List patch comments

GET /api/patches/73540/comments/?format=api&order=submitter
HTTP 200 OK
Allow: GET, HEAD, OPTIONS
Content-Type: application/json
Link: 
<https://patches.dpdk.org/api/patches/73540/comments/?format=api&order=submitter&page=1>; rel="first",
<https://patches.dpdk.org/api/patches/73540/comments/?format=api&order=submitter&page=1>; rel="last"
Vary: Accept
[ { "id": 115627, "web_url": "https://patches.dpdk.org/comment/115627/", "msgid": "<CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com>", "list_archive_url": "https://inbox.dpdk.org/dev/CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com", "date": "2020-07-09T13:48:21", "subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "submitter": { "id": 1173, "url": "https://patches.dpdk.org/api/people/1173/?format=api", "name": "David Marchand", "email": "david.marchand@redhat.com" }, "content": "On Wed, Jul 8, 2020 at 10:17 PM Vladimir Medvedkin\n<vladimir.medvedkin@intel.com> wrote:\n>\n> New data type to manipulate 512 bit AVX values.\n\nThe title mentions a \"zmm\" type that is not added by this patch.\n\nMaybe instead, \"eal/x86: introduce AVX 512-bit type\"\n\n\n>\n> Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>\n> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>\n> ---\n> lib/librte_eal/x86/include/rte_vect.h | 21 +++++++++++++++++++++\n> 1 file changed, 21 insertions(+)\n>\n> diff --git a/lib/librte_eal/x86/include/rte_vect.h b/lib/librte_eal/x86/include/rte_vect.h\n> index df5a60762..ae59126bc 100644\n> --- a/lib/librte_eal/x86/include/rte_vect.h\n> +++ b/lib/librte_eal/x86/include/rte_vect.h\n> @@ -13,6 +13,7 @@\n>\n> #include <stdint.h>\n> #include <rte_config.h>\n> +#include <rte_common.h>\n> #include \"generic/rte_vect.h\"\n>\n> #if (defined(__ICC) || \\\n> @@ -90,6 +91,26 @@ __extension__ ({ \\\n> })\n> #endif /* (defined(__ICC) && __ICC < 1210) */\n>\n> +#ifdef __AVX512F__\n> +\n> +typedef __m512i __x86_zmm_t;\n\nWe don't need this interim type, using the native __m512 is enough afaics.\n\nLooking at the whole applied series:\n$ git grep -lw __x86_zmm_t\nlib/librte_eal/x86/include/rte_vect.h\n\n\n> +\n> +#define ZMM_SIZE (sizeof(__x86_zmm_t))\n> +#define ZMM_MASK (ZMM_SIZE - 1)\n\nMacros in a public header need a RTE_ prefix + this is x86 specific,\nthen RTE_X86_.\n\nLooking at the whole applied series:\n$ git grep -lw ZMM_SIZE\nlib/librte_eal/x86/include/rte_vect.h\n$ git grep -lw ZMM_MASK\nlib/librte_eal/x86/include/rte_vect.h\n\nSo I wonder if we need to export it or we can instead just #undef\nafter the struct definition.\n\n\n> +\n> +typedef union __rte_x86_zmm {\n> + __x86_zmm_t z;\n> + ymm_t y[ZMM_SIZE / sizeof(ymm_t)];\n> + xmm_t x[ZMM_SIZE / sizeof(xmm_t)];\n> + uint8_t u8[ZMM_SIZE / sizeof(uint8_t)];\n> + uint16_t u16[ZMM_SIZE / sizeof(uint16_t)];\n> + uint32_t u32[ZMM_SIZE / sizeof(uint32_t)];\n> + uint64_t u64[ZMM_SIZE / sizeof(uint64_t)];\n> + double pd[ZMM_SIZE / sizeof(double)];\n> +} __rte_aligned(ZMM_SIZE) __rte_x86_zmm_t;\n\nI don't understand this forced alignment statement.