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/tail-dup-repeat.ll

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

79 lines
2.7 KiB
LLVM
Raw Normal View History

[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
Revert "Reland "[DebugInfo] Enable the debug entry values feature by default"" This reverts commit rGfaff707db82d. A failure found on an ARM 2-stage buildbot. The investigation is needed.
2020-02-20 20:43:01 +08:00
; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -debug-entry-values | FileCheck %s
Codegen: Tail-duplicate during placement. The tail duplication pass uses an assumed layout when making duplication decisions. This is fine, but passes up duplication opportunities that may arise when blocks are outlined. Because we want the updated CFG to affect subsequent placement decisions, this change must occur during placement. In order to achieve this goal, TailDuplicationPass is split into a utility class, TailDuplicator, and the pass itself. The pass delegates nearly everything to the TailDuplicator object, except for looping over the blocks in a function. This allows the same code to be used for tail duplication in both places. This change, in concert with outlining optional branches, allows triangle shaped code to perform much better, esepecially when the taken/untaken branches are correlated, as it creates a second spine when the tests are small enough. Issue from previous rollback fixed, and a new test was added for that case as well. Issue was worklist/scheduling/taildup issue in layout. Issue from 2nd rollback fixed, with 2 additional tests. Issue was tail merging/loop info/tail-duplication causing issue with loops that share a header block. Issue with early tail-duplication of blocks that branch to a fallthrough predecessor fixed with test case: tail-dup-branch-to-fallthrough.ll Differential revision: https://reviews.llvm.org/D18226 llvm-svn: 283934
2016-10-12 04:36:43 +08:00
; Function Attrs: uwtable
; When tail-duplicating during placement, we work backward from blocks with
; multiple successors. In this case, the block dup1 gets duplicated into dup2
; and if.then64, and then the block dup2 gets duplicated into land.lhs.true
; and if.end70
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
[X86] Remove icmp undef from reduced tests Pre-commit for D59363 (Add icmp UNDEF handling to SelectionDAG::FoldSetCC) Approved by @spatel (Sanjay Patel) llvm-svn: 356859
2019-03-25 01:02:08 +08:00
define void @repeated_tail_dup(i1 %a1, i1 %a2, i32* %a4, i32* %a5, i8* %a6, i32 %a7) #0 align 2 {
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-LABEL: repeated_tail_dup:
; CHECK: # %bb.0: # %entry
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: jmp .LBB0_1
[MBP] Disable aggressive loop rotate in plain mode Patch https://reviews.llvm.org/D43256 introduced more aggressive loop layout optimization which depends on profile information. If profile information is not available, the statically estimated profile information(generated by BranchProbabilityInfo.cpp) is used. If user program doesn't behave as BranchProbabilityInfo.cpp expected, the layout may be worse. To be conservative this patch restores the original layout algorithm in plain mode. But user can still try the aggressive layout optimization with -force-precise-rotation-cost=true. Differential Revision: https://reviews.llvm.org/D65673 llvm-svn: 369664
2019-08-23 00:21:32 +08:00
; CHECK-NEXT: .p2align 4, 0x90
Revert [MBP] Disable aggressive loop rotate in plain mode This reverts r369664 (git commit 51f48295cbe8fa3a44db263b528dd9f7bae7bf9a) It causes many benchmark regressions, internally and in llvm's benchmark suite. llvm-svn: 370398
2019-08-30 03:03:58 +08:00
; CHECK-NEXT: .LBB0_2: # %land.lhs.true
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # in Loop: Header=BB0_1 Depth=1
Revert [MBP] Disable aggressive loop rotate in plain mode This reverts r369664 (git commit 51f48295cbe8fa3a44db263b528dd9f7bae7bf9a) It causes many benchmark regressions, internally and in llvm's benchmark suite. llvm-svn: 370398
2019-08-30 03:03:58 +08:00
; CHECK-NEXT: movl $10, (%rdx)
; CHECK-NEXT: .LBB0_6: # %dup2
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # in Loop: Header=BB0_1 Depth=1
Revert [MBP] Disable aggressive loop rotate in plain mode This reverts r369664 (git commit 51f48295cbe8fa3a44db263b528dd9f7bae7bf9a) It causes many benchmark regressions, internally and in llvm's benchmark suite. llvm-svn: 370398
2019-08-30 03:03:58 +08:00
; CHECK-NEXT: movl $2, (%rcx)
; CHECK-NEXT: testl %r9d, %r9d
; CHECK-NEXT: jne .LBB0_8
; CHECK-NEXT: .LBB0_1: # %for.cond
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
Revert [MBP] Disable aggressive loop rotate in plain mode This reverts r369664 (git commit 51f48295cbe8fa3a44db263b528dd9f7bae7bf9a) It causes many benchmark regressions, internally and in llvm's benchmark suite. llvm-svn: 370398
2019-08-30 03:03:58 +08:00
; CHECK-NEXT: testb $1, %dil
; CHECK-NEXT: jne .LBB0_2
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # %bb.3: # %if.end56
; CHECK-NEXT: # in Loop: Header=BB0_1 Depth=1
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-NEXT: testb $1, %sil
; CHECK-NEXT: je .LBB0_5
; CHECK-NEXT: # %bb.4: # %if.then64
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # in Loop: Header=BB0_1 Depth=1
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-NEXT: movb $1, (%r8)
