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/sink-addsub-of-const.ll

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

476 lines
14 KiB
LLVM
Raw Normal View History

[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+slow-lea,+slow-3ops-lea,+sse,+sse2 | FileCheck %s --check-prefixes=ALL,X32
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+slow-lea,+slow-3ops-lea,+sse,+sse2 | FileCheck %s --check-prefixes=ALL,X64
; Scalar tests. Trying to avoid LEA here, so the output is actually readable..
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; add (add %x, C), %y
; Outer 'add' is commutative - 2 variants.
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_add_of_const_to_add0(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_add_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl %ecx, %eax
; X32-NEXT: addl $32, %eax
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: sink_add_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal 32(%rdi,%rsi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add i32 %a, 32 ; constant always on RHS
%r = add i32 %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_add_of_const_to_add1(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_add_of_const_to_add1:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl %ecx, %eax
; X32-NEXT: addl $32, %eax
; X32-NEXT: retl
;
; X64-LABEL: sink_add_of_const_to_add1:
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal 32(%rdi,%rsi), %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add i32 %a, 32 ; constant always on RHS
%r = add i32 %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
; add (sub %x, C), %y
; Outer 'add' is commutative - 2 variants.
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_of_const_to_add0(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl %ecx, %eax
; X32-NEXT: addl $-32, %eax
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: sink_sub_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal -32(%rdi,%rsi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 %a, 32
%r = add i32 %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_of_const_to_add1(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_of_const_to_add1:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl %ecx, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: addl $-32, %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: sink_sub_of_const_to_add1:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal -32(%rdi,%rsi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 %a, 32
%r = add i32 %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; add (sub C, %x), %y
; Outer 'add' is commutative - 2 variants.
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_from_const_to_add0(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_from_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: addl $32, %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: sink_sub_from_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
; X64-NEXT: subl %edi, %esi
; X64-NEXT: leal 32(%rsi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 32, %a
%r = add i32 %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_from_const_to_add1(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_from_const_to_add1:
; X32: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: addl $32, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_sub_from_const_to_add1:
; X64: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
; X64-NEXT: subl %edi, %esi
; X64-NEXT: leal 32(%rsi), %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 32, %a
%r = add i32 %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; sub (add %x, C), %y
; sub %y, (add %x, C)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_add_of_const_to_sub(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_add_of_const_to_sub:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombiner][X86][AArch64][AMDGPU] (x + C) - y -> (x - y) + C fold. Try 3 Summary: The main motivation is shown by all these `neg` instructions that are now created. In particular, the `@reg32_lshr_by_negated_unfolded_sub_b` test. AArch64 test changes all look good (`neg` created), or neutral. X86 changes look neutral (vectors), or good (`neg` / `xor eax, eax` created). I'm not sure about `X86/ragreedy-hoist-spill.ll`, it looks like the spill is now hoisted into preheader (which should still be good?), 2 4-byte reloads become 1 8-byte reload, and are elsewhere, but i'm not sure how that affects that loop. I'm unable to interpret AMDGPU change, looks neutral-ish? This is hopefully a step towards solving [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/pkdq (we are missing more patterns, i'll submit them later) This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: craig.topper, RKSimon, spatel, arsenm Reviewed By: RKSimon Subscribers: bjope, qcolombet, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62223 llvm-svn: 362142
2019-05-31 04:36:54 +08:00
; X32-NEXT: addl $32, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_add_of_const_to_sub:
; X64: # %bb.0:
; X64-NEXT: # kill: def $edi killed $edi def $rdi
; X64-NEXT: subl %esi, %edi
; X64-NEXT: leal 32(%rdi), %eax
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add i32 %a, 32 ; constant always on RHS
%r = sub i32 %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_add_of_const_to_sub2(i32 %a, i32 %b) {
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-LABEL: sink_add_of_const_to_sub2:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombiner][X86][AArch64][SPARC][SystemZ] y - (x + C) -> (y - x) - C fold. Try 3 Summary: Direct sibling of D62223 patch. While i don't have a direct motivational pattern for this, it would seem to make sense to handle both patterns (or none), for symmetry? The aarch64 changes look neutral; sparc and systemz look like improvement (one less instruction each); x86 changes - 32bit case improves, 64bit case shows that LEA no longer gets constructed, which may be because that whole test is `-mattr=+slow-lea,+slow-3ops-lea` https://rise4fun.com/Alive/ffh This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel, t.p.northover Reviewed By: t.p.northover Subscribers: t.p.northover, jyknight, javed.absar, kristof.beyls, fedor.sergeev, jrtc27, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62252 llvm-svn: 362143
2019-05-31 04:37:18 +08:00
; X32-NEXT: addl $-32, %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_add_of_const_to_sub2:
; X64: # %bb.0:
[DAGCombiner][X86][AArch64][SPARC][SystemZ] y - (x + C) -> (y - x) - C fold. Try 3 Summary: Direct sibling of D62223 patch. While i don't have a direct motivational pattern for this, it would seem to make sense to handle both patterns (or none), for symmetry? The aarch64 changes look neutral; sparc and systemz look like improvement (one less instruction each); x86 changes - 32bit case improves, 64bit case shows that LEA no longer gets constructed, which may be because that whole test is `-mattr=+slow-lea,+slow-3ops-lea` https://rise4fun.com/Alive/ffh This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel, t.p.northover Reviewed By: t.p.northover Subscribers: t.p.northover, jyknight, javed.absar, kristof.beyls, fedor.sergeev, jrtc27, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62252 llvm-svn: 362143
2019-05-31 04:37:18 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
; X64-NEXT: subl %edi, %esi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal -32(%rsi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add i32 %a, 32 ; constant always on RHS
%r = sub i32 %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; sub (sub %x, C), %y
; sub %y, (sub %x, C)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_of_const_to_sub(i32 %a, i32 %b) {
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-LABEL: sink_sub_of_const_to_sub:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombiner][X86][AArch64][AMDGPU] (x + C) - y -> (x - y) + C fold. Try 3 Summary: The main motivation is shown by all these `neg` instructions that are now created. In particular, the `@reg32_lshr_by_negated_unfolded_sub_b` test. AArch64 test changes all look good (`neg` created), or neutral. X86 changes look neutral (vectors), or good (`neg` / `xor eax, eax` created). I'm not sure about `X86/ragreedy-hoist-spill.ll`, it looks like the spill is now hoisted into preheader (which should still be good?), 2 4-byte reloads become 1 8-byte reload, and are elsewhere, but i'm not sure how that affects that loop. I'm unable to interpret AMDGPU change, looks neutral-ish? This is hopefully a step towards solving [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/pkdq (we are missing more patterns, i'll submit them later) This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: craig.topper, RKSimon, spatel, arsenm Reviewed By: RKSimon Subscribers: bjope, qcolombet, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62223 llvm-svn: 362142
2019-05-31 04:36:54 +08:00
; X32-NEXT: addl $-32, %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_sub_of_const_to_sub:
; X64: # %bb.0:
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: subl %esi, %edi
; X64-NEXT: leal -32(%rdi), %eax
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 %a, 32
%r = sub i32 %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_of_const_to_sub2(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_of_const_to_sub2:
; X32: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: subl {{[0-9]+}}(%esp), %eax
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-NEXT: addl $32, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_sub_of_const_to_sub2:
; X64: # %bb.0:
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
; X64-NEXT: subl %edi, %esi
; X64-NEXT: leal 32(%rsi), %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 %a, 32
%r = sub i32 %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
; sub (sub C, %x), %y
; sub %y, (sub C, %x)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_from_const_to_sub(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_from_const_to_sub:
; X32: # %bb.0:
[DAGCombine][X86][AArch64][MIPS][LANAI] (C - x) - y -> C - (x + y) fold (PR41952) Summary: This *might* be the last fold for `sink-addsub-of-const.ll`, but i'm not sure yet. As far as i can tell, there are no regressions here (ignoring x86-32), all changes are either good or neutral. This, almost surprisingly to me, fixes the motivational tests (in `shift-amount-mod.ll`) `@reg32_lshr_by_sub_from_negated` from [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/vMd3 Reviewers: RKSimon, t.p.northover, craig.topper, spatel, efriedma Reviewed By: RKSimon Subscribers: sdardis, javed.absar, arichardson, kristof.beyls, jrtc27, atanasyan, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62774 llvm-svn: 362488
2019-06-04 19:06:21 +08:00
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl {{[0-9]+}}(%esp), %ecx
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: movl $32, %eax
[DAGCombine][X86][AArch64][MIPS][LANAI] (C - x) - y -> C - (x + y) fold (PR41952) Summary: This *might* be the last fold for `sink-addsub-of-const.ll`, but i'm not sure yet. As far as i can tell, there are no regressions here (ignoring x86-32), all changes are either good or neutral. This, almost surprisingly to me, fixes the motivational tests (in `shift-amount-mod.ll`) `@reg32_lshr_by_sub_from_negated` from [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/vMd3 Reviewers: RKSimon, t.p.northover, craig.topper, spatel, efriedma Reviewed By: RKSimon Subscribers: sdardis, javed.absar, arichardson, kristof.beyls, jrtc27, atanasyan, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62774 llvm-svn: 362488
2019-06-04 19:06:21 +08:00
; X32-NEXT: subl %ecx, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: sink_sub_from_const_to_sub:
; X64: # %bb.0:
[DAGCombine][X86][AArch64][MIPS][LANAI] (C - x) - y -> C - (x + y) fold (PR41952) Summary: This *might* be the last fold for `sink-addsub-of-const.ll`, but i'm not sure yet. As far as i can tell, there are no regressions here (ignoring x86-32), all changes are either good or neutral. This, almost surprisingly to me, fixes the motivational tests (in `shift-amount-mod.ll`) `@reg32_lshr_by_sub_from_negated` from [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/vMd3 Reviewers: RKSimon, t.p.northover, craig.topper, spatel, efriedma Reviewed By: RKSimon Subscribers: sdardis, javed.absar, arichardson, kristof.beyls, jrtc27, atanasyan, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62774 llvm-svn: 362488
2019-06-04 19:06:21 +08:00
; X64-NEXT: addl %esi, %edi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: movl $32, %eax
; X64-NEXT: subl %edi, %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 32, %a
%r = sub i32 %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define i32 @sink_sub_from_const_to_sub2(i32 %a, i32 %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: sink_sub_from_const_to_sub2:
; X32: # %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT: addl %ecx, %eax
; X32-NEXT: addl $-32, %eax
; X32-NEXT: retl
;
; X64-LABEL: sink_sub_from_const_to_sub2:
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: # kill: def $esi killed $esi def $rsi
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: # kill: def $edi killed $edi def $rdi
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: leal -32(%rdi,%rsi), %eax
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub i32 32, %a
%r = sub i32 %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret i32 %r
}
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
;------------------------------------------------------------------------------;
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; Basic vector tests. Here it is easier to see where the constant operand is.
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
;------------------------------------------------------------------------------;
; add (add %x, C), %y
; Outer 'add' is commutative - 2 variants.
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_add_of_const_to_add0(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_add_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: paddd %xmm1, %xmm0
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm0
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: vec_sink_add_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
; X64-NEXT: paddd %xmm1, %xmm0
; X64-NEXT: paddd {{.*}}(%rip), %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46> ; constant always on RHS
%r = add <4 x i32> %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_add_of_const_to_add1(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_add_of_const_to_add1:
; X32: # %bb.0:
; X32-NEXT: paddd %xmm1, %xmm0
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm0
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_add_of_const_to_add1:
; X64: # %bb.0:
; X64-NEXT: paddd %xmm1, %xmm0
; X64-NEXT: paddd {{.*}}(%rip), %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46> ; constant always on RHS
%r = add <4 x i32> %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}
; add (sub %x, C), %y
; Outer 'add' is commutative - 2 variants.
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_of_const_to_add0(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_sub_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: paddd %xmm1, %xmm0
; X32-NEXT: psubd {{\.LCPI.*}}, %xmm0
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: vec_sink_sub_of_const_to_add0:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
; X64-NEXT: paddd %xmm1, %xmm0
; X64-NEXT: psubd {{.*}}(%rip), %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46>
%r = add <4 x i32> %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_of_const_to_add1(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_sub_of_const_to_add1:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32: # %bb.0:
; X32-NEXT: paddd %xmm1, %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: psubd {{\.LCPI.*}}, %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-LABEL: vec_sink_sub_of_const_to_add1:
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64: # %bb.0:
; X64-NEXT: paddd %xmm1, %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: psubd {{.*}}(%rip), %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46>
%r = add <4 x i32> %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; add (sub C, %x), %y
; Outer 'add' is commutative - 2 variants.
