From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53379) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a6x8v-0001vs-62 for qemu-devel@nongnu.org; Thu, 10 Dec 2015 04:04:06 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1a6x8r-0002iv-3f for qemu-devel@nongnu.org; Thu, 10 Dec 2015 04:04:05 -0500 Received: from mail-qg0-x22f.google.com ([2607:f8b0:400d:c04::22f]:35756) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a6x8r-0002iq-0M for qemu-devel@nongnu.org; Thu, 10 Dec 2015 04:04:01 -0500 Received: by qgec40 with SMTP id c40so129374306qge.2 for ; Thu, 10 Dec 2015 01:04:00 -0800 (PST) Sender: Paolo Bonzini References: <1449576535-3369-1-git-send-email-liang.z.li@intel.com> <1449576535-3369-2-git-send-email-liang.z.li@intel.com> <566700CA.1080408@twiddle.net> <56684141.5020504@twiddle.net> From: Paolo Bonzini Message-ID: <56693FF6.6020807@redhat.com> Date: Thu, 10 Dec 2015 10:03:50 +0100 MIME-Version: 1.0 In-Reply-To: <56684141.5020504@twiddle.net> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [v3 1/3] cutils: add avx2 instruction optimization List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Richard Henderson , "Li, Liang Z" , "qemu-devel@nongnu.org" Cc: "amit.shah@redhat.com" , "quintela@redhat.com" , "dgilbert@redhat.com" , "stefanha@redhat.com" , "mst@redhat.com" On 09/12/2015 15:57, Richard Henderson wrote: >> I think you means the ' __attribute__((target("avx2")))', I have tried >> this way, the issue here is: >> 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 // #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