maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/PowerPC/fmf-propagation.ll

495 lines
17 KiB
LLVM
Raw Normal View History

[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; REQUIRES: asserts
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; RUN: llc < %s -mtriple=powerpc64le -debug-only=isel -o /dev/null 2>&1 | FileCheck %s --check-prefix=FMFDEBUG
; RUN: llc < %s -mtriple=powerpc64le | FileCheck %s --check-prefix=FMF
[PowerPC] add more tests for FMF propagation; NFC llvm-svn: 332295
2018-05-15 05:17:49 +08:00
; RUN: llc < %s -mtriple=powerpc64le -debug-only=isel -o /dev/null 2>&1 -enable-unsafe-fp-math -enable-no-nans-fp-math | FileCheck %s --check-prefix=GLOBALDEBUG
; RUN: llc < %s -mtriple=powerpc64le -enable-unsafe-fp-math -enable-no-nans-fp-math | FileCheck %s --check-prefix=GLOBAL
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; Test FP transforms using instruction/node-level fast-math-flags.
; We're also checking debug output to verify that FMF is propagated to the newly created nodes.
; The run with the global unsafe param tests the pre-FMF behavior using regular instructions/nodes.
declare float @llvm.fma.f32(float, float, float)
declare float @llvm.sqrt.f32(float)
; X * Y + Z --> fma(X, Y, Z)
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_contract1:'
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMFDEBUG: fma contract {{t[0-9]+}}, {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_contract1:'
define float @fmul_fadd_contract1(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_contract1:
; FMF: # %bb.0:
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_contract1:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul float %x, %y
%add = fadd contract float %mul, %z
ret float %add
}
; This shouldn't change anything - the intermediate fmul result is now also flagged.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_contract2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fma contract {{t[0-9]+}}, {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_contract2:'
define float @fmul_fadd_contract2(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_contract2:
; FMF: # %bb.0:
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_contract2:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul contract float %x, %y
%add = fadd contract float %mul, %z
ret float %add
}
; Reassociation implies that FMA contraction is allowed.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_reassoc1:'
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMFDEBUG: fma reassoc {{t[0-9]+}}, {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_reassoc1:'
define float @fmul_fadd_reassoc1(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_reassoc1:
; FMF: # %bb.0:
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_reassoc1:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul float %x, %y
%add = fadd reassoc float %mul, %z
ret float %add
}
; This shouldn't change anything - the intermediate fmul result is now also flagged.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_reassoc2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fma reassoc {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_reassoc2:'
define float @fmul_fadd_reassoc2(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_reassoc2:
; FMF: # %bb.0:
Fast Math Flag mapping into SDNode Summary: Adding support for Fast flags in the SDNode to leverage fast math sub flag usage. Reviewers: spatel, arsenm, jbhateja, hfinkel, escha, qcolombet, echristo, wristow, javed.absar Reviewed By: spatel Subscribers: llvm-commits, rampitec, nhaehnle, tstellar, FarhanaAleen, nemanjai, javed.absar, jbhateja, hfinkel, wdng Differential Revision: https://reviews.llvm.org/D45710 llvm-svn: 331547
