target/6569: sparc-sun-solaris2.7 C testsuite regression in compile/20011119-2.c
ghazi@caip.rutgers.edu
ghazi@caip.rutgers.edu
Sun May 5 21:56:00 GMT 2002
>Number: 6569
>Category: target
>Synopsis: sparc-sun-solaris2.7 C testsuite regression in compile/20011119-2.c
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: rejects-legal
>Submitter-Id: net
>Arrival-Date: Sun May 05 21:56:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator: Kaveh Ghazi
>Release: gcc version 3.1 20020505 (prerelease)
>Organization:
>Environment:
sparc-sun-solaris2.7 native as/ld
>Description:
Sometime between 4/26 and 4/29, compile/20011119-2.c regressed. It worked here:
http://gcc.gnu.org/ml/gcc-testresults/2002-04/msg01050.html
http://gcc.gnu.org/ml/gcc-testresults/2002-04/msg01168.html
It fails here:
http://gcc.gnu.org/ml/gcc-testresults/2002-04/msg01165.html
http://gcc.gnu.org/ml/gcc-testresults/2002-04/msg01120.html
When I compile it by hand using today's checkout, I get:
compile/20011119-2.c:1: warning: weak declaration of `foo' after first use results in unspecified behavior
/usr/ccs/bin/as: "/var/tmp//ccrCbkea.s", line 51: error: invalid operand
I've attached the .s output.
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
----gnatsweb-attachment----
Content-Type: text/plain; name="20011119-2.s"
Content-Disposition: inline; filename="20011119-2.s"
.file "20011119-2.c"
.section ".text"
.align 4
.global bar
.type bar,#function
.proc 04
bar:
!#PROLOGUE# 0
save %sp, -112, %sp
!#PROLOGUE# 1
call foo, 0
nop
mov %o0, %i0
nop
ret
restore
.LLfe1:
.size bar,.LLfe1-bar
.weak foo
foo = xxx
.align 4
.global baz
.type baz,#function
.proc 04
baz:
!#PROLOGUE# 0
save %sp, -112, %sp
!#PROLOGUE# 1
call foo, 0
nop
mov %o0, %i0
nop
ret
restore
.LLfe2:
.size baz,.LLfe2-baz
.align 4
.global xxx
.type xxx,#function
.proc 04
xxx:
!#PROLOGUE# 0
save %sp, -112, %sp
!#PROLOGUE# 1
mov 23, %i0
nop
ret
restore
.LLfe3:
.size xxx,.LLfe3-xxx
.weak foo
.ident "GCC: (GNU) 3.1 20020505 (prerelease)"
More information about the Gcc-bugs
mailing list