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/avx-basic.ll

136 lines
5.2 KiB
LLVM
Raw Normal View History

[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=avx | FileCheck %s
Begin to support some vector operations for AVX 256-bit intructions. The long term goal here is to be able to match enough of vector_shuffle and build_vector so all avx intrinsics which aren't mapped to their own built-ins but to shufflevector calls can be codegen'd. This is the first (baby) step, support building zeroed vectors. llvm-svn: 110897
2010-08-12 10:06:36 +08:00
@x = common global <8 x float> zeroinitializer, align 32
@y = common global <4 x double> zeroinitializer, align 32
Rename and tidy up tests llvm-svn: 137103
2011-08-09 11:04:23 +08:00
@z = common global <4 x float> zeroinitializer, align 16
Begin to support some vector operations for AVX 256-bit intructions. The long term goal here is to be able to match enough of vector_shuffle and build_vector so all avx intrinsics which aren't mapped to their own built-ins but to shufflevector calls can be codegen'd. This is the first (baby) step, support building zeroed vectors. llvm-svn: 110897
2010-08-12 10:06:36 +08:00
Rename and tidy up tests llvm-svn: 137103
2011-08-09 11:04:23 +08:00
define void @zero128() nounwind ssp {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: zero128:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
; CHECK-NEXT: movq _z@{{.*}}(%rip), %rax
; CHECK-NEXT: vmovaps %xmm0, (%rax)
; CHECK-NEXT: retq
Rename and tidy up tests llvm-svn: 137103
2011-08-09 11:04:23 +08:00
store <4 x float> zeroinitializer, <4 x float>* @z, align 16
ret void
}
define void @zero256() nounwind ssp {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: zero256:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: movq _x@{{.*}}(%rip), %rax
[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
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vmovaps %ymm0, (%rax)
; CHECK-NEXT: movq _y@{{.*}}(%rip), %rax
; CHECK-NEXT: vmovaps %ymm0, (%rax)
; CHECK-NEXT: vzeroupper
; CHECK-NEXT: retq
Begin to support some vector operations for AVX 256-bit intructions. The long term goal here is to be able to match enough of vector_shuffle and build_vector so all avx intrinsics which aren't mapped to their own built-ins but to shufflevector calls can be codegen'd. This is the first (baby) step, support building zeroed vectors. llvm-svn: 110897
2010-08-12 10:06:36 +08:00
store <8 x float> zeroinitializer, <8 x float>* @x, align 32
store <4 x double> zeroinitializer, <4 x double>* @y, align 32
ret void
}
Codegen allonesvector better while using AVX: vpcmpeqd + vinsertf128 This also fixes PR10452 llvm-svn: 136004
2011-07-26 07:05:32 +08:00
define void @ones([0 x float]* nocapture %RET, [0 x float]* nocapture %aFOO) nounwind {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: ones:
[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
; CHECK: ## %bb.0: ## %allocas
[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
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
[x86, SSE] AVX1 PR28129 (256-bit all-ones rematerialization) Further perf tests on Jaguar indicate that: vxorps %ymm0, %ymm0, %ymm0 vcmpps $15, %ymm0, %ymm0, %ymm0 is consistently faster (by about 9%) than: vpcmpeqd %xmm0, %xmm0, %xmm0 vinsertf128 $1, %xmm0, %ymm0, %ymm0 Testing equivalent code on a SandyBridge (E5-2640) puts it slightly (~3%) faster as well. Committed on behalf of @dtemirbulatov Differential Revision: https://reviews.llvm.org/D32416 llvm-svn: 302989
2017-05-13 21:42:35 +08:00
; CHECK-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vmovaps %ymm0, (%rdi)
; CHECK-NEXT: vzeroupper
; CHECK-NEXT: retq
Codegen allonesvector better while using AVX: vpcmpeqd + vinsertf128 This also fixes PR10452 llvm-svn: 136004
2011-07-26 07:05:32 +08:00
allocas:
%ptr2vec615 = bitcast [0 x float]* %RET to <8 x float>*
store <8 x float> <float 0xFFFFFFFFE0000000, float 0xFFFFFFFFE0000000, float
0xFFFFFFFFE0000000, float 0xFFFFFFFFE0000000, float 0xFFFFFFFFE0000000, float
0xFFFFFFFFE0000000, float 0xFFFFFFFFE0000000, float 0xFFFFFFFFE0000000>, <8 x
float>* %ptr2vec615, align 32
ret void
}
Since vectors with all ones can't be created with a 256-bit instruction, avoid returning early for v8i32 types, which would only be valid for vector with all zeros. Also split the handling of zeros and ones into separate checking logic since they are handled differently. This fixes PR10547 llvm-svn: 136642
2011-08-02 03:51:53 +08:00
define void @ones2([0 x i32]* nocapture %RET, [0 x i32]* nocapture %aFOO) nounwind {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: ones2:
[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
; CHECK: ## %bb.0: ## %allocas
[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
; CHECK-NEXT: vxorps %xmm0, %xmm0, %xmm0
[x86, SSE] AVX1 PR28129 (256-bit all-ones rematerialization) Further perf tests on Jaguar indicate that: vxorps %ymm0, %ymm0, %ymm0 vcmpps $15, %ymm0, %ymm0, %ymm0 is consistently faster (by about 9%) than: vpcmpeqd %xmm0, %xmm0, %xmm0 vinsertf128 $1, %xmm0, %ymm0, %ymm0 Testing equivalent code on a SandyBridge (E5-2640) puts it slightly (~3%) faster as well. Committed on behalf of @dtemirbulatov Differential Revision: https://reviews.llvm.org/D32416 llvm-svn: 302989
2017-05-13 21:42:35 +08:00
; CHECK-NEXT: vcmptrueps %ymm0, %ymm0, %ymm0
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vmovaps %ymm0, (%rdi)
; CHECK-NEXT: vzeroupper
; CHECK-NEXT: retq
Since vectors with all ones can't be created with a 256-bit instruction, avoid returning early for v8i32 types, which would only be valid for vector with all zeros. Also split the handling of zeros and ones into separate checking logic since they are handled differently. This fixes PR10547 llvm-svn: 136642
2011-08-02 03:51:53 +08:00
allocas:
%ptr2vec615 = bitcast [0 x i32]* %RET to <8 x i32>*
store <8 x i32> <i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1>, <8 x i32>* %ptr2vec615, align 32
ret void
}
Fix PR10492 by teaching MOVHLPS and MOVLPS mask matching to be more strict. llvm-svn: 137324
2011-08-12 02:59:13 +08:00
;;; Just make sure this doesn't crash
define <4 x i64> @ISelCrash(<4 x i64> %a) nounwind uwtable readnone ssp {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: ISelCrash:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vextractf128 $1, %ymm0, %xmm0
; CHECK-NEXT: retq
Fix PR10492 by teaching MOVHLPS and MOVLPS mask matching to be more strict. llvm-svn: 137324
2011-08-12 02:59:13 +08:00
%shuffle = shufflevector <4 x i64> %a, <4 x i64> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 4>
ret <4 x i64> %shuffle
}
Instead of always leaving the work to the generic legalizer when there is no support for native 256-bit shuffles, be more smart in some cases, for example, when you can extract specific 128-bit parts and use regular 128-bit shuffles for them. Example: For this shuffle: shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 1, i32 0, i32 7, i32 6> This was expanded to: vextractf128 $1, %ymm1, %xmm2 vpextrq $0, %xmm2, %rax vmovd %rax, %xmm1 vpextrq $1, %xmm2, %rax vmovd %rax, %xmm2 vpunpcklqdq %xmm1, %xmm2, %xmm1 vpextrq $0, %xmm0, %rax vmovd %rax, %xmm2 vpextrq $1, %xmm0, %rax vmovd %rax, %xmm0 vpunpcklqdq %xmm2, %xmm0, %xmm0 vinsertf128 $1, %xmm1, %ymm0, %ymm0 ret Now we get: vshufpd $1, %xmm0, %xmm0, %xmm0 vextractf128 $1, %ymm1, %xmm1 vshufpd $1, %xmm1, %xmm1, %xmm1 vinsertf128 $1, %xmm1, %ymm0, %ymm0 llvm-svn: 137733
2011-08-17 02:21:54 +08:00
Fix PR10845. SUBREG_TO_REG shouldn't be used when the input and destination types are equal! llvm-svn: 139553
2011-09-13 06:59:23 +08:00
;;; Don't crash on movd
define <8 x i32> @VMOVZQI2PQI([0 x float]* nocapture %aFOO) nounwind {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: VMOVZQI2PQI:
[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
; CHECK: ## %bb.0:
[X86] Teach execution domain fixing to convert between VPERMILPS and VPSHUFD. llvm-svn: 313507