2018-05-05 02:48:20 +08:00
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_reassoc2:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul reassoc float %x, %y
%add = fadd reassoc float %mul, %z
ret float %add
}
; The fadd is now fully 'fast'. This implies that contraction is allowed.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_fast1:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fma nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}, {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_fast1:'
define float @fmul_fadd_fast1(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_fast1:
; FMF: # %bb.0:
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_fast1:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul fast float %x, %y
%add = fadd fast float %mul, %z
ret float %add
}
; This shouldn't change anything - the intermediate fmul result is now also flagged.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fadd_fast2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fma nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}, {{t[0-9]+}}, {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fadd_fast2:'
define float @fmul_fadd_fast2(float %x, float %y, float %z) {
; FMF-LABEL: fmul_fadd_fast2:
; FMF: # %bb.0:
; FMF-NEXT: xsmaddasp 3, 1, 2
; FMF-NEXT: fmr 1, 3
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fadd_fast2:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xsmaddasp 3, 1, 2
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%mul = fmul fast float %x, %y
%add = fadd fast float %mul, %z
ret float %add
}
; fma(X, 7.0, X * 42.0) --> X * 49.0
; This is the minimum FMF needed for this transform - the FMA allows reassociation.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_reassoc1:'
[PowerPC] Replace the Post RA List Scheduler with the Machine Scheduler We want to run the Machine Scheduler instead of the List Scheduler after RA. Checked with a performance run on a Power 9 machine with SPEC 2006 and while some benchmarks improved and others degraded the geomean was slightly improved with the Machine Scheduler. Differential Revision: https://reviews.llvm.org/D45265 llvm-svn: 336295
2018-07-05 02:54:25 +08:00
; FMFDEBUG: fmul reassoc {{t[0-9]+}},
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_reassoc1:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_reassoc1:'
Fast Math Flag mapping into SDNode Summary: Adding support for Fast flags in the SDNode to leverage fast math sub flag usage. Reviewers: spatel, arsenm, jbhateja, hfinkel, escha, qcolombet, echristo, wristow, javed.absar Reviewed By: spatel Subscribers: llvm-commits, rampitec, nhaehnle, tstellar, FarhanaAleen, nemanjai, javed.absar, jbhateja, hfinkel, wdng Differential Revision: https://reviews.llvm.org/D45710 llvm-svn: 331547
2018-05-05 02:48:20 +08:00
; GLOBALDEBUG: fmul reassoc {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_reassoc1:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @fmul_fma_reassoc1(float %x) {
; FMF-LABEL: fmul_fma_reassoc1:
; FMF: # %bb.0:
; FMF-NEXT: addis 3, 2, .LCPI6_0@toc@ha
; FMF-NEXT: addi 3, 3, .LCPI6_0@toc@l
; FMF-NEXT: lfsx 0, 0, 3
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmulsp 1, 1, 0
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fma_reassoc1:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: addis 3, 2, .LCPI6_0@toc@ha
; GLOBAL-NEXT: addi 3, 3, .LCPI6_0@toc@l
; GLOBAL-NEXT: lfsx 0, 0, 3
; GLOBAL-NEXT: xsmulsp 1, 1, 0
; GLOBAL-NEXT: blr
%mul = fmul float %x, 42.0
%fma = call reassoc float @llvm.fma.f32(float %x, float 7.0, float %mul)
ret float %fma
}
