mirrors/gcc
0
0
Fork 0
mirror of https://gcc.gnu.org/git/gcc.git synced 2025-01-20 19:33:58 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

vect, aarch64: Fix alignment units for IFN_MASK* [PR95401]

Browse Source 
The IFN_MASK* functions take two leading arguments: a load or
store pointer and a “cookie”.  The type of the cookie is the
type of the access for TBAA purposes (like for MEM_REFs)
while the value of the cookie is the alignment of the access.
This PR was caused by a disagreement about whether the alignment
is measured in bits or bytes.

It looks like this goes back to PR68786, which made the
vectoriser create its own cookie argument rather than reusing
the one created by ifcvt.  The alignment value of the new cookie
was measured in bytes (as needed by set_ptr_info_alignment)
while the existing code expected it to be measured in bits.
The folds I added for IFN_MASK_LOAD and STORE then made
things worse.

gcc/
	PR tree-optimization/95401
	* config/aarch64/aarch64-sve-builtins.cc
	(gimple_folder::load_store_cookie): Use bits rather than bytes
	for the alignment argument to IFN_MASK_LOAD and IFN_MASK_STORE.
	* gimple-fold.c (gimple_fold_mask_load_store_mem_ref): Likewise.
	* tree-vect-stmts.c (vectorizable_store): Likewise.
	(vectorizable_load): Likewise.

gcc/testsuite/
	PR tree-optimization/95401
	* g++.dg/vect/pr95401.cc: New test.
	* g++.dg/vect/pr95401a.cc: Likewise.
This commit is contained in:
Richard Sandiford 2021-01-04 14:44:21 +00:00
parent 6288183377
commit aa204d5118
5 changed files with 36 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
gcc/config/aarch64/aarch64-sve-builtins.cc
View File

@ -2580,7 +2580,7 @@ gimple_folder::fold_contiguous_base (gimple_seq &stmts, tree vectype)
tree
gimple_folder::load_store_cookie (tree type)
{
return build_int_cst (build_pointer_type (type), TYPE_ALIGN_UNIT (type));
return build_int_cst (build_pointer_type (type), TYPE_ALIGN (type));
}
/* Fold the call to a call to INSTANCE, with the same arguments. */

2
gcc/gimple-fold.c
View File

@ -5201,7 +5201,7 @@ gimple_fold_mask_load_store_mem_ref (gcall *call, tree vectype)
if (!tree_fits_uhwi_p (alias_align) || !integer_all_onesp (mask))
return NULL_TREE;
unsigned HOST_WIDE_INT align = tree_to_uhwi (alias_align) * BITS_PER_UNIT;
unsigned HOST_WIDE_INT align = tree_to_uhwi (alias_align);
if (TYPE_ALIGN (vectype) != align)
vectype = build_aligned_type (vectype, align);
tree offset = build_zero_cst (TREE_TYPE (alias_align));

13
gcc/testsuite/g++.dg/vect/pr95401.cc Normal file
View File

@ -0,0 +1,13 @@
// { dg-additional-options "-mavx2 -O3" { target avx2_runtime } }
// { dg-additional-sources pr95401a.cc }
extern int var_9;
extern unsigned var_14;
extern int arr_16[];
#include <algorithm>
void test() {
for (short a = 0; a < (short)var_9; a += 12140)
for (short b = 0; b < 8; b++)
if (std::max(var_14, 1U))
arr_16[a + b] = 0;
}

13
gcc/testsuite/g++.dg/vect/pr95401a.cc Normal file
View File

@ -0,0 +1,13 @@
// { dg-do compile }
#include "../../gcc.dg/vect/tree-vect.h"
int var_9 = 1693986256, var_14;
int arr_16[11];
void test();
int main()
{
check_vect();
test();
return 0;
}

14
gcc/tree-vect-stmts.c
View File

@ -7964,7 +7964,7 @@ vectorizable_store (vec_info *vinfo,
/* Emit:
MASK_STORE_LANES (DATAREF_PTR, ALIAS_PTR, VEC_MASK,
VEC_ARRAY). */
unsigned int align = TYPE_ALIGN_UNIT (TREE_TYPE (vectype));
unsigned int align = TYPE_ALIGN (TREE_TYPE (vectype));
tree alias_ptr = build_int_cst (ref_type, align);
call = gimple_build_call_internal (IFN_MASK_STORE_LANES, 4,
dataref_ptr, alias_ptr,
@ -8079,7 +8079,7 @@ vectorizable_store (vec_info *vinfo,
if (final_mask)
{
align = least_bit_hwi (misalign | align);
tree ptr = build_int_cst (ref_type, align);
tree ptr = build_int_cst (ref_type, align * BITS_PER_UNIT);
gcall *call
= gimple_build_call_internal (IFN_MASK_STORE, 4,
dataref_ptr, ptr,
@ -8094,7 +8094,7 @@ vectorizable_store (vec_info *vinfo,
= vect_get_loop_len (loop_vinfo, loop_lens,
vec_num * ncopies, vec_num * j + i);
align = least_bit_hwi (misalign | align);
tree ptr = build_int_cst (ref_type, align);
tree ptr = build_int_cst (ref_type, align * BITS_PER_UNIT);
machine_mode vmode = TYPE_MODE (vectype);
opt_machine_mode new_ovmode
= get_len_load_store_mode (vmode, false);
@ -9246,7 +9246,7 @@ vectorizable_load (vec_info *vinfo,
/* Emit:
VEC_ARRAY = MASK_LOAD_LANES (DATAREF_PTR, ALIAS_PTR,
VEC_MASK). */
unsigned int align = TYPE_ALIGN_UNIT (TREE_TYPE (vectype));
unsigned int align = TYPE_ALIGN (TREE_TYPE (vectype));
tree alias_ptr = build_int_cst (ref_type, align);
call = gimple_build_call_internal (IFN_MASK_LOAD_LANES, 3,
dataref_ptr, alias_ptr,
@ -9347,7 +9347,8 @@ vectorizable_load (vec_info *vinfo,
if (final_mask)
{
align = least_bit_hwi (misalign | align);
tree ptr = build_int_cst (ref_type, align);
tree ptr = build_int_cst (ref_type,
align * BITS_PER_UNIT);
gcall *call
= gimple_build_call_internal (IFN_MASK_LOAD, 3,
dataref_ptr, ptr,
@ -9363,7 +9364,8 @@ vectorizable_load (vec_info *vinfo,
vec_num * ncopies,
vec_num * j + i);
align = least_bit_hwi (misalign | align);
tree ptr = build_int_cst (ref_type, align);
tree ptr = build_int_cst (ref_type,
align * BITS_PER_UNIT);
gcall *call
= gimple_build_call_internal (IFN_LEN_LOAD, 3,
dataref_ptr, ptr,