diff options
| author | Richard Sandiford <richard.sandiford@arm.com> | 2019-12-10 11:30:49 +0000 |
|---|---|---|
| committer | Richard Sandiford <rsandifo@gcc.gnu.org> | 2019-12-10 11:30:49 +0000 |
| commit | ca49c831ca9c042db6cea746c21040c1ada18abb (patch) | |
| tree | 9e6852083c60a649efeb63abd2e0bc9a964348ef /gcc/tree-vect-loop.c | |
| parent | 9995ce07f190d963ec52f52697ccc155dd079568 (diff) | |
| download | gcc-ca49c831ca9c042db6cea746c21040c1ada18abb.zip gcc-ca49c831ca9c042db6cea746c21040c1ada18abb.tar.gz gcc-ca49c831ca9c042db6cea746c21040c1ada18abb.tar.bz2 | |
Disallow EXTRACT_LAST_REDUCTION for reduction chains
gcc.dg/vect/vect-cond-reduc-5.c was ICEing for SVE because we
tried to use an extract-last reduction for a chain of COND_EXPRs.
Adding support for the chained case would be too invasive for stage 3
so this patch explicitly forbids it instead. I've filed PR92884 for
the possible future work.
2019-12-10 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-vect-loop.c (vectorizable_reduction): Don't use
EXTRACT_LAST_REDUCTION for chained reductions.
From-SVN: r279161
Diffstat (limited to 'gcc/tree-vect-loop.c')
| -rw-r--r-- | gcc/tree-vect-loop.c | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/gcc/tree-vect-loop.c b/gcc/tree-vect-loop.c index ddc1d83..50f1a22 100644 --- a/gcc/tree-vect-loop.c +++ b/gcc/tree-vect-loop.c @@ -6196,8 +6196,9 @@ vectorizable_reduction (stmt_vec_info stmt_info, slp_tree slp_node, return false; } - if (direct_internal_fn_supported_p (IFN_FOLD_EXTRACT_LAST, - vectype_in, OPTIMIZE_FOR_SPEED)) + if (reduc_chain_length == 1 + && direct_internal_fn_supported_p (IFN_FOLD_EXTRACT_LAST, + vectype_in, OPTIMIZE_FOR_SPEED)) { if (dump_enabled_p ()) dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, |