; This shouldn't change anything - the intermediate fmul result is now also flagged.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_reassoc2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fmul reassoc {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_reassoc2:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_reassoc2:'
Fast Math Flag mapping into SDNode Summary: Adding support for Fast flags in the SDNode to leverage fast math sub flag usage. Reviewers: spatel, arsenm, jbhateja, hfinkel, escha, qcolombet, echristo, wristow, javed.absar Reviewed By: spatel Subscribers: llvm-commits, rampitec, nhaehnle, tstellar, FarhanaAleen, nemanjai, javed.absar, jbhateja, hfinkel, wdng Differential Revision: https://reviews.llvm.org/D45710 llvm-svn: 331547
2018-05-05 02:48:20 +08:00
; GLOBALDEBUG: fmul reassoc {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_reassoc2:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @fmul_fma_reassoc2(float %x) {
; FMF-LABEL: fmul_fma_reassoc2:
; FMF: # %bb.0:
; FMF-NEXT: addis 3, 2, .LCPI7_0@toc@ha
; FMF-NEXT: addi 3, 3, .LCPI7_0@toc@l
; FMF-NEXT: lfsx 0, 0, 3
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmulsp 1, 1, 0
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fma_reassoc2:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: addis 3, 2, .LCPI7_0@toc@ha
; GLOBAL-NEXT: addi 3, 3, .LCPI7_0@toc@l
; GLOBAL-NEXT: lfsx 0, 0, 3
; GLOBAL-NEXT: xsmulsp 1, 1, 0
; GLOBAL-NEXT: blr
%mul = fmul reassoc float %x, 42.0
%fma = call reassoc float @llvm.fma.f32(float %x, float 7.0, float %mul)
ret float %fma
}
; The FMA is now fully 'fast'. This implies that reassociation is allowed.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_fast1:'
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMFDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_fast1:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_fast1:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; GLOBALDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_fast1:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @fmul_fma_fast1(float %x) {
; FMF-LABEL: fmul_fma_fast1:
; FMF: # %bb.0:
; FMF-NEXT: addis 3, 2, .LCPI8_0@toc@ha
; FMF-NEXT: addi 3, 3, .LCPI8_0@toc@l
; FMF-NEXT: lfsx 0, 0, 3
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmulsp 1, 1, 0
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fma_fast1:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: addis 3, 2, .LCPI8_0@toc@ha
; GLOBAL-NEXT: addi 3, 3, .LCPI8_0@toc@l
; GLOBAL-NEXT: lfsx 0, 0, 3
; GLOBAL-NEXT: xsmulsp 1, 1, 0
; GLOBAL-NEXT: blr
%mul = fmul float %x, 42.0
%fma = call fast float @llvm.fma.f32(float %x, float 7.0, float %mul)
ret float %fma
}
; This shouldn't change anything - the intermediate fmul result is now also flagged.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_fast2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; FMFDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_fast2:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fmul_fma_fast2:'
propagate fast math flags via IR on fma and sub expressions Summary: This change uses fmf subflags to guard fma optimizations as well as unsafe. These changes originated from D46483 and have been simplified via getNode. Reviewers: spatel, arsenm, hfinkel, javed.absar Reviewed By: spatel Subscribers: nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47388 llvm-svn: 334242
2018-06-08 06:49:09 +08:00
; GLOBALDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fmul_fma_fast2:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @fmul_fma_fast2(float %x) {
; FMF-LABEL: fmul_fma_fast2:
; FMF: # %bb.0:
; FMF-NEXT: addis 3, 2, .LCPI9_0@toc@ha
; FMF-NEXT: addi 3, 3, .LCPI9_0@toc@l
; FMF-NEXT: lfsx 0, 0, 3
Utilize new SDNode flag functionality to expand current support for fma Summary: This patch originated from D47388 and is a proper subset of the originating changes, containing only the fmf optimization guard extensions. Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar, rampitec, nhaehnle, nemanjai Reviewed By: rampitec, nhaehnle Subscribers: tpr, nemanjai, wdng Differential Revision: https://reviews.llvm.org/D47918 llvm-svn: 334876
2018-06-16 08:03:06 +08:00
; FMF-NEXT: xsmulsp 1, 1, 0
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fmul_fma_fast2:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: addis 3, 2, .LCPI9_0@toc@ha
; GLOBAL-NEXT: addi 3, 3, .LCPI9_0@toc@l
; GLOBAL-NEXT: lfsx 0, 0, 3
; GLOBAL-NEXT: xsmulsp 1, 1, 0
; GLOBAL-NEXT: blr
%mul = fmul fast float %x, 42.0
%fma = call fast float @llvm.fma.f32(float %x, float 7.0, float %mul)
ret float %fma
}
; Reduced precision for sqrt is allowed - should use estimate and NR iterations.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'sqrt_afn:'
guard fsqrt with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. It contains only context for fsqrt. Reviewers: spatel, hfinkel, arsenm Reviewed By: spatel Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai Differential Revision: https://reviews.llvm.org/D47749 llvm-svn: 334113
2018-06-07 02:47:55 +08:00
; FMFDEBUG: fmul afn {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'sqrt_afn:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'sqrt_afn:'
guard fsqrt with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. It contains only context for fsqrt. Reviewers: spatel, hfinkel, arsenm Reviewed By: spatel Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai Differential Revision: https://reviews.llvm.org/D47749 llvm-svn: 334113
2018-06-07 02:47:55 +08:00
; GLOBALDEBUG: fmul afn {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'sqrt_afn:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @sqrt_afn(float %x) {
; FMF-LABEL: sqrt_afn:
[PowerPC] Replace the Post RA List Scheduler with the Machine Scheduler We want to run the Machine Scheduler instead of the List Scheduler after RA. Checked with a performance run on a Power 9 machine with SPEC 2006 and while some benchmarks improved and others degraded the geomean was slightly improved with the Machine Scheduler. Differential Revision: https://reviews.llvm.org/D45265 llvm-svn: 336295
2018-07-05 02:54:25 +08:00
; FMF: # %bb.0:
; FMF-NEXT: xxlxor 0, 0, 0
; FMF-NEXT: fcmpu 0, 1, 0
; FMF-NEXT: beq 0, .LBB10_2
; FMF-NEXT: # %bb.1:
; FMF-NEXT: addis 3, 2, .LCPI10_0@toc@ha
; FMF-NEXT: xsrsqrtesp 3, 1
; FMF-NEXT: addi 3, 3, .LCPI10_0@toc@l
; FMF-NEXT: lfsx 0, 0, 3
; FMF-NEXT: xsmulsp 2, 1, 0
; FMF-NEXT: xsmulsp 4, 3, 3
; FMF-NEXT: xssubsp 2, 2, 1
; FMF-NEXT: xsmulsp 2, 2, 4
; FMF-NEXT: xssubsp 0, 0, 2
; FMF-NEXT: xsmulsp 0, 3, 0
; FMF-NEXT: xsmulsp 0, 0, 1
; FMF-NEXT: .LBB10_2:
; FMF-NEXT: fmr 1, 0
; FMF-NEXT: blr
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
;
; GLOBAL-LABEL: sqrt_afn:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xxlxor 0, 0, 0
; GLOBAL-NEXT: fcmpu 0, 1, 0
; GLOBAL-NEXT: beq 0, .LBB10_2
; GLOBAL-NEXT: # %bb.1:
; GLOBAL-NEXT: xsrsqrtesp 2, 1
; GLOBAL-NEXT: fneg 0, 1
[PowerPC] Replace the Post RA List Scheduler with the Machine Scheduler We want to run the Machine Scheduler instead of the List Scheduler after RA. Checked with a performance run on a Power 9 machine with SPEC 2006 and while some benchmarks improved and others degraded the geomean was slightly improved with the Machine Scheduler. Differential Revision: https://reviews.llvm.org/D45265 llvm-svn: 336295
2018-07-05 02:54:25 +08:00
; GLOBAL-NEXT: addis 3, 2, .LCPI10_0@toc@ha
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; GLOBAL-NEXT: fmr 4, 1
; GLOBAL-NEXT: addi 3, 3, .LCPI10_0@toc@l
; GLOBAL-NEXT: lfsx 3, 0, 3
; GLOBAL-NEXT: xsmaddasp 4, 0, 3
; GLOBAL-NEXT: xsmulsp 0, 2, 2
; GLOBAL-NEXT: xsmaddasp 3, 4, 0
; GLOBAL-NEXT: xsmulsp 0, 2, 3
; GLOBAL-NEXT: xsmulsp 0, 0, 1
; GLOBAL-NEXT: .LBB10_2:
; GLOBAL-NEXT: fmr 1, 0
; GLOBAL-NEXT: blr
%rt = call afn float @llvm.sqrt.f32(float %x)
ret float %rt
}
; The call is now fully 'fast'. This implies that approximation is allowed.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'sqrt_fast:'
guard fsqrt with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. It contains only context for fsqrt. Reviewers: spatel, hfinkel, arsenm Reviewed By: spatel Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai Differential Revision: https://reviews.llvm.org/D47749 llvm-svn: 334113
2018-06-07 02:47:55 +08:00
; FMFDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'sqrt_fast:'
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'sqrt_fast:'
guard fsqrt with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. It contains only context for fsqrt. Reviewers: spatel, hfinkel, arsenm Reviewed By: spatel Subscribers: hfinkel, wdng, andrew.w.kaylor, wristow, efriedma, nemanjai Differential Revision: https://reviews.llvm.org/D47749 llvm-svn: 334113
2018-06-07 02:47:55 +08:00
; GLOBALDEBUG: fmul nnan ninf nsz arcp contract afn reassoc {{t[0-9]+}}
[PowerPC] add more FMF debug output; NFC We can't see all of the problems currently unless we look at debug output when the global 'unsafe' is on. It's a mess. This is another attempt to make sure that D45710 is not making changes unintentionally. llvm-svn: 331476
2018-05-04 02:49:35 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'sqrt_fast:'
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
define float @sqrt_fast(float %x) {
; FMF-LABEL: sqrt_fast:
[PowerPC] Replace the Post RA List Scheduler with the Machine Scheduler We want to run the Machine Scheduler instead of the List Scheduler after RA. Checked with a performance run on a Power 9 machine with SPEC 2006 and while some benchmarks improved and others degraded the geomean was slightly improved with the Machine Scheduler. Differential Revision: https://reviews.llvm.org/D45265 llvm-svn: 336295
2018-07-05 02:54:25 +08:00
; FMF: # %bb.0:
; FMF-NEXT: xxlxor 0, 0, 0
; FMF-NEXT: fcmpu 0, 1, 0
; FMF-NEXT: beq 0, .LBB11_2
; FMF-NEXT: # %bb.1:
; FMF-NEXT: xsrsqrtesp 2, 1
; FMF-NEXT: fneg 0, 1
; FMF-NEXT: addis 3, 2, .LCPI11_0@toc@ha
; FMF-NEXT: fmr 4, 1
; FMF-NEXT: addi 3, 3, .LCPI11_0@toc@l
; FMF-NEXT: lfsx 3, 0, 3
; FMF-NEXT: xsmaddasp 4, 0, 3
; FMF-NEXT: xsmulsp 0, 2, 2
; FMF-NEXT: xsmaddasp 3, 4, 0
; FMF-NEXT: xsmulsp 0, 2, 3
; FMF-NEXT: xsmulsp 0, 0, 1
; FMF-NEXT: .LBB11_2:
; FMF-NEXT: fmr 1, 0
; FMF-NEXT: blr
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
;
; GLOBAL-LABEL: sqrt_fast:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xxlxor 0, 0, 0
; GLOBAL-NEXT: fcmpu 0, 1, 0
; GLOBAL-NEXT: beq 0, .LBB11_2
; GLOBAL-NEXT: # %bb.1:
; GLOBAL-NEXT: xsrsqrtesp 2, 1
; GLOBAL-NEXT: fneg 0, 1
[PowerPC] Replace the Post RA List Scheduler with the Machine Scheduler We want to run the Machine Scheduler instead of the List Scheduler after RA. Checked with a performance run on a Power 9 machine with SPEC 2006 and while some benchmarks improved and others degraded the geomean was slightly improved with the Machine Scheduler. Differential Revision: https://reviews.llvm.org/D45265 llvm-svn: 336295
2018-07-05 02:54:25 +08:00
; GLOBAL-NEXT: addis 3, 2, .LCPI11_0@toc@ha
[PowerPC] add tests for FMF propagation; NFC I'm choosing PPC out of convenience because it does all of the transforms of interest in these tests by default. There are multiple FMF problems shown in the current checks. D45710 is proposing to fix part of that. llvm-svn: 331471
2018-05-04 01:41:37 +08:00
; GLOBAL-NEXT: fmr 4, 1
; GLOBAL-NEXT: addi 3, 3, .LCPI11_0@toc@l
; GLOBAL-NEXT: lfsx 3, 0, 3
; GLOBAL-NEXT: xsmaddasp 4, 0, 3
; GLOBAL-NEXT: xsmulsp 0, 2, 2
; GLOBAL-NEXT: xsmaddasp 3, 4, 0
; GLOBAL-NEXT: xsmulsp 0, 2, 3
; GLOBAL-NEXT: xsmulsp 0, 0, 1
; GLOBAL-NEXT: .LBB11_2:
; GLOBAL-NEXT: fmr 1, 0
; GLOBAL-NEXT: blr
%rt = call fast float @llvm.sqrt.f32(float %x)
ret float %rt
}
[PowerPC] add more tests for FMF propagation; NFC llvm-svn: 332295
2018-05-15 05:17:49 +08:00
; fcmp can have fast-math-flags.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fcmp_nnan:'
; FMFDEBUG: select_cc {{t[0-9]+}}
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fcmp_nnan:'
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fcmp_nnan:'
; GLOBALDEBUG: select_cc {{t[0-9]+}}
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fcmp_nnan:'
define double @fcmp_nnan(double %a, double %y, double %z) {
; FMF-LABEL: fcmp_nnan:
; FMF: # %bb.0:
; FMF-NEXT: xxlxor 0, 0, 0
[DAG] propagate FMF for all FPMathOperators This is a simple hack based on what's proposed in D37686, but we can extend it if needed in follow-ups. It gets us most of the FMF functionality that we want without adding any state bits to the flags. It also intentionally leaves out non-FMF flags (nsw, etc) to minimize the patch. It should provide a superset of the functionality from D46563 - the extra tests show propagation and codegen diffs for fcmp, vecreduce, and FP libcalls. The PPC log2() test shows the limits of this most basic approach - we only applied 'afn' to the last node created for the call. AFAIK, there aren't any libcall optimizations based on the flags currently, so that shouldn't make any difference. Differential Revision: https://reviews.llvm.org/D46854 llvm-svn: 332358
2018-05-15 22:16:24 +08:00
; FMF-NEXT: xscmpudp 0, 1, 0
; FMF-NEXT: blt 0, .LBB12_2
[PowerPC] add more tests for FMF propagation; NFC llvm-svn: 332295
2018-05-15 05:17:49 +08:00
; FMF-NEXT: # %bb.1:
; FMF-NEXT: fmr 3, 2
; FMF-NEXT: .LBB12_2:
; FMF-NEXT: fmr 1, 3
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fcmp_nnan:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: xxlxor 0, 0, 0
; GLOBAL-NEXT: xscmpudp 0, 1, 0
; GLOBAL-NEXT: blt 0, .LBB12_2
; GLOBAL-NEXT: # %bb.1:
; GLOBAL-NEXT: fmr 3, 2
; GLOBAL-NEXT: .LBB12_2:
; GLOBAL-NEXT: fmr 1, 3
; GLOBAL-NEXT: blr
%cmp = fcmp nnan ult double %a, 0.0
%z.y = select i1 %cmp, double %z, double %y
ret double %z.y
}
; FP library calls can have fast-math-flags.
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'log2_approx:'
; FMFDEBUG: ch,glue = PPCISD::CALL_NOP t11, TargetGlobalAddress:i64<double (double)* @log2>
; FMFDEBUG: ch,glue = callseq_end t15, TargetConstant:i64<32>, TargetConstant:i64<0>, t15:1
[DAG] propagate FMF for all FPMathOperators This is a simple hack based on what's proposed in D37686, but we can extend it if needed in follow-ups. It gets us most of the FMF functionality that we want without adding any state bits to the flags. It also intentionally leaves out non-FMF flags (nsw, etc) to minimize the patch. It should provide a superset of the functionality from D46563 - the extra tests show propagation and codegen diffs for fcmp, vecreduce, and FP libcalls. The PPC log2() test shows the limits of this most basic approach - we only applied 'afn' to the last node created for the call. AFAIK, there aren't any libcall optimizations based on the flags currently, so that shouldn't make any difference. Differential Revision: https://reviews.llvm.org/D46854 llvm-svn: 332358
2018-05-15 22:16:24 +08:00
; FMFDEBUG: f64,ch,glue = CopyFromReg afn t16, Register:f64 $f1, t16:1
[PowerPC] add more tests for FMF propagation; NFC llvm-svn: 332295
