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/X86/vec_ss_load_fold.ll

416 lines
16 KiB
LLVM
Raw Normal View History

[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
[X86] Teach selectScalarSSELoad to accept full 128-bit vector loads and the X86ISD::VZEXT_LOAD opcode. Disable peephole on some of the tests that no longer require it to properly fold scalar intrinsics. llvm-svn: 289424
2016-12-12 15:57:24 +08:00
; RUN: llc < %s -disable-peephole -mtriple=i686-apple-darwin9 -mattr=+sse,+sse2,+sse4.1 | FileCheck %s --check-prefix=X32
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin9 -mattr=+sse,+sse2,+sse4.1 | FileCheck %s --check-prefix=X64
; RUN: llc < %s -disable-peephole -mtriple=i686-apple-darwin9 -mattr=+avx | FileCheck %s --check-prefix=X32_AVX --check-prefix=X32_AVX1
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin9 -mattr=+avx | FileCheck %s --check-prefix=X64_AVX --check-prefix=X64_AVX1
; RUN: llc < %s -disable-peephole -mtriple=i686-apple-darwin9 -mattr=+avx512f | FileCheck %s --check-prefix=X32_AVX --check-prefix=X32_AVX512
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin9 -mattr=+avx512f | FileCheck %s --check-prefix=X64_AVX --check-prefix=X64_AVX512
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
Add nounwind. llvm-svn: 50837
2008-05-08 06:59:08 +08:00
define i16 @test1(float %f) nounwind {
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32-NEXT: addss LCPI0_0, %xmm0
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: mulss LCPI0_1, %xmm0
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32-NEXT: xorps %xmm1, %xmm1
; X32-NEXT: blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X32-NEXT: minss LCPI0_2, %xmm0
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: maxss %xmm1, %xmm0
; X32-NEXT: cvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X32-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: retl
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
;
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0:
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X64-NEXT: addss {{.*}}(%rip), %xmm0
; X64-NEXT: mulss {{.*}}(%rip), %xmm0
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: xorps %xmm1, %xmm1
; X64-NEXT: blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X64-NEXT: minss {{.*}}(%rip), %xmm0
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: maxss %xmm1, %xmm0
; X64-NEXT: cvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X64-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX1-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX1: ## %bb.0:
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX1-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32_AVX1-NEXT: vaddss LCPI0_0, %xmm0, %xmm0
; X32_AVX1-NEXT: vmulss LCPI0_1, %xmm0, %xmm0
[X86] Remove unnecessary (v4f32 (X86vzmovl (v4f32 (scalar_to_vector FR32X)))) patterns We had already disabled the pattern for SSE4.1 and SSE4.2. But it got re-enabled for AVX and AVX512. With SSE41 we rely on a separate (v4f32 (X86vzmovl VR128)) pattern to select blendps with a xorps to create zeroess. And a separate (v4f32 (scalar_to_vector FR32X)) to select a COPY_TO_REG_CLASS to move FR32 to VR128 The same thing can happen for AVX with vblendps and those separate patterns already exist. For AVX512, (v4f32 (X86vzmov VR128)) will select a VMOVSS instruction instead of VBLENDPS due to their not being a EVEX VBLENDPS. This is what we were getting out of the larger pattern anyway. So the larger pattern is unneeded for AVX512 too. For SSE1-SSSE3 we can rely on (v4f32 (X86vzmov VR128)) selecting a MOVSS similar to AVX512. Again this is what the larger pattern did too. So the only real change here is that AVX1/2 now properly outputs a VBLENDPS during isel instead of a VMOVSS to match SSE41. Most tests didn't notice because the two address instruction pass knows how to turn VMOVSS into VBLENDPS to get an independent destination register. llvm-svn: 312564
2017-09-06 03:09:02 +08:00
; X32_AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX1-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; X32_AVX1-NEXT: vminss LCPI0_2, %xmm0, %xmm0
; X32_AVX1-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X32_AVX1-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X32_AVX1-NEXT: ## kill: def $ax killed $ax killed $eax
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX1-NEXT: retl
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX1-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX1: ## %bb.0:
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X64_AVX1-NEXT: vaddss {{.*}}(%rip), %xmm0, %xmm0
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX1-NEXT: vmulss {{.*}}(%rip), %xmm0, %xmm0
[X86] Remove unnecessary (v4f32 (X86vzmovl (v4f32 (scalar_to_vector FR32X)))) patterns We had already disabled the pattern for SSE4.1 and SSE4.2. But it got re-enabled for AVX and AVX512. With SSE41 we rely on a separate (v4f32 (X86vzmovl VR128)) pattern to select blendps with a xorps to create zeroess. And a separate (v4f32 (scalar_to_vector FR32X)) to select a COPY_TO_REG_CLASS to move FR32 to VR128 The same thing can happen for AVX with vblendps and those separate patterns already exist. For AVX512, (v4f32 (X86vzmov VR128)) will select a VMOVSS instruction instead of VBLENDPS due to their not being a EVEX VBLENDPS. This is what we were getting out of the larger pattern anyway. So the larger pattern is unneeded for AVX512 too. For SSE1-SSSE3 we can rely on (v4f32 (X86vzmov VR128)) selecting a MOVSS similar to AVX512. Again this is what the larger pattern did too. So the only real change here is that AVX1/2 now properly outputs a VBLENDPS during isel instead of a VMOVSS to match SSE41. Most tests didn't notice because the two address instruction pass knows how to turn VMOVSS into VBLENDPS to get an independent destination register. llvm-svn: 312564
2017-09-06 03:09:02 +08:00
; X64_AVX1-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X64_AVX1-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX1-NEXT: vminss {{.*}}(%rip), %xmm0, %xmm0
; X64_AVX1-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X64_AVX1-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X64_AVX1-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX1-NEXT: retq
;
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX512-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX512: ## %bb.0:
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX512-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32_AVX512-NEXT: vaddss LCPI0_0, %xmm0, %xmm0
; X32_AVX512-NEXT: vmulss LCPI0_1, %xmm0, %xmm0
[X86] Remove unnecessary (v4f32 (X86vzmovl (v4f32 (scalar_to_vector FR32X)))) patterns We had already disabled the pattern for SSE4.1 and SSE4.2. But it got re-enabled for AVX and AVX512. With SSE41 we rely on a separate (v4f32 (X86vzmovl VR128)) pattern to select blendps with a xorps to create zeroess. And a separate (v4f32 (scalar_to_vector FR32X)) to select a COPY_TO_REG_CLASS to move FR32 to VR128 The same thing can happen for AVX with vblendps and those separate patterns already exist. For AVX512, (v4f32 (X86vzmov VR128)) will select a VMOVSS instruction instead of VBLENDPS due to their not being a EVEX VBLENDPS. This is what we were getting out of the larger pattern anyway. So the larger pattern is unneeded for AVX512 too. For SSE1-SSSE3 we can rely on (v4f32 (X86vzmov VR128)) selecting a MOVSS similar to AVX512. Again this is what the larger pattern did too. So the only real change here is that AVX1/2 now properly outputs a VBLENDPS during isel instead of a VMOVSS to match SSE41. Most tests didn't notice because the two address instruction pass knows how to turn VMOVSS into VBLENDPS to get an independent destination register. llvm-svn: 312564
2017-09-06 03:09:02 +08:00
; X32_AVX512-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86] Prefer blendi over movss/sd when avx512 is enabled unless optimizing for size. AVX512 doesn't have an immediate controlled blend instruction. But blend throughput is still better than movss/sd on SKX. This commit changes AVX512 to use the AVX blend instructions instead of MOVSS/MOVSD. This constrains the register allocation since it won't be able to use XMM16-31, but hopefully the increased throughput and reduced port 5 pressure makes up for that. llvm-svn: 337083
2018-07-14 10:05:08 +08:00
; X32_AVX512-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; X32_AVX512-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX512-NEXT: vminss LCPI0_2, %xmm0, %xmm0
; X32_AVX512-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X32_AVX512-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X32_AVX512-NEXT: ## kill: def $ax killed $ax killed $eax
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X32_AVX512-NEXT: retl
;
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX512-LABEL: test1:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX512: ## %bb.0:
[X86] Remove the scalar intrinsics for fadd/fsub/fdiv/fmul Summary: These intrinsics have been unused for clang for a while. This patch removes them. We auto upgrade them to extractelements, a scalar operation and then an insertelement. This matches the sequence used by clangs intrinsic file. Reviewers: zvi, delena, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D26660 llvm-svn: 287083
2016-11-16 13:24:10 +08:00
; X64_AVX512-NEXT: vaddss {{.*}}(%rip), %xmm0, %xmm0
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX512-NEXT: vmulss {{.*}}(%rip), %xmm0, %xmm0
