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/not-and-simplify.ll

58 lines
2.1 KiB
LLVM
Raw Normal View History

[x86] add vector tests for demanded bits; NFC llvm-svn: 302949
2017-05-13 04:53:48 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
[x86] add common check prefix to reduce duplication; NFC llvm-svn: 288522
2016-12-03 01:58:26 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=-bmi | FileCheck %s --check-prefix=ALL --check-prefix=NO_BMI
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+bmi | FileCheck %s --check-prefix=ALL --check-prefix=BMI
[x86] add tests to show missing demanded bits analysis; NFC llvm-svn: 288515
2016-12-03 01:48:48 +08:00
; Clear high bits via shift, set them with xor (not), then mask them off.
define i32 @shrink_xor_constant1(i32 %x) {
[TargetLowering] add special-case for demanded bits analysis of 'not' We treat bitwise 'not' as a special operation and try not to reduce its all-ones mask. Presumably, this is because a 'not' may be cheaper than a generic 'xor' or it may get folded into another logic op if the target has those. However, if we can remove a logic instruction by changing the xor's constant mask value, that should always be a win. Note that the IR version of SimplifyDemandedBits() does not treat 'not' as a special-case currently (although that's marked with a FIXME). So if you run this IR through -instcombine, you should get the same end result. I'm hoping to add a different backend transform that will expose this problem though, so I need to solve this first. Differential Revision: https://reviews.llvm.org/D27356 llvm-svn: 288676
2016-12-05 23:58:21 +08:00
; ALL-LABEL: shrink_xor_constant1:
[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
; ALL: # %bb.0:
[TargetLowering] add special-case for demanded bits analysis of 'not' We treat bitwise 'not' as a special operation and try not to reduce its all-ones mask. Presumably, this is because a 'not' may be cheaper than a generic 'xor' or it may get folded into another logic op if the target has those. However, if we can remove a logic instruction by changing the xor's constant mask value, that should always be a win. Note that the IR version of SimplifyDemandedBits() does not treat 'not' as a special-case currently (although that's marked with a FIXME). So if you run this IR through -instcombine, you should get the same end result. I'm hoping to add a different backend transform that will expose this problem though, so I need to solve this first. Differential Revision: https://reviews.llvm.org/D27356 llvm-svn: 288676
2016-12-05 23:58:21 +08:00
; ALL-NEXT: shrl $31, %edi
; ALL-NEXT: xorl $1, %edi
; ALL-NEXT: movl %edi, %eax
; ALL-NEXT: retq
[x86] add tests to show missing demanded bits analysis; NFC llvm-svn: 288515
2016-12-03 01:48:48 +08:00
%sh = lshr i32 %x, 31
%not = xor i32 %sh, -1
%and = and i32 %not, 1
ret i32 %and
}
[x86] add vector tests for demanded bits; NFC llvm-svn: 302949
2017-05-13 04:53:48 +08:00
define <4 x i32> @shrink_xor_constant1_splat(<4 x i32> %x) {
; ALL-LABEL: shrink_xor_constant1_splat:
[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
; ALL: # %bb.0:
[x86] add vector tests for demanded bits; NFC llvm-svn: 302949
2017-05-13 04:53:48 +08:00
; ALL-NEXT: psrld $31, %xmm0
; ALL-NEXT: pandn {{.*}}(%rip), %xmm0
; ALL-NEXT: retq
%sh = lshr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
%not = xor <4 x i32> %sh, <i32 -1, i32 -1, i32 -1, i32 -1>
%and = and <4 x i32> %not, <i32 1, i32 1, i32 1, i32 1>
ret <4 x i32> %and
}
[x86] add tests to show missing demanded bits analysis; NFC llvm-svn: 288515
2016-12-03 01:48:48 +08:00
; Clear low bits via shift, set them with xor (not), then mask them off.
define i8 @shrink_xor_constant2(i8 %x) {
[x86] add common check prefix to reduce duplication; NFC llvm-svn: 288522
2016-12-03 01:58:26 +08:00
; ALL-LABEL: shrink_xor_constant2:
[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
; ALL: # %bb.0:
[x86] add common check prefix to reduce duplication; NFC llvm-svn: 288522
2016-12-03 01:58:26 +08:00
; ALL-NEXT: shlb $5, %dil
[TargetLowering] add special-case for demanded bits analysis of 'not' We treat bitwise 'not' as a special operation and try not to reduce its all-ones mask. Presumably, this is because a 'not' may be cheaper than a generic 'xor' or it may get folded into another logic op if the target has those. However, if we can remove a logic instruction by changing the xor's constant mask value, that should always be a win. Note that the IR version of SimplifyDemandedBits() does not treat 'not' as a special-case currently (although that's marked with a FIXME). So if you run this IR through -instcombine, you should get the same end result. I'm hoping to add a different backend transform that will expose this problem though, so I need to solve this first. Differential Revision: https://reviews.llvm.org/D27356 llvm-svn: 288676
2016-12-05 23:58:21 +08:00
; ALL-NEXT: xorb $-32, %dil
[x86] add common check prefix to reduce duplication; NFC llvm-svn: 288522
2016-12-03 01:58:26 +08:00
; ALL-NEXT: movl %edi, %eax
; ALL-NEXT: retq
[x86] add tests to show missing demanded bits analysis; NFC llvm-svn: 288515
2016-12-03 01:48:48 +08:00
%sh = shl i8 %x, 5
%not = xor i8 %sh, -1
%and = and i8 %not, 224 ; 0xE0
ret i8 %and
}
[x86] add vector tests for demanded bits; NFC llvm-svn: 302949
2017-05-13 04:53:48 +08:00
define <16 x i8> @shrink_xor_constant2_splat(<16 x i8> %x) {
; ALL-LABEL: shrink_xor_constant2_splat:
[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
; ALL: # %bb.0:
[SelectionDAG] Teach simplifyDemandedBits to handle shifts by constant splat vectors This teach simplifyDemandedBits to handle constant splat vector shifts. This required changing some uses of getZExtValue to getLimitedValue since we can't rely on legalization using getShiftAmountTy for the shift amount. I believe there may have been a bug in the ((X << C1) >>u ShAmt) handling where we didn't check if the inner shift was too large. I've fixed that here. I had to add new patterns to ARM because the zext/sext the patterns were trying to look for got turned into an any_extend with this patch. Happy to split that out too, but not sure how to test without this change. Differential Revision: https://reviews.llvm.org/D37665 llvm-svn: 314139
2017-09-26 03:26:08 +08:00
; ALL-NEXT: movaps {{.*#+}} xmm0 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
[x86] add vector tests for demanded bits; NFC llvm-svn: 302949
2017-05-13 04:53:48 +08:00
; ALL-NEXT: retq
%sh = shl <16 x i8> %x, <i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5, i8 5>
%not = xor <16 x i8> %sh, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
%and = and <16 x i8> %not, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
ret <16 x i8> %and
}