diff options
| author | Robin Dapp <rdapp@ventanamicro.com> | 2023-11-06 11:24:37 +0100 |
|---|---|---|
| committer | Robin Dapp <rdapp@ventanamicro.com> | 2023-11-07 22:33:52 +0100 |
| commit | fd940d248bfccb6994794152681dc4c693160919 (patch) | |
| tree | 92ef48d71aa921bbff8525139d4e5b5308c35e25 /gcc/tree-vect-loop.cc | |
| parent | 8f2a59c2629f634e0ea7b2bcc4443fd57c2a0e84 (diff) | |
| download | gcc-fd940d248bfccb6994794152681dc4c693160919.zip gcc-fd940d248bfccb6994794152681dc4c693160919.tar.gz gcc-fd940d248bfccb6994794152681dc4c693160919.tar.bz2 | |
vect/ifcvt: Add vec_cond fallback and check for vector versioning.
This restricts tree-ifcvt to only create COND_OPs when we versioned the
loop for vectorization. Apart from that it re-creates a VEC_COND_EXPR
in vect_expand_fold_left if we emitted a COND_OP.
gcc/ChangeLog:
PR tree-optimization/112361
PR target/112359
PR middle-end/112406
* tree-if-conv.cc (convert_scalar_cond_reduction): Remember if
loop was versioned and only then create COND_OPs.
(predicate_scalar_phi): Do not create COND_OP when not
vectorizing.
* tree-vect-loop.cc (vect_expand_fold_left): Re-create
VEC_COND_EXPR.
(vectorize_fold_left_reduction): Pass mask to
vect_expand_fold_left.
gcc/testsuite/ChangeLog:
* gcc.dg/pr112359.c: New test.
Diffstat (limited to 'gcc/tree-vect-loop.cc')
| -rw-r--r-- | gcc/tree-vect-loop.cc | 22 |
1 files changed, 20 insertions, 2 deletions
diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index 5213aa0..a544bc9 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -6908,11 +6908,13 @@ merge_with_identity (gimple_stmt_iterator *gsi, tree mask, tree vectype, /* Successively apply CODE to each element of VECTOR_RHS, in left-to-right order, starting with LHS. Insert the extraction statements before GSI and associate the new scalar SSA names with variable SCALAR_DEST. + If MASK is nonzero mask the input and then operate on it unconditionally. Return the SSA name for the result. */ static tree vect_expand_fold_left (gimple_stmt_iterator *gsi, tree scalar_dest, - tree_code code, tree lhs, tree vector_rhs) + tree_code code, tree lhs, tree vector_rhs, + tree mask) { tree vectype = TREE_TYPE (vector_rhs); tree scalar_type = TREE_TYPE (vectype); @@ -6920,6 +6922,21 @@ vect_expand_fold_left (gimple_stmt_iterator *gsi, tree scalar_dest, unsigned HOST_WIDE_INT vec_size_in_bits = tree_to_uhwi (TYPE_SIZE (vectype)); unsigned HOST_WIDE_INT element_bitsize = tree_to_uhwi (bitsize); + /* Re-create a VEC_COND_EXPR to mask the input here in order to be able + to perform an unconditional element-wise reduction of it. */ + if (mask) + { + tree masked_vector_rhs = make_temp_ssa_name (vectype, NULL, + "masked_vector_rhs"); + tree neutral_op = neutral_op_for_reduction (scalar_type, code, NULL_TREE, + false); + tree vector_identity = build_vector_from_val (vectype, neutral_op); + gassign *select = gimple_build_assign (masked_vector_rhs, VEC_COND_EXPR, + mask, vector_rhs, vector_identity); + gsi_insert_before (gsi, select, GSI_SAME_STMT); + vector_rhs = masked_vector_rhs; + } + for (unsigned HOST_WIDE_INT bit_offset = 0; bit_offset < vec_size_in_bits; bit_offset += element_bitsize) @@ -7156,7 +7173,8 @@ vectorize_fold_left_reduction (loop_vec_info loop_vinfo, else { reduc_var = vect_expand_fold_left (gsi, scalar_dest_var, - tree_code (code), reduc_var, def0); + tree_code (code), reduc_var, def0, + mask); new_stmt = SSA_NAME_DEF_STMT (reduc_var); /* Remove the statement, so that we can use the same code paths as for statements that we've just created. */ |