diff options
| author | Tamar Christina <tamar.christina@arm.com> | 2023-10-18 09:33:30 +0100 |
|---|---|---|
| committer | Tamar Christina <tamar.christina@arm.com> | 2023-10-18 09:53:47 +0100 |
| commit | b0fe8f2f960d746e61debd61655f231f503bccaa (patch) | |
| tree | aa9dcd1d435b37c2d27ff4f3f8afc263e5842420 /gcc | |
| parent | 4b39aeef594f311e2c1715f15608f1d7ebc2d868 (diff) | |
| download | gcc-b0fe8f2f960d746e61debd61655f231f503bccaa.zip gcc-b0fe8f2f960d746e61debd61655f231f503bccaa.tar.gz gcc-b0fe8f2f960d746e61debd61655f231f503bccaa.tar.bz2 | |
middle-end: ifcvt: Allow any const IFN in conditional blocks
When ifcvt was initially added masking was not a thing and as such it was
rather conservative in what it supported.
For builtins it only allowed C99 builtin functions which it knew it can fold
away.
These days the vectorizer is able to deal with needing to mask IFNs itself.
vectorizable_call is able vectorize the IFN by emitting a VEC_PERM_EXPR after
the operation to emulate the masking.
This is then used by match.pd to conver the IFN into a masked variant if it's
available.
For these reasons the restriction in ifconvert is no longer require and we
needless block vectorization when we can effectively handle the operations.
Bootstrapped Regtested on aarch64-none-linux-gnu and no issues.
Note: This patch is part of a testseries and tests for it are added in the
AArch64 patch that adds supports for the optab.
gcc/ChangeLog:
PR tree-optimization/109154
* tree-if-conv.cc (if_convertible_stmt_p): Allow any const IFN.
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/tree-if-conv.cc | 15 |
1 files changed, 10 insertions, 5 deletions
diff --git a/gcc/tree-if-conv.cc b/gcc/tree-if-conv.cc index 3b42388..edce842 100644 --- a/gcc/tree-if-conv.cc +++ b/gcc/tree-if-conv.cc @@ -1107,17 +1107,21 @@ if_convertible_stmt_p (gimple *stmt, vec<data_reference_p> refs) case GIMPLE_CALL: { + /* There are some IFN_s that are used to replace builtins but have the + same semantics. Even if MASK_CALL cannot handle them vectorable_call + will insert the proper selection, so do not block conversion. */ + int flags = gimple_call_flags (stmt); + if ((flags & ECF_CONST) + && !(flags & ECF_LOOPING_CONST_OR_PURE) + && gimple_call_combined_fn (stmt) != CFN_LAST) + return true; + tree fndecl = gimple_call_fndecl (stmt); if (fndecl) { /* We can vectorize some builtins and functions with SIMD "inbranch" clones. */ - int flags = gimple_call_flags (stmt); struct cgraph_node *node = cgraph_node::get (fndecl); - if ((flags & ECF_CONST) - && !(flags & ECF_LOOPING_CONST_OR_PURE) - && fndecl_built_in_p (fndecl)) - return true; if (node && node->simd_clones != NULL) /* Ensure that at least one clone can be "inbranch". */ for (struct cgraph_node *n = node->simd_clones; n != NULL; @@ -1129,6 +1133,7 @@ if_convertible_stmt_p (gimple *stmt, vec<data_reference_p> refs) return true; } } + return false; } |