Jakub Jelinek <ja...@redhat.com> writes: > On Tue, Apr 17, 2012 at 09:11:37PM +0100, Richard Sandiford wrote: >> >> * rtl.def (ADDRESS): Turn operand into a HOST_WIDE_INT. >> >> * alias.c (reg_base_value): Expand and update comment. >> >> (arg_base_value): New variable. >> >> (unique_id): Move up file. >> >> (unique_base_value, unique_base_value_p, known_base_value_p): New. >> >> (find_base_value): Use arg_base_value and known_base_value_p. >> >> (record_set): Document REG_NOALIAS handling. Use >> >> unique_base_value. >> >> (find_base_term): Use known_base_value_p. >> >> (base_alias_check): Use unique_base_value_p. >> >> (init_alias_target): Initialize arg_base_value. Use >> >> unique_base_value. >> >> (init_alias_analysis): Use 1 as the first id for REG_NOALIAS bases. > > This also broke RTL checking bootstrap everywhere (and broke DSE silently > otherwise). > > Here is a fix, bootstrapped/regtested on x86_64-linux and i686-linux, > ok for trunk? > > 2012-04-20 Jakub Jelinek <ja...@redhat.com> > > PR bootstrap/53021 > * alias.h (UNIQUE_BASE_VALUE_SP, UNIQUE_BASE_VALUE_ARGP, > UNIQUE_BASE_VALUE_FP, UNIQUE_BASE_VALUE_HFP): Define. > * alias.c (init_alias_targets): Use UNIQUE_BASE_VALUE_* > macros instead of constants. > * dse.c (record_store): Check for SP ADDRESS by comparing > XWINT to UNIQUE_BASE_VALUE_SP instead of expecting > XEXP to be stack_pointer_rtx.
Thanks, and sorry for messing this one up rather badly. Richard
