = fcmp oge <4 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fcmp oge <8 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F64 = fcmp oge <16 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX-LABEL: 'cmp_float_oge'
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oge float undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oge <2 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oge <4 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp oge <8 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = fcmp oge <16 x float> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oge double undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oge <2 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp oge <4 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = fcmp oge <8 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F64 = fcmp oge <16 x double> undef, undef
-; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-; AVX512-LABEL: 'cmp_float_oge'
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp oge float undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp oge <2 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp oge <4 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F32 = fcmp oge <8 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16F32 = fcmp oge <16 x float> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp oge double undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp oge <2 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F64 = fcmp oge <4 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8F64 = fcmp oge <8 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F64 = fcmp oge <16 x double> undef, undef
-; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
-;
-  %F32 = fcmp oge float undef, undef
-  %V2F32 = fcmp oge <2 x float> undef, undef
-  %V4F32 = fcmp oge <4 x float> undef, undef
-  %V8F32 = fcmp oge <8 x float> undef, undef
-  %V16F32 = fcmp oge <16 x float> undef, undef
-
-  %F64 = fcmp oge double undef, undef
-  %V2F64 = fcmp oge <2 x double> undef, undef
-  %V4F64 = fcmp oge <4 x double> undef, undef
-  %V8F64 = fcmp oge <8 x double> undef, undef
-  %V16F64 = fcmp oge <16 x double> undef, undef
-
-  ret i32 undef
-}
-
-define i32 @cmp_float_ogt(i32 %arg) {
-; SSE-LABEL: 'cmp_float_ogt'
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F32 = fcmp ogt float undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F32 = fcmp ogt <2 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4F32 = fcmp ogt <4 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = fcmp ogt <8 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fcmp ogt <16 x float> undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %F64 = fcmp ogt double undef, undef
-; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2F64 = fcmp ogt <2 x double> undef, undef
-; SSE-NEXT:  Cost Model: Found a...
[truncated]

@RKSimon RKSimon force-pushed the x86-fcmp-costs-all branch from 3c33306 to c570058 Compare March 18, 2025 21:47
@RKSimon RKSimon merged commit 841d6c4 into llvm:main Mar 19, 2025
11 checks passed
@RKSimon RKSimon deleted the x86-fcmp-costs-all branch March 19, 2025 10:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backend:X86 llvm:analysis Includes value tracking, cost tables and constant folding
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants