[{"id":3675235,"web_url":"http://patchwork.ozlabs.org/comment/3675235/","msgid":"","list_archive_url":null,"date":"2026-04-09T10:11:34","subject":"Re: [ovs-dev] [PATCH] dpif-netdev: Remove pmd-stats-show in favor\n of pmd-perf-show.","submitter":{"id":70613,"url":"http://patchwork.ozlabs.org/api/people/70613/","name":"Eelco Chaudron","email":"echaudro@redhat.com"},"content":"On 8 Apr 2026, at 19:16, Ilya Maximets wrote:\n\n> The 'pmd-perf-show' command provides all the same information and more.\n> It is also better visually structured and easier to read as a result.\n>\n> Let's remove the old 'pmd-stats-show' command, as there is no real need\n> to have two commands reporting the same data.\n>\n> The only difference until now was that 'pmd-perf-show' didn't provide\n> information for the \"main\" thread. This change makes it report the\n> statistics for the aggregated \"main\" thread as well, omitting things\n> related to CPU cycles, as we can't collect those for threads that are\n> not pinned. For the same reason histograms are also always disabled.\n> Omission is done by checking the total number of iterations to be zero.\n> \"main\" thread doesn't start/end iterations.\n>\n> The actual unixctl command is preserved undocumented and serves as an\n> alias for 'pmd-perf-show'. This should allow old scripts that are just\n> capturing the output for humans (or LLMs?) to read to keep working.\n> Note, however, that the exact output format for unixctl commands was\n> never a guarantee, so scripts that attempt to parse the output may\n> still break.\n>\n> Signed-off-by: Ilya Maximets \n> ---\n>\n> Note: I believe the change in system-dpdk-offloads.at is correct,\n> but I didn't run the testsuite, as I have no hardware for it.\n>\n\nNot a review, but I want to confirm that the offload testsuite runs fine.\n\n//Eelco","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","ovs-dev@openvswitch.org"],"Delivered-To":["patchwork-incoming@legolas.ozlabs.org","ovs-dev@lists.linuxfoundation.org"],"Authentication-Results":["legolas.ozlabs.org;\n\tdkim=fail reason=\"signature verification failed\" (1024-bit key;\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=gAKzbkQp;\n\tdkim-atps=neutral","legolas.ozlabs.org;\n spf=pass (sender SPF authorized) smtp.mailfrom=openvswitch.org\n (client-ip=2605:bc80:3010::133; helo=smtp2.osuosl.org;\n envelope-from=ovs-dev-bounces@openvswitch.org; receiver=patchwork.ozlabs.org)","smtp2.osuosl.org;\n\tdkim=fail reason=\"signature verification failed\" (1024-bit key)\n header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=gAKzbkQp","smtp1.osuosl.org; dmarc=pass (p=quarantine dis=none)\n header.from=redhat.com","smtp1.osuosl.org; dkim=pass (1024-bit key,\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=gAKzbkQp"],"Received":["from smtp2.osuosl.org (smtp2.osuosl.org [IPv6:2605:bc80:3010::133])\n\t(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)\n\t key-exchange x25519 server-signature ECDSA (secp384r1) server-digest SHA384)\n\t(No client certificate requested)\n\tby legolas.ozlabs.org (Postfix) with ESMTPS id 4frwhN0H4dz1yCv\n\tfor ; Thu, 09 Apr 2026 20:11:47 +1000 (AEST)","from localhost (localhost [127.0.0.1])\n\tby smtp2.osuosl.org (Postfix) with ESMTP id 718F5403F1;\n\tThu, 9 Apr 2026 10:11:46 +0000 (UTC)","from smtp2.osuosl.org ([127.0.0.1])\n by localhost (smtp2.osuosl.org [127.0.0.1]) (amavis, port 10024) with ESMTP\n id k8U76Sx7B9xt; Thu, 9 Apr 2026 10:11:45 +0000 (UTC)","from lists.linuxfoundation.org (lf-lists.osuosl.org\n [IPv6:2605:bc80:3010:104::8cd3:938])\n\tby smtp2.osuosl.org (Postfix) with ESMTPS id 9409E400BA;\n\tThu, 9 Apr 2026 10:11:45 +0000 (UTC)","from lf-lists.osuosl.org (localhost [127.0.0.1])\n\tby lists.linuxfoundation.org (Postfix) with ESMTP id 58A49C054A;\n\tThu, 9 Apr 2026 10:11:45 +0000 (UTC)","from smtp1.osuosl.org (smtp1.osuosl.org [IPv6:2605:bc80:3010::138])\n by lists.linuxfoundation.org (Postfix) with ESMTP id 218FFC0549\n for ; Thu, 9 Apr 2026 10:11:44 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n by smtp1.osuosl.org (Postfix) with ESMTP id 0037B81036\n for ; Thu, 9 Apr 2026 10:11:43 +0000 (UTC)","from smtp1.osuosl.org ([127.0.0.1])\n by localhost (smtp1.osuosl.org [127.0.0.1]) (amavis, port 10024) with ESMTP\n id sBcY1xJ6RPTQ for ;\n Thu, 9 Apr 2026 10:11:43 +0000 (UTC)","from us-smtp-delivery-124.mimecast.com\n (us-smtp-delivery-124.mimecast.com [170.10.133.124])\n by smtp1.osuosl.org (Postfix) with ESMTPS id CB53D8100B\n for ; Thu, 9 Apr 2026 10:11:42 +0000 (UTC)","from mail-ej1-f70.google.com (mail-ej1-f70.google.com\n [209.85.218.70]) by relay.mimecast.com with ESMTP with STARTTLS\n (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id\n us-mta-180-_F2cCYfQMlWhK8ZdY2p6CQ-1; Thu, 09 Apr 2026 06:11:39 -0400","by mail-ej1-f70.google.com with SMTP id\n a640c23a62f3a-b9c0bdea9faso59839466b.2\n for ; Thu, 09 Apr 2026 03:11:39 -0700 (PDT)","from [10.44.32.212] (5920ab7b.static.cust.trined.nl.\n [89.32.171.123])\n by smtp.gmail.com with ESMTPSA id\n a640c23a62f3a-b9c3c97218esm720715566b.9.2026.04.09.03.11.37\n (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128);\n Thu, 09 Apr 2026 03:11:37 -0700 (PDT)"],"X-Virus-Scanned":["amavis at osuosl.org","amavis at osuosl.org"],"X-Comment":"SPF check N/A for local connections -\n client-ip=2605:bc80:3010:104::8cd3:938; helo=lists.linuxfoundation.org;\n envelope-from=ovs-dev-bounces@openvswitch.org; receiver= ","DKIM-Filter":["OpenDKIM Filter v2.11.0 smtp2.osuosl.org 9409E400BA","OpenDKIM Filter v2.11.0 smtp1.osuosl.org CB53D8100B"],"Received-SPF":"Pass (mailfrom) identity=mailfrom; client-ip=170.10.133.124;\n helo=us-smtp-delivery-124.mimecast.com; envelope-from=echaudro@redhat.com;\n receiver=","DMARC-Filter":"OpenDMARC Filter v1.4.2 smtp1.osuosl.org CB53D8100B","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n s=mimecast20190719; t=1775729501;\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=sNyT5LxfXOp3SwRfxhmPVIVOwgy1HvZ9o5dY0HsLuIQ=;\n b=gAKzbkQpEF1FlEPboBBotSfQdPy8Pb0Oa0P91Mkmxjdxt+i2q5zedMWLTi2aIgaFOuPCX4\n 0/dCUG0pWpCOqoD9Yc7iPgP9IdVoiEtjWNdl0AZ1iBknc8Ur5ze83GGJ93YTgFzRkgpOlm\n yfeBpR/xMfoiRKhuuqEQX8aRQnoezV0=","X-MC-Unique":"_F2cCYfQMlWhK8ZdY2p6CQ-1","X-Mimecast-MFC-AGG-ID":"_F2cCYfQMlWhK8ZdY2p6CQ_1775729498","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n d=1e100.net; s=20251104; t=1775729498; x=1776334298;\n h=mime-version:references:in-reply-to:message-id:date:subject:cc:to\n :from:x-gm-gg:x-gm-message-state:from:to:cc:subject:date:message-id\n :reply-to;\n bh=sNyT5LxfXOp3SwRfxhmPVIVOwgy1HvZ9o5dY0HsLuIQ=;\n b=XFcMsuC2GVV29qPc0JwbQgS6J14SP0pABhK98BnNzpnefJPhhFwfUXi9vjEn7jEk4V\n voF19ujB0R/tEaME+JDPwnsZ3EL40m9nXvMKJrs0VxsDZJPSwC+rXdVHuAVDx1IFkD85\n FQbiUbaQbJFhqEBM7oRlbwqRocl0VpofH/B3HMk3/N7LkIt3eqJVpWz2h1Ez7G3U1m5D\n ziCr/rDVxhZvGHyEpBt/CI4cbS4J8DimHIBTUEKX3shFyXNMyWjbo1BQOw8D0rSWbB0z\n 57c0hqLDpW1M9e88PVZ0e9fUo6g/uo37rRD2bpJspqCP+HWGhipcpGGFkUK2//lQ4Hb4\n f7XQ==","X-Gm-Message-State":"AOJu0YzvNGp3cArOsz1g2FSXipdMSw48BnDqUJ3YKFdueutaH9tAj0Wb\n bQFfip773cXPk7PaBcOJsGilygFab2DAczuv6hK56WKA6o1ShzZOPEmmqLmKabv4RVRLv4bvw0U\n sLj+BZgEAAsBy9TzxkmPkDRFWYWLBuig1Xps/XFaKNnZt26RPI6MHsmCdlHA=","X-Gm-Gg":"AeBDieupgnXF1YY7/PlswnpllifENTt02m++G+aHNEGOSgN4yPRNM7vNH/5L3qr/jUR\n 153LBZlsEgbYyrtxbxJP6ISpSqvPNhq8ubtmya8OkPlbyLbq/uwEUzXWbMFvG75yWQO9pGiKyS0\n MoNiVz2rGLCQgaMkWU1Kg1Uv7eprpA42Y5VkyJmpX56nfsp49ENqCfEmoUXp3nZLOwC3KWMjZxb\n slk2KRQPrh5htc1cVwPNsNjlcmXjLIYtbSEcTP5I/t/9nmqvIx2AashnSvNh9z5M2aq2lYvSUyx\n eS5nWv7HOYTS08/sBE7T0JPjP4A/2E0gF/uAnMESAsVvEjvExDf0LE5XNrroo4VOO1FQutVebZa\n CFFFjr2LhO+p7FMJzUJbKr+BRreOpbjr4y4xrsGvsA3WXB9rq73Slg0DCZfLZgBLwKAXDhx4=","X-Received":["by 2002:a17:906:9f88:b0:b9d:14e9:8416 with SMTP id\n a640c23a62f3a-b9d476a401emr150532766b.37.1775729498356;\n Thu, 09 Apr 2026 03:11:38 -0700 (PDT)","by 2002:a17:906:9f88:b0:b9d:14e9:8416 with SMTP id\n a640c23a62f3a-b9d476a401emr150530766b.37.1775729497883;\n Thu, 09 Apr 2026 03:11:37 -0700 (PDT)"],"To":"Ilya Maximets ","Date":"Thu, 09 Apr 2026 12:11:34 +0200","X-Mailer":"MailMate (2.0r6290)","Message-ID":"","In-Reply-To":"<20260408171644.1404735-1-i.maximets@ovn.org>","References":"<20260408171644.1404735-1-i.maximets@ovn.org>","MIME-Version":"1.0","X-Mimecast-Spam-Score":"0","X-Mimecast-MFC-PROC-ID":"rg83ncKHPUqAsRm8snMci-HPPbS77r_e4Z8aib2UcP0_1775729498","X-Mimecast-Originator":"redhat.com","Subject":"Re: [ovs-dev] [PATCH] dpif-netdev: Remove pmd-stats-show in favor\n of pmd-perf-show.","X-BeenThere":"ovs-dev@openvswitch.org","X-Mailman-Version":"2.1.30","Precedence":"list","List-Id":"","List-Unsubscribe":",\n ","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n ","From":"Eelco Chaudron via dev ","Reply-To":"Eelco Chaudron ","Cc":"ovs-dev@openvswitch.org","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"ovs-dev-bounces@openvswitch.org","Sender":"\"dev\" "}},{"id":3678085,"web_url":"http://patchwork.ozlabs.org/comment/3678085/","msgid":"<3b04299a-f445-45da-8007-70320ccd1c51@redhat.com>","list_archive_url":null,"date":"2026-04-16T10:40:58","subject":"Re: [ovs-dev] [PATCH] dpif-netdev: Remove pmd-stats-show in favor\n of pmd-perf-show.","submitter":{"id":70482,"url":"http://patchwork.ozlabs.org/api/people/70482/","name":"Kevin Traynor","email":"ktraynor@redhat.com"},"content":"On 4/8/26 6:16 PM, Ilya Maximets wrote:\n> The 'pmd-perf-show' command provides all the same information and more.\n> It is also better visually structured and easier to read as a result.\n> \n> Let's remove the old 'pmd-stats-show' command, as there is no real need\n> to have two commands reporting the same data.\n> \n> The only difference until now was that 'pmd-perf-show' didn't provide\n> information for the \"main\" thread. This change makes it report the\n> statistics for the aggregated \"main\" thread as well, omitting things\n> related to CPU cycles, as we can't collect those for threads that are\n> not pinned. For the same reason histograms are also always disabled.\n> Omission is done by checking the total number of iterations to be zero.\n> \"main\" thread doesn't start/end iterations.\n> \n> The actual unixctl command is preserved undocumented and serves as an\n> alias for 'pmd-perf-show'. This should allow old scripts that are just\n> capturing the output for humans (or LLMs?) to read to keep working.\n> Note, however, that the exact output format for unixctl commands was\n> never a guarantee, so scripts that attempt to parse the output may\n> still break.\n> \n\nHi Ilya,\n\nthanks for this, couple of comment below, otherwise LGTM.\n\n> Signed-off-by: Ilya Maximets \n> ---\n> \n> Note: I believe the change in system-dpdk-offloads.at is correct,\n> but I didn't run the testsuite, as I have no hardware for it.\n> \n> Documentation/intro/install/afxdp.rst | 2 +-\n> Documentation/intro/install/dpdk.rst | 2 +-\n> Documentation/topics/dpdk/bridge.rst | 4 +-\n> Documentation/topics/dpdk/pmd.rst | 4 -\n> NEWS | 4 +\n> lib/dpif-netdev-perf.c | 39 +++--\n> lib/dpif-netdev-perf.h | 2 +-\n> lib/dpif-netdev-unixctl.man | 62 ++++---\n> lib/dpif-netdev.c | 158 ++++--------------\n> tests/dpif-netdev.at | 20 +--\n> tests/pmd.at | 47 +++---\n> tests/system-dpdk-offloads.at | 8 +-\n> .../plugins/system-logs/openvswitch.xml | 2 +-\n> 13 files changed, 136 insertions(+), 218 deletions(-)\n> \n> diff --git a/Documentation/intro/install/afxdp.rst b/Documentation/intro/install/afxdp.rst\n> index 63a10e328..07225a885 100644\n> --- a/Documentation/intro/install/afxdp.rst\n> +++ b/Documentation/intro/install/afxdp.rst\n> @@ -273,7 +273,7 @@ Measure your system call rate by doing::\n> \n> Or, use OVS pmd tool::\n> \n> - ovs-appctl dpif-netdev/pmd-stats-show\n> + ovs-appctl dpif-netdev/pmd-perf-show\n> \n> \n> Example Script\n> diff --git a/Documentation/intro/install/dpdk.rst b/Documentation/intro/install/dpdk.rst\n> index 6f4687bde..d5c897e8b 100644\n> --- a/Documentation/intro/install/dpdk.rst\n> +++ b/Documentation/intro/install/dpdk.rst\n> @@ -709,7 +709,7 @@ level:\n> \n> The average number of packets per output batch can be checked in PMD stats::\n> \n> - $ ovs-appctl dpif-netdev/pmd-stats-show\n> + $ ovs-appctl dpif-netdev/pmd-perf-show\n> \n> Limitations\n> ------------\n> diff --git a/Documentation/topics/dpdk/bridge.rst b/Documentation/topics/dpdk/bridge.rst\n> index 03c4dd4e3..4468b904b 100644\n> --- a/Documentation/topics/dpdk/bridge.rst\n> +++ b/Documentation/topics/dpdk/bridge.rst\n> @@ -103,7 +103,7 @@ the packet itself and others (for example, VLAN tag or Ethernet type) can be\n> extracted without fully parsing the packet. This allows OVS to significantly\n> speed up packet forwarding for these flows with simple match criteria.\n> Statistics on the number of packets matched in this way can be found in a\n> -`simple match hits` counter of `ovs-appctl dpif-netdev/pmd-stats-show` command.\n> +`Simple Match hits` counter of `ovs-appctl dpif-netdev/pmd-perf-show` command.\n> \n> EMC Insertion Probability\n> -------------------------\n> @@ -127,7 +127,7 @@ If ``N`` is set to 1, an insertion will be performed for every flow. If set to\n> With default ``N`` set to 100, higher megaflow hits will occur initially as\n> observed with pmd stats::\n> \n> - $ ovs-appctl dpif-netdev/pmd-stats-show\n> + $ ovs-appctl dpif-netdev/pmd-perf-show\n> \n> For certain traffic profiles with many parallel flows, it's recommended to set\n> ``N`` to '0' to achieve higher forwarding performance.\n> diff --git a/Documentation/topics/dpdk/pmd.rst b/Documentation/topics/dpdk/pmd.rst\n> index 2e8cf5edb..1589d521c 100644\n> --- a/Documentation/topics/dpdk/pmd.rst\n> +++ b/Documentation/topics/dpdk/pmd.rst\n> @@ -57,10 +57,6 @@ PMD Thread Statistics\n> \n> To show current stats::\n> \n> - $ ovs-appctl dpif-netdev/pmd-stats-show\n> -\n> -or::\n> -\n> $ ovs-appctl dpif-netdev/pmd-perf-show\n> \n> Detailed performance metrics for ``pmd-perf-show`` can also be enabled::\n> diff --git a/NEWS b/NEWS\n> index 1a3044cbf..b35bcff6e 100644\n> --- a/NEWS\n> +++ b/NEWS\n> @@ -3,6 +3,10 @@ Post-v3.7.0\n> - Userspace datapath:\n> * ARP/ND lookups for native tunnel are now rate limited. The holdout\n> timer can be configured with 'tnl/neigh/retrans_time'.\n> + - ovs-appctl:\n> + * 'dpif-netdev/pmd-stats-show' command was removed in favor of the more\n> + informative and better structured 'dpif-netdev/pmd-perf-show', which\n> + now also provides statistics for the \"main\" thread.\n> \n> \n> v3.7.0 - 16 Feb 2026\n> diff --git a/lib/dpif-netdev-perf.c b/lib/dpif-netdev-perf.c\n> index 1cd4ee084..ba370d7c1 100644\n> --- a/lib/dpif-netdev-perf.c\n> +++ b/lib/dpif-netdev-perf.c\n> @@ -233,7 +233,8 @@ pmd_perf_format_overall_stats(struct ds *str, struct pmd_perf_stats *s,\n> uint64_t sleep_iter = stats[PMD_SLEEP_ITER];\n> uint64_t tot_sleep_cycles = stats[PMD_CYCLES_SLEEP];\n> \n> - ds_put_format(str,\n> + if (tot_iter) {\n\nWhile the change is to cater for main thread, it also changes display\nfor pmd cores with no iterations. Previously there was:\n\npmd thread numa_id 1 core_id 9:\n\n Iterations: 0 (0.00 us/it)\n - Used TSC cycles: 0 ( 0.0 % of total cycles)\n - idle iterations: 0 ( 0.0 % of used cycles)\n - busy iterations: 0 ( 0.0 % of used cycles)\n - sleep iterations: 0 ( 0.0 % of iterations)\n Sleep time (us): 0 ( 0 us/iteration avg.)\n Rx packets: 0\n Tx packets: 0\n\nThat is now changed so that iterations of zero are implicit.\n\npmd thread numa_id 1 core_id 11:\n\n Rx packets: 0\n Tx packets: 0\n\n\nConsidering we can also have a case where there are iterations but no\npackets like below, i think we should leave zero iterations explicit for\npmd thread as it currently is.\n\npmd thread numa_id 0 core_id 8:\n\n Iterations: 17369522 (0.15 us/it)\n - Used TSC cycles: 6816002305 ( 80.1 % of total cycles)\n - idle iterations: 17369522 (100.0 % of used cycles)\n - busy iterations: 0 ( 0.0 % of used cycles)\n - sleep iterations: 0 ( 0.0 % of iterations)\n Sleep time (us): 0 ( 0 us/iteration avg.)\n Rx packets: 0\n Tx packets: 0\n\n\nI agree that for main thread it doesn't make sense to display it, so\nmaybe we could add a bool to pmd_perf_format_overall_stats() args to\nmake showing the iteration section conditional, then set differently\nwhen the non pmd core in pmd_info_show_perf().\n\n> + ds_put_format(str,\n> \" Iterations: %12\"PRIu64\" (%.2f us/it)\\n\"\n\n> \" - Used TSC cycles: %12\"PRIu64\" (%5.1f %% of total cycles)\\n\"\n\naside from comments above, with current version there is no need to\nre-check for tot_iter (not shown in diff)\n\n> \" - idle iterations: %12\"PRIu64\" (%5.1f %% of used cycles)\\n\"\n> @@ -252,9 +253,18 @@ pmd_perf_format_overall_stats(struct ds *str, struct pmd_perf_stats *s,\n> sleep_iter, tot_iter ? 100.0 * sleep_iter / tot_iter : 0,\n> tot_sleep_cycles * us_per_cycle,\n> sleep_iter ? (tot_sleep_cycles * us_per_cycle) / sleep_iter : 0);\n> + }\n> if (rx_packets > 0) {\n> ds_put_format(str,\n> - \" Rx packets: %12\"PRIu64\" (%.0f Kpps, %.0f cycles/pkt)\\n\"\n> + \" Rx packets: %12\"PRIu64\" (%.0f Kpps\",\n> + rx_packets, (rx_packets / duration) / 1000);\n> + if (tot_iter) {\n> + ds_put_format(str, \", %.0f cycles/pkt\",\n> + 1.0 * stats[PMD_CYCLES_ITER_BUSY] / rx_packets);\n> + }\n> + ds_put_cstr(str, \")\\n\");\n> +\n> + ds_put_format(str,\n> \" Datapath passes: %12\"PRIu64\" (%.2f passes/pkt)\\n\"\n> \" - PHWOL hits: %12\"PRIu64\" (%5.1f %%)\\n\"\n> \" - MFEX Opt hits: %12\"PRIu64\" (%5.1f %%)\\n\"\n> @@ -262,11 +272,7 @@ pmd_perf_format_overall_stats(struct ds *str, struct pmd_perf_stats *s,\n> \" - EMC hits: %12\"PRIu64\" (%5.1f %%)\\n\"\n> \" - SMC hits: %12\"PRIu64\" (%5.1f %%)\\n\"\n> \" - Megaflow hits: %12\"PRIu64\" (%5.1f %%, %.2f \"\n> - \"subtbl lookups/hit)\\n\"\n> - \" - Upcalls: %12\"PRIu64\" (%5.1f %%, %.1f us/upcall)\\n\"\n> - \" - Lost upcalls: %12\"PRIu64\" (%5.1f %%)\\n\",\n> - rx_packets, (rx_packets / duration) / 1000,\n> - 1.0 * stats[PMD_CYCLES_ITER_BUSY] / rx_packets,\n> + \"subtbl lookups/hit)\\n\",\n> passes, 1.0 * passes / rx_packets,\n> stats[PMD_STAT_PHWOL_HIT],\n> 100.0 * stats[PMD_STAT_PHWOL_HIT] / passes,\n> @@ -282,11 +288,20 @@ pmd_perf_format_overall_stats(struct ds *str, struct pmd_perf_stats *s,\n> 100.0 * stats[PMD_STAT_MASKED_HIT] / passes,\n> stats[PMD_STAT_MASKED_HIT]\n> ? 1.0 * stats[PMD_STAT_MASKED_LOOKUP] / stats[PMD_STAT_MASKED_HIT]\n> - : 0,\n> - upcalls, 100.0 * upcalls / passes,\n> - upcalls ? (upcall_cycles * us_per_cycle) / upcalls : 0,\n> - stats[PMD_STAT_LOST],\n> - 100.0 * stats[PMD_STAT_LOST] / passes);\n> + : 0);\n> +\n> + ds_put_format(str,\n> + \" - Upcalls: %12\"PRIu64\" (%5.1f %%\",\n> + upcalls, 100.0 * upcalls / passes);\n> + if (tot_iter) {\n> + ds_put_format(str, \", %.1f us/upcall\",\n> + upcalls ? (upcall_cycles * us_per_cycle) / upcalls : 0);\n> + }\n> + ds_put_cstr(str, \")\\n\");\n> +\n> + ds_put_format(str,\n> + \" - Lost upcalls: %12\"PRIu64\" (%5.1f %%)\\n\",\n> + stats[PMD_STAT_LOST], 100.0 * stats[PMD_STAT_LOST] / passes);\n> } else {\n> ds_put_format(str,\n> \" Rx packets: %12d\\n\", 0);\n> diff --git a/lib/dpif-netdev-perf.h b/lib/dpif-netdev-perf.h\n> index 84beced15..8a41afa8a 100644\n> --- a/lib/dpif-netdev-perf.h\n> +++ b/lib/dpif-netdev-perf.h\n> @@ -317,7 +317,7 @@ void pmd_perf_read_counters(struct pmd_perf_stats *s,\n> * NON-PMD they might be updated from multiple threads, but we can live\n> * with losing a rare update as 100% accuracy is not required.\n> * However, as counters are read for display from outside the PMD thread\n> - * with e.g. pmd-stats-show, we make sure that the 64-bit read and store\n> + * with e.g. pmd-perf-show, we make sure that the 64-bit read and store\n> * operations are atomic also on 32-bit systems so that readers cannot\n> * not read garbage. On 64-bit systems this incurs no overhead. */\n> \n> diff --git a/lib/dpif-netdev-unixctl.man b/lib/dpif-netdev-unixctl.man\n> index 8cd847416..3d5ab437c 100644\n> --- a/lib/dpif-netdev-unixctl.man\n> +++ b/lib/dpif-netdev-unixctl.man\n> @@ -6,44 +6,22 @@ argument can be omitted. By default the commands present data for all pmd\n> threads in the datapath. By specifying the \"-pmd Core\" option one can filter\n> the output for a single pmd in the datapath.\n> .\n> -.IP \"\\fBdpif-netdev/pmd-stats-show\\fR [\\fB-pmd\\fR \\fIcore\\fR] [\\fIdp\\fR]\"\n> -Shows performance statistics for one or all pmd threads of the datapath\n> -\\fIdp\\fR. The special thread \"main\" sums up the statistics of every non pmd\n> -thread.\n> -\n> -The sum of \"phwol hits\", \"simple match hits\", \"emc hits\", \"smc hits\",\n> -\"megaflow hits\" and \"miss\" is the number of packet lookups performed by the\n> -datapath. Beware that a recirculated packet experiences one additional lookup\n> -per recirculation, so there may be more lookups than forwarded packets in the\n> -datapath.\n> -\n> -The MFEX Opt hits displays the number of packets that are processed by the\n> -optimized miniflow extract implementations.\n> -\n> -Cycles are counted using the TSC or similar facilities (when available on\n> -the platform). The duration of one cycle depends on the processing platform.\n> -\n> -\"idle cycles\" refers to cycles spent in PMD iterations not forwarding any\n> -any packets. \"processing cycles\" refers to cycles spent in PMD iterations\n> -forwarding at least one packet, including the cost for polling, processing and\n> -transmitting said packets.\n> -\n> -To reset these counters use \\fBdpif-netdev/pmd-stats-clear\\fR.\n> -.\n> .IP \"\\fBdpif-netdev/pmd-stats-clear\\fR [\\fIdp\\fR]\"\n> Resets to zero the per pmd thread performance numbers shown by the\n> -\\fBdpif-netdev/pmd-stats-show\\fR and \\fBdpif-netdev/pmd-perf-show\\fR commands.\n> -It will NOT reset datapath or bridge statistics, only the values shown by\n> -the above commands.\n> +\\fBdpif-netdev/pmd-perf-show\\fR command. It will NOT reset datapath or bridge\n> +statistics, only the values shown by the above command.\n> .\n> .IP \"\\fBdpif-netdev/pmd-perf-show\\fR [\\fB-nh\\fR] [\\fB-it\\fR \\fIiter_len\\fR] \\\n> [\\fB-ms\\fR \\fIms_len\\fR] [\\fB-pmd\\fR \\fIcore\\fR] [\\fIdp\\fR]\"\n> Shows detailed performance metrics for one or all pmds threads of the\n> -user space datapath.\n> +user space datapath. The special thread \"main\" sums up the statistics of every\n> +non pmd thread.\n> \n> -The collection of detailed statistics can be controlled by a new\n> -configuration parameter \"other_config:pmd-perf-metrics\". By default it\n> -is disabled. The run-time overhead, when enabled, is in the order of 1%.\n> +The collection of additional detailed statistics can be controlled by a\n> +configuration parameter \\fBother-config:pmd-perf-metrics\\fR. By default it is\n> +disabled. The run-time overhead, when enabled, is in the order of 1%.