https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122675
Bug ID: 122675
Summary: BB reordering for optimizing jump fallthrough is not
working
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: dimitar at gcc dot gnu.org
Target Milestone: ---
Created attachment 62787
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62787&action=edit
test.c
Starting with r16-4438-ga93f80feeef744, the attached test case does not
optimize the likely jump as a fallthrough BB edge.
The following code for x86 is expected:
test:
testl %edi, %edi
je .L3
subq $8, %rsp
call f1
addq $8, %rsp
ret
.L3:
movl %esi, %eax
ret
Instead this code is generated, with unwanted jump for the likely edge (call to
f1):
test:
.LFB0:
testl %edi, %edi
jne .L8
movl %esi, %eax
ret
.L8:
subq $8, %rsp
call f1
addq $8, %rsp
ret