2017-09-18 11:29:47 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,0,1,1]
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: retq
Fix PR10845. SUBREG_TO_REG shouldn't be used when the input and destination types are equal! llvm-svn: 139553
2011-09-13 06:59:23 +08:00
%ptrcast.i33.i = bitcast [0 x float]* %aFOO to i32*
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val.i34.i = load i32, i32* %ptrcast.i33.i, align 4
[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction One of several parallel first steps to remove the target type of pointers, replacing them with a single opaque pointer type. This adds an explicit type parameter to the gep instruction so that when the first parameter becomes an opaque pointer type, the type to gep through is still available to the instructions. * This doesn't modify gep operators, only instructions (operators will be handled separately) * Textual IR changes only. Bitcode (including upgrade) and changing the in-memory representation will be in separate changes. * geps of vectors are transformed as: getelementptr <4 x float*> %x, ... ->getelementptr float, <4 x float*> %x, ... Then, once the opaque pointer type is introduced, this will ultimately look like: getelementptr float, <4 x ptr> %x with the unambiguous interpretation that it is a vector of pointers to float. * address spaces remain on the pointer, not the type: getelementptr float addrspace(1)* %x ->getelementptr float, float addrspace(1)* %x Then, eventually: getelementptr float, ptr addrspace(1) %x Importantly, the massive amount of test case churn has been automated by same crappy python code. I had to manually update a few test cases that wouldn't fit the script's model (r228970,r229196,r229197,r229198). The python script just massages stdin and writes the result to stdout, I then wrapped that in a shell script to handle replacing files, then using the usual find+xargs to migrate all the files. update.py: import fileinput import sys import re ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") normrep = re.compile( r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))") def conv(match, line): if not match: return line line = match.groups()[0] if len(match.groups()[5]) == 0: line += match.groups()[2] line += match.groups()[3] line += ", " line += match.groups()[1] line += "\n" return line for line in sys.stdin: if line.find("getelementptr ") == line.find("getelementptr inbounds"): if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("): line = conv(re.match(ibrep, line), line) elif line.find("getelementptr ") != line.find("getelementptr ("): line = conv(re.match(normrep, line), line) sys.stdout.write(line) apply.sh: for name in "$@" do python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name" rm -f "$name.tmp" done The actual commands: From llvm/src: find test/ -name *.ll | xargs ./apply.sh From llvm/src/tools/clang: find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}" From llvm/src/tools/polly: find test/ -name *.ll | xargs ./apply.sh After that, check-all (with llvm, clang, clang-tools-extra, lld, compiler-rt, and polly all checked out). The extra 'rm' in the apply.sh script is due to a few files in clang's test suite using interesting unicode stuff that my python script was throwing exceptions on. None of those files needed to be migrated, so it seemed sufficient to ignore those cases. Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7636 llvm-svn: 230786
2015-02-28 03:29:02 +08:00
%ptroffset.i22.i992 = getelementptr [0 x float], [0 x float]* %aFOO, i64 0, i64 1
Fix PR10845. SUBREG_TO_REG shouldn't be used when the input and destination types are equal! llvm-svn: 139553
2011-09-13 06:59:23 +08:00
%ptrcast.i23.i = bitcast float* %ptroffset.i22.i992 to i32*
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val.i24.i = load i32, i32* %ptrcast.i23.i, align 4
Fix PR10845. SUBREG_TO_REG shouldn't be used when the input and destination types are equal! llvm-svn: 139553
