This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] alias_invariant
On May 21, 2004, at 4:37 PM, Richard Henderson wrote:
On Fri, May 21, 2004 at 04:32:42PM -0700, Devang Patel wrote:
! alias_invariant = ggc_realloc (alias_invariant,
! maxreg * sizeof (rtx));
memset (alias_invariant, 0, maxreg * sizeof (rtx));
Not ok. Use ggc_calloc, and...
- free (alias_invariant);
ggc_free. The realloc-ness is stupid and useless.
It seems someone stopped short of removing realloc but left comment...
How about this one? I am restarting bootstrap and DejaGN test run...
--
Devang
2004-05-21 Devang Patel <dpatel@apple.com>
* alias.c (init_alias_analysis): Use ggc_calloc instead of
xrealloc.
(end_alias_analysis): Use ggc_free instead of free.
Index: alias.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/alias.c,v
retrieving revision 1.224
diff -c -3 -p -r1.224 alias.c
*** alias.c 30 Mar 2004 19:18:48 -0000 1.224
--- alias.c 21 May 2004 23:49:10 -0000
*************** init_alias_analysis (void)
*** 2827,2836 ****
reg_seen = xmalloc (maxreg);
if (! reload_completed && flag_old_unroll_loops)
{
! /* ??? Why are we realloc'ing if we're just going to zero it?
*/
! alias_invariant = xrealloc (alias_invariant,
! maxreg * sizeof (rtx));
! memset (alias_invariant, 0, maxreg * sizeof (rtx));
alias_invariant_size = maxreg;
}
--- 2827,2833 ----
reg_seen = xmalloc (maxreg);
if (! reload_completed && flag_old_unroll_loops)
{
! alias_invariant = ggc_calloc (maxreg, sizeof (rtx));
alias_invariant_size = maxreg;
}
*************** end_alias_analysis (void)
*** 3031,3037 ****
reg_known_equiv_p = 0;
if (alias_invariant)
{
! free (alias_invariant);
alias_invariant = 0;
alias_invariant_size = 0;
}
--- 3028,3034 ----
reg_known_equiv_p = 0;
if (alias_invariant)
{
! ggc_free (alias_invariant);
alias_invariant = 0;
alias_invariant_size = 0;
}