diff options
| author | Bill Schmidt <wschmidt@linux.vnet.ibm.com> | 2016-08-12 15:23:34 +0000 |
|---|---|---|
| committer | William Schmidt <wschmidt@gcc.gnu.org> | 2016-08-12 15:23:34 +0000 |
| commit | 42b5ebf32cc8da61acd04925d8736c7827549dfa (patch) | |
| tree | bf06d0240cfb2afe03fadf214b8cbd141e4c1fb6 /gcc | |
| parent | 8eb414aa6c71225d941e24572d5cdd9ce6feee50 (diff) | |
| download | gcc-42b5ebf32cc8da61acd04925d8736c7827549dfa.zip gcc-42b5ebf32cc8da61acd04925d8736c7827549dfa.tar.gz gcc-42b5ebf32cc8da61acd04925d8736c7827549dfa.tar.bz2 | |
rs6000.c (rs6000_builtin_vectorization_cost): Correct costs for vec_construct.
2016-08-12 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* config/rs6000/rs6000.c (rs6000_builtin_vectorization_cost):
Correct costs for vec_construct.
From-SVN: r239417
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/ChangeLog | 5 | ||||
| -rw-r--r-- | gcc/config/rs6000/rs6000.c | 14 |
2 files changed, 14 insertions, 5 deletions
diff --git a/gcc/ChangeLog b/gcc/ChangeLog index e0dab59..49f0b6d 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,8 @@ +2016-08-12 Bill Schmidt <wschmidt@linux.vnet.ibm.com> + + * config/rs6000/rs6000.c (rs6000_builtin_vectorization_cost): + Correct costs for vec_construct. + 2016-08-12 Bin Cheng <bin.cheng@arm.com> PR tree-optimization/69848 diff --git a/gcc/config/rs6000/rs6000.c b/gcc/config/rs6000/rs6000.c index 756e69d..e58ede1 100644 --- a/gcc/config/rs6000/rs6000.c +++ b/gcc/config/rs6000/rs6000.c @@ -5266,16 +5266,20 @@ rs6000_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, return 2; case vec_construct: - elements = TYPE_VECTOR_SUBPARTS (vectype); + /* This is a rough approximation assuming non-constant elements + constructed into a vector via element insertion. FIXME: + vec_construct is not granular enough for uniformly good + decisions. If the initialization is a splat, this is + cheaper than we estimate. Improve this someday. */ 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. */ + precision, so we need 2 permutes, 2 converts, and 1 merge + to construct a vector of short floats from them. */ if (SCALAR_FLOAT_TYPE_P (elem_type) && TYPE_PRECISION (elem_type) == 32) - return elements + 1; + return 5; else - return elements / 2 + 1; + return max (2, TYPE_VECTOR_SUBPARTS (vectype) - 1); default: gcc_unreachable (); |