This removes --param vect-inner-loop-cost-factor in favor of looking at the basic block counts of the outer vs. the inner loop header when they are comparable and otherwise just assumes a single inner iteration which is conservative on the side of not vectorizing.
Bootstrap and regtest running on x86_64-unknown-linux-gnu. Does this look sane? I will be looking into computing the scalar loop cost on the original, not if-converted body where I'd have to do sth similar for the different branches. From there I can see to fix the issue with BB vectorization of if-converted loop bodies. Thanks, Richard. 2021-08-23 Richard Biener <rguent...@suse.de> * doc/invoke.texi (vect-inner-loop-cost-factor): Remove documentation. * params.opt (--param vect-inner-loop-cost-factor): Remove. * tree-vect-loop.c (_loop_vec_info::_loop_vec_info): Initialize inner_loop_cost_factor to 1. (vect_analyze_loop_form): Initialize inner_loop_cost_factor from inner and outer loop header counts. --- gcc/doc/invoke.texi | 5 ----- gcc/params.opt | 4 ---- gcc/tree-vect-loop.c | 9 ++++++++- 3 files changed, 8 insertions(+), 10 deletions(-) diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index d8a6b0b60c9..c7819917dce 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -14385,11 +14385,6 @@ code to iterate. 2 allows partial vector loads and stores in all loops. The parameter only has an effect on targets that support partial vector loads and stores. -@item vect-inner-loop-cost-factor -The factor which the loop vectorizer applies to the cost of statements -in an inner loop relative to the loop being vectorized. The default -value is 50. - @item avoid-fma-max-bits Maximum number of bits for which we avoid creating FMAs. diff --git a/gcc/params.opt b/gcc/params.opt index f9264887b40..f7b19fa430d 100644 --- a/gcc/params.opt +++ b/gcc/params.opt @@ -1113,8 +1113,4 @@ Bound on number of runtime checks inserted by the vectorizer's loop versioning f Common Joined UInteger Var(param_vect_partial_vector_usage) Init(2) IntegerRange(0, 2) Param Optimization Controls how loop vectorizer uses partial vectors. 0 means never, 1 means only for loops whose need to iterate can be removed, 2 means for all loops. The default value is 2. --param=vect-inner-loop-cost-factor= -Common Joined UInteger Var(param_vect_inner_loop_cost_factor) Init(50) IntegerRange(1, 999999) Param Optimization -The factor which the loop vectorizer applies to the cost of statements in an inner loop relative to the loop being vectorized. - ; This comment is to ensure we retain the blank line above. diff --git a/gcc/tree-vect-loop.c b/gcc/tree-vect-loop.c index c521b43a47c..86cd5d8f730 100644 --- a/gcc/tree-vect-loop.c +++ b/gcc/tree-vect-loop.c @@ -841,7 +841,7 @@ _loop_vec_info::_loop_vec_info (class loop *loop_in, vec_info_shared *shared) single_scalar_iteration_cost (0), vec_outside_cost (0), vec_inside_cost (0), - inner_loop_cost_factor (param_vect_inner_loop_cost_factor), + inner_loop_cost_factor (1), vectorizable (false), can_use_partial_vectors_p (param_vect_partial_vector_usage != 0), using_partial_vectors_p (false), @@ -1519,6 +1519,13 @@ vect_analyze_loop_form (class loop *loop, vec_info_shared *shared) stmt_vec_info inner_loop_cond_info = loop_vinfo->lookup_stmt (inner_loop_cond); STMT_VINFO_TYPE (inner_loop_cond_info) = loop_exit_ctrl_vec_info_type; + /* If we can compare profile counts of the outer and inner loop header + compute the scale based on that, otherwise conservatively assume + a single inner iteration. */ + if (loop->inner->header->count > loop->header->count) + LOOP_VINFO_INNER_LOOP_COST_FACTOR (loop_vinfo) + = CEIL (loop->inner->header->count.value (), + loop->header->count.value ()); } gcc_assert (!loop->aux); -- 2.31.1