summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/X86/cmov-into-branch.ll
diff options
context:
space:
mode:
authorKyle Butt <kyle+llvm@iteratee.net>2016-10-04 00:00:09 +0000
committerKyle Butt <kyle+llvm@iteratee.net>2016-10-04 00:00:09 +0000
commit396bfdd7076325be148e1e95c433b78849c1083d (patch)
tree0728afd074400877d223036b746cd4bc265ff8fd /llvm/test/CodeGen/X86/cmov-into-branch.ll
parentda68203678e198ef806b40ac23e4375733097592 (diff)
downloadbcm5719-llvm-396bfdd7076325be148e1e95c433b78849c1083d.tar.gz
bcm5719-llvm-396bfdd7076325be148e1e95c433b78849c1083d.zip
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. llvm-svn: 283164
Diffstat (limited to 'llvm/test/CodeGen/X86/cmov-into-branch.ll')
-rw-r--r--llvm/test/CodeGen/X86/cmov-into-branch.ll4
1 files changed, 3 insertions, 1 deletions
diff --git a/llvm/test/CodeGen/X86/cmov-into-branch.ll b/llvm/test/CodeGen/X86/cmov-into-branch.ll
index c0c6fc4ac22..6e4762b2e79 100644
--- a/llvm/test/CodeGen/X86/cmov-into-branch.ll
+++ b/llvm/test/CodeGen/X86/cmov-into-branch.ll
@@ -105,9 +105,11 @@ define i32 @weighted_select3(i32 %a, i32 %b) {
; CHECK-NEXT: testl %edi, %edi
; CHECK-NEXT: je [[LABEL_BB6:.*]]
; CHECK: movl %edi, %eax
+; CHECK-NEXT: retq
; CHECK: [[LABEL_BB6]]
; CHECK-NEXT: movl %esi, %edi
-; CHECK-NEXT: jmp
+; CHECK-NEXT: movl %edi, %eax
+; CHECK-NEXT: retq
;
%cmp = icmp ne i32 %a, 0
%sel = select i1 %cmp, i32 %a, i32 %b, !prof !2
OpenPOWER on IntegriCloud