qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: "Li, Liang Z" <liang.z.li@intel.com>,
	Richard Henderson <rth@twiddle.net>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: "amit.shah@redhat.com" <amit.shah@redhat.com>,
	"quintela@redhat.com" <quintela@redhat.com>,
	"dgilbert@redhat.com" <dgilbert@redhat.com>,
	"stefanha@redhat.com" <stefanha@redhat.com>,
	"mst@redhat.com" <mst@redhat.com>
Subject: Re: [Qemu-devel] [v3 1/3] cutils: add avx2 instruction optimization
Date: Thu, 10 Dec 2015 10:51:13 +0100	[thread overview]
Message-ID: <56694B11.1050106@redhat.com> (raw)
In-Reply-To: <F2CBF3009FA73547804AE4C663CAB28E0238AC9B@SHSMSX101.ccr.corp.intel.com>



On 10/12/2015 10:22, Li, Liang Z wrote:
>>>>   without the ' -mavx2' option for gcc, there are compiling error:
>>>> '__m256i undeclared', the __attribute__((target("avx2"))) can't solve
>>>> this issue.  Any idea?
>>>
>>> You're right that you can't use the normal __m256i, as it doesn't get
>>> declared.
>>
>> It should be declared.  *intrin.h uses #pragma GCC target and always defines
>> all vector types.
>>
>> In fact, the following compiles for me with just "gcc foo.c" under GCC 5.x:
>>
>> #include <immintrin.h>
>>
>> // #if defined CONFIG_IFUNC && defined CONFIG_AVX2 #pragma GCC
>> push_options #pragma GCC target("avx2")
>> #define AVX2_VECTYPE        __m256i
>> #define AVX2_SPLAT(p)       _mm256_set1_epi8(*(p))
>> #define AVX2_ALL_EQ(v1, v2) \
>>     (_mm256_movemask_epi8(_mm256_cmpeq_epi8(v1, v2)) == 0xFFFFFFFF)
>> #define AVX2_VEC_OR(v1, v2) (_mm256_or_si256(v1, v2))
>>
>> size_t buffer_find_nonzero_offset_avx2(const void *buf, size_t len) {
>>     const AVX2_VECTYPE *p = buf;
>>     const AVX2_VECTYPE zero = (AVX2_VECTYPE){0};
>>     size_t i;
>>
>>     if (!len) {
>>         return 0;
>>     }
>>
>>     for (i = 0; i < 4; i++) {
>>         if (!AVX2_ALL_EQ(p[i], zero)) {
>>             return i * sizeof(AVX2_VECTYPE);
>>         }
>>     }
>>
>>     for (i = 4; i < len / sizeof(AVX2_VECTYPE); i += 4) {
>>         AVX2_VECTYPE tmp0 = AVX2_VEC_OR(p[i + 0], p[i + 1]);
>>         AVX2_VECTYPE tmp1 = AVX2_VEC_OR(p[i + 2], p[i + 3]);
>>         AVX2_VECTYPE tmp2 = AVX2_VEC_OR(p[i + 4], p[i + 5]);
>>         AVX2_VECTYPE tmp3 = AVX2_VEC_OR(p[i + 6], p[i + 7]);
>>         AVX2_VECTYPE tmp01 = AVX2_VEC_OR(tmp0, tmp1);
>>         AVX2_VECTYPE tmp23 = AVX2_VEC_OR(tmp2, tmp3);
>>         if (!AVX2_ALL_EQ(AVX2_VEC_OR(tmp01, tmp23), zero)) {
>>             break;
>>         }
>>     }
>>
>>     return i * sizeof(AVX2_VECTYPE);
>> }
>>
>> #pragma GCC pop_options
>> // #endif
>>
>> so perhaps the configure test is testing the wrong thing?
>>
>> Paolo
> 
> Hi Paolo,
> 
> what's your opinion?  putting the AVX2 related code to util/cutils.c and use the "#pragma ..." you referred?

Yes, that's best.  And you can keep using __m256i if you prefer that.

Paolo

  reply	other threads:[~2015-12-10  9:51 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-08 12:08 [Qemu-devel] [v3 0/3] add avx2 instruction optimization Liang Li
2015-12-08 12:08 ` [Qemu-devel] [v3 1/3] cutils: " Liang Li
2015-12-08 16:09   ` Richard Henderson
2015-12-09  9:32     ` Li, Liang Z
2015-12-09 14:57       ` Richard Henderson
2015-12-10  1:10         ` Li, Liang Z
2015-12-10  9:03         ` Paolo Bonzini
2015-12-10  9:22           ` Li, Liang Z
2015-12-10  9:51             ` Paolo Bonzini [this message]
2015-12-08 12:08 ` [Qemu-devel] [v3 2/3] configure: detect ifunc attribute Liang Li
2015-12-08 12:08 ` [Qemu-devel] [v3 3/3] configure: add options to config avx2 Liang Li
2015-12-08 12:54   ` Peter Maydell
2015-12-08 14:18     ` Li, Liang Z

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=56694B11.1050106@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=amit.shah@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=liang.z.li@intel.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=rth@twiddle.net \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).