Re: gcc/gcc ChangeLog tree-ssa-loop-im.c

[Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>]

Hello,

> On May 10, 2005, at 4:04 PM, rakdver@gcc.gnu.org wrote:
> >	* tree-ssa-loop-im.c: Include hashtab.h.
> >	(struct mem_ref_loc): New.
> >	(struct mem_ref): Describe the set of references with the same
> >	shape.
> >	(max_stmt_uid, get_stmt_uid, record_mem_ref, free_mem_refs,
> >	maybe_queue_var, fem_single_reachable_address,
> >	for_each_memref, single_reachable_address,
> >	is_call_clobbered_ref, determine_lsm_reg): Removed.
> >	(record_mem_ref_loc, free_mem_ref_locs, determine_lsm_ref,
> >	hoist_memory_reference, memref_hash, memref_eq, memref_del,
> >	gather_mem_refs_stmt, gather_mem_refs, find_more_ref_vops):
> >	New functions.
> >	(rewrite_mem_refs, schedule_sm): Use mem_ref_loc list.
> >	(determine_lsm_loop): Rewritten.
> >	(determine_lsm): Do not set stmt uids.
> 
> This patch causes a bootstrap failure on powerpc-darwin.
> The problem is that we get the following differences for emit-rtl.c
> between stage1 and stage2, this usually means that the hastable is
> traversed but the hastable is not stable across the compiling:

I cannot reproduce the problem (using powerpc-linux and
powerpc64-linux).

Zdenek