This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Update in regclass.c breaks bootstrap.
- To: Toon Moene <toon at moene dot indiv dot nluug dot nl>
- Subject: Re: Update in regclass.c breaks bootstrap.
- From: Richard Henderson <rth at cygnus dot com>
- Date: Sun, 28 May 2000 13:40:56 -0700
- Cc: gcc at gcc dot gnu dot org, gcc-patches at gcc dot gnu dot org
- References: <3930F88C.ECF95C38@moene.indiv.nluug.nl>
On Sun, May 28, 2000 at 12:44:28PM +0200, Toon Moene wrote:
> I'm sorry I have to report that someone on the Left Side of the Atlantic
> went to bed without performing the required bootstrap:
Not exactly. Geoff bootstrapped on two of the systems
in which `reg_changes_size' does get defined. ;-)
r~
* regclass.c (record_operand_costs): Protect reg_changes_size
with CLASS_CANNOT_CHANGE_SIZE.
Index: regclass.c
===================================================================
RCS file: /cvs/gcc/egcs/gcc/regclass.c,v
retrieving revision 1.99
diff -c -p -d -r1.99 regclass.c
*** regclass.c 2000/05/28 05:51:16 1.99
--- regclass.c 2000/05/28 20:36:37
*************** record_operand_costs (insn, op_costs, re
*** 859,867 ****
--- 859,869 ----
if (GET_CODE (recog_data.operand[i]) == SUBREG)
{
rtx inner = SUBREG_REG (recog_data.operand[i]);
+ #ifdef CLASS_CANNOT_CHANGE_SIZE
if (GET_MODE_SIZE (modes[i]) != GET_MODE_SIZE (GET_MODE (inner))
&& GET_CODE (inner) == REG)
SET_REGNO_REG_SET (reg_changes_size, REGNO (inner));
+ #endif
recog_data.operand[i] = inner;
}