This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug rtl-optimization/15921] [3.5 Regression] gcc.c-torture/compile/920501-4.c fails on powerpc-*
- From: "paolo dot bonzini at polimi dot it" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 11 Jun 2004 07:17:04 -0000
- Subject: [Bug rtl-optimization/15921] [3.5 Regression] gcc.c-torture/compile/920501-4.c fails on powerpc-*
- References: <20040610225843.15921.pinskia@gcc.gnu.org>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From paolo dot bonzini at polimi dot it 2004-06-11 07:16 -------
Subject: Re: New: [3.5 Regression] gcc.c-torture/compile/920501-4.c
fails on powerpc-*
pinskia at gcc dot gnu dot org wrote:
> /Users/regress/tbox/cvs-gcc/gcc/gcc/testsuite/gcc.c-torture/compile/920501-4.c: In function `foo':
> ^M
> /Users/regress/tbox/cvs-gcc/gcc/gcc/testsuite/gcc.c-torture/compile/920501-4.c:53: error: insn
> does not satisfy its constraints:^M
> (insn 171 169 173 0 (parallel [^M
> (set (reg:SI 6 r6 [186])^M
> (lshiftrt:SI (neg:SI (abs:SI (reg/v:SI 0 r0 [orig:132 m_max ] [132])))^M
> (const_int 31 [0x1f])))^M
> (clobber (reg:SI 0 r0))^M
> ]) 419 {*ne0} (nil)^M
> (nil))^M
It seems more likely that Vladimir's patch causes it since it is post
global register allocation (and similarly for x86's FAIL on
gcc.c-torture/compile/950613-1.c).
Paolo
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15921