https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105617
Hongtao Liu <liuhongt at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |liuhongt at gcc dot gnu.org
--- Comment #25 from Hongtao Liu <liuhongt at gcc dot gnu.org> ---
in vectorization, store shouldn't have such high cost because once the address
is computed, the store instruction doesn't stall the pipeline (except for
memory dependencies that are hard for compilers to detect). I'm experimenting
with adjusting store costs: setting integer stores to COST_N_INSNS (1) - 1
(minus 1 since integer can benefit from renaming and have lower STLF costs),
while keeping vector/floating-point stores at COST_N_INSNS (1). This approach
should discourage vectorization for integer vector construction + vector store
patterns, while still promoting vectorization for floating-point operations in
vector construct + vector store scenarios.
I'm testing below patch to see if there's any surprise.
diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc
index 52f82185e32..420104a04b2 100644
--- a/gcc/config/i386/i386.cc
+++ b/gcc/config/i386/i386.cc
@@ -25363,8 +25363,7 @@ ix86_default_vector_cost (enum vect_cost_for_stmt
type_of_cost,
: ix86_cost->int_load [2]) / 2;
case scalar_store:
- return COSTS_N_INSNS (fp ? ix86_cost->sse_store[0]
- : ix86_cost->int_store [2]) / 2;
+ return fp ? COSTS_N_INSNS (1) : COSTS_N_INSNS (1) - 1;
case vector_stmt:
return ix86_vec_cost (mode,
@@ -25378,11 +25377,7 @@ ix86_default_vector_cost (enum vect_cost_for_stmt
type_of_cost,
return COSTS_N_INSNS (ix86_cost->sse_load[index]) / 2;
case vector_store:
- index = sse_store_index (mode);
- /* See PR82713 - we may end up being called on non-vector type. */
- if (index < 0)
- index = 2;
- return COSTS_N_INSNS (ix86_cost->sse_store[index]) / 2;
+ return ix86_vec_cost (mode, ix86_cost->sse_op);
case vec_to_scalar:
case scalar_to_vec:
@@ -25398,11 +25393,7 @@ ix86_default_vector_cost (enum vect_cost_for_stmt
type_of_cost,
return COSTS_N_INSNS (ix86_cost->sse_unaligned_load[index]) / 2;
case unaligned_store:
- index = sse_store_index (mode);
- /* See PR82713 - we may end up being called on non-vector type. */
- if (index < 0)
- index = 2;
- return COSTS_N_INSNS (ix86_cost->sse_unaligned_store[index]) / 2;
+ return ix86_vec_cost (mode, ix86_cost->sse_op);
case vector_gather_load:
return ix86_vec_cost (mode,