[X86] Remove unnecessary (v4f32 (X86vzmovl (v4f32 (scalar_to_vector FR32X)))) patterns We had already disabled the pattern for SSE4.1 and SSE4.2. But it got re-enabled for AVX and AVX512. With SSE41 we rely on a separate (v4f32 (X86vzmovl VR128)) pattern to select blendps with a xorps to create zeroess. And a separate (v4f32 (scalar_to_vector FR32X)) to select a COPY_TO_REG_CLASS to move FR32 to VR128 The same thing can happen for AVX with vblendps and those separate patterns already exist. For AVX512, (v4f32 (X86vzmov VR128)) will select a VMOVSS instruction instead of VBLENDPS due to their not being a EVEX VBLENDPS. This is what we were getting out of the larger pattern anyway. So the larger pattern is unneeded for AVX512 too. For SSE1-SSSE3 we can rely on (v4f32 (X86vzmov VR128)) selecting a MOVSS similar to AVX512. Again this is what the larger pattern did too. So the only real change here is that AVX1/2 now properly outputs a VBLENDPS during isel instead of a VMOVSS to match SSE41. Most tests didn't notice because the two address instruction pass knows how to turn VMOVSS into VBLENDPS to get an independent destination register. llvm-svn: 312564
2017-09-06 03:09:02 +08:00
; X64_AVX512-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86] Prefer blendi over movss/sd when avx512 is enabled unless optimizing for size. AVX512 doesn't have an immediate controlled blend instruction. But blend throughput is still better than movss/sd on SKX. This commit changes AVX512 to use the AVX blend instructions instead of MOVSS/MOVSD. This constrains the register allocation since it won't be able to use XMM16-31, but hopefully the increased throughput and reduced port 5 pressure makes up for that. llvm-svn: 337083
2018-07-14 10:05:08 +08:00
; X64_AVX512-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
; X64_AVX512-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX512-NEXT: vminss {{.*}}(%rip), %xmm0, %xmm0
; X64_AVX512-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X64_AVX512-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X64_AVX512-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements. We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets llvm-svn: 283037
2016-10-01 22:26:11 +08:00
; X64_AVX512-NEXT: retq
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
%tmp = insertelement <4 x float> undef, float %f, i32 0 ; <<4 x float>> [#uses=1]
%tmp10 = insertelement <4 x float> %tmp, float 0.000000e+00, i32 1 ; <<4 x float>> [#uses=1]
%tmp11 = insertelement <4 x float> %tmp10, float 0.000000e+00, i32 2 ; <<4 x float>> [#uses=1]
%tmp12 = insertelement <4 x float> %tmp11, float 0.000000e+00, i32 3 ; <<4 x float>> [#uses=1]
%tmp28 = tail call <4 x float> @llvm.x86.sse.sub.ss( <4 x float> %tmp12, <4 x float> < float 1.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1]
%tmp37 = tail call <4 x float> @llvm.x86.sse.mul.ss( <4 x float> %tmp28, <4 x float> < float 5.000000e-01, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1]
%tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp37, <4 x float> < float 6.553500e+04, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1]
%tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> zeroinitializer ) ; <<4 x float>> [#uses=1]
%tmp.upgrd.1 = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1]
%tmp69 = trunc i32 %tmp.upgrd.1 to i16 ; <i16> [#uses=1]
ret i16 %tmp69
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
}
Add nounwind. llvm-svn: 50837
2008-05-08 06:59:08 +08:00
define i16 @test2(float %f) nounwind {
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-LABEL: test2:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32-NEXT: addss LCPI1_0, %xmm0
; X32-NEXT: mulss LCPI1_1, %xmm0
; X32-NEXT: minss LCPI1_2, %xmm0
; X32-NEXT: xorps %xmm1, %xmm1
; X32-NEXT: maxss %xmm1, %xmm0
; X32-NEXT: cvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X32-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: retl
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
;
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-LABEL: test2:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: addss {{.*}}(%rip), %xmm0
; X64-NEXT: mulss {{.*}}(%rip), %xmm0
; X64-NEXT: minss {{.*}}(%rip), %xmm0
; X64-NEXT: xorps %xmm1, %xmm1
; X64-NEXT: maxss %xmm1, %xmm0
; X64-NEXT: cvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X64-NEXT: ## kill: def $ax killed $ax killed $eax
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
; X32_AVX-LABEL: test2:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0:
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32_AVX-NEXT: vaddss LCPI1_0, %xmm0, %xmm0
; X32_AVX-NEXT: vmulss LCPI1_1, %xmm0, %xmm0
; X32_AVX-NEXT: vminss LCPI1_2, %xmm0, %xmm0
; X32_AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; X32_AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X32_AVX-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X32_AVX-NEXT: ## kill: def $ax killed $ax killed $eax
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: test2:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0:
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: vaddss {{.*}}(%rip), %xmm0, %xmm0
; X64_AVX-NEXT: vmulss {{.*}}(%rip), %xmm0, %xmm0
; X64_AVX-NEXT: vminss {{.*}}(%rip), %xmm0, %xmm0
