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

tree-optimization/115868 - ICE with .MASK_CALL in simdclone

Browse source 
The following adjusts mask recording which didn't take into account
that we can merge call arguments from two vectors like

  _50 = {vect_d_1.253_41, vect_d_1.254_43};
  _51 = VIEW_CONVERT_EXPR<unsigned char>(mask__19.257_49);
  _52 = (unsigned int) _51;
  _53 = _Z3bazd.simdclone.7 (_50, _52);
  _54 = BIT_FIELD_REF <_53, 256, 0>;
  _55 = BIT_FIELD_REF <_53, 256, 256>;

The testcase g++.dg/vect/pr68762-2.cc exercises this on x86_64 with
partial vector usage enabled and AVX512 support.

	PR tree-optimization/115868
	* tree-vect-stmts.cc (vectorizable_simd_clone_call): Correctly
	compute the number of mask copies required for vect_record_loop_mask.
This commit is contained in:
Richard Biener 2024-07-11 09:56:56 +02:00 • committed by Richard Biener
parent 2ee5b58be6
commit abf3964711
1 changed files with 8 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
gcc/tree-vect-stmts.cc
View file

@ -4349,9 +4349,14 @@ vectorizable_simd_clone_call (vec_info *vinfo, stmt_vec_info stmt_info,
case SIMD_CLONE_ARG_TYPE_MASK:
if (loop_vinfo
&& LOOP_VINFO_CAN_USE_PARTIAL_VECTORS_P (loop_vinfo))
vect_record_loop_mask (loop_vinfo,
&LOOP_VINFO_MASKS (loop_vinfo),
ncopies, vectype, op);
{
unsigned nmasks
= exact_div (ncopies * bestn->simdclone->simdlen,
TYPE_VECTOR_SUBPARTS (vectype)).to_constant ();
vect_record_loop_mask (loop_vinfo,
&LOOP_VINFO_MASKS (loop_vinfo),
nmasks, vectype, op);
}
break;
}