This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: bugfix for cselib_invalidate_regno
- From: Richard Earnshaw <rearnsha at arm dot com>
- To: Denis Chertykov <denisc at overta dot ru>
- Cc: gcc-patches at gcc dot gnu dot org, Mike Panetta <mike dot panetta at netplanner dot com>, AVR GCC List <avr-gcc-list at avr1 dot org>, Richard dot Earnshaw at arm dot com
- Date: Thu, 22 May 2003 16:09:27 +0100
- Subject: Re: bugfix for cselib_invalidate_regno
- Organization: ARM Ltd.
- Reply-to: Richard dot Earnshaw at arm dot com
> Denis Chertykov <denisc@overta.ru> writes:
>
> > I have founded a bug in cselib.c:cselib_invalidate_regno.
> > The avr port triggers this bug.
> > Current version of cselib_invalidate_regno didn't invalidate lower
> > register numbers if they contain values that overlap REGNO in case
> > that (regno < FIRST_PSEUDO_REGISTER && mode == VOIDmode). This happens
> > if cselib_invalidate_regno called from cselib_process_insn which
> > handle CALL_INSN.
> > IE
> >
> > /* If this is a call instruction, forget anything stored in a
> > call clobbered register, or, if this is not a const call, in
> > memory. */
> > if (GET_CODE (insn) == CALL_INSN)
> > {
> > for (i = 0; i < FIRST_PSEUDO_REGISTER; i++)
> > if (call_used_regs[i])
> > cselib_invalidate_regno (i, VOIDmode);
> >
> >
> > Can I apply the following patch to mainline and 3.3 branches ?
> >
> > Denis.
> >
> >
> > Index: cselib.c
> > ===================================================================
> > RCS file: /cvsroot/gcc/egcs/gcc/cselib.c,v
> > retrieving revision 1.27
> > diff -c -3 -p -r1.27 cselib.c
> > *** cselib.c 17 Apr 2003 01:07:12 -0000 1.27
> > --- cselib.c 19 May 2003 11:09:38 -0000
> > *************** cselib_invalidate_regno (regno, mode)
> > *** 1019,1032 ****
> > pseudos, only REGNO is affected. For hard regs, we must take MODE
> > into account, and we must also invalidate lower register numbers
> > if they contain values that overlap REGNO. */
> > ! if (regno < FIRST_PSEUDO_REGISTER && mode != VOIDmode)
> > {
> > if (regno < max_value_regs)
> > i = 0;
> > else
> > i = regno - max_value_regs;
> >
> > ! endregno = regno + HARD_REGNO_NREGS (regno, mode);
> > }
> > else
> > {
> > --- 1019,1033 ----
> > pseudos, only REGNO is affected. For hard regs, we must take MODE
> > into account, and we must also invalidate lower register numbers
> > if they contain values that overlap REGNO. */
> > ! if (regno < FIRST_PSEUDO_REGISTER)
> > {
> > if (regno < max_value_regs)
> > i = 0;
> > else
> > i = regno - max_value_regs;
> >
> > ! endregno = regno
> > ! + (mode != VOIDmode ? HARD_REGNO_NREGS (regno, mode) : 1);
> > }
> > else
> > {
>
> Anybody, please, review my tiny patch.
>
> Denis.
>
Why not change the caller to pass reg_raw_mode[regno]. That seems much
cleaner to me than pretending a hard register can have no mode.
R.