[X86][FMA4] Prefer FMA4 to FMA

We currently output FMA instructions on targets which support both FMA4 + FMA (i.e. later Bulldozer CPUS bdver2/bdver3/bdver4).

This patch flips this so FMA4 is preferred; this is for several reasons:

1 - FMA4 is non-destructive reducing the need for mov instructions.
2 - Its more straighforward to commute and fold inputs (although the recent work on FMA has reduced this difference).
3 - All supported targets have FMA4 performance equal or better to FMA - Piledriver (bdver2) in particular has half the throughput when executing FMA instructions.

Its looks like no future AMD processor lines will support FMA4 after the Bulldozer series so we're not causing problems for later CPUs.

Differential Revision: http://reviews.llvm.org/D14997

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254339 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86Subtarget.h b/lib/Target/X86/X86Subtarget.h
index 425bc2482e9b8530a941b0b41b385625fc49a1df..eb0199aecbebd57bfa978a5505da8700f963b166 100644 (file)
--- a/lib/Target/X86/X86Subtarget.h
+++ b/lib/Target/X86/X86Subtarget.h
@@ -354,9 +354,10 @@ public:
   bool hasXSAVEC() const { return HasXSAVEC; }
   bool hasXSAVES() const { return HasXSAVES; }
   bool hasPCLMUL() const { return HasPCLMUL; }
-  bool hasFMA() const { return HasFMA; }
-  // FIXME: Favor FMA when both are enabled. Is this the right thing to do?
-  bool hasFMA4() const { return HasFMA4 && !HasFMA; }
+  // Prefer FMA4 to FMA - its better for commutation/memory folding and
+  // has equal or better performance on all supported targets.
+  bool hasFMA() const { return HasFMA && !HasFMA4; }
+  bool hasFMA4() const { return HasFMA4; }
   bool hasXOP() const { return HasXOP; }
   bool hasTBM() const { return HasTBM; }
   bool hasMOVBE() const { return HasMOVBE; }

diff --git a/test/CodeGen/X86/fma-commute-x86.ll b/test/CodeGen/X86/fma-commute-x86.ll
index 9e57b00bc0b49da2b80736f4afc122aa271c4fc5..9a368792133bbeb0edd8abb282f5c5cc465a439a 100644 (file)
--- a/test/CodeGen/X86/fma-commute-x86.ll
+++ b/test/CodeGen/X86/fma-commute-x86.ll
@@ -1,5 +1,5 @@
 ; RUN: llc < %s -mtriple=x86_64-pc-win32 -mcpu=core-avx2 | FileCheck %s
-; RUN: llc < %s -mtriple=x86_64-pc-win32 -mattr=+fma,+fma4 | FileCheck %s
+; RUN: llc < %s -mtriple=x86_64-pc-win32 -mattr=+fma | FileCheck %s
 ; RUN: llc < %s -mcpu=bdver2 -mtriple=x86_64-pc-win32 -mattr=-fma4 | FileCheck %s
 
 attributes #0 = { nounwind }

diff --git a/test/CodeGen/X86/fma_patterns.ll b/test/CodeGen/X86/fma_patterns.ll
index 942d799d97616b7b320fbac58bb44ccae3451104..e3295e45823c6ba334cf51bce33eb4e72cc4d583 100644 (file)
--- a/test/CodeGen/X86/fma_patterns.ll
+++ b/test/CodeGen/X86/fma_patterns.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma -fp-contract=fast | FileCheck %s --check-prefix=ALL --check-prefix=FMA
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4,+fma -fp-contract=fast | FileCheck %s --check-prefix=ALL --check-prefix=FMA
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4,+fma -fp-contract=fast | FileCheck %s --check-prefix=ALL --check-prefix=FMA4
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4 -fp-contract=fast | FileCheck %s --check-prefix=ALL --check-prefix=FMA4
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl -fp-contract=fast | FileCheck %s --check-prefix=ALL --check-prefix=AVX512
 

diff --git a/test/CodeGen/X86/fma_patterns_wide.ll b/test/CodeGen/X86/fma_patterns_wide.ll
index de77a27ad2b69afef1a8c92ed9215ceb4df9833d..f412c174fe376f9fee641533356efe39dec43943 100644 (file)
--- a/test/CodeGen/X86/fma_patterns_wide.ll
+++ b/test/CodeGen/X86/fma_patterns_wide.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma -fp-contract=fast | FileCheck %s --check-prefix=FMA
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4,+fma -fp-contract=fast | FileCheck %s --check-prefix=FMA
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4,+fma -fp-contract=fast | FileCheck %s --check-prefix=FMA4
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4 -fp-contract=fast | FileCheck %s --check-prefix=FMA4
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f -fp-contract=fast | FileCheck %s --check-prefix=AVX512