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/insertelement-zero.ll

502 lines
17 KiB
LLVM
Raw Normal View History

[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
[X86][SSE] Dropped -mcpu from zero insertion tests Use triple and attribute only for consistency llvm-svn: 306661
2017-06-29 19:08:11 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE3
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ssse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
X86 Tests: Update more isel tests with FastVariableShuffle feature Summary: Added the FastVariableShuffle feature to cases that resembled processors for which this fearure is on. For AVX2 there are processors with and w/o this fearue enable. For AVX512 only KNL does enable this feature so cases which only have +avx512f were left without the FastVariableShuffle enabled. Reviewers: RKSimon, craig.topper Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D41851 llvm-svn: 322090
2018-01-10 00:26:06 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=ALL,AVX,AVX2,AVX2-SLOW
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-shuffle | FileCheck %s --check-prefixes=ALL,AVX,AVX2,AVX2-FAST
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
define <2 x double> @insert_v2f64_z1(<2 x double> %a) {
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-LABEL: insert_v2f64_z1:
[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
; SSE2: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-NEXT: xorpd %xmm1, %xmm1
; SSE2-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v2f64_z1:
[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
; SSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE3-NEXT: xorpd %xmm1, %xmm1
; SSE3-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v2f64_z1:
[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
; SSSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSSE3-NEXT: xorpd %xmm1, %xmm1
; SSSE3-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v2f64_z1:
[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
; SSE41: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm1, %xmm1
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
;
; AVX-LABEL: insert_v2f64_z1:
[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
; AVX: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: retq
%1 = insertelement <2 x double> %a, double 0.0, i32 0
ret <2 x double> %1
}
define <4 x double> @insert_v4f64_0zz3(<4 x double> %a) {
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-LABEL: insert_v4f64_0zz3:
[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
; SSE2: # %bb.0:
[X86][SSE] Improve shuffle combining with zero insertions Add support for handling shuffles with scalar_to_vector(0) llvm-svn: 292766
2017-01-23 06:21:44 +08:00
; SSE2-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-NEXT: xorpd %xmm2, %xmm2
; SSE2-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4f64_0zz3:
[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
; SSE3: # %bb.0:
[X86][SSE] Improve shuffle combining with zero insertions Add support for handling shuffles with scalar_to_vector(0) llvm-svn: 292766
2017-01-23 06:21:44 +08:00
; SSE3-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE3-NEXT: xorpd %xmm2, %xmm2
; SSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4f64_0zz3:
[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
; SSSE3: # %bb.0:
[X86][SSE] Improve shuffle combining with zero insertions Add support for handling shuffles with scalar_to_vector(0) llvm-svn: 292766
2017-01-23 06:21:44 +08:00
; SSSE3-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSSE3-NEXT: xorpd %xmm2, %xmm2
; SSSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v4f64_0zz3:
[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
; SSE41: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: movq {{.*#+}} xmm0 = xmm0[0],zero
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm2, %xmm2
; SSE41-NEXT: blendps {{.*#+}} xmm1 = xmm2[0,1],xmm1[2,3]
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
;
; AVX-LABEL: insert_v4f64_0zz3:
[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
; AVX: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: retq
%1 = insertelement <4 x double> %a, double 0.0, i32 1
%2 = insertelement <4 x double> %1, double 0.0, i32 2
ret <4 x double> %2
}
define <2 x i64> @insert_v2i64_z1(<2 x i64> %a) {
; SSE2-LABEL: insert_v2i64_z1:
[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
; SSE2: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: xorpd %xmm1, %xmm1
; SSE2-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v2i64_z1:
[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
; SSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: xorpd %xmm1, %xmm1
; SSE3-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v2i64_z1:
[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
; SSSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: xorpd %xmm1, %xmm1
; SSSE3-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v2i64_z1:
[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
; SSE41: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm1, %xmm1
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; AVX-LABEL: insert_v2i64_z1:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
; AVX-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <2 x i64> %a, i64 0, i32 0
ret <2 x i64> %1
}
define <4 x i64> @insert_v4i64_01z3(<4 x i64> %a) {
; SSE2-LABEL: insert_v4i64_01z3:
[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
; SSE2: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: xorpd %xmm2, %xmm2
; SSE2-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4i64_01z3:
[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
; SSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: xorpd %xmm2, %xmm2
; SSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4i64_01z3:
[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
; SSSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: xorpd %xmm2, %xmm2
; SSSE3-NEXT: movsd {{.*#+}} xmm1 = xmm2[0],xmm1[1]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v4i64_01z3:
[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
; SSE41: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm2, %xmm2
; SSE41-NEXT: blendps {{.*#+}} xmm1 = xmm2[0,1],xmm1[2,3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; AVX-LABEL: insert_v4i64_01z3:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
; AVX-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <4 x i64> %a, i64 0, i32 2
ret <4 x i64> %1
}
define <4 x float> @insert_v4f32_01z3(<4 x float> %a) {
; SSE2-LABEL: insert_v4f32_01z3:
[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
; SSE2: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: xorps %xmm1, %xmm1
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4f32_01z3:
[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
; SSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: xorps %xmm1, %xmm1
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4f32_01z3:
[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
; SSSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: xorps %xmm1, %xmm1
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v4f32_01z3:
[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
; SSE41: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: xorps %xmm1, %xmm1
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
; AVX-LABEL: insert_v4f32_01z3:
[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
; AVX: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; AVX-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: retq
%1 = insertelement <4 x float> %a, float 0.0, i32 2
ret <4 x float> %1
}
define <8 x float> @insert_v8f32_z12345z7(<8 x float> %a) {
; SSE2-LABEL: insert_v8f32_z12345z7:
[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
; SSE2: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: xorps %xmm2, %xmm2
; SSE2-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v8f32_z12345z7:
[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
; SSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: xorps %xmm2, %xmm2
; SSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
; SSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v8f32_z12345z7:
[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
; SSSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: xorps %xmm2, %xmm2
; SSSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
; SSSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v8f32_z12345z7:
[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
; SSE41: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: xorps %xmm2, %xmm2
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: blendps {{.*#+}} xmm1 = xmm1[0,1],xmm2[2],xmm1[3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
