procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix another thinko in peeling for gap compute of get_group_load_store_type

Browse source 
There's inconsistent handling of the cpart_size == cnunits which
currently avoids reporting peeling for gaps being insufficient, but
the following condition which is enough to trigger it,
cremain + group_size < cpart_size with cpart_size == cnunits is
equal to the condition that brings us here in the first place,
maybe_lt (remain + group_size, nunits).  The following fixes this
by not checking cpart_size against special values.

	* tree-vect-stmts.cc (get_group_load_store_type): Do not
	exempt cpart_size == cnunits from failing.
This commit is contained in:
Richard Biener 2024-11-14 14:22:01 +01:00 • committed by Richard Biener
parent faf594ce21
commit 119c1b76e2
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
gcc/tree-vect-stmts.cc
View file

@ -2196,7 +2196,7 @@ get_group_load_store_type (vec_info *vinfo, stmt_vec_info stmt_info,
if (!nunits.is_constant (&cnunits)
|| !LOOP_VINFO_VECT_FACTOR (loop_vinfo).is_constant (&cvf)
|| (((cremain = (group_size * cvf - gap) % cnunits), true)
&& ((cpart_size = (1 << ceil_log2 (cremain))) != cnunits)
&& ((cpart_size = (1 << ceil_log2 (cremain))), true)
&& (cremain + group_size < cpart_size
|| vector_vector_composition_type
(vectype, cnunits / cpart_size,