diff options
| author | Ju-Zhe Zhong <juzhe.zhong@rivai.ai> | 2023-07-06 14:51:35 +0800 |
|---|---|---|
| committer | Pan Li <pan2.li@intel.com> | 2023-07-06 15:12:13 +0800 |
| commit | f2b37c8a38e44696287b19856bcbbf52e8b004f8 (patch) | |
| tree | 6ee905f3e12addf1ed66ad67d2edb7d2456e210e /gcc/tree-vect-stmts.cc | |
| parent | 2ab065c3e6950bc4995fcb16cebc5164d6213708 (diff) | |
| download | gcc-f2b37c8a38e44696287b19856bcbbf52e8b004f8.zip gcc-f2b37c8a38e44696287b19856bcbbf52e8b004f8.tar.gz gcc-f2b37c8a38e44696287b19856bcbbf52e8b004f8.tar.bz2 | |
VECT: Fix ICE of variable stride on strieded load/store with SELECT_VL loop control.
Hi, Richi.
Sorry for making mistake on LEN_MASK_GATHER_LOAD/LEN_MASK_SCATTER_STORE
with SELECT_VL loop control.
Consider this following case:
void __attribute__ ((noinline, noclone)) \
f_##DATA_TYPE##_##BITS (DATA_TYPE *restrict dest, DATA_TYPE *restrict src, \
INDEX##BITS stride, INDEX##BITS n) \
{ \
for (INDEX##BITS i = 0; i < n; ++i) \
dest[i] += src[i * stride]; \
}
When "stride" is a constant, current flow works fine.
However, when "stride" is a variable. It causes an ICE:
...
_96 = .SELECT_VL (ivtmp_94, 4);
...
ivtmp_78 = ((sizetype) _39 * (sizetype) _96) * 4;
vect__11.69_87 = .LEN_MASK_GATHER_LOAD (vectp_src.67_85, _84, 4, { 0, 0, 0, 0 }, { -1, -1, -1, -1 }, _96, 0);
...
vectp_src.67_86 = vectp_src.67_85 + ivtmp_78;
Becase the IR: ivtmp_78 = ((sizetype) _39 * (sizetype) _96) * 4;
Instead, I split the IR into:
step_stride = _39
step = step_stride * 4
ivtmp_78 = step * _96
Thanks.
gcc/ChangeLog:
* tree-vect-stmts.cc (vect_get_strided_load_store_ops): Fix ICE.
Diffstat (limited to 'gcc/tree-vect-stmts.cc')
| -rw-r--r-- | gcc/tree-vect-stmts.cc | 6 |
1 files changed, 2 insertions, 4 deletions
diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index c10a4be..10e7117 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -3176,10 +3176,8 @@ vect_get_strided_load_store_ops (stmt_vec_info stmt_info, = fold_build2 (MULT_EXPR, sizetype, fold_convert (sizetype, unshare_expr (DR_STEP (dr))), loop_len); - tree bump = make_temp_ssa_name (sizetype, NULL, "ivtmp"); - gassign *assign = gimple_build_assign (bump, tmp); - gsi_insert_before (gsi, assign, GSI_SAME_STMT); - *dataref_bump = bump; + *dataref_bump = force_gimple_operand_gsi (gsi, tmp, true, NULL_TREE, true, + GSI_SAME_STMT); } else { |