diff options
| author | Bill Schmidt <wschmidt@linux.ibm.com> | 2012-06-13 12:27:57 +0000 |
|---|---|---|
| committer | William Schmidt <wschmidt@gcc.gnu.org> | 2012-06-13 12:27:57 +0000 |
| commit | a21892add39f50fadabff622b086e8b8faa7e93b (patch) | |
| tree | a90486546faa1a3cb017da25e97444b6cabc379f /gcc/config | |
| parent | 2be13164c11f39884bc26e941e5c573eacfbe0ac (diff) | |
| download | gcc-a21892add39f50fadabff622b086e8b8faa7e93b.zip gcc-a21892add39f50fadabff622b086e8b8faa7e93b.tar.gz gcc-a21892add39f50fadabff622b086e8b8faa7e93b.tar.bz2 | |
targhooks.c (default_builtin_vectorized_conversion): Handle vec_construct, using vectype to base cost on subparts.
2012-06-13 Bill Schmidt <wschmidt@linux.ibm.com>
* targhooks.c (default_builtin_vectorized_conversion): Handle
vec_construct, using vectype to base cost on subparts.
* target.h (enum vect_cost_for_stmt): Add vec_construct.
* tree-vect-stmts.c (vect_model_load_cost): Use vec_construct
instead of scalar_to-vec.
* config/spu/spu.c (spu_builtin_vectorization_cost): Handle
vec_construct in same way as default for now.
* config/i386/i386.c (ix86_builtin_vectorization_cost): Likewise.
* config/rs6000/rs6000.c (rs6000_builtin_vectorization_cost):
Handle vec_construct, including special case for 32-bit loads.
From-SVN: r188508
Diffstat (limited to 'gcc/config')
| -rw-r--r-- | gcc/config/i386/i386.c | 8 | ||||
| -rw-r--r-- | gcc/config/rs6000/rs6000.c | 13 | ||||
| -rw-r--r-- | gcc/config/spu/spu.c | 8 |
3 files changed, 27 insertions, 2 deletions
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c index 13755f4..d10bf16 100644 --- a/gcc/config/i386/i386.c +++ b/gcc/config/i386/i386.c @@ -36072,9 +36072,11 @@ static const struct attribute_spec ix86_attribute_table[] = /* Implement targetm.vectorize.builtin_vectorization_cost. */ static int ix86_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, - tree vectype ATTRIBUTE_UNUSED, + tree vectype, int misalign ATTRIBUTE_UNUSED) { + unsigned elements; + switch (type_of_cost) { case scalar_stmt: @@ -36115,6 +36117,10 @@ ix86_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, case vec_promote_demote: return ix86_cost->vec_stmt_cost; + case vec_construct: + elements = TYPE_VECTOR_SUBPARTS (vectype); + return elements / 2 + 1; + default: gcc_unreachable (); } diff --git a/gcc/config/rs6000/rs6000.c b/gcc/config/rs6000/rs6000.c index 11c4bf7..ad9d438 100644 --- a/gcc/config/rs6000/rs6000.c +++ b/gcc/config/rs6000/rs6000.c @@ -3405,6 +3405,7 @@ rs6000_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, tree vectype, int misalign) { unsigned elements; + tree elem_type; switch (type_of_cost) { @@ -3504,6 +3505,18 @@ rs6000_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, return 2; + case vec_construct: + elements = TYPE_VECTOR_SUBPARTS (vectype); + elem_type = TREE_TYPE (vectype); + /* 32-bit vectors loaded into registers are stored as double + precision, so we need n/2 converts in addition to the usual + n/2 merges to construct a vector of short floats from them. */ + if (SCALAR_FLOAT_TYPE_P (elem_type) + && TYPE_PRECISION (elem_type) == 32) + return elements + 1; + else + return elements / 2 + 1; + default: gcc_unreachable (); } diff --git a/gcc/config/spu/spu.c b/gcc/config/spu/spu.c index b81bf5e..5310ba7 100644 --- a/gcc/config/spu/spu.c +++ b/gcc/config/spu/spu.c @@ -6908,9 +6908,11 @@ spu_builtin_mask_for_load (void) /* Implement targetm.vectorize.builtin_vectorization_cost. */ static int spu_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, - tree vectype ATTRIBUTE_UNUSED, + tree vectype, int misalign ATTRIBUTE_UNUSED) { + unsigned elements; + switch (type_of_cost) { case scalar_stmt: @@ -6937,6 +6939,10 @@ spu_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, case cond_branch_taken: return 6; + case vec_construct: + elements = TYPE_VECTOR_SUBPARTS (vectype); + return elements / 2 + 1; + default: gcc_unreachable (); } |