\n> +\n> +Collected statistics include:\n> \n> .RS\n> .IP\n> @@ -153,8 +131,26 @@ pmd thread numa_id 0 core_id 1:\n> .RE\n> .IP\n> Here \"Rx packets\" actually reflects the number of packets forwarded by the\n> -datapath. \"Datapath passes\" matches the number of packet lookups as\n> -reported by the \\fBdpif-netdev/pmd-stats-show\\fR command.\n> +datapath.\n> +\n> +The sum of \"PHWOL hits\", \"Simple Match hits\", \"EMC hits\", \"SMC hits\",\n> +\"Megaflow hits\" and \"Upcalls\" is the number of packet lookups performed by the\n> +datapath and it is reported as \"Datapath passes\". Beware that a recirculated\n> +packet experiences one additional lookup per recirculation, so there may be\n> +more lookups than forwarded packets in the datapath.\n> +\n> +The \"MFEX Opt hits\" displays the number of packets that are processed by the\n> +optimized miniflow extract implementations.\n> +\n> +Cycles are counted using the TSC or similar facilities (when available on\n> +the platform). The duration of one cycle depends on the processing platform.\n> +Statistics based on cycles are not reported for the \"main\" thread, since the\n> +accurate accounting of CPU cycles is not possible in this case.\n> +\n> +\"idle iterations\" refers to PMD iterations that didn't not result in processing\n\ntypo, \"didn't not\"\n\nthanks,\nKevin.\n\n> +any packets. \"busy iterations\" refers to PMD iterations that included\n> +processing of at least one packet. The reported used TSC cycles include the\n> +cost for polling, processing and transmitting said packets.\n> \n> To reset the counters and start a new measurement use\n> \\fBdpif-netdev/pmd-stats-clear\\fR.\n> diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c\n> index 9df05c4c2..db5823a91 100644\n> --- a/lib/dpif-netdev.c\n> +++ b/lib/dpif-netdev.c\n> @@ -620,8 +620,7 @@ get_dp_netdev(const struct dpif *dpif)\n> }\n> \f\n> enum pmd_info_type {\n> - PMD_INFO_SHOW_STATS, /* Show how cpu cycles are spent. */\n> - PMD_INFO_CLEAR_STATS, /* Set the cycles count to 0. */\n> + PMD_INFO_CLEAR_STATS, /* Set the cycle and the packet counters to 0. */\n> PMD_INFO_SHOW_RXQ, /* Show poll lists of pmd threads. */\n> PMD_INFO_PERF_SHOW, /* Show pmd performance details. */\n> PMD_INFO_SLEEP_SHOW, /* Show max sleep configuration details. */\n> @@ -641,127 +640,42 @@ format_pmd_thread(struct ds *reply, struct dp_netdev_pmd_thread *pmd)\n> ds_put_cstr(reply, \":\\n\");\n> }\n> \n> -static void\n> -pmd_info_show_stats(struct ds *reply,\n> - struct dp_netdev_pmd_thread *pmd)\n> -{\n> - uint64_t stats[PMD_N_STATS];\n> - uint64_t total_cycles, total_packets;\n> - double passes_per_pkt = 0;\n> - double lookups_per_hit = 0;\n> - double packets_per_batch = 0;\n> -\n> - pmd_perf_read_counters(&pmd->perf_stats, stats);\n> - total_cycles = stats[PMD_CYCLES_ITER_IDLE]\n> - + stats[PMD_CYCLES_ITER_BUSY];\n> - total_packets = stats[PMD_STAT_RECV];\n> -\n> - format_pmd_thread(reply, pmd);\n> -\n> - if (total_packets > 0) {\n> - passes_per_pkt = (total_packets + stats[PMD_STAT_RECIRC])\n> - / (double) total_packets;\n> - }\n> - if (stats[PMD_STAT_MASKED_HIT] > 0) {\n> - lookups_per_hit = stats[PMD_STAT_MASKED_LOOKUP]\n> - / (double) stats[PMD_STAT_MASKED_HIT];\n> - }\n> - if (stats[PMD_STAT_SENT_BATCHES] > 0) {\n> - packets_per_batch = stats[PMD_STAT_SENT_PKTS]\n> - / (double) stats[PMD_STAT_SENT_BATCHES];\n> - }\n> -\n> - ds_put_format(reply,\n> - \" packets received: %\"PRIu64\"\\n\"\n> - \" packet recirculations: %\"PRIu64\"\\n\"\n> - \" avg. datapath passes per packet: %.02f\\n\"\n> - \" phwol hits: %\"PRIu64\"\\n\"\n> - \" mfex opt hits: %\"PRIu64\"\\n\"\n> - \" simple match hits: %\"PRIu64\"\\n\"\n> - \" emc hits: %\"PRIu64\"\\n\"\n> - \" smc hits: %\"PRIu64\"\\n\"\n> - \" megaflow hits: %\"PRIu64\"\\n\"\n> - \" avg. subtable lookups per megaflow hit: %.02f\\n\"\n> - \" miss with success upcall: %\"PRIu64\"\\n\"\n> - \" miss with failed upcall: %\"PRIu64\"\\n\"\n> - \" avg. packets per output batch: %.02f\\n\",\n> - total_packets, stats[PMD_STAT_RECIRC],\n> - passes_per_pkt, stats[PMD_STAT_PHWOL_HIT],\n> - stats[PMD_STAT_MFEX_OPT_HIT],\n> - stats[PMD_STAT_SIMPLE_HIT],\n> - stats[PMD_STAT_EXACT_HIT],\n> - stats[PMD_STAT_SMC_HIT],\n> - stats[PMD_STAT_MASKED_HIT],\n> - lookups_per_hit, stats[PMD_STAT_MISS], stats[PMD_STAT_LOST],\n> - packets_per_batch);\n> -\n> - if (total_cycles == 0) {\n> - return;\n> - }\n> -\n> - ds_put_format(reply,\n> - \" idle cycles: %\"PRIu64\" (%.02f%%)\\n\"\n> - \" processing cycles: %\"PRIu64\" (%.02f%%)\\n\",\n> - stats[PMD_CYCLES_ITER_IDLE],\n> - stats[PMD_CYCLES_ITER_IDLE] / (double) total_cycles * 100,\n> - stats[PMD_CYCLES_ITER_BUSY],\n> - stats[PMD_CYCLES_ITER_BUSY] / (double) total_cycles * 100);\n> -\n> - if (total_packets == 0) {\n> - return;\n> - }\n> -\n> - ds_put_format(reply,\n> - \" avg cycles per packet: %.02f (%\"PRIu64\"/%\"PRIu64\")\\n\",\n> - total_cycles / (double) total_packets,\n> - total_cycles, total_packets);\n> -\n> - ds_put_format(reply,\n> - \" avg processing cycles per packet: \"\n> - \"%.02f (%\"PRIu64\"/%\"PRIu64\")\\n\",\n> - stats[PMD_CYCLES_ITER_BUSY] / (double) total_packets,\n> - stats[PMD_CYCLES_ITER_BUSY], total_packets);\n> -}\n> -\n> static void\n> pmd_info_show_perf(struct ds *reply,\n> struct dp_netdev_pmd_thread *pmd,\n> struct pmd_perf_params *par)\n> {\n> - if (pmd->core_id != NON_PMD_CORE_ID) {\n\n\n\n\n\n> - char *time_str =\n> - xastrftime_msec(\"%H:%M:%S.