This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug rtl-optimization/39543] [4.4 Regression] Reload failure on mplayer from SVN
- From: "jakub at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 25 Mar 2009 13:10:33 -0000
- Subject: [Bug rtl-optimization/39543] [4.4 Regression] Reload failure on mplayer from SVN
- References: <bug-39543-87@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #5 from jakub at gcc dot gnu dot org 2009-03-25 13:10 -------
That can be solved easily, just compare whether new_rtx doesn't need more
registers than old_rtx and only propagate into asm_noperands () >= 0 insns
those that require the same number or fewer registers.
Consider e.g.:
int s[128];
void
f1 (void)
{
int i;
asm volatile ("# %0 %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16
%17"
: "=r" (i)
: "m" (s[0]), "m" (s[2]), "m" (s[4]), "m" (s[6]), "m" (s[8]),
"m" (s[10]), "m" (s[12]), "m" (s[14]), "m" (s[16]), "m" (s[18]),
"m" (s[20]), "m" (s[22]), "m" (s[24]), "m" (s[26]), "m" (s[28]),
"m" (s[30]), "m" (s[32]));
asm volatile ("# %0 %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16
%17"
: "=r" (i)
: "m" (s[0]), "m" (s[2]), "m" (s[4]), "m" (s[6]), "m" (s[8]),
"m" (s[10]), "m" (s[12]), "m" (s[14]), "m" (s[16]), "m" (s[18]),
"m" (s[20]), "m" (s[22]), "m" (s[24]), "m" (s[26]), "m" (s[28]),
"m" (s[30]), "m" (s[32]));
}
void
f2 (int *q)
{
int i;
int *p = q + 32;
asm volatile ("# %0 %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16
%17"
: "=r" (i)
: "m" (p[0]), "m" (p[2]), "m" (p[4]), "m" (p[6]), "m" (p[8]),
"m" (p[10]), "m" (p[12]), "m" (p[14]), "m" (p[16]), "m" (p[18]),
"m" (p[20]), "m" (p[22]), "m" (p[24]), "m" (p[26]), "m" (p[28]),
"m" (p[30]), "m" (p[32]));
asm volatile ("# %0 %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16
%17"
: "=r" (i)
: "m" (p[0]), "m" (p[2]), "m" (p[4]), "m" (p[6]), "m" (p[8]),
"m" (p[10]), "m" (p[12]), "m" (p[14]), "m" (p[16]), "m" (p[18]),
"m" (p[20]), "m" (p[22]), "m" (p[24]), "m" (p[26]), "m" (p[28]),
"m" (p[30]), "m" (p[32]));
}
at -O2, here cse_not_expected hack in stmt.c doesn't help, but I'd say fwprop
could fix it up.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39543