2018-05-15 05:17:49 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'log2_approx:'
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'log2_approx:'
; GLOBALDEBUG: ch,glue = PPCISD::CALL_NOP t11, TargetGlobalAddress:i64<double (double)* @log2>
; GLOBALDEBUG: ch,glue = callseq_end t15, TargetConstant:i64<32>, TargetConstant:i64<0>, t15:1
[DAG] propagate FMF for all FPMathOperators This is a simple hack based on what's proposed in D37686, but we can extend it if needed in follow-ups. It gets us most of the FMF functionality that we want without adding any state bits to the flags. It also intentionally leaves out non-FMF flags (nsw, etc) to minimize the patch. It should provide a superset of the functionality from D46563 - the extra tests show propagation and codegen diffs for fcmp, vecreduce, and FP libcalls. The PPC log2() test shows the limits of this most basic approach - we only applied 'afn' to the last node created for the call. AFAIK, there aren't any libcall optimizations based on the flags currently, so that shouldn't make any difference. Differential Revision: https://reviews.llvm.org/D46854 llvm-svn: 332358
2018-05-15 22:16:24 +08:00
; GLOBALDEBUG: f64,ch,glue = CopyFromReg afn t16, Register:f64 $f1, t16:1
[PowerPC] add more tests for FMF propagation; NFC llvm-svn: 332295
2018-05-15 05:17:49 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'log2_approx:'
declare double @log2(double)
define double @log2_approx(double %x) nounwind {
; FMF-LABEL: log2_approx:
; FMF: # %bb.0:
; FMF-NEXT: mflr 0
; FMF-NEXT: std 0, 16(1)
; FMF-NEXT: stdu 1, -32(1)
; FMF-NEXT: bl log2
; FMF-NEXT: nop
; FMF-NEXT: addi 1, 1, 32
; FMF-NEXT: ld 0, 16(1)
; FMF-NEXT: mtlr 0
; FMF-NEXT: blr
;
; GLOBAL-LABEL: log2_approx:
; GLOBAL: # %bb.0:
; GLOBAL-NEXT: mflr 0
; GLOBAL-NEXT: std 0, 16(1)
; GLOBAL-NEXT: stdu 1, -32(1)
; GLOBAL-NEXT: bl log2
; GLOBAL-NEXT: nop
; GLOBAL-NEXT: addi 1, 1, 32
; GLOBAL-NEXT: ld 0, 16(1)
; GLOBAL-NEXT: mtlr 0
; GLOBAL-NEXT: blr
%r = call afn double @log2(double %x)
ret double %r
}
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
; -(X - Y) --> (Y - X)
; FMFDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fneg_fsub_nozeros_1:'
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00
; FMFDEBUG: fsub nsz {{t[0-9]+}}, {{t[0-9]+}}
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
; FMFDEBUG: Type-legalized selection DAG: %bb.0 'fneg_fsub_nozeros_1:'
; GLOBALDEBUG-LABEL: Optimized lowered selection DAG: %bb.0 'fneg_fsub_nozeros_1:'
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00
; GLOBALDEBUG: fsub nsz {{t[0-9]+}}, {{t[0-9]+}}
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
; GLOBALDEBUG: Type-legalized selection DAG: %bb.0 'fneg_fsub_nozeros_1:'
define float @fneg_fsub_nozeros_1(float %x, float %y, float %z) {
; FMF-LABEL: fneg_fsub_nozeros_1:
; FMF: # %bb.0:
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00
; FMF-NEXT: xssubsp 1, 2, 1
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
; FMF-NEXT: blr
;
; GLOBAL-LABEL: fneg_fsub_nozeros_1:
; GLOBAL: # %bb.0:
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00
; GLOBAL-NEXT: xssubsp 1, 2, 1
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
; GLOBAL-NEXT: blr
%neg = fsub float %x, %y
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00
%add = fsub nsz float 0.0, %neg
NFC: adding baseline fneg case for fmf llvm-svn: 334035
2018-06-06 02:12:25 +08:00
ret float %add
}
guard fneg with fmf sub flags Summary: This change uses fmf subflags to guard optimizations as well as unsafe. These changes originated from D46483. Reviewers: spatel, hfinkel Reviewed By: spatel Subscribers: nemanjai Differential Revision: https://reviews.llvm.org/D47389 llvm-svn: 334037
2018-06-06 02:49:47 +08:00