###\", time_wall_msec(), true);\n> - long long now = time_msec();\n> - double duration = (now - pmd->perf_stats.start_ms) / 1000.0;\n> -\n> - ds_put_cstr(reply, \"\\n\");\n> - ds_put_format(reply, \"Time: %s\\n\", time_str);\n> - ds_put_format(reply, \"Measurement duration: %.3f s\\n\", duration);\n> - ds_put_cstr(reply, \"\\n\");\n> - format_pmd_thread(reply, pmd);\n> - ds_put_cstr(reply, \"\\n\");\n> - pmd_perf_format_overall_stats(reply, &pmd->perf_stats, duration);\n> - if (pmd_perf_metrics_enabled(pmd)) {\n> - /* Prevent parallel clearing of perf metrics. */\n> - ovs_mutex_lock(&pmd->perf_stats.clear_mutex);\n> - if (par->histograms) {\n> - ds_put_cstr(reply, \"\\n\");\n> - pmd_perf_format_histograms(reply, &pmd->perf_stats);\n> - }\n> - if (par->iter_hist_len > 0) {\n> - ds_put_cstr(reply, \"\\n\");\n> - pmd_perf_format_iteration_history(reply, &pmd->perf_stats,\n> - par->iter_hist_len);\n> - }\n> - if (par->ms_hist_len > 0) {\n> - ds_put_cstr(reply, \"\\n\");\n> - pmd_perf_format_ms_history(reply, &pmd->perf_stats,\n> - par->ms_hist_len);\n> - }\n> - ovs_mutex_unlock(&pmd->perf_stats.clear_mutex);\n> + char *time_str = xastrftime_msec(\"%H:%M:%S.###\", time_wall_msec(), true);\n> + long long now = time_msec();\n> + double duration = (now - pmd->perf_stats.start_ms) / 1000.0;\n> +\n> + ds_put_cstr(reply, \"\\n\");\n> + ds_put_format(reply, \"Time: %s\\n\", time_str);\n> + ds_put_format(reply, \"Measurement duration: %.3f s\\n\", duration);\n> + ds_put_cstr(reply, \"\\n\");\n> + format_pmd_thread(reply, pmd);\n> + ds_put_cstr(reply, \"\\n\");\n> + pmd_perf_format_overall_stats(reply, &pmd->perf_stats, duration);\n> + if (pmd_perf_metrics_enabled(pmd) && pmd->core_id != NON_PMD_CORE_ID) {\n> + /* Prevent parallel clearing of perf metrics. */\n> + ovs_mutex_lock(&pmd->perf_stats.clear_mutex);\n> + if (par->histograms) {\n> + ds_put_cstr(reply, \"\\n\");\n> + pmd_perf_format_histograms(reply, &pmd->perf_stats);\n> }\n> - free(time_str);\n> + if (par->iter_hist_len > 0) {\n> + ds_put_cstr(reply, \"\\n\");\n> + pmd_perf_format_iteration_history(reply, &pmd->perf_stats,\n> + par->iter_hist_len);\n> + }\n> + if (par->ms_hist_len > 0) {\n> + ds_put_cstr(reply, \"\\n\");\n> + pmd_perf_format_ms_history(reply, &pmd->perf_stats,\n> + par->ms_hist_len);\n> + }\n> + ovs_mutex_unlock(&pmd->perf_stats.clear_mutex);\n> }\n> + free(time_str);\n> }\n> \n> static int\n> @@ -1443,8 +1357,6 @@ dpif_netdev_pmd_info(struct unixctl_conn *conn, int argc, const char *argv[],\n> pmd_info_show_rxq(&reply, pmd, secs);\n> } else if (type == PMD_INFO_CLEAR_STATS) {\n> pmd_perf_stats_clear(&pmd->perf_stats);\n> - } else if (type == PMD_INFO_SHOW_STATS) {\n> - pmd_info_show_stats(&reply, pmd);\n> } else if (type == PMD_INFO_PERF_SHOW) {\n> pmd_info_show_perf(&reply, pmd, (struct pmd_perf_params *)aux);\n> } else if (type == PMD_INFO_SLEEP_SHOW) {\n> @@ -1554,14 +1466,10 @@ dpif_netdev_bond_show(struct unixctl_conn *conn, int argc,\n> static int\n> dpif_netdev_init(void)\n> {\n> - static enum pmd_info_type show_aux = PMD_INFO_SHOW_STATS,\n> - clear_aux = PMD_INFO_CLEAR_STATS,\n> + static enum pmd_info_type clear_aux = PMD_INFO_CLEAR_STATS,\n> poll_aux = PMD_INFO_SHOW_RXQ,\n> sleep_aux = PMD_INFO_SLEEP_SHOW;\n> \n> - unixctl_command_register(\"dpif-netdev/pmd-stats-show\", \"[-pmd core] [dp]\",\n> - 0, 3, dpif_netdev_pmd_info,\n> - (void *)&show_aux);\n> unixctl_command_register(\"dpif-netdev/pmd-stats-clear\", \"[-pmd core] [dp]\",\n> 0, 3, dpif_netdev_pmd_info,\n> (void *)&clear_aux);\n> @@ -1578,6 +1486,10 @@ dpif_netdev_init(void)\n> \" [-pmd core] [dp]\",\n> 0, 8, pmd_perf_show_cmd,\n> NULL);\n> + /* 'pmd-stats-show' is just an undocumented alias for 'pmd-perf-show',\n> + * for compatibility with old muscle memory. */\n> + unixctl_command_register(\"dpif-netdev/pmd-stats-show\", NULL,\n> + 0, 8, pmd_perf_show_cmd, NULL);\n> unixctl_command_register(\"dpif-netdev/pmd-rxq-rebalance\", \"[dp]\",\n> 0, 1, dpif_netdev_pmd_rebalance,\n> NULL);\n> diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at\n> index 231197970..405094856 100644\n> --- a/tests/dpif-netdev.at\n> +++ b/tests/dpif-netdev.at\n> @@ -979,32 +979,32 @@ AT_CHECK([cat good_frame | sed -e \"s/6b72/dead/\" > bad_frame])\n> \n> CHECK_FWD_PACKET(p1, p2, , [bad_frame], [bad_frame])\n> dnl First packet, no simple matching.\n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | grep 'simple match hits'], [0], [dnl\n> - simple match hits: 0\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | grep 'Simple Match hits'], [0], [dnl\n> + - Simple Match hits: 0 ( 0.0 %)\n> ])\n> \n> dnl No Rx flag.\n> CHECK_FWD_PACKET(p1, p2, , [bad_frame], [bad_frame])\n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | grep 'simple match hits'], [0], [dnl\n> - simple match hits: 1\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | grep 'Simple Match hits'], [0], [dnl\n> + - Simple Match hits: 1 ( 50.0 %)\n> ])\n> \n> dnl Flag as Rx good.\n> CHECK_FWD_PACKET(p1, p2, ol_l4_rx_csum_set_good, [bad_frame], [bad_frame])\n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | grep 'simple match hits'], [0], [dnl\n> - simple match hits: 2\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | grep 'Simple Match hits'], [0], [dnl\n> + - Simple Match hits: 2 ( 66.7 %)\n> ])\n> \n> dnl Flag as Rx bad.\n> CHECK_FWD_PACKET(p1, p2, ol_l4_rx_csum_set_bad, [bad_frame], [bad_frame])\n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | grep 'simple match hits'], [0], [dnl\n> - simple match hits: 3\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | grep 'Simple Match hits'], [0], [dnl\n> + - Simple Match hits: 3 ( 75.0 %)\n> ])\n> \n> dnl Flag as Rx partial.