2011-09-13 06:59:23 +08:00
%updatedret.i30.i = insertelement <8 x i32> undef, i32 %val.i34.i, i32 1
ret <8 x i32> %updatedret.i30.i
}
Add support for lowering fneg when AVX is enabled. rdar://10566486 llvm-svn: 146625
2011-12-15 09:02:25 +08:00
;;;; Don't crash on fneg
; rdar://10566486
Teach DAG combiner to constant fold fneg of a BUILD_VECTOR of constants. llvm-svn: 163483
2012-09-10 06:58:45 +08:00
define <16 x float> @fneg(<16 x float> %a) nounwind {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: fneg:
[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
; CHECK: ## %bb.0:
[X86] Don't lower FABS/FNEG masking directly to a ConstantPool load. Just create a ConstantFPSDNode and let that be lowered. This allows broadcast loads to used when available. llvm-svn: 279958
2016-08-29 12:49:31 +08:00
; CHECK-NEXT: vmovaps {{.*#+}} ymm2 = [-0.000000e+00,-0.000000e+00,-0.000000e+00,-0.000000e+00,-0.000000e+00,-0.000000e+00,-0.000000e+00,-0.000000e+00]
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vxorps %ymm2, %ymm0, %ymm0
; CHECK-NEXT: vxorps %ymm2, %ymm1, %ymm1
; CHECK-NEXT: retq
Teach DAG combiner to constant fold fneg of a BUILD_VECTOR of constants. llvm-svn: 163483
2012-09-10 06:58:45 +08:00
%1 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %a
Add support for lowering fneg when AVX is enabled. rdar://10566486 llvm-svn: 146625
2011-12-15 09:02:25 +08:00
ret <16 x float> %1
}
Fix assert in LowerBUILD_VECTOR for v16i16 type on AVX. Patch by Elena Demikhovsky <elena.demikhovsky@intel.com>! llvm-svn: 146684
2011-12-16 05:34:44 +08:00
;;; Don't crash on build vector
define <16 x i16> @build_vec_16x16(i16 %a) nounwind readonly {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: build_vec_16x16:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: movzwl %di, %eax
; CHECK-NEXT: vmovd %eax, %xmm0
; CHECK-NEXT: retq
Fix assert in LowerBUILD_VECTOR for v16i16 type on AVX. Patch by Elena Demikhovsky <elena.demikhovsky@intel.com>! llvm-svn: 146684
2011-12-16 05:34:44 +08:00
%res = insertelement <16 x i16> <i16 undef, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>, i16 %a, i32 0
ret <16 x i16> %res
}
Put VMOVPQIto64rr in the VRPDI class. Patch by Joshua Magee. llvm-svn: 180842
2013-05-01 21:00:16 +08:00
Remove some instructions that existed to provide aliases to the assembler. Can be done with InstAlias instead. Unfortunately, this was causing printer to use 'vmovq' or 'vmovd' based on what was parsed. To cleanup the inconsistencies convert all 'vmovd' with 64-bit registers to 'vmovq', but provide an alias so that 'vmovd' will still parse. llvm-svn: 192171
2013-10-08 13:53:50 +08:00
;;; Check that VMOVPQIto64rr generates the assembly string "vmovq". Previously
Put VMOVPQIto64rr in the VRPDI class. Patch by Joshua Magee. llvm-svn: 180842
2013-05-01 21:00:16 +08:00
;;; an incorrect mnemonic of "movd" was printed for this instruction.
define i64 @VMOVPQIto64rr(<2 x i64> %a) {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: VMOVPQIto64rr:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vmovq %xmm0, %rax
; CHECK-NEXT: retq
Put VMOVPQIto64rr in the VRPDI class. Patch by Joshua Magee. llvm-svn: 180842
2013-05-01 21:00:16 +08:00
%vecext.i = extractelement <2 x i64> %a, i32 0
ret i64 %vecext.i
}
[X86] Use vmovss to handle inserting an element into index 0 of a v8f32 vector of zeros. llvm-svn: 231354
2015-03-05 14:38:42 +08:00
; PR22685
define <8 x float> @mov00_8f32(float* %ptr) {
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-LABEL: mov00_8f32:
[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
; CHECK: ## %bb.0:
[x86, AVX] tighten checks llvm-svn: 258828
2016-01-27 02:22:50 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: retq
[X86] Use vmovss to handle inserting an element into index 0 of a v8f32 vector of zeros. llvm-svn: 231354
2015-03-05 14:38:42 +08:00
%val = load float, float* %ptr
%vec = insertelement <8 x float> zeroinitializer, float %val, i32 0
ret <8 x float> %vec
}