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_from_const_to_add0(<4 x i32> %a, <4 x i32> %b) {
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-LABEL: vec_sink_sub_from_const_to_add0:
; X32: # %bb.0:
; X32-NEXT: psubd %xmm0, %xmm1
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm1
; X32-NEXT: movdqa %xmm1, %xmm0
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_sub_from_const_to_add0:
; X64: # %bb.0:
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: paddd {{.*}}(%rip), %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> <i32 42, i32 24, i32 undef, i32 46>, %a
%r = add <4 x i32> %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_from_const_to_add1(<4 x i32> %a, <4 x i32> %b) {
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-LABEL: vec_sink_sub_from_const_to_add1:
; X32: # %bb.0:
; X32-NEXT: psubd %xmm0, %xmm1
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm1
; X32-NEXT: movdqa %xmm1, %xmm0
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_sub_from_const_to_add1:
; X64: # %bb.0:
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: paddd {{.*}}(%rip), %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> <i32 42, i32 24, i32 undef, i32 46>, %a
%r = add <4 x i32> %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; sub (add %x, C), %y
; sub %y, (add %x, C)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_add_of_const_to_sub(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_add_of_const_to_sub:
; X32: # %bb.0:
; X32-NEXT: psubd %xmm1, %xmm0
[DAGCombiner][X86][AArch64][AMDGPU] (x + C) - y -> (x - y) + C fold. Try 3 Summary: The main motivation is shown by all these `neg` instructions that are now created. In particular, the `@reg32_lshr_by_negated_unfolded_sub_b` test. AArch64 test changes all look good (`neg` created), or neutral. X86 changes look neutral (vectors), or good (`neg` / `xor eax, eax` created). I'm not sure about `X86/ragreedy-hoist-spill.ll`, it looks like the spill is now hoisted into preheader (which should still be good?), 2 4-byte reloads become 1 8-byte reload, and are elsewhere, but i'm not sure how that affects that loop. I'm unable to interpret AMDGPU change, looks neutral-ish? This is hopefully a step towards solving [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/pkdq (we are missing more patterns, i'll submit them later) This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: craig.topper, RKSimon, spatel, arsenm Reviewed By: RKSimon Subscribers: bjope, qcolombet, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62223 llvm-svn: 362142
2019-05-31 04:36:54 +08:00
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_add_of_const_to_sub:
; X64: # %bb.0:
; X64-NEXT: psubd %xmm1, %xmm0
[DAGCombiner][X86][AArch64][AMDGPU] (x + C) - y -> (x - y) + C fold. Try 3 Summary: The main motivation is shown by all these `neg` instructions that are now created. In particular, the `@reg32_lshr_by_negated_unfolded_sub_b` test. AArch64 test changes all look good (`neg` created), or neutral. X86 changes look neutral (vectors), or good (`neg` / `xor eax, eax` created). I'm not sure about `X86/ragreedy-hoist-spill.ll`, it looks like the spill is now hoisted into preheader (which should still be good?), 2 4-byte reloads become 1 8-byte reload, and are elsewhere, but i'm not sure how that affects that loop. I'm unable to interpret AMDGPU change, looks neutral-ish? This is hopefully a step towards solving [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/pkdq (we are missing more patterns, i'll submit them later) This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: craig.topper, RKSimon, spatel, arsenm Reviewed By: RKSimon Subscribers: bjope, qcolombet, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, javed.absar, dstuttard, tpr, t-tye, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62223 llvm-svn: 362142
2019-05-31 04:36:54 +08:00
; X64-NEXT: paddd {{.*}}(%rip), %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46> ; constant always on RHS
%r = sub <4 x i32> %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_add_of_const_to_sub2(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-LABEL: vec_sink_add_of_const_to_sub2:
; X32: # %bb.0:
[DAGCombiner][X86][AArch64][SPARC][SystemZ] y - (x + C) -> (y - x) - C fold. Try 3 Summary: Direct sibling of D62223 patch. While i don't have a direct motivational pattern for this, it would seem to make sense to handle both patterns (or none), for symmetry? The aarch64 changes look neutral; sparc and systemz look like improvement (one less instruction each); x86 changes - 32bit case improves, 64bit case shows that LEA no longer gets constructed, which may be because that whole test is `-mattr=+slow-lea,+slow-3ops-lea` https://rise4fun.com/Alive/ffh This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel, t.p.northover Reviewed By: t.p.northover Subscribers: t.p.northover, jyknight, javed.absar, kristof.beyls, fedor.sergeev, jrtc27, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62252 llvm-svn: 362143
2019-05-31 04:37:18 +08:00
; X32-NEXT: psubd %xmm0, %xmm1
; X32-NEXT: psubd {{\.LCPI.*}}, %xmm1
; X32-NEXT: movdqa %xmm1, %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_add_of_const_to_sub2:
; X64: # %bb.0:
[DAGCombiner][X86][AArch64][SPARC][SystemZ] y - (x + C) -> (y - x) - C fold. Try 3 Summary: Direct sibling of D62223 patch. While i don't have a direct motivational pattern for this, it would seem to make sense to handle both patterns (or none), for symmetry? The aarch64 changes look neutral; sparc and systemz look like improvement (one less instruction each); x86 changes - 32bit case improves, 64bit case shows that LEA no longer gets constructed, which may be because that whole test is `-mattr=+slow-lea,+slow-3ops-lea` https://rise4fun.com/Alive/ffh This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel, t.p.northover Reviewed By: t.p.northover Subscribers: t.p.northover, jyknight, javed.absar, kristof.beyls, fedor.sergeev, jrtc27, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62252 llvm-svn: 362143
2019-05-31 04:37:18 +08:00
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: psubd {{.*}}(%rip), %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = add <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46> ; constant always on RHS
%r = sub <4 x i32> %b, %t0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; sub (sub %x, C), %y
; sub %y, (sub %x, C)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_of_const_to_sub(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-LABEL: vec_sink_sub_of_const_to_sub:
; X32: # %bb.0:
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: psubd %xmm1, %xmm0
[DAGCombine] (x - C) - y -> (x - y) - C fold. Try 3 Summary: Again only vectors affected. Frustrating. Let me take a look into that.. https://rise4fun.com/Alive/AAq This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel Reviewed By: RKSimon Subscribers: javed.absar, JDevlieghere, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62294 llvm-svn: 362145
2019-05-31 04:37:39 +08:00
; X32-NEXT: psubd {{\.LCPI.*}}, %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_sub_of_const_to_sub:
; X64: # %bb.0:
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: psubd %xmm1, %xmm0
[DAGCombine] (x - C) - y -> (x - y) - C fold. Try 3 Summary: Again only vectors affected. Frustrating. Let me take a look into that.. https://rise4fun.com/Alive/AAq This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs, and then reverted in rL362109 to fix missing constant folds that were causing endless combine loops. Reviewers: RKSimon, craig.topper, spatel Reviewed By: RKSimon Subscribers: javed.absar, JDevlieghere, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62294 llvm-svn: 362145
2019-05-31 04:37:39 +08:00
; X64-NEXT: psubd {{.*}}(%rip), %xmm0
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46>
%r = sub <4 x i32> %t0, %b
[NFC][X86][AArch64] Add tests for sinking of add/sub by constant through add/sub Looks we can transform all 8 variants of the pattern: https://rise4fun.com/Alive/auH This comes up as an issue on the path towards https://bugs.llvm.org/show_bug.cgi?id=41952 llvm-svn: 361303
2019-05-22 04:14:54 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_of_const_to_sub2(<4 x i32> %a, <4 x i32> %b) {
[DAGCombine][X86][AArch64][ARM] (C - x) + y -> (y - x) + C fold Summary: All changes except ARM look **great**. https://rise4fun.com/Alive/R2M The regression `test/CodeGen/ARM/addsubcarry-promotion.ll` is recovered fully by D62392 + D62450. Reviewers: RKSimon, craig.topper, spatel, rogfer01, efriedma Reviewed By: efriedma Subscribers: dmgreen, javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62266 llvm-svn: 362487