\n> CHECK_FWD_PACKET(p1, p2, ol_l4_rx_csum_set_partial, [bad_frame], [good_frame])\n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | grep 'simple match hits'], [0], [dnl\n> - simple match hits: 4\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | grep 'Simple Match hits'], [0], [dnl\n> + - Simple Match hits: 4 ( 80.0 %)\n> ])\n> \n> OVS_VSWITCHD_STOP\n> diff --git a/tests/pmd.at b/tests/pmd.at\n> index 8254ac3b0..e8590044a 100644\n> --- a/tests/pmd.at\n> +++ b/tests/pmd.at\n> @@ -440,20 +440,12 @@ dummy@ovs-dummy: hit:0 missed:0\n> p0 7/1: (dummy-pmd: n_rxq=4, n_txq=1, numa_id=0)\n> ])\n> \n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | sed SED_NUMA_CORE_PATTERN | sed '/cycles/d' | grep pmd -A 12], [0], [dnl\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | sed SED_NUMA_CORE_PATTERN \\\n> + | sed '/cycles/d' | sed '/[[Ii]]teration/d' | grep pmd -A 3], [0], [dnl\n> pmd thread numa_id core_id :\n> - packets received: 0\n> - packet recirculations: 0\n> - avg. datapath passes per packet: 0.00\n> - phwol hits: 0\n> - mfex opt hits: 0\n> - simple match hits: 0\n> - emc hits: 0\n> - smc hits: 0\n> - megaflow hits: 0\n> - avg. subtable lookups per megaflow hit: 0.00\n> - miss with success upcall: 0\n> - miss with failed upcall: 0\n> +\n> + Rx packets: 0\n> + Tx packets: 0\n> ])\n> \n> ovs-appctl time/stop\n> @@ -474,20 +466,23 @@ AT_CHECK([cat ovs-vswitchd.log | filter_flow_install | strip_xout], [0], [dnl\n> recirc_id(0),in_port(1),packet_type(ns=0,id=0),eth(src=50:54:00:00:00:77,dst=50:54:00:00:01:78),eth_type(0x0800),ipv4(frag=no), actions: \n> ])\n> \n> -AT_CHECK([ovs-appctl dpif-netdev/pmd-stats-show | sed SED_NUMA_CORE_PATTERN | sed '/cycles/d' | grep pmd -A 12], [0], [dnl\n> +AT_CHECK([ovs-appctl dpif-netdev/pmd-perf-show | sed SED_NUMA_CORE_PATTERN \\\n> + | sed '/cycles/d' | sed '/[[Ii]]teration/d' \\\n> + | sed 's/, .* us/, us/' \\\n> + | sed 's/[[0-9]]* Kpps/ Kpps/' | grep pmd -A 12], [0], [dnl\n> pmd thread numa_id core_id :\n> - packets received: 20\n> - packet recirculations: 0\n> - avg. datapath passes per packet: 1.00\n> - phwol hits: 0\n> - mfex opt hits: 0\n> - simple match hits: 0\n> - emc hits: 19\n> - smc hits: 0\n> - megaflow hits: 0\n> - avg. subtable lookups per megaflow hit: 0.00\n> - miss with success upcall: 1\n> - miss with failed upcall: 0\n> +\n> + Datapath passes: 20 (1.00 passes/pkt)\n> + - PHWOL hits: 0 ( 0.0 %)\n> + - MFEX Opt hits: 0 ( 0.0 %)\n> + - Simple Match hits: 0 ( 0.0 %)\n> + - EMC hits: 19 ( 95.0 %)\n> + - SMC hits: 0 ( 0.0 %)\n> + - Megaflow hits: 0 ( 0.0 %, 0.00 subtbl lookups/hit)\n> + - Upcalls: 1 ( 5.0 %, us/upcall)\n> + - Lost upcalls: 0 ( 0.0 %)\n> + Tx packets: 20 ( Kpps)\n> + Tx batches: 20 (1.00 pkts/batch)\n> ])\n> \n> OVS_VSWITCHD_STOP\n> diff --git a/tests/system-dpdk-offloads.at b/tests/system-dpdk-offloads.at\n> index 81ab89b2f..09bdaf639 100644\n> --- a/tests/system-dpdk-offloads.at\n> +++ b/tests/system-dpdk-offloads.at\n> @@ -144,8 +144,8 @@ AT_CHECK([ovs-appctl dpctl/dump-flows type=dpdk,partially-offloaded \\\n> in_port(2),eth_type(0x0800),ipv4(frag=no), packets:9, bytes:954, used:0.0s, actions:check_pkt_len(size=200,gt(4),le(5)),3\n> ])\n> \n> -AT_CHECK([test $(ovs-appctl dpif-netdev/pmd-stats-show | \\\n> - awk '/phwol hits:/ {sum += $3} END {print sum}') -ge 8])\n> +AT_CHECK([test $(ovs-appctl dpif-netdev/pmd-perf-show | \\\n> + awk '/PHWOL hits:/ {sum += $4} END {print sum}') -ge 8])\n> \n> OVS_TRAFFIC_VSWITCHD_STOP\n> AT_CLEANUP\n> @@ -216,8 +216,8 @@ OVS_WAIT_UNTIL_EQUAL(\n> in_port(ovs-p0),eth(macs),eth_type(0x0800),ipv4(frag=no), packets:50, bytes:3000, used:0.0s, actions:ovs-p1\n> in_port(ovs-p1),eth(macs),eth_type(0x0800),ipv4(frag=no), packets:50, bytes:3000, used:0.0s, actions:ovs-p0])\n> \n> -AT_CHECK([test $(ovs-appctl dpif-netdev/pmd-stats-show | \\\n> - awk '/packets received:/ {sum += $3} END {print sum}') -lt 10])\n> +AT_CHECK([test $(ovs-appctl dpif-netdev/pmd-perf-show | \\\n> + awk '/Rx packets:/ {sum += $3} END {print sum}') -lt 10])\n> \n> OVS_TRAFFIC_VSWITCHD_STOP\n> AT_CLEANUP\n> diff --git a/utilities/bugtool/plugins/system-logs/openvswitch.xml b/utilities/bugtool/plugins/system-logs/openvswitch.xml\n> index 46c731812..0f17add75 100644\n> --- a/utilities/bugtool/plugins/system-logs/openvswitch.xml\n> +++ b/utilities/bugtool/plugins/system-logs/openvswitch.xml\n> @@ -20,7 +20,7 @@\n> /etc/openvswitch\n> /var/lib/openvswitch\n> df -h\n> - ovs-appctl dpif-netdev/pmd-stats-show | grep pmd\n> + ovs-appctl dpif-netdev/pmd-perf-show | grep pmd\n> ovs-appctl vlog/list\n> journalctl\n> ulimit -a","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","ovs-dev@openvswitch.org"],"Delivered-To":["patchwork-incoming@legolas.ozlabs.org","ovs-dev@lists.linuxfoundation.org"],"Authentication-Results":["legolas.ozlabs.org;\n\tdkim=fail reason=\"signature verification failed\" (1024-bit key;\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=aXJhYtH5;\n\tdkim-atps=neutral","legolas.ozlabs.org;\n spf=pass (sender SPF authorized) smtp.mailfrom=openvswitch.org\n (client-ip=2605:bc80:3010::136; helo=smtp3.osuosl.org;\n envelope-from=ovs-dev-bounces@openvswitch.org; receiver=patchwork.ozlabs.org)","smtp3.osuosl.org;\n\tdkim=fail reason=\"signature verification failed\" (1024-bit key,\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=aXJhYtH5","smtp1.osuosl.org; dmarc=pass (p=quarantine dis=none)\n header.from=redhat.com","smtp1.osuosl.org;\n dkim=fail reason=\"signature verification failed\" (1024-bit key,\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=aXJhYtH5"],"Received":["from smtp3.osuosl.org (smtp3.osuosl.org [IPv6:2605:bc80:3010::136])\n\t(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)\n\t key-exchange x25519 server-signature ECDSA (secp384r1) server-digest SHA384)\n\t(No client certificate requested)\n\tby legolas.ozlabs.org (Postfix) with ESMTPS id 4fxF192MFkz1yCv\n\tfor ; Thu, 16 Apr 2026 20:41:17 +1000 (AEST)","from localhost (localhost [127.0.0.1])\n\tby smtp3.osuosl.org (Postfix) with ESMTP id 048A760702;\n\tThu, 16 Apr 2026 10:41:15 +0000 (UTC)","from smtp3.osuosl.org ([127.0.0.1])\n by localhost (smtp3.osuosl.org [127.0.0.1]) (amavis, port 10024) with ESMTP\n id SQhGAi8G-g7B; Thu, 16 Apr 2026 10:41:12 +0000 (UTC)","from lists.linuxfoundation.org (lf-lists.osuosl.org [140.211.9.56])\n\tby smtp3.osuosl.org (Postfix) with ESMTPS id 39FF9606E5;\n\tThu, 16 Apr 2026 10:41:12 +0000 (UTC)","from lf-lists.osuosl.org (localhost [127.0.0.1])\n\tby lists.linuxfoundation.org (Postfix) with ESMTP id E36A0C054A;\n\tThu, 16 Apr 2026 10:41:11 +0000 (UTC)","from smtp1.osuosl.org (smtp1.osuosl.org [IPv6:2605:bc80:3010::138])\n by lists.linuxfoundation.org (Postfix) with ESMTP id 08E93C0549\n for ; Thu, 16 Apr 2026 10:41:11 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n by smtp1.osuosl.org (Postfix) with ESMTP id D2F9380C92\n for ; Thu, 16 Apr 2026 10:41:10 +0000 (UTC)","from smtp1.osuosl.org ([127.0.0.1])\n by localhost (smtp1.osuosl.org [127.0.0.1]) (amavis, port 10024) with ESMTP\n id BoIqrhJ-tRTk for ;\n Thu, 16 Apr 2026 10:41:09 +0000 (UTC)","from us-smtp-delivery-124.mimecast.com\n (us-smtp-delivery-124.mimecast.com [170.10.129.124])\n by smtp1.osuosl.org (Postfix) with ESMTPS id F39CD80C7E\n for ; Thu, 16 Apr 2026 10:41:08 +0000 (UTC)","from mail-wm1-f70.google.com (mail-wm1-f70.google.com\n [209.85.128.70]) by relay.mimecast.com with ESMTP with STARTTLS\n (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id\n us-mta-607-OQVDGwT3PnOt3CD8HlGJGg-1; Thu, 16 Apr 2026 06:41:05 -0400","by mail-wm1-f70.google.com with SMTP id\n 5b1f17b1804b1-488bd1ee9e7so49384735e9.1\n for ; Thu, 16 Apr 2026 03:41:05 -0700 (PDT)","from [192.168.0.65] ([78.16.130.11])\n by smtp.gmail.com with ESMTPSA id\n 5b1f17b1804b1-488f58165e1sm41520245e9.2.2026.04.16.03.40.59\n (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128);\n Thu, 16 Apr 2026 03:41:00 -0700 (PDT)"],"X-Virus-Scanned":["amavis at osuosl.org","amavis at osuosl.org"],"X-Comment":"SPF check N/A for local connections - client-ip=140.211.9.56;\n helo=lists.linuxfoundation.org;\n envelope-from=ovs-dev-bounces@openvswitch.org; receiver= ","DKIM-Filter":["OpenDKIM Filter v2.11.0 smtp3.osuosl.org 39FF9606E5","OpenDKIM Filter v2.11.0 smtp1.osuosl.org F39CD80C7E"],"Received-SPF":"Pass (mailfrom) identity=mailfrom; client-ip=170.10.129.124;\n helo=us-smtp-delivery-124.mimecast.com; envelope-from=ktraynor@redhat.com;\n receiver=","DMARC-Filter":"OpenDMARC Filter v1.4.2 smtp1.osuosl.org F39CD80C7E","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n s=mimecast20190719; t=1776336067;\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 content-transfer-encoding:content-transfer-encoding:\n in-reply-to:in-reply-to:references:references:autocrypt:autocrypt;\n bh=uRHDbsA9WJOj7/3ZabjQ2DsWNjjrNknTYUl2Pydb0AY=;\n b=aXJhYtH55FTZNbpaDdoS46tLpM5eEP4639T7BkIoQ8XV3nShwFx/9OMwmM4tawFqeaRIjJ\n uRLRUiH1JpFiF89xLOV0fkPCOU6b4oRTalNP8IxUxqj5PHUeuonuqioFgqcMgtA512il3j\n xQBn4o/SLHm9MDiH2eWUaztOj5x4URA=","X-MC-Unique":"OQVDGwT3PnOt3CD8HlGJGg-1","X-Mimecast-MFC-AGG-ID":"OQVDGwT3PnOt3CD8HlGJGg_1776336064","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n d=1e100.net; s=20251104; t=1776336062; x=1776940862;\n h=content-transfer-encoding:in-reply-to:cc:to:autocrypt:from\n :content-language:references:subject:user-agent:mime-version:date\n :message-id:x-gm-gg:x-gm-message-state:from:to:cc:subject:date\n :message-id:reply-to;\n bh=jWdXX0ZRTlQ7ui0gbT/N3uO88zjpYoBsDJ4FgL/kfPk=;\n b=GNVjTyOr+RfA+h81mSr+pz0kaU8LNIs/p7yJ6CiIx/N+em9PbOE/HAnED4Fv54Hwrb\n mwwTiJzBsEXrHRJ6QIkFlf/yaikDyw/tHA8+0QUB7GcKnASEE2llbuK8cUP0yV5Uu4xB\n 2SceOHGncQ5cKQIhaQGDs44dNa6bSxvKwVzaHLDsdl9WKp49ALMlweNvQdsf25jI0b6g\n +19v9qKkFp+Y83A72gVFnoB3rTZX1+SnARWkzZa4SK45YWaH3GeJXq2VrQmtwxmnHXGD\n kVhJ/qRfCkeZCP4B7cKOZ6/LiSMQcuPAWlz7IqWHEHB8DW59sstWSKZXeCmlyCLF2dhj\n chPA==","X-Forwarded-Encrypted":"i=1;\n AFNElJ+TXBEN6/WaHcbhcIFo6hhICoae0tPSYNTI8hLultO53AWaQ/SSTYYPFxWGfhGmpZwWBwQvg/ZR@openvswitch.org","X-Gm-Message-State":"AOJu0Yzpn8IeWWMpMAPfMreV5K01C83UEP1o+QKR6KqcaKBUHFTn5zcx\n krB4/WrkIDdG+9t8QMT1Po5YChXhIcGi3njCBPWxUtkMqznyXBJ+MWLlfM6WhL/NFy5VgBnLUzs\n C+GxsVC2AdcztePP2YRmo5C1xvyc3RykFLgrccQjI3BePDdWcXhEzwQbg+xo=","X-Gm-Gg":"AeBDiet9CeMsmbA05DjsfUjqzlYI67MabS76l1Om08viP0lgHSuT5RHb2fa3z5r+oD8\n yJV3LSg/o3KBDqeXQU41M2RdL9hpzNRWLJUd6OOVIToRT15BIfCUoQVXigLjCJKuYYlX40e+AA4\n WxUvL2cBIOEhcI7v/aDPSvz3Nq6HVV3frai1k97+pkasxLbekWiluL9LGhsWa320Ti8fFSTQw1s\n ImE6OLz41z4L+HVynMgPlphbqMof1knqb3FCH1EYWUqOTZQw5cuwcvdpffFvbGLKqj80FQUHLpj\n 86kNCihzxNB9eNTSgC9CEjnF4iwmwDGiQf99ddSKVpAKqaXAASopZxDFlDAf7UNUmLiMFj7kyuO\n KEkqCdY46Qnk+kQg6ECysGnHPXCsAOCNfHg==","X-Received":["by 2002:a05:600c:a105:b0:486:d76c:fa57 with SMTP id\n 5b1f17b1804b1-488d685b75emr245599405e9.17.1776336062047;\n Thu, 16 Apr 2026 03:41:02 -0700 (PDT)","by 2002:a05:600c:a105:b0:486:d76c:fa57 with SMTP id\n 5b1f17b1804b1-488d685b75emr245599025e9.17.1776336061244;\n Thu, 16 Apr 2026 03:41:01 -0700 (PDT)"],"Message-ID":"<3b04299a-f445-45da-8007-70320ccd1c51@redhat.com>","Date":"Thu, 16 Apr 2026 11:40:58 +0100","MIME-Version":"1.0","User-Agent":"Mozilla Thunderbird","References":"<20260408171644.1404735-1-i.maximets@ovn.org>","Autocrypt":"addr=ktraynor@redhat.com; keydata=\n xsFNBF2J2awBEADUEPNhgNI+nJNgiTAUcw4YIgVXEoHlsNPyyzG1BEXkWXALy0Y3fNTiw6+r\n ltWDkF9jzL9kfkecgQ67itGfk1OaBXgSGKuw1PUpxAwX2Bi76LAR6M5OsyGM9TSVVQwARalz\n hMwRBIZPzPc7or6Pw7jAOJ8SQGJ1Zlp1YJCjrvpe87V1tH/LY8Wnxn/EuoseFmWILAQZAtYS\n tGjcrAgYn3SPMLR1B0BP5bTBY06vWQjiufH8drenfDnMJAzuBdG1mqjnTqCjULZ3Hunv4xqZ\n aMnkvL/K5Tj1c12Oe4930EE53LrXIBUltRg5mBudSWHnC7twjH0082HH9f963Z/2UI63SFIT\n iUvRvAzJYytgy7XnWLQ0+goZBADKYfolOuC0H8VgCaux8u8KFF28Dy+N6TV2KI58jTlyg1Zu\n l7QwykZpnOkJFiy37Gfbu3YEOzO72cP/S7/A+zvuqkxi63jyEkd+FY99vLt/HN2MUZwRmKDw\n UPbLkmