On Wed, Jun 27, 2012 at 02:36:14PM -0700, Richard Henderson wrote:
> As noticed by Igor Zamyatin. Committed.
> PR target/53749
> * config/i386/i386.c (ix86_rtx_costs): Fix typo vs UNITS_PER_WORD
> in 2012-06-23 change. Adjust two other DImode tests as well.
This change broke cost computation for vector PLUS/MINUS/AND/IOR/XOR,
the vector modes are all wider than word, but they are now all handled as
double word integer arithmetics.
> ---
> gcc/ChangeLog | 6 ++++++
> gcc/config/i386/i386.c | 8 +++-----
> 2 files changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
> index edfc649..aae8a4d 100644
> --- a/gcc/config/i386/i386.c
> +++ b/gcc/config/i386/i386.c
...
> @@ -32441,7 +32440,7 @@ ix86_rtx_costs (rtx x, int code_i, int outer_code_i,
> int opno, int *total,
> case AND:
> case IOR:
> case XOR:
> - if (!TARGET_64BIT && mode == DImode)
> + if (GET_MODE_SIZE (mode) > UNITS_PER_WORD)
> {
> *total = (cost->add * 2
> + (rtx_cost (XEXP (x, 0), outer_code, opno, speed)
Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok for
trunk?
2012-09-18 Jakub Jelinek <ja...@redhat.com>
PR target/54592
* config/i386/i386.c (ix86_rtx_costs): Limit > UNITS_PER_WORD
AND/IOR/XOR cost calculation to MODE_INT class modes.
* gcc.target/i386/pr54592.c: New test.
--- gcc/config/i386/i386.c.jj 2012-09-13 18:29:08.000000000 +0200
+++ gcc/config/i386/i386.c 2012-09-18 08:55:08.747028184 +0200
@@ -32792,7 +32792,8 @@ ix86_rtx_costs (rtx x, int code_i, int o
case AND:
case IOR:
case XOR:
- if (GET_MODE_SIZE (mode) > UNITS_PER_WORD)
+ if (GET_MODE_CLASS (mode) == MODE_INT
+ && GET_MODE_SIZE (mode) > UNITS_PER_WORD)
{
*total = (cost->add * 2
+ (rtx_cost (XEXP (x, 0), outer_code, opno, speed)
--- gcc/testsuite/gcc.target/i386/pr54592.c.jj 2012-09-18 09:06:09.399013382
+0200
+++ gcc/testsuite/gcc.target/i386/pr54592.c 2012-09-18 09:13:04.482914236
+0200
@@ -0,0 +1,17 @@
+/* PR target/54592 */
+/* { dg-do compile } */
+/* { dg-options "-Os -msse2" } */
+/* { dg-require-effective-target sse2 } */
+
+#include <emmintrin.h>
+
+void
+func (__m128i * foo, size_t a, size_t b, int *dst)
+{
+ __m128i x = foo[a];
+ __m128i y = foo[b];
+ __m128i sum = _mm_add_epi32 (x, y);
+ *dst = _mm_cvtsi128_si32 (sum);
+}
+
+/* { dg-final { scan-assembler "paddd\[^\n\r\]*(\\(\[^\n\r\]*\\)|XMMWORD PTR)"
} } */
Jakub