2019-06-04 19:06:08 +08:00
; X32-LABEL: vec_sink_sub_of_const_to_sub2:
; X32: # %bb.0:
; X32-NEXT: psubd %xmm0, %xmm1
; X32-NEXT: paddd {{\.LCPI.*}}, %xmm1
; X32-NEXT: movdqa %xmm1, %xmm0
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_sub_of_const_to_sub2:
; X64: # %bb.0:
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: paddd {{.*}}(%rip), %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> %a, <i32 42, i32 24, i32 undef, i32 46>
%r = sub <4 x i32> %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}
; sub (sub C, %x), %y
; sub %y, (sub C, %x)
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_from_const_to_sub(<4 x i32> %a, <4 x i32> %b) {
; ALL-LABEL: vec_sink_sub_from_const_to_sub:
; ALL: # %bb.0:
; ALL-NEXT: movdqa {{.*#+}} xmm2 = <42,24,u,46>
[DAGCombine][X86][AArch64][MIPS][LANAI] (C - x) - y -> C - (x + y) fold (PR41952) Summary: This *might* be the last fold for `sink-addsub-of-const.ll`, but i'm not sure yet. As far as i can tell, there are no regressions here (ignoring x86-32), all changes are either good or neutral. This, almost surprisingly to me, fixes the motivational tests (in `shift-amount-mod.ll`) `@reg32_lshr_by_sub_from_negated` from [[ https://bugs.llvm.org/show_bug.cgi?id=41952 | PR41952 ]]. https://rise4fun.com/Alive/vMd3 Reviewers: RKSimon, t.p.northover, craig.topper, spatel, efriedma Reviewed By: RKSimon Subscribers: sdardis, javed.absar, arichardson, kristof.beyls, jrtc27, atanasyan, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62774 llvm-svn: 362488
2019-06-04 19:06:21 +08:00
; ALL-NEXT: paddd %xmm1, %xmm0
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; ALL-NEXT: psubd %xmm0, %xmm2
; ALL-NEXT: movdqa %xmm2, %xmm0
; ALL-NEXT: ret{{[l|q]}}
%t0 = sub <4 x i32> <i32 42, i32 24, i32 undef, i32 46>, %a
%r = sub <4 x i32> %t0, %b
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
define <4 x i32> @vec_sink_sub_from_const_to_sub2(<4 x i32> %a, <4 x i32> %b) {
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-LABEL: vec_sink_sub_from_const_to_sub2:
; X32: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X32-NEXT: paddd %xmm1, %xmm0
[DAGCombiner][X86][AArch64] (x - C) + y -> (x + y) - C fold. Try 2 Summary: Only vector tests are being affected here, since subtraction by scalar constant is rewritten as addition by negated constant. No surprising test changes. https://rise4fun.com/Alive/pbT This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs. Reviewers: RKSimon, craig.topper, spatel Reviewed By: RKSimon Subscribers: javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62257 llvm-svn: 362146
2019-05-31 04:37:49 +08:00
; X32-NEXT: psubd {{\.LCPI.*}}, %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X32-NEXT: retl
;
; X64-LABEL: vec_sink_sub_from_const_to_sub2:
; X64: # %bb.0:
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
; X64-NEXT: paddd %xmm1, %xmm0
[DAGCombiner][X86][AArch64] (x - C) + y -> (x + y) - C fold. Try 2 Summary: Only vector tests are being affected here, since subtraction by scalar constant is rewritten as addition by negated constant. No surprising test changes. https://rise4fun.com/Alive/pbT This is a recommit, originally committed in rL361852, but reverted to investigate test-suite compile-time hangs. Reviewers: RKSimon, craig.topper, spatel Reviewed By: RKSimon Subscribers: javed.absar, kristof.beyls, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D62257 llvm-svn: 362146
2019-05-31 04:37:49 +08:00
; X64-NEXT: psubd {{.*}}(%rip), %xmm0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
; X64-NEXT: retq
[NFC][Codegen] shift-amount-mod.ll: drop innermost operation I have initially added it in for test to display both whether the binop w/ constant is sinked or hoisted. But as it can be seen from the 'sub (sub C, %x), %y' test, that actually conceals the issues it is supposed to test. At least two more patterns are unhandled: * 'add (sub C, %x), %y' - D62266 * 'sub (sub C, %x), %y' llvm-svn: 362295
2019-06-01 19:08:29 +08:00
%t0 = sub <4 x i32> <i32 42, i32 24, i32 undef, i32 46>, %a
%r = sub <4 x i32> %b, %t0
[NFC][X86][AArch64] Rewrite sink-addsub-of-const.ll tests to have full permutation coverage Somehow missed some patterns initially.. While there, add comments. llvm-svn: 361390
2019-05-22 22:42:41 +08:00
ret <4 x i32> %r
}