[X86] Remove icmp undef from reduced tests Pre-commit for D59363 (Add icmp UNDEF handling to SelectionDAG::FoldSetCC) Approved by @spatel (Sanjay Patel) llvm-svn: 356859
2019-03-25 01:02:08 +08:00
; CHECK-NEXT: testl %r9d, %r9d
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-NEXT: je .LBB0_1
; CHECK-NEXT: jmp .LBB0_8
; CHECK-NEXT: .p2align 4, 0x90
; CHECK-NEXT: .LBB0_5: # %if.end70
[MBP] Avoid tail duplication if it can't bring benefit Current tail duplication integrated in bb layout is designed to increase the fallthrough from a BB's predecessor to its successor, but we have observed cases that duplication doesn't increase fallthrough, or it brings too much size overhead. To overcome these two issues in function canTailDuplicateUnplacedPreds I add two checks: make sure there is at least one duplication in current work set. the number of duplication should not exceed the number of successors. The modification in hasBetterLayoutPredecessor fixes a bug that potential predecessor must be at the bottom of a chain. Differential Revision: https://reviews.llvm.org/D64376
2019-12-05 08:01:20 +08:00
; CHECK-NEXT: # in Loop: Header=BB0_1 Depth=1
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-NEXT: movl $12, (%rdx)
Revert [MBP] Disable aggressive loop rotate in plain mode This reverts r369664 (git commit 51f48295cbe8fa3a44db263b528dd9f7bae7bf9a) It causes many benchmark regressions, internally and in llvm's benchmark suite. llvm-svn: 370398
2019-08-30 03:03:58 +08:00
; CHECK-NEXT: jmp .LBB0_6
[X86] Regenerate tail call tests llvm-svn: 356083
2019-03-14 03:04:45 +08:00
; CHECK-NEXT: .LBB0_8: # %for.end
; CHECK-NEXT: retq
Codegen: Tail-duplicate during placement. The tail duplication pass uses an assumed layout when making duplication decisions. This is fine, but passes up duplication opportunities that may arise when blocks are outlined. Because we want the updated CFG to affect subsequent placement decisions, this change must occur during placement. In order to achieve this goal, TailDuplicationPass is split into a utility class, TailDuplicator, and the pass itself. The pass delegates nearly everything to the TailDuplicator object, except for looping over the blocks in a function. This allows the same code to be used for tail duplication in both places. This change, in concert with outlining optional branches, allows triangle shaped code to perform much better, esepecially when the taken/untaken branches are correlated, as it creates a second spine when the tests are small enough. Issue from previous rollback fixed, and a new test was added for that case as well. Issue was worklist/scheduling/taildup issue in layout. Issue from 2nd rollback fixed, with 2 additional tests. Issue was tail merging/loop info/tail-duplication causing issue with loops that share a header block. Issue with early tail-duplication of blocks that branch to a fallthrough predecessor fixed with test case: tail-dup-branch-to-fallthrough.ll Differential revision: https://reviews.llvm.org/D18226 llvm-svn: 283934
2016-10-12 04:36:43 +08:00
entry:
br label %for.cond
for.cond: ; preds = %dup1, %entry
br i1 %a1, label %land.lhs.true, label %if.end56
land.lhs.true: ; preds = %for.cond
store i32 10, i32* %a4, align 8
br label %dup2
if.end56: ; preds = %for.cond
br i1 %a2, label %if.then64, label %if.end70
if.then64: ; preds = %if.end56
store i8 1, i8* %a6, align 1
br label %dup1
if.end70: ; preds = %if.end56
store i32 12, i32* %a4, align 8
br label %dup2
dup2: ; preds = %if.end70, %land.lhs.true
store i32 2, i32* %a5, align 4
br label %dup1
dup1: ; preds = %dup2, %if.then64
%val = load i32, i32* %a4, align 8
[X86] Remove icmp undef from reduced tests Pre-commit for D59363 (Add icmp UNDEF handling to SelectionDAG::FoldSetCC) Approved by @spatel (Sanjay Patel) llvm-svn: 356859
2019-03-25 01:02:08 +08:00
%switch = icmp ult i32 %a7, 1
Codegen: Tail-duplicate during placement. The tail duplication pass uses an assumed layout when making duplication decisions. This is fine, but passes up duplication opportunities that may arise when blocks are outlined. Because we want the updated CFG to affect subsequent placement decisions, this change must occur during placement. In order to achieve this goal, TailDuplicationPass is split into a utility class, TailDuplicator, and the pass itself. The pass delegates nearly everything to the TailDuplicator object, except for looping over the blocks in a function. This allows the same code to be used for tail duplication in both places. This change, in concert with outlining optional branches, allows triangle shaped code to perform much better, esepecially when the taken/untaken branches are correlated, as it creates a second spine when the tests are small enough. Issue from previous rollback fixed, and a new test was added for that case as well. Issue was worklist/scheduling/taildup issue in layout. Issue from 2nd rollback fixed, with 2 additional tests. Issue was tail merging/loop info/tail-duplication causing issue with loops that share a header block. Issue with early tail-duplication of blocks that branch to a fallthrough predecessor fixed with test case: tail-dup-branch-to-fallthrough.ll Differential revision: https://reviews.llvm.org/D18226 llvm-svn: 283934
2016-10-12 04:36:43 +08:00
br i1 %switch, label %for.cond, label %for.end
for.end: ; preds = %dup1
ret void
}
attributes #0 = { uwtable }