[Qemu-devel] [PATCH] cutils: fix compiling with icc

[Qemu-devel] [PATCH] cutils: fix compiling with icc

[Artyom Tarasenko]

Intel C Compiler version 15.0.3.187 Build 20150407 doesn't support
'|' function for non floating-point simd operands.
Use instead _mm_or_si128 which is supported both in icc and gcc.

Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
---
 util/cutils.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/util/cutils.c b/util/cutils.c
index 144b25c..ffc9ddb 100644
--- a/util/cutils.c
+++ b/util/cutils.c
@@ -207,13 +207,13 @@ size_t buffer_find_nonzero_offset(const void *buf, size_t len)
     for (i = BUFFER_FIND_NONZERO_OFFSET_UNROLL_FACTOR;
          i < len / sizeof(VECTYPE);
          i += BUFFER_FIND_NONZERO_OFFSET_UNROLL_FACTOR) {
-        VECTYPE tmp0 = p[i + 0] | p[i + 1];
-        VECTYPE tmp1 = p[i + 2] | p[i + 3];
-        VECTYPE tmp2 = p[i + 4] | p[i + 5];
-        VECTYPE tmp3 = p[i + 6] | p[i + 7];
-        VECTYPE tmp01 = tmp0 | tmp1;
-        VECTYPE tmp23 = tmp2 | tmp3;
-        if (!ALL_EQ(tmp01 | tmp23, zero)) {
+        VECTYPE tmp0 = _mm_or_si128(p[i + 0], p[i + 1]);
+        VECTYPE tmp1 = _mm_or_si128(p[i + 2], p[i + 3]);
+        VECTYPE tmp2 = _mm_or_si128(p[i + 4], p[i + 5]);
+        VECTYPE tmp3 = _mm_or_si128(p[i + 6], p[i + 7]);
+        VECTYPE tmp01 = _mm_or_si128(tmp0, tmp1);
+        VECTYPE tmp23 = _mm_or_si128(tmp2, tmp3);
+        if (!ALL_EQ(_mm_or_si128(tmp01, tmp23), zero)) {
             break;
         }
     }
-- 
1.8.3.1

Re: [Qemu-devel] [PATCH] cutils: fix compiling with icc

[Paolo Bonzini]

On 22/06/2015 18:11, Artyom Tarasenko wrote:
> Intel C Compiler version 15.0.3.187 Build 20150407 doesn't support
> '|' function for non floating-point simd operands.
> Use instead _mm_or_si128 which is supported both in icc and gcc.

This would work for x86, but not for Altivec or non-SIMD.  Please define
a new macro like VEC_OR instead.

Paolo

> Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
> ---
>  util/cutils.c | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/util/cutils.c b/util/cutils.c
> index 144b25c..ffc9ddb 100644
> --- a/util/cutils.c
> +++ b/util/cutils.c
> @@ -207,13 +207,13 @@ size_t buffer_find_nonzero_offset(const void *buf, size_t len)
>      for (i = BUFFER_FIND_NONZERO_OFFSET_UNROLL_FACTOR;
>           i < len / sizeof(VECTYPE);
>           i += BUFFER_FIND_NONZERO_OFFSET_UNROLL_FACTOR) {
> -        VECTYPE tmp0 = p[i + 0] | p[i + 1];
> -        VECTYPE tmp1 = p[i + 2] | p[i + 3];
> -        VECTYPE tmp2 = p[i + 4] | p[i + 5];
> -        VECTYPE tmp3 = p[i + 6] | p[i + 7];
> -        VECTYPE tmp01 = tmp0 | tmp1;
> -        VECTYPE tmp23 = tmp2 | tmp3;
> -        if (!ALL_EQ(tmp01 | tmp23, zero)) {
> +        VECTYPE tmp0 = _mm_or_si128(p[i + 0], p[i + 1]);
> +        VECTYPE tmp1 = _mm_or_si128(p[i + 2], p[i + 3]);
> +        VECTYPE tmp2 = _mm_or_si128(p[i + 4], p[i + 5]);
> +        VECTYPE tmp3 = _mm_or_si128(p[i + 6], p[i + 7]);
> +        VECTYPE tmp01 = _mm_or_si128(tmp0, tmp1);
> +        VECTYPE tmp23 = _mm_or_si128(tmp2, tmp3);
> +        if (!ALL_EQ(_mm_or_si128(tmp01, tmp23), zero)) {
>              break;
>          }