procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Disable some VEC_COND_EXPR transformations after vector lowering

Browse source 
ARM understands VEC_COND_EXPR<v == w, -1, 0> but not a plain v == w which is
fed to something other than VEC_COND_EXPR (say BIT_IOR_EXPR). This patch avoids
introducing the second kind of statement after the vector lowering pass, which
is the last chance to turn v == w back into something the target handles.

This is just a workaround to avoid ICEs, a v == w produced before vector
lowering will yield pretty bad code. Either the arm target needs to learn to
handle vector comparisons (aarch64 already does), or the middle-end needs to
fall back to vcond when plain comparisons are not supported (or ...).

2020-08-07  Marc Glisse  <marc.glisse@inria.fr>

	* generic-match-head.c (optimize_vectors_before_lowering_p): New
	function.
	* gimple-match-head.c (optimize_vectors_before_lowering_p):
	Likewise.
	* match.pd ((v ? w : 0) ? a : b, c1 ? c2 ? a : b : b): Use it.
This commit is contained in:
Marc Glisse 2020-08-07 18:49:04 +02:00
parent 6c3ae88d1e
commit a1ee6d507b
3 changed files with 31 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
gcc/generic-match-head.c
View file

@ -80,6 +80,16 @@ canonicalize_math_after_vectorization_p ()
return false;
}
/* Return true if we can still perform transformations that may introduce
vector operations that are not supported by the target. Vector lowering
normally handles those, but after that pass, it becomes unsafe. */
static inline bool
optimize_vectors_before_lowering_p ()
{
return true;
}
/* Return true if successive divisions can be optimized.
Defer to GIMPLE opts. */

10
gcc/gimple-match-head.c
View file

@ -1158,6 +1158,16 @@ canonicalize_math_after_vectorization_p ()
return !cfun || (cfun->curr_properties & PROP_gimple_lvec) != 0;
}
/* Return true if we can still perform transformations that may introduce
vector operations that are not supported by the target. Vector lowering
normally handles those, but after that pass, it becomes unsafe. */
static inline bool
optimize_vectors_before_lowering_p ()
{
return !cfun || (cfun->curr_properties & PROP_gimple_lvec) == 0;
}
/* Return true if pow(cst, x) should be optimized into exp(log(cst) * x).
As a workaround for SPEC CPU2017 628.pop2_s, don't do it if arg0
is an exact integer, arg1 = phi_res +/- cst1 and phi_res = PHI <cst2, ...>

20
gcc/match.pd
View file

@ -3461,40 +3461,42 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
(vec_cond @0 (op! @3 @1) (op! @3 @2))))
#endif
/* (v ? w : 0) ? a : b is just (v & w) ? a : b */
/* (v ? w : 0) ? a : b is just (v & w) ? a : b
Currently disabled after pass lvec because ARM understands
VEC_COND_EXPR<v==w,-1,0> but not a plain v==w fed to BIT_IOR_EXPR. */
(simplify
(vec_cond (vec_cond:s @0 @3 integer_zerop) @1 @2)
(if (types_match (@0, @3))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @3))
(vec_cond (bit_and @0 @3) @1 @2)))
(simplify
(vec_cond (vec_cond:s @0 integer_all_onesp @3) @1 @2)
(if (types_match (@0, @3))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @3))
(vec_cond (bit_ior @0 @3) @1 @2)))
(simplify
(vec_cond (vec_cond:s @0 integer_zerop @3) @1 @2)
(if (types_match (@0, @3))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @3))
(vec_cond (bit_ior @0 (bit_not @3)) @2 @1)))
(simplify
(vec_cond (vec_cond:s @0 @3 integer_all_onesp) @1 @2)
(if (types_match (@0, @3))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @3))
(vec_cond (bit_and @0 (bit_not @3)) @2 @1)))
/* c1 ? c2 ? a : b : b --> (c1 & c2) ? a : b */
(simplify
(vec_cond @0 (vec_cond:s @1 @2 @3) @3)
(if (types_match (@0, @1))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @1))
(vec_cond (bit_and @0 @1) @2 @3)))
(simplify
(vec_cond @0 @2 (vec_cond:s @1 @2 @3))
(if (types_match (@0, @1))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @1))
(vec_cond (bit_ior @0 @1) @2 @3)))
(simplify
(vec_cond @0 (vec_cond:s @1 @2 @3) @2)
(if (types_match (@0, @1))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @1))
(vec_cond (bit_ior (bit_not @0) @1) @2 @3)))
(simplify
(vec_cond @0 @3 (vec_cond:s @1 @2 @3))
(if (types_match (@0, @1))
(if (optimize_vectors_before_lowering_p () && types_match (@0, @1))
(vec_cond (bit_and (bit_not @0) @1) @2 @3)))
/* Simplification moved from fold_cond_expr_with_comparison. It may also