; AVX-LABEL: insert_v8f32_z12345z7:
[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
; AVX: # %bb.0:
[X86] SET0 to use XMM registers where possible PR26018 PR32862 Differential Revision: https://reviews.llvm.org/D35839 llvm-svn: 309298
2017-07-28 01:47:01 +08:00
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
[X86][SSE41] Combine vector blends with zero Part 2 of 2 This patch add support for combining target shuffles into blends-with-zero. Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261745
2016-02-24 23:14:21 +08:00
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5],ymm1[6],ymm0[7]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: retq
%1 = insertelement <8 x float> %a, float 0.0, i32 0
%2 = insertelement <8 x float> %1, float 0.0, i32 6
ret <8 x float> %2
}
define <4 x i32> @insert_v4i32_01z3(<4 x i32> %a) {
; SSE2-LABEL: insert_v4i32_01z3:
[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
; SSE2: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE2-NEXT: xorps %xmm1, %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v4i32_01z3:
[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
; SSE3: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE3-NEXT: xorps %xmm1, %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v4i32_01z3:
[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
; SSSE3: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSSE3-NEXT: xorps %xmm1, %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,2]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v4i32_01z3:
[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
; SSE41: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm1, %xmm1
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; AVX-LABEL: insert_v4i32_01z3:
; AVX: # %bb.0:
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3]
; AVX-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <4 x i32> %a, i32 0, i32 2
ret <4 x i32> %1
}
define <8 x i32> @insert_v8i32_z12345z7(<8 x i32> %a) {
; SSE2-LABEL: insert_v8i32_z12345z7:
[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
; SSE2: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: xorps %xmm2, %xmm2
; SSE2-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE2-NEXT: xorps %xmm2, %xmm2
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v8i32_z12345z7:
[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
; SSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: xorps %xmm2, %xmm2
; SSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE3-NEXT: xorps %xmm2, %xmm2
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v8i32_z12345z7:
[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
; SSSE3: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: xorps %xmm2, %xmm2
; SSSE3-NEXT: movss {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSSE3-NEXT: xorps %xmm2, %xmm2
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,0],xmm1[3,0]
; SSSE3-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm2[0,2]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v8i32_z12345z7:
[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
; SSE41: # %bb.0:
[X86][SSE] Add custom execution domain fixing for BLENDPD/BLENDPS/PBLENDD/PBLENDW (PR34873) Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW. Differential Revision: https://reviews.llvm.org/D42042 llvm-svn: 322524
2018-01-16 06:18:45 +08:00
; SSE41-NEXT: xorps %xmm2, %xmm2
; SSE41-NEXT: blendps {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3]
; SSE41-NEXT: blendps {{.*#+}} xmm1 = xmm1[0,1],xmm2[2],xmm1[3]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
[X86] Add VBLENDPS/VPBLENDD to the execution domain fixing tables. llvm-svn: 312449
2017-09-04 01:52:23 +08:00
; AVX-LABEL: insert_v8i32_z12345z7:
[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
; AVX: # %bb.0:
[X86] Add VBLENDPS/VPBLENDD to the execution domain fixing tables. llvm-svn: 312449
2017-09-04 01:52:23 +08:00
; AVX-NEXT: vxorps %xmm1, %xmm1, %xmm1
; AVX-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5],ymm1[6],ymm0[7]
; AVX-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <8 x i32> %a, i32 0, i32 0
%2 = insertelement <8 x i32> %1, i32 0, i32 6
ret <8 x i32> %2
}
define <8 x i16> @insert_v8i16_z12345z7(<8 x i16> %a) {
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-LABEL: insert_v8i16_z12345z7:
[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