; X64_AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; X64_AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X64_AVX-NEXT: vcvttss2si %xmm0, %eax
Followup on Proposal to move MIR physical register namespace to '$' sigil. Discussed here: http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html In preparation for adding support for named vregs we are changing the sigil for physical registers in MIR to '$' from '%'. This will prevent name clashes of named physical register with named vregs. llvm-svn: 323922
2018-02-01 06:04:26 +08:00
; X64_AVX-NEXT: ## kill: def $ax killed $ax killed $eax
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: retq
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
%tmp28 = fsub float %f, 1.000000e+00 ; <float> [#uses=1]
%tmp37 = fmul float %tmp28, 5.000000e-01 ; <float> [#uses=1]
%tmp375 = insertelement <4 x float> undef, float %tmp37, i32 0 ; <<4 x float>> [#uses=1]
%tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp375, <4 x float> < float 6.553500e+04, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1]
%tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> < float 0.000000e+00, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1]
%tmp = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1]
%tmp69 = trunc i32 %tmp to i16 ; <i16> [#uses=1]
ret i16 %tmp69
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
}
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>)
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
declare <4 x float> @llvm.x86.sse.mul.ss(<4 x float>, <4 x float>)
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
declare <4 x float> @llvm.x86.sse.min.ss(<4 x float>, <4 x float>)
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
declare <4 x float> @llvm.x86.sse.max.ss(<4 x float>, <4 x float>)
new testcase llvm-svn: 30893
2006-10-12 06:07:38 +08:00
Remove llvm-upgrade and update tests. llvm-svn: 47432
2008-02-21 15:42:26 +08:00
declare i32 @llvm.x86.sse.cvttss2si(<4 x float>)
fix rdar://7653908, a crash on a case where we would fold a load into a roundss intrinsic, producing a cyclic dag. The root cause of this is badness handling ComplexPattern nodes in the old dagisel that I noticed through inspection. Eliminate a copy of the of the code that handled ComplexPatterns by making EmitChildMatchCode call into EmitMatchCode. llvm-svn: 96408
2010-02-17 06:35:06 +08:00
declare <4 x float> @llvm.x86.sse41.round.ss(<4 x float>, <4 x float>, i32)
[X86][SSE] Regenerated scalar load folding tests llvm-svn: 270431
2016-05-23 20:53:09 +08:00
fix rdar://7653908, a crash on a case where we would fold a load into a roundss intrinsic, producing a cyclic dag. The root cause of this is badness handling ComplexPattern nodes in the old dagisel that I noticed through inspection. Eliminate a copy of the of the code that handled ComplexPatterns by making EmitChildMatchCode call into EmitMatchCode. llvm-svn: 96408
2010-02-17 06:35:06 +08:00
declare <4 x float> @f()
define <4 x float> @test3(<4 x float> %A, float *%b, i32 %C) nounwind {
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-LABEL: test3:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: roundss $4, (%eax), %xmm0
; X32-NEXT: retl
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
;
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-LABEL: test3:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: roundss $4, (%rdi), %xmm0
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
[X86] Teach EVEX->VEX pass to turn VRNDSCALE into VROUND when bits 7:4 of the immediate are 0 and the regular EVEX->VEX checks pass. Bits 7:4 control the scale part of the operation. If the scale is 0 the behavior is equivalent to VROUND. Fixes PR36246 llvm-svn: 324985
2018-02-13 12:19:26 +08:00
; X32_AVX-LABEL: test3:
; X32_AVX: ## %bb.0:
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32_AVX-NEXT: vroundss $4, (%eax), %xmm0, %xmm0
; X32_AVX-NEXT: retl
[X86] Use EVEX encoded VRNDSCALE instructions to implement the legacy round intrinsics. The VRNDSCALE instructions implement a superset of the (V)ROUND instructions. They are equivalent if the upper 4-bits of the immediate are 0. This patch lowers the legacy intrinsics to the VRNDSCALE ISD node and masks the upper bits of the immediate to 0. This allows us to take advantage of the larger register encoding space. We should maybe consider converting VRNDSCALE back to VROUND in the EVEX to VEX pass if the extended registers are not being used. I notice some load folding opportunities being missed for the VRNDSCALESS/SD instructions that I'll try to fix in future patches. llvm-svn: 318008
2017-11-13 10:03:00 +08:00
;
[X86] Teach EVEX->VEX pass to turn VRNDSCALE into VROUND when bits 7:4 of the immediate are 0 and the regular EVEX->VEX checks pass. Bits 7:4 control the scale part of the operation. If the scale is 0 the behavior is equivalent to VROUND. Fixes PR36246 llvm-svn: 324985
2018-02-13 12:19:26 +08:00
; X64_AVX-LABEL: test3:
; X64_AVX: ## %bb.0:
; X64_AVX-NEXT: vroundss $4, (%rdi), %xmm0, %xmm0
; X64_AVX-NEXT: retq
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%a = load float , float *%b
fix rdar://7653908, a crash on a case where we would fold a load into a roundss intrinsic, producing a cyclic dag. The root cause of this is badness handling ComplexPattern nodes in the old dagisel that I noticed through inspection. Eliminate a copy of the of the code that handled ComplexPatterns by making EmitChildMatchCode call into EmitMatchCode. llvm-svn: 96408
2010-02-17 06:35:06 +08:00
%B = insertelement <4 x float> undef, float %a, i32 0
%X = call <4 x float> @llvm.x86.sse41.round.ss(<4 x float> %A, <4 x float> %B, i32 4)
ret <4 x float> %X
}
define <4 x float> @test4(<4 x float> %A, float *%b, i32 %C) nounwind {
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-LABEL: test4:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: subl $28, %esp
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32-NEXT: movaps %xmm0, (%esp) ## 16-byte Spill
; X32-NEXT: calll _f
[X86] Remove some intrinsic instructions from hasPartialRegUpdate Summary: These intrinsic instructions are all selected from intrinsics that have well defined behavior for where the upper bits come from. It's not the same place as the lower bits. As you can see we were suppressing load folding for these instructions in some cases. In none of the cases was the separate load helping avoid a partial dependency on the destination register. So we should just go ahead and allow the load to be folded. Only foldMemoryOperand was suppressing folding for these. They all have patterns for folding sse_load_f32/f64 that aren't gated with OptForSize, but sse_load_f32/f64 doesn't allow 128-bit vector loads. It only allows scalar_to_vector and vzmovl of scalar loads to match. There's no reason we can't allow a 128-bit vector load to be narrowed so I would like to fix sse_load_f32/f64 to allow that. And if I do that it changes some of these same test cases to fold the load too. Reviewers: spatel, zvi, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D27611 llvm-svn: 289419
2016-12-12 13:07:17 +08:00
; X32-NEXT: roundss $4, (%esp), %xmm0 ## 16-byte Folded Reload
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: addl $28, %esp
; X32-NEXT: retl
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
;
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-LABEL: test4:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: subq $24, %rsp
; X64-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64-NEXT: movaps %xmm0, (%rsp) ## 16-byte Spill
; X64-NEXT: callq _f
[X86] Remove some intrinsic instructions from hasPartialRegUpdate Summary: These intrinsic instructions are all selected from intrinsics that have well defined behavior for where the upper bits come from. It's not the same place as the lower bits. As you can see we were suppressing load folding for these instructions in some cases. In none of the cases was the separate load helping avoid a partial dependency on the destination register. So we should just go ahead and allow the load to be folded. Only foldMemoryOperand was suppressing folding for these. They all have patterns for folding sse_load_f32/f64 that aren't gated with OptForSize, but sse_load_f32/f64 doesn't allow 128-bit vector loads. It only allows scalar_to_vector and vzmovl of scalar loads to match. There's no reason we can't allow a 128-bit vector load to be narrowed so I would like to fix sse_load_f32/f64 to allow that. And if I do that it changes some of these same test cases to fold the load too. Reviewers: spatel, zvi, RKSimon Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D27611 llvm-svn: 289419
2016-12-12 13:07:17 +08:00
; X64-NEXT: roundss $4, (%rsp), %xmm0 ## 16-byte Folded Reload
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: addq $24, %rsp
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
[X86] Add more vector instructions to the memory folding table using the autogenerated table as a guide. The test cahnge is because we now fold stack reload into RNDSCALE and RNDSCALE can be turned into ROUND by EVEX->VEX. llvm-svn: 334728
2018-06-14 23:40:31 +08:00
; X32_AVX-LABEL: test4:
; X32_AVX: ## %bb.0:
; X32_AVX-NEXT: subl $28, %esp
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32_AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X32_AVX-NEXT: vmovaps %xmm0, (%esp) ## 16-byte Spill
; X32_AVX-NEXT: calll _f
; X32_AVX-NEXT: vroundss $4, (%esp), %xmm0, %xmm0 ## 16-byte Folded Reload
; X32_AVX-NEXT: addl $28, %esp
; X32_AVX-NEXT: retl
[X86] Use EVEX encoded VRNDSCALE instructions to implement the legacy round intrinsics. The VRNDSCALE instructions implement a superset of the (V)ROUND instructions. They are equivalent if the upper 4-bits of the immediate are 0. This patch lowers the legacy intrinsics to the VRNDSCALE ISD node and masks the upper bits of the immediate to 0. This allows us to take advantage of the larger register encoding space. We should maybe consider converting VRNDSCALE back to VROUND in the EVEX to VEX pass if the extended registers are not being used. I notice some load folding opportunities being missed for the VRNDSCALESS/SD instructions that I'll try to fix in future patches. llvm-svn: 318008
2017-11-13 10:03:00 +08:00
;
[X86] Add more vector instructions to the memory folding table using the autogenerated table as a guide. The test cahnge is because we now fold stack reload into RNDSCALE and RNDSCALE can be turned into ROUND by EVEX->VEX. llvm-svn: 334728
2018-06-14 23:40:31 +08:00
; X64_AVX-LABEL: test4:
; X64_AVX: ## %bb.0:
; X64_AVX-NEXT: subq $24, %rsp
; X64_AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64_AVX-NEXT: vmovaps %xmm0, (%rsp) ## 16-byte Spill
; X64_AVX-NEXT: callq _f
; X64_AVX-NEXT: vroundss $4, (%rsp), %xmm0, %xmm0 ## 16-byte Folded Reload
; X64_AVX-NEXT: addq $24, %rsp
; X64_AVX-NEXT: retq
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%a = load float , float *%b
fix rdar://7653908, a crash on a case where we would fold a load into a roundss intrinsic, producing a cyclic dag. The root cause of this is badness handling ComplexPattern nodes in the old dagisel that I noticed through inspection. Eliminate a copy of the of the code that handled ComplexPatterns by making EmitChildMatchCode call into EmitMatchCode. llvm-svn: 96408
2010-02-17 06:35:06 +08:00
%B = insertelement <4 x float> undef, float %a, i32 0
%q = call <4 x float> @f()
%X = call <4 x float> @llvm.x86.sse41.round.ss(<4 x float> %q, <4 x float> %B, i32 4)
ret <4 x float> %X
temporarily disable this. llvm-svn: 96717
2010-02-21 11:24:41 +08:00
}
X86: move Int_CVTSD2SSrr, Int_CVTSI2SSrr, Int_CVTSI2SDrr, Int_CVTSS2SDrr from OpTbl1 to OpTbl2 since they have 3 operands and the last operand can be changed to a memory operand. PR13576 llvm-svn: 161769
2012-08-14 02:29:41 +08:00
[X86][SSE] Regenerated scalar load folding tests llvm-svn: 270431
2016-05-23 20:53:09 +08:00
; PR13576
X86: move Int_CVTSD2SSrr, Int_CVTSI2SSrr, Int_CVTSI2SDrr, Int_CVTSS2SDrr from OpTbl1 to OpTbl2 since they have 3 operands and the last operand can be changed to a memory operand. PR13576 llvm-svn: 161769
2012-08-14 02:29:41 +08:00
define <2 x double> @test5() nounwind uwtable readnone noinline {
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-LABEL: test5:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0: ## %entry
[X86] Remove and autoupgrade cvtsi2ss/cvtsi2sd intrinsics to match what clang has used for a very long time. llvm-svn: 332186
2018-05-13 07:14:39 +08:00
; X32-NEXT: movaps {{.*#+}} xmm0 = [1.280000e+02,1.233210e+02]
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X32-NEXT: retl
VirtRegMap: Replace some identity copies with KILL instructions. An identity COPY like this: %AL = COPY %AL, %EAX<imp-def> has no semantic effect, but encodes liveness information: Further users of %EAX only depend on this instruction even though it does not define the full register. Replace the COPY with a KILL instruction in those cases to maintain this liveness information. (This reverts a small part of r238588 but this time adds a comment explaining why a KILL instruction is useful). llvm-svn: 274952
2016-07-09 08:19:07 +08:00
;
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-LABEL: test5:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0: ## %entry
[X86] Remove and autoupgrade cvtsi2ss/cvtsi2sd intrinsics to match what clang has used for a very long time. llvm-svn: 332186
2018-05-13 07:14:39 +08:00
; X64-NEXT: movaps {{.*#+}} xmm0 = [1.280000e+02,1.233210e+02]
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
; X32_AVX-LABEL: test5:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0: ## %entry
[X86] Remove and autoupgrade cvtsi2ss/cvtsi2sd intrinsics to match what clang has used for a very long time. llvm-svn: 332186
2018-05-13 07:14:39 +08:00
; X32_AVX-NEXT: vmovaps {{.*#+}} xmm0 = [1.280000e+02,1.233210e+02]
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: test5:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0: ## %entry
[X86] Remove and autoupgrade cvtsi2ss/cvtsi2sd intrinsics to match what clang has used for a very long time. llvm-svn: 332186
2018-05-13 07:14:39 +08:00
; X64_AVX-NEXT: vmovaps {{.*#+}} xmm0 = [1.280000e+02,1.233210e+02]
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: retq
X86: move Int_CVTSD2SSrr, Int_CVTSI2SSrr, Int_CVTSI2SDrr, Int_CVTSS2SDrr from OpTbl1 to OpTbl2 since they have 3 operands and the last operand can be changed to a memory operand. PR13576 llvm-svn: 161769
2012-08-14 02:29:41 +08:00
entry:
[X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets llvm-svn: 279630
2016-08-24 23:07:11 +08:00
%0 = tail call <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double> <double 4.569870e+02, double 1.233210e+02>, i32 128) nounwind readnone
X86: move Int_CVTSD2SSrr, Int_CVTSI2SSrr, Int_CVTSI2SDrr, Int_CVTSS2SDrr from OpTbl1 to OpTbl2 since they have 3 operands and the last operand can be changed to a memory operand. PR13576 llvm-svn: 161769
2012-08-14 02:29:41 +08:00
ret <2 x double> %0
}
declare <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double>, i32) nounwind readnone
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
define <4 x float> @minss_fold(float* %x, <4 x float> %y) {
; X32-LABEL: minss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0: ## %entry
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: minss (%eax), %xmm0
; X32-NEXT: retl
;
; X64-LABEL: minss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0: ## %entry
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X64-NEXT: minss (%rdi), %xmm0
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
; X32_AVX-LABEL: minss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32_AVX-NEXT: vminss (%eax), %xmm0, %xmm0
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: minss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: vminss (%rdi), %xmm0, %xmm0
; X64_AVX-NEXT: retq
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
entry:
%0 = load float, float* %x, align 1
%vecinit.i = insertelement <4 x float> undef, float %0, i32 0
%vecinit2.i = insertelement <4 x float> %vecinit.i, float 0.000000e+00, i32 1
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float 0.000000e+00, i32 2
%vecinit4.i = insertelement <4 x float> %vecinit3.i, float 0.000000e+00, i32 3
%1 = tail call <4 x float> @llvm.x86.sse.min.ss(<4 x float> %y, <4 x float> %vecinit4.i)
ret <4 x float> %1
}
define <4 x float> @maxss_fold(float* %x, <4 x float> %y) {
; X32-LABEL: maxss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0: ## %entry
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: maxss (%eax), %xmm0
; X32-NEXT: retl
;
; X64-LABEL: maxss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0: ## %entry
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
; X64-NEXT: maxss (%rdi), %xmm0
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
; X32_AVX-LABEL: maxss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32_AVX-NEXT: vmaxss (%eax), %xmm0, %xmm0
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: maxss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: vmaxss (%rdi), %xmm0, %xmm0
; X64_AVX-NEXT: retq
[X86][SSE] Add MINSD/MAXSD/MINSS/MAXSS intrinsic scalar load folding support These are no different in load behaviour to the existing ADD/SUB/MUL/DIV scalar ops but were missing from isNonFoldablePartialRegisterLoad llvm-svn: 279652
2016-08-25 02:40:53 +08:00
entry:
%0 = load float, float* %x, align 1
%vecinit.i = insertelement <4 x float> undef, float %0, i32 0
%vecinit2.i = insertelement <4 x float> %vecinit.i, float 0.000000e+00, i32 1
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float 0.000000e+00, i32 2
%vecinit4.i = insertelement <4 x float> %vecinit3.i, float 0.000000e+00, i32 3
%1 = tail call <4 x float> @llvm.x86.sse.max.ss(<4 x float> %y, <4 x float> %vecinit4.i)
ret <4 x float> %1
}
[X86][SSE] Add CMPSS/CMPSD intrinsic scalar load folding support. llvm-svn: 279806
2016-08-26 15:08:00 +08:00
define <4 x float> @cmpss_fold(float* %x, <4 x float> %y) {
; X32-LABEL: cmpss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0: ## %entry
[X86][SSE] Add CMPSS/CMPSD intrinsic scalar load folding support. llvm-svn: 279806
2016-08-26 15:08:00 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: cmpeqss (%eax), %xmm0
; X32-NEXT: retl
;
; X64-LABEL: cmpss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0: ## %entry
[X86][SSE] Add CMPSS/CMPSD intrinsic scalar load folding support. llvm-svn: 279806
2016-08-26 15:08:00 +08:00
; X64-NEXT: cmpeqss (%rdi), %xmm0
; X64-NEXT: retq
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
;
; X32_AVX-LABEL: cmpss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32_AVX-NEXT: vcmpeqss (%eax), %xmm0, %xmm0
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: cmpss_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0: ## %entry
[X86] Add AVX and AVX512 command lines to the vec_ss_load_fold test. llvm-svn: 280645
2016-09-05 10:20:53 +08:00
; X64_AVX-NEXT: vcmpeqss (%rdi), %xmm0, %xmm0
; X64_AVX-NEXT: retq
[X86][SSE] Add CMPSS/CMPSD intrinsic scalar load folding support. llvm-svn: 279806
2016-08-26 15:08:00 +08:00
entry:
%0 = load float, float* %x, align 1
%vecinit.i = insertelement <4 x float> undef, float %0, i32 0
%vecinit2.i = insertelement <4 x float> %vecinit.i, float 0.000000e+00, i32 1
%vecinit3.i = insertelement <4 x float> %vecinit2.i, float 0.000000e+00, i32 2
%vecinit4.i = insertelement <4 x float> %vecinit3.i, float 0.000000e+00, i32 3
%1 = tail call <4 x float> @llvm.x86.sse.cmp.ss(<4 x float> %y, <4 x float> %vecinit4.i, i8 0)
ret <4 x float> %1
}
declare <4 x float> @llvm.x86.sse.cmp.ss(<4 x float>, <4 x float>, i8) nounwind readnone
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
define <4 x float> @double_fold(float* %x, <4 x float> %y) {
; X32-LABEL: double_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32: ## %bb.0: ## %entry
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[X86] Add a hasOneUse check to selectScalarSSELoad to keep the same load from being folded multiple times. Summary: When selectScalarSSELoad is looking for a scalar_to_vector of a scalar load, it makes sure the load is only used by the scalar_to_vector. But it doesn't make sure the scalar_to_vector is only used once. This can cause the same load to be folded multiple times. This can be bad for performance. This also causes the chain output to be duplicated, but not connected to anything so chain dependencies will not be satisfied. Reviewers: RKSimon, zvi, delena, spatel Subscribers: andreadb, llvm-commits Differential Revision: https://reviews.llvm.org/D26790 llvm-svn: 287983
2016-11-27 01:29:25 +08:00
; X32-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X32-NEXT: movaps %xmm0, %xmm2
; X32-NEXT: minss %xmm1, %xmm2
; X32-NEXT: maxss %xmm1, %xmm0
; X32-NEXT: addps %xmm2, %xmm0
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X32-NEXT: retl
;
; X64-LABEL: double_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64: ## %bb.0: ## %entry
[X86] Add a hasOneUse check to selectScalarSSELoad to keep the same load from being folded multiple times. Summary: When selectScalarSSELoad is looking for a scalar_to_vector of a scalar load, it makes sure the load is only used by the scalar_to_vector. But it doesn't make sure the scalar_to_vector is only used once. This can cause the same load to be folded multiple times. This can be bad for performance. This also causes the chain output to be duplicated, but not connected to anything so chain dependencies will not be satisfied. Reviewers: RKSimon, zvi, delena, spatel Subscribers: andreadb, llvm-commits Differential Revision: https://reviews.llvm.org/D26790 llvm-svn: 287983
2016-11-27 01:29:25 +08:00
; X64-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X64-NEXT: movaps %xmm0, %xmm2
; X64-NEXT: minss %xmm1, %xmm2
; X64-NEXT: maxss %xmm1, %xmm0
; X64-NEXT: addps %xmm2, %xmm0
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X64-NEXT: retq
;
; X32_AVX-LABEL: double_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X32_AVX: ## %bb.0: ## %entry
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X32_AVX-NEXT: movl {{[0-9]+}}(%esp), %eax
[X86] Add a hasOneUse check to selectScalarSSELoad to keep the same load from being folded multiple times. Summary: When selectScalarSSELoad is looking for a scalar_to_vector of a scalar load, it makes sure the load is only used by the scalar_to_vector. But it doesn't make sure the scalar_to_vector is only used once. This can cause the same load to be folded multiple times. This can be bad for performance. This also causes the chain output to be duplicated, but not connected to anything so chain dependencies will not be satisfied. Reviewers: RKSimon, zvi, delena, spatel Subscribers: andreadb, llvm-commits Differential Revision: https://reviews.llvm.org/D26790 llvm-svn: 287983
2016-11-27 01:29:25 +08:00
; X32_AVX-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X32_AVX-NEXT: vminss %xmm1, %xmm0, %xmm2
; X32_AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X32_AVX-NEXT: vaddps %xmm0, %xmm2, %xmm0
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X32_AVX-NEXT: retl
;
; X64_AVX-LABEL: double_fold:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; X64_AVX: ## %bb.0: ## %entry
[X86] Add a hasOneUse check to selectScalarSSELoad to keep the same load from being folded multiple times. Summary: When selectScalarSSELoad is looking for a scalar_to_vector of a scalar load, it makes sure the load is only used by the scalar_to_vector. But it doesn't make sure the scalar_to_vector is only used once. This can cause the same load to be folded multiple times. This can be bad for performance. This also causes the chain output to be duplicated, but not connected to anything so chain dependencies will not be satisfied. Reviewers: RKSimon, zvi, delena, spatel Subscribers: andreadb, llvm-commits Differential Revision: https://reviews.llvm.org/D26790 llvm-svn: 287983
2016-11-27 01:29:25 +08:00
; X64_AVX-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X64_AVX-NEXT: vminss %xmm1, %xmm0, %xmm2
; X64_AVX-NEXT: vmaxss %xmm1, %xmm0, %xmm0
; X64_AVX-NEXT: vaddps %xmm0, %xmm2, %xmm0
[X86] Add a test case where, due to a bug in selectScalarSSELoad, we fold the same load twice. llvm-svn: 287210
2016-11-17 13:37:39 +08:00
; X64_AVX-NEXT: retq
entry:
%0 = load float, float* %x, align 1
%vecinit.i = insertelement <4 x float> undef, float %0, i32 0
%1 = tail call <4 x float> @llvm.x86.sse.min.ss(<4 x float> %y, <4 x float> %vecinit.i)
%2 = tail call <4 x float> @llvm.x86.sse.max.ss(<4 x float> %y, <4 x float> %vecinit.i)
%3 = fadd <4 x float> %1, %2
ret <4 x float> %3
}