\nWould not natural alignment be enough, since all fields in this union\nhave the same size?", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@inbox.dpdk.org", "Delivered-To": "patchwork@inbox.dpdk.org", "Received": [ "from dpdk.org (dpdk.org [92.243.14.124])\n\tby inbox.dpdk.org (Postfix) with ESMTP id 5CA01A0528;\n\tThu, 9 Jul 2020 15:48:38 +0200 (CEST)", "from [92.243.14.124] (localhost [127.0.0.1])\n\tby dpdk.org (Postfix) with ESMTP id 72E931E906;\n\tThu, 9 Jul 2020 15:48:37 +0200 (CEST)", "from us-smtp-1.mimecast.com (us-smtp-delivery-1.mimecast.com\n [207.211.31.120]) by dpdk.org (Postfix) with ESMTP id 457041E900\n for <dev@dpdk.org>; Thu, 9 Jul 2020 15:48:36 +0200 (CEST)", "from mail-ua1-f69.google.com (mail-ua1-f69.google.com\n [209.85.222.69]) (Using TLS) by relay.mimecast.com with ESMTP id\n us-mta-207-HgRr_htWM1eIqdvuzwpNBQ-1; Thu, 09 Jul 2020 09:48:34 -0400", "by mail-ua1-f69.google.com with SMTP id c13so887876uae.0\n for <dev@dpdk.org>; Thu, 09 Jul 2020 06:48:33 -0700 (PDT)" ], "DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n s=mimecast20190719; t=1594302515;\n h=from:from:reply-to:subject:subject:date:date:message-id:message-id:\n to:to:cc:cc:mime-version:mime-version:content-type:content-type:\n in-reply-to:in-reply-to:references:references;\n bh=1MrYzYF8j1k2jo94PGCoJR0izIR2jtrSSy5Y+mbQ7KU=;\n b=FQuHt/UpG0fT2e4jhXqywU/Cdj+xLZ7oLNFa6DY4/D7iFKQAePgAd6zjXejGaksppHAr8G\n jGcYU9kN+5uA859jg+VEBX4uWuSwE4Fbhd6i+HQC9TQKEXyCjm42v+zN7Uxze5KoPv/DpB\n DlsjCCmUzvc7LFcAdsxPlC9mM8DOgDs=", "X-MC-Unique": "HgRr_htWM1eIqdvuzwpNBQ-1", "X-Google-DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed;\n d=1e100.net; s=20161025;\n h=x-gm-message-state:mime-version:references:in-reply-to:from:date\n :message-id:subject:to:cc;\n bh=1MrYzYF8j1k2jo94PGCoJR0izIR2jtrSSy5Y+mbQ7KU=;\n b=Br8XIIJzNB6EaeqO3uwBslYCNbQtqyzE9uof8eF+es4+3YBDAMnSDgFuigAirlivsr\n 5Aut5XbHoDslzPFwzTQ3aj+BQYnVV+JwEu+XwL5VIofhS3P0wrwyBzlqDH0CybY8gosi\n EDQNdu+9Q9s3v104LCajx6okZkutJ+bFwbv3RR6xPABIoI0te1ap7+AqVjtycIC+Ay4v\n w5UOTzYdJ30C8R/vBe3wfLvgBSWKUd8BZ3UVv4bUNuMch7xUp6dyiTey6JzSVSyHkc0h\n 4FooHlLCjREyI41c41YuE3Yfg+K6XRrEtHCenkoHOptTVc+OrDE2KucGTwyue4jgceqs\n e82g==", "X-Gm-Message-State": "AOAM531R4wbZ6E+4LmxZLmDULNgLqTXTwlB0JGyqsQQCakZvJdEK19p5\n vdEhjMRH1k6AN2t2GE2pbP13KQl3T+yAiPwIDeuxU+G8ccPJN4nk02tGqgOayjX2VmoFpFR5aRL\n VJb6oHe0RnhM/H9J8Jjw=", "X-Received": [ "by 2002:a67:26c2:: with SMTP id\n m185mr46141343vsm.39.1594302513032;\n Thu, 09 Jul 2020 06:48:33 -0700 (PDT)", "by 2002:a67:26c2:: with SMTP id\n m185mr46141325vsm.39.1594302512734;\n Thu, 09 Jul 2020 06:48:32 -0700 (PDT)" ], "X-Google-Smtp-Source": "\n ABdhPJzRFqwCuZP2Z6Ib9hMbqmrmcOtMn0d4fgqypckTW3/ZR43cjG7LIIyE08br6Asq0pvprG8rnZv48j4XlpLamfo=", "MIME-Version": "1.0", "References": "<cover.1589890262.git.vladimir.medvedkin@intel.com>\n <cover.1594238609.git.vladimir.medvedkin@intel.com>\n <98b10e12eb46cff65494a94eaf0f04b2dcefd245.1594238610.git.vladimir.medvedkin@intel.com>", "In-Reply-To": "\n <98b10e12eb46cff65494a94eaf0f04b2dcefd245.1594238610.git.vladimir.medvedkin@intel.com>", "From": "David Marchand <david.marchand@redhat.com>", "Date": "Thu, 9 Jul 2020 15:48:21 +0200", "Message-ID": "\n <CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com>", "To": "Vladimir Medvedkin <vladimir.medvedkin@intel.com>", "Cc": "dev <dev@dpdk.org>, \"Ananyev, Konstantin\" <konstantin.ananyev@intel.com>,\n Bruce Richardson <bruce.richardson@intel.com>", "X-Mimecast-Spam-Score": "0", "X-Mimecast-Originator": "redhat.com", "Content-Type": "text/plain; charset=\"UTF-8\"", "Subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n <mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://mails.dpdk.org/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<https://mails.dpdk.org/listinfo/dev>,\n <mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "addressed": null }, { "id": 115644, "web_url": "https://patches.dpdk.org/comment/115644/", "msgid": "<CAJFAV8yh8-s0wjtOeA5J9=a1pZ0P7+s7oRpumB2f=6VpgL2A5Q@mail.gmail.com>", "list_archive_url": "https://inbox.dpdk.org/dev/CAJFAV8yh8-s0wjtOeA5J9=a1pZ0P7+s7oRpumB2f=6VpgL2A5Q@mail.gmail.com", "date": "2020-07-09T15:20:09", "subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "submitter": { "id": 1173, "url": "https://patches.dpdk.org/api/people/1173/?format=api", "name": "David Marchand", "email": "david.marchand@redhat.com" }, "content": "On Thu, Jul 9, 2020 at 4:52 PM Medvedkin, Vladimir\n<vladimir.medvedkin@intel.com> wrote:\n> >> +\n> >> +#define ZMM_SIZE (sizeof(__x86_zmm_t))\n> >> +#define ZMM_MASK (ZMM_SIZE - 1)\n> >\n> > Macros in a public header need a RTE_ prefix + this is x86 specific,\n> > then RTE_X86_.\n> >\n> > Looking at the whole applied series:\n> > $ git grep -lw ZMM_SIZE\n> > lib/librte_eal/x86/include/rte_vect.h\n> > $ git grep -lw ZMM_MASK\n> > lib/librte_eal/x86/include/rte_vect.h\n> >\n> > So I wonder if we need to export it or we can instead just #undef\n> > after the struct definition.\n>\n> I think it's better to undef it\n\nEven if you undef the macro, please still prefix it.\nThis is to avoid conflicts with macros defined before including this\nrte_vect.h header.\n\n\n>\n> >\n> >\n> >> +\n> >> +typedef union __rte_x86_zmm {\n> >> + __x86_zmm_t z;\n> >> + ymm_t y[ZMM_SIZE / sizeof(ymm_t)];\n> >> + xmm_t x[ZMM_SIZE / sizeof(xmm_t)];\n> >> + uint8_t u8[ZMM_SIZE / sizeof(uint8_t)];\n> >> + uint16_t u16[ZMM_SIZE / sizeof(uint16_t)];\n> >> + uint32_t u32[ZMM_SIZE / sizeof(uint32_t)];\n> >> + uint64_t u64[ZMM_SIZE / sizeof(uint64_t)];\n> >> + double pd[ZMM_SIZE / sizeof(double)];\n> >> +} __rte_aligned(ZMM_SIZE) __rte_x86_zmm_t;\n> >\n> > I don't understand this forced alignment statement.\n> > Would not natural alignment be enough, since all fields in this union\n> > have the same size?\n> >\n>\n> Some compilers won't align this union\n> https://mails.dpdk.org/archives/dev/2020-March/159591.html\n\nOk, interesting, I will try to keep in mind.", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@inbox.dpdk.org", "Delivered-To": "patchwork@inbox.dpdk.org", "Received": [ "from dpdk.org (dpdk.org [92.243.14.124])\n\tby inbox.dpdk.org (Postfix) with ESMTP id 0D317A0528;\n\tThu, 9 Jul 2020 17:20:42 +0200 (CEST)", "from [92.243.14.124] (localhost [127.0.0.1])\n\tby dpdk.org (Postfix) with ESMTP id DABB51DB74;\n\tThu, 9 Jul 2020 17:20:40 +0200 (CEST)", "from us-smtp-1.mimecast.com (us-smtp-delivery-1.mimecast.com\n [205.139.110.120]) by dpdk.org (Postfix) with ESMTP id 588921DB65\n for <dev@dpdk.org>; Thu, 9 Jul 2020 17:20:39 +0200 (CEST)", "from mail-ua1-f70.google.com (mail-ua1-f70.google.com\n [209.85.222.70]) (Using TLS) by relay.mimecast.com with ESMTP id\n us-mta-364-dzlT5GLBMnWaSWFgzJ50KQ-1; Thu, 09 Jul 2020 11:20:21 -0400", "by mail-ua1-f70.google.com with SMTP id c14so989705uap.16\n for <dev@dpdk.org>; Thu, 09 Jul 2020 08:20:21 -0700 (PDT)" ], "DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n s=mimecast20190719; t=1594308038;\n h=from:from:reply-to:subject:subject:date:date:message-id:message-id:\n to:to:cc:cc:mime-version:mime-version:content-type:content-type:\n in-reply-to:in-reply-to:references:references;\n bh=qQm0rRCubEjLN/ZTkibXmFaoHxEtTd7JgJcBki7auYQ=;\n b=Jci6uiXfjuco3hjx8U2lj0i6C6yax9fdMVX5r592L/TAaXhSZjyHXccVUYaqnG14dYiUdo\n gTC6i7B+LQgfZTof4xdESrUEA9Re5ULwnjr7SPTk4Rd6k3RLM7yYjX6zwfHoLjGUoNh4Xt\n QnjhOf0DTisY4n+dBH0T+yN+xw8OFn4=", "X-MC-Unique": "dzlT5GLBMnWaSWFgzJ50KQ-1", "X-Google-DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed;\n d=1e100.net; s=20161025;\n h=x-gm-message-state:mime-version:references:in-reply-to:from:date\n :message-id:subject:to:cc;\n bh=qQm0rRCubEjLN/ZTkibXmFaoHxEtTd7JgJcBki7auYQ=;\n b=BwRoV4MIWwXG6YZyAQP6KGxMZNpK0xEVGMFWlwNBaYoNVMIpydYB16rD2RvLYM5kmF\n tWg2QufcMB6ULDQjUAFQ+qn6RwXE5Lg//IZ90gklyodAj6pcgU4b38N8+vqvv1lsnwCh\n l8qz6T0iTqO8CQbuwk2TvD75ECHv9/hpFVrC+p8j8MuUMsyjUwNbZ/7NMu2eS5Elze4a\n gQr3e+RS199J/T647PcZ2ZFPOGPFsyzjvWFRPib8u2FMkspdUaOpIAiLTkIzjdVrLyHm\n 3h/fozHBqf4/5I0kzcH9dBfi5Hbsh/WYAWbRPL6mcQVwuz5dcAmuy2rlCpvkmWjPBLIR\n fiCA==", "X-Gm-Message-State": "AOAM530OnswQ2hOx4ZCC72bUjr/xkO8plHldbzBV5DLoVCROIAZX1fme\n uL8N0wrxAiZg2q8aOHxiY5paKcFi0RAv9iTW53AwnQY+B9f1SMH3Yx+gDCmhBMCNaVulhEsvEkB\n gb/m1TzEICdU+uSS4NQc=", "X-Received": [ "by 2002:a05:6102:249:: with SMTP id\n a9mr18632192vsq.198.1594308020872;\n Thu, 09 Jul 2020 08:20:20 -0700 (PDT)", "by 2002:a05:6102:249:: with SMTP id\n a9mr18632169vsq.198.1594308020595;\n Thu, 09 Jul 2020 08:20:20 -0700 (PDT)" ], "X-Google-Smtp-Source": "\n ABdhPJy46ss9utIH+pOv93ZKjSXd5oug/Xbb6kHLQR5fg7BWSq5QVyMKHLnKWPa4k6Zb9Lm2EZVG18JD5kO1HuXckRE=", "MIME-Version": "1.0", "References": "<cover.1589890262.git.vladimir.medvedkin@intel.com>\n <cover.1594238609.git.vladimir.medvedkin@intel.com>\n <98b10e12eb46cff65494a94eaf0f04b2dcefd245.1594238610.git.vladimir.medvedkin@intel.com>\n <CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com>\n <3f68ec75-417b-1587-e13a-07cdeb5056e9@intel.com>", "In-Reply-To": "<3f68ec75-417b-1587-e13a-07cdeb5056e9@intel.com>", "From": "David Marchand <david.marchand@redhat.com>", "Date": "Thu, 9 Jul 2020 17:20:09 +0200", "Message-ID": "\n <CAJFAV8yh8-s0wjtOeA5J9=a1pZ0P7+s7oRpumB2f=6VpgL2A5Q@mail.gmail.com>", "To": "\"Medvedkin, Vladimir\" <vladimir.medvedkin@intel.com>", "Cc": "dev <dev@dpdk.org>, \"Ananyev, Konstantin\" <konstantin.ananyev@intel.com>,\n Bruce Richardson <bruce.richardson@intel.com>", "X-Mimecast-Spam-Score": "0", "X-Mimecast-Originator": "redhat.com", "Content-Type": "text/plain; charset=\"UTF-8\"", "Subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n <mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://mails.dpdk.org/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<https://mails.dpdk.org/listinfo/dev>,\n <mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "addressed": null }, { "id": 115641, "web_url": "https://patches.dpdk.org/comment/115641/", "msgid": "<3f68ec75-417b-1587-e13a-07cdeb5056e9@intel.com>", "list_archive_url": "https://inbox.dpdk.org/dev/3f68ec75-417b-1587-e13a-07cdeb5056e9@intel.com", "date": "2020-07-09T14:52:50", "subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "submitter": { "id": 1216, "url": "https://patches.dpdk.org/api/people/1216/?format=api", "name": "Vladimir Medvedkin", "email": "vladimir.medvedkin@intel.com" }, "content": "Hi David,\n\nThanks for review\n\nOn 09/07/2020 14:48, David Marchand wrote:\n> On Wed, Jul 8, 2020 at 10:17 PM Vladimir Medvedkin\n> <vladimir.medvedkin@intel.com> wrote:\n>>\n>> New data type to manipulate 512 bit AVX values.\n> \n> The title mentions a \"zmm\" type that is not added by this patch.\n> \n> Maybe instead, \"eal/x86: introduce AVX 512-bit type\"\n> \n\nAgree\n\n> \n>>\n>> Signed-off-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>\n>> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>\n>> ---\n>> lib/librte_eal/x86/include/rte_vect.h | 21 +++++++++++++++++++++\n>> 1 file changed, 21 insertions(+)\n>>\n>> diff --git a/lib/librte_eal/x86/include/rte_vect.h b/lib/librte_eal/x86/include/rte_vect.h\n>> index df5a60762..ae59126bc 100644\n>> --- a/lib/librte_eal/x86/include/rte_vect.h\n>> +++ b/lib/librte_eal/x86/include/rte_vect.h\n>> @@ -13,6 +13,7 @@\n>>\n>> #include <stdint.h>\n>> #include <rte_config.h>\n>> +#include <rte_common.h>\n>> #include \"generic/rte_vect.h\"\n>>\n>> #if (defined(__ICC) || \\\n>> @@ -90,6 +91,26 @@ __extension__ ({ \\\n>> })\n>> #endif /* (defined(__ICC) && __ICC < 1210) */\n>>\n>> +#ifdef __AVX512F__\n>> +\n>> +typedef __m512i __x86_zmm_t;\n> \n> We don't need this interim type, using the native __m512 is enough afaics.\n> \n\nAgree\n\n> Looking at the whole applied series:\n> $ git grep -lw __x86_zmm_t\n> lib/librte_eal/x86/include/rte_vect.h\n> \n> \n>> +\n>> +#define ZMM_SIZE (sizeof(__x86_zmm_t))\n>> +#define ZMM_MASK (ZMM_SIZE - 1)\n> \n> Macros in a public header need a RTE_ prefix + this is x86 specific,\n> then RTE_X86_.\n> \n> Looking at the whole applied series:\n> $ git grep -lw ZMM_SIZE\n> lib/librte_eal/x86/include/rte_vect.h\n> $ git grep -lw ZMM_MASK\n> lib/librte_eal/x86/include/rte_vect.h\n> \n> So I wonder if we need to export it or we can instead just #undef\n> after the struct definition.\n\nI think it's better to undef it\n\n> \n> \n>> +\n>> +typedef union __rte_x86_zmm {\n>> + __x86_zmm_t z;\n>> + ymm_t y[ZMM_SIZE / sizeof(ymm_t)];\n>> + xmm_t x[ZMM_SIZE / sizeof(xmm_t)];\n>> + uint8_t u8[ZMM_SIZE / sizeof(uint8_t)];\n>> + uint16_t u16[ZMM_SIZE / sizeof(uint16_t)];\n>> + uint32_t u32[ZMM_SIZE / sizeof(uint32_t)];\n>> + uint64_t u64[ZMM_SIZE / sizeof(uint64_t)];\n>> + double pd[ZMM_SIZE / sizeof(double)];\n>> +} __rte_aligned(ZMM_SIZE) __rte_x86_zmm_t;\n> \n> I don't understand this forced alignment statement.\n> Would not natural alignment be enough, since all fields in this union\n> have the same size?\n> \n\nSome compilers won't align this union\nhttps://mails.dpdk.org/archives/dev/2020-March/159591.html\n\n>", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@inbox.dpdk.org", "Delivered-To": "patchwork@inbox.dpdk.org", "Received": [ "from dpdk.org (dpdk.org [92.243.14.124])\n\tby inbox.dpdk.org (Postfix) with ESMTP id E3588A0528;\n\tThu, 9 Jul 2020 16:52:56 +0200 (CEST)", "from [92.243.14.124] (localhost [127.0.0.1])\n\tby dpdk.org (Postfix) with ESMTP id A633B1E8EC;\n\tThu, 9 Jul 2020 16:52:55 +0200 (CEST)", "from mga18.intel.com (mga18.intel.com [134.134.136.126])\n by dpdk.org (Postfix) with ESMTP id C1BA11E880\n for <dev@dpdk.org>; Thu, 9 Jul 2020 16:52:53 +0200 (CEST)", "from orsmga007.jf.intel.com ([10.7.209.58])\n by orsmga106.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384;\n 09 Jul 2020 07:52:52 -0700", "from vmedvedk-mobl.ger.corp.intel.com (HELO [10.213.225.93])\n ([10.213.225.93])\n by orsmga007.jf.intel.com with ESMTP; 09 Jul 2020 07:52:50 -0700" ], "IronPort-SDR": [ "\n gf46O67hWJ0qQOHLs7wmwLBorSxwF4Z3sP45/Xu/raSPWH5TgvPhE5/S5gaqiAZZUeWgB/oyKT\n LozYoCixjqFg==", "\n oRckbs5NLBbOK8GxnXno3gHpBSDWu7xEEPwOeVrHxlPCULbq5CtLMCADLXVKnEeTsnDciUtgWD\n a67YGbG2y7FA==" ], "X-IronPort-AV": [ "E=McAfee;i=\"6000,8403,9677\"; a=\"135466191\"", "E=Sophos;i=\"5.75,331,1589266800\"; d=\"scan'208\";a=\"135466191\"", "E=Sophos;i=\"5.75,331,1589266800\"; d=\"scan'208\";a=\"324244020\"" ], "X-Amp-Result": "SKIPPED(no attachment in message)", "X-Amp-File-Uploaded": "False", "X-ExtLoop1": "1", "To": "David Marchand <david.marchand@redhat.com>", "Cc": "dev <dev@dpdk.org>, \"Ananyev, Konstantin\" <konstantin.ananyev@intel.com>,\n Bruce Richardson <bruce.richardson@intel.com>", "References": "<cover.1589890262.git.vladimir.medvedkin@intel.com>\n <cover.1594238609.git.vladimir.medvedkin@intel.com>\n <98b10e12eb46cff65494a94eaf0f04b2dcefd245.1594238610.git.vladimir.medvedkin@intel.com>\n <CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com>", "From": "\"Medvedkin, Vladimir\" <vladimir.medvedkin@intel.com>", "Message-ID": "<3f68ec75-417b-1587-e13a-07cdeb5056e9@intel.com>", "Date": "Thu, 9 Jul 2020 15:52:50 +0100", "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64; rv:68.0) Gecko/20100101\n Thunderbird/68.10.0", "MIME-Version": "1.0", "In-Reply-To": "\n <CAJFAV8z5Ft=7HaEukKoapWfiOhLLO0Rf6TVtHcdnvnsF2GEGcA@mail.gmail.com>", "Content-Type": "text/plain; charset=utf-8; format=flowed", "Content-Language": "en-US", "Content-Transfer-Encoding": "7bit", "Subject": "Re: [dpdk-dev] [PATCH v4 1/8] eal: introduce zmm type for AVX\n\t512-bit", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n <mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://mails.dpdk.org/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<https://mails.dpdk.org/listinfo/dev>,\n <mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "addressed": null } ]