; SSE2: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-NEXT: xorl %eax, %eax
; SSE2-NEXT: pinsrw $0, %eax, %xmm0
; SSE2-NEXT: pinsrw $6, %eax, %xmm0
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v8i16_z12345z7:
[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
; SSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE3-NEXT: xorl %eax, %eax
; SSE3-NEXT: pinsrw $0, %eax, %xmm0
; SSE3-NEXT: pinsrw $6, %eax, %xmm0
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v8i16_z12345z7:
[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
; SSSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSSE3-NEXT: xorl %eax, %eax
; SSSE3-NEXT: pinsrw $0, %eax, %xmm0
; SSSE3-NEXT: pinsrw $6, %eax, %xmm0
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v8i16_z12345z7:
[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
; SSE41: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: pxor %xmm1, %xmm1
[X86][SSE41] Combine vector blends with zero Part 2 of 2 This patch add support for combining target shuffles into blends-with-zero. Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261745
2016-02-24 23:14:21 +08:00
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3,4,5],xmm1[6],xmm0[7]
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
;
; AVX-LABEL: insert_v8i16_z12345z7:
[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
; AVX: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
[X86][SSE41] Combine vector blends with zero Part 2 of 2 This patch add support for combining target shuffles into blends-with-zero. Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261745
2016-02-24 23:14:21 +08:00
; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3,4,5],xmm1[6],xmm0[7]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX-NEXT: retq
%1 = insertelement <8 x i16> %a, i16 0, i32 0
%2 = insertelement <8 x i16> %1, i16 0, i32 6
ret <8 x i16> %2
}
Fixed test name. NFCI. llvm-svn: 305787
2017-06-20 18:24:06 +08:00
define <16 x i16> @insert_v16i16_z12345z789ABCDEz(<16 x i16> %a) {
; SSE2-LABEL: insert_v16i16_z12345z789ABCDEz:
[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
; SSE2: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE2-NEXT: xorl %eax, %eax
; SSE2-NEXT: pinsrw $0, %eax, %xmm0
; SSE2-NEXT: pinsrw $6, %eax, %xmm0
; SSE2-NEXT: pinsrw $7, %eax, %xmm1
; SSE2-NEXT: retq
;
Fixed test name. NFCI. llvm-svn: 305787
2017-06-20 18:24:06 +08:00
; SSE3-LABEL: insert_v16i16_z12345z789ABCDEz:
[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
; SSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE3-NEXT: xorl %eax, %eax
; SSE3-NEXT: pinsrw $0, %eax, %xmm0
; SSE3-NEXT: pinsrw $6, %eax, %xmm0
; SSE3-NEXT: pinsrw $7, %eax, %xmm1
; SSE3-NEXT: retq
;
Fixed test name. NFCI. llvm-svn: 305787
2017-06-20 18:24:06 +08:00
; SSSE3-LABEL: insert_v16i16_z12345z789ABCDEz:
[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
; SSSE3: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSSE3-NEXT: xorl %eax, %eax
; SSSE3-NEXT: pinsrw $0, %eax, %xmm0
; SSSE3-NEXT: pinsrw $6, %eax, %xmm0
; SSSE3-NEXT: pinsrw $7, %eax, %xmm1
; SSSE3-NEXT: retq
;
Fixed test name. NFCI. llvm-svn: 305787
2017-06-20 18:24:06 +08:00
; SSE41-LABEL: insert_v16i16_z12345z789ABCDEz:
[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
; SSE41: # %bb.0:
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: pxor %xmm2, %xmm2
[X86][SSE41] Combine vector blends with zero Part 2 of 2 This patch add support for combining target shuffles into blends-with-zero. Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261745
2016-02-24 23:14:21 +08:00
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm2[0],xmm0[1,2,3,4,5],xmm2[6],xmm0[7]
[X86][SSE41] Combine insertion of zero scalars into vector blends with zero Part 1 of 2 This patch attempts to replace the insertion of zero scalars with a vector blend with zero, avoiding the use of the integer insertion instructions (which are particularly slow on many targets). (Part 2 will add support for combining multiple blends-with-zero). Differential Revision: http://reviews.llvm.org/D17483 llvm-svn: 261743
2016-02-24 22:53:27 +08:00
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6],xmm2[7]
; SSE41-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
;
[X86][SSE] Relax 0/-1 vector element insertion to work for any vector with >=16bit elements Shuffle lowering/combining now does a good job for 256/512-bit vectors - we don't need to prevent this llvm-svn: 305801
2017-06-20 23:19:02 +08:00
; AVX-LABEL: insert_v16i16_z12345z789ABCDEz:
[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
; AVX: # %bb.0:
[X86][SSE] Relax 0/-1 vector element insertion to work for any vector with >=16bit elements Shuffle lowering/combining now does a good job for 256/512-bit vectors - we don't need to prevent this llvm-svn: 305801
2017-06-20 23:19:02 +08:00
; AVX-NEXT: vandps {{.*}}(%rip), %ymm0, %ymm0
; AVX-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <16 x i16> %a, i16 0, i32 0
%2 = insertelement <16 x i16> %1, i16 0, i32 6
%3 = insertelement <16 x i16> %2, i16 0, i32 15
ret <16 x i16> %3
}
Fix test name. NFCI. llvm-svn: 292763
2017-01-23 05:06:28 +08:00
define <16 x i8> @insert_v16i8_z123456789ABCDEz(<16 x i8> %a) {
; SSE2-LABEL: insert_v16i8_z123456789ABCDEz:
[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
; SSE2: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE2-NEXT: andps {{.*}}(%rip), %xmm0
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: retq
;
Fix test name. NFCI. llvm-svn: 292763
2017-01-23 05:06:28 +08:00
; SSE3-LABEL: insert_v16i8_z123456789ABCDEz:
[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
; SSE3: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE3-NEXT: andps {{.*}}(%rip), %xmm0
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: retq
;
Fix test name. NFCI. llvm-svn: 292763
2017-01-23 05:06:28 +08:00
; SSSE3-LABEL: insert_v16i8_z123456789ABCDEz:
[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
; SSSE3: # %bb.0:
[X86][SSE] Improve shuffle combining with zero insertions Add support for handling shuffles with scalar_to_vector(0) llvm-svn: 292766
2017-01-23 06:21:44 +08:00
; SSSE3-NEXT: andps {{.*}}(%rip), %xmm0
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: retq
;
Fix test name. NFCI. llvm-svn: 292763
2017-01-23 05:06:28 +08:00
; SSE41-LABEL: insert_v16i8_z123456789ABCDEz:
[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
; SSE41: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: xorl %eax, %eax
; SSE41-NEXT: pinsrb $0, %eax, %xmm0
; SSE41-NEXT: pinsrb $15, %eax, %xmm0
; SSE41-NEXT: retq
;
X86 Tests: Update more isel tests with FastVariableShuffle feature Summary: Added the FastVariableShuffle feature to cases that resembled processors for which this fearure is on. For AVX2 there are processors with and w/o this fearue enable. For AVX512 only KNL does enable this feature so cases which only have +avx512f were left without the FastVariableShuffle enabled. Reviewers: RKSimon, craig.topper Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D41851 llvm-svn: 322090
2018-01-10 00:26:06 +08:00
; AVX1-LABEL: insert_v16i8_z123456789ABCDEz:
; AVX1: # %bb.0:
; AVX1-NEXT: xorl %eax, %eax
; AVX1-NEXT: vpinsrb $0, %eax, %xmm0, %xmm0
; AVX1-NEXT: vpinsrb $15, %eax, %xmm0, %xmm0
; AVX1-NEXT: retq
;
; AVX2-SLOW-LABEL: insert_v16i8_z123456789ABCDEz:
; AVX2-SLOW: # %bb.0:
; AVX2-SLOW-NEXT: xorl %eax, %eax
; AVX2-SLOW-NEXT: vpinsrb $0, %eax, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpinsrb $15, %eax, %xmm0, %xmm0
; AVX2-SLOW-NEXT: retq
;
; AVX2-FAST-LABEL: insert_v16i8_z123456789ABCDEz:
; AVX2-FAST: # %bb.0:
; AVX2-FAST-NEXT: vandps {{.*}}(%rip), %xmm0, %xmm0
; AVX2-FAST-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <16 x i8> %a, i8 0, i32 0
%2 = insertelement <16 x i8> %1, i8 0, i32 15
ret <16 x i8> %2
}
define <32 x i8> @insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz(<32 x i8> %a) {
; SSE2-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
[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
; SSE2: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE2-NEXT: andps {{.*}}(%rip), %xmm0
[X86][SSE] Add missing X86ISD::ANDNP combines. llvm-svn: 292767
2017-01-23 06:45:23 +08:00
; SSE2-NEXT: andps {{.*}}(%rip), %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE2-NEXT: retq
;
; SSE3-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
[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
; SSE3: # %bb.0:
[X86][SSE] Lower scalar_to_vector(0) to zero vector Replaces an xor+movd/movq with an xorps which will be shorter in codesize, avoid an int-fpu transfer, allow modern cores to fast path the result during decode and helps other combines recognise an all-zero vector. The only reason I can think of that we'd want to keep scalar_to_vector in this case is to help recognise the upper elts are undef but this doesn't seem to be a problem. Differential Revision: https://reviews.llvm.org/D29097 llvm-svn: 293438
2017-01-30 02:13:37 +08:00
; SSE3-NEXT: andps {{.*}}(%rip), %xmm0
[X86][SSE] Add missing X86ISD::ANDNP combines. llvm-svn: 292767
2017-01-23 06:45:23 +08:00
; SSE3-NEXT: andps {{.*}}(%rip), %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE3-NEXT: retq
;
; SSSE3-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
[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
; SSSE3: # %bb.0:
[X86][SSE] Improve shuffle combining with zero insertions Add support for handling shuffles with scalar_to_vector(0) llvm-svn: 292766
2017-01-23 06:21:44 +08:00
; SSSE3-NEXT: andps {{.*}}(%rip), %xmm0
; SSSE3-NEXT: andps {{.*}}(%rip), %xmm1
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSSE3-NEXT: retq
;
; SSE41-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
[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
; SSE41: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: xorl %eax, %eax
; SSE41-NEXT: pinsrb $0, %eax, %xmm0
; SSE41-NEXT: pinsrb $15, %eax, %xmm0
[X86][SSE] Add support for combining PINSRB into a target shuffle. llvm-svn: 293637
2017-01-31 22:59:44 +08:00
; SSE41-NEXT: pxor %xmm2, %xmm2
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6],xmm2[7]
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; SSE41-NEXT: retq
;
; AVX1-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
[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
; AVX1: # %bb.0:
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX1-NEXT: xorl %eax, %eax
; AVX1-NEXT: vpinsrb $0, %eax, %xmm0, %xmm1
[X86] Move turning 256-bit INSERT_SUBVECTORS into BLENDI from legalize to DAG combine. On one test this seems to have given more chance for DAG combine to do other INSERT_SUBVECTOR/EXTRACT_SUBVECTOR combines before the BLENDI was created. Looks like we can still improve more by teaching DAG combine to optimize INSERT_SUBVECTOR/EXTRACT_SUBVECTOR with BLENDI. llvm-svn: 293944
2017-02-03 06:02:57 +08:00
; AVX1-NEXT: vpinsrb $15, %eax, %xmm1, %xmm1
[X86] Move code for using blendi for insert_subvector out to an isel pattern. This gives the DAG combiner more opportunity to optimize without needing to dig through the blend. llvm-svn: 294876
2017-02-12 06:57:12 +08:00
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
[X86][SSE] Add support for combining PINSRB into a target shuffle. llvm-svn: 293637
2017-01-31 22:59:44 +08:00
; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
[X86] Move code for using blendi for insert_subvector out to an isel pattern. This gives the DAG combiner more opportunity to optimize without needing to dig through the blend. llvm-svn: 294876
2017-02-12 06:57:12 +08:00
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
; AVX1-NEXT: retq
;
X86 Tests: Update more isel tests with FastVariableShuffle feature Summary: Added the FastVariableShuffle feature to cases that resembled processors for which this fearure is on. For AVX2 there are processors with and w/o this fearue enable. For AVX512 only KNL does enable this feature so cases which only have +avx512f were left without the FastVariableShuffle enabled. Reviewers: RKSimon, craig.topper Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D41851 llvm-svn: 322090
2018-01-10 00:26:06 +08:00
; AVX2-SLOW-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; AVX2-SLOW: # %bb.0:
; AVX2-SLOW-NEXT: xorl %eax, %eax
; AVX2-SLOW-NEXT: vpinsrb $0, %eax, %xmm0, %xmm1
; AVX2-SLOW-NEXT: vpinsrb $15, %eax, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-SLOW-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: retq
;
; AVX2-FAST-LABEL: insert_v32i8_z123456789ABCDEzGHIJKLMNOPQRSTzz:
; AVX2-FAST: # %bb.0:
; AVX2-FAST-NEXT: vpand {{.*}}(%rip), %xmm0, %xmm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-FAST-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm2[7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: retq
[X86][SSE] Added tests for insertion of zero elements into vectors Many of these could be much better if we just lowered them all as shuffles - especially for the 256-bit vectors. llvm-svn: 256708
2016-01-04 01:33:32 +08:00
%1 = insertelement <32 x i8> %a, i8 0, i32 0
%2 = insertelement <32 x i8> %1, i8 0, i32 15
%3 = insertelement <32 x i8> %2, i8 0, i32 30
%4 = insertelement <32 x i8> %3, i8 0, i32 31
ret <32 x i8> %4
}