diff options
| author | Andrew Stubbs <ams@codesourcery.com> | 2022-10-21 14:19:31 +0100 |
|---|---|---|
| committer | Sandra Loosemore <sloosemore@baylibre.com> | 2025-05-15 20:25:47 +0000 |
| commit | 103758a6d5a021b8d703887bbdc990e2b6986c73 (patch) | |
| tree | 5885b15a588af3394c77d9c816140c67a619ac71 /gcc | |
| parent | 41e11972a51b683823ca7f0edd72a19e31957a1e (diff) | |
| download | gcc-103758a6d5a021b8d703887bbdc990e2b6986c73.zip gcc-103758a6d5a021b8d703887bbdc990e2b6986c73.tar.gz gcc-103758a6d5a021b8d703887bbdc990e2b6986c73.tar.bz2 | |
vect: WORKAROUND vectorizer bug
This patch disables vectorization of memory accesses to non-default address
spaces where the pointer size is different to the usual pointer size. This
condition typically occurs in OpenACC programs on amdgcn, where LDS memory is
used for broadcasting gang-private variables between threads. In particular,
see libgomp.oacc-c-c++-common/private-variables.c
The problem is that the address space information is dropped from the various
types in the middle-end and eventually it triggers an ICE trying to do an
address conversion. That ICE can be avoided by defining
POINTERS_EXTEND_UNSIGNED, but that just produces wrong RTL code later on.
A correct solution would ensure that all the vectypes have the correct address
spaces, but I don't have time for that right now.
gcc/ChangeLog:
* tree-vect-data-refs.cc (vect_analyze_data_refs): Workaround an
address-space bug.
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/tree-vect-data-refs.cc | 16 |
1 files changed, 15 insertions, 1 deletions
diff --git a/gcc/tree-vect-data-refs.cc b/gcc/tree-vect-data-refs.cc index 3ba271b..d6cd93a 100644 --- a/gcc/tree-vect-data-refs.cc +++ b/gcc/tree-vect-data-refs.cc @@ -5035,7 +5035,21 @@ vect_analyze_data_refs (vec_info *vinfo, poly_uint64 *min_vf, bool *fatal) /* Set vectype for STMT. */ scalar_type = TREE_TYPE (DR_REF (dr)); tree vectype = get_vectype_for_scalar_type (vinfo, scalar_type); - if (!vectype) + + /* FIXME: If the object is in an address-space in which the pointer size + is different to the default address space then vectorizing here will + lead to an ICE down the road because the address space information + gets lost. This work-around fixes the problem until we have a proper + solution. */ + tree base_object = DR_REF (dr); + tree op = (TREE_CODE (base_object) == COMPONENT_REF + || TREE_CODE (base_object) == ARRAY_REF + ? TREE_OPERAND (base_object, 0) : base_object); + addr_space_t as = TYPE_ADDR_SPACE (TREE_TYPE (op)); + bool addr_space_bug = (!ADDR_SPACE_GENERIC_P (as) + && targetm.addr_space.pointer_mode (as) != Pmode); + + if (!vectype || addr_space_bug) { if (dump_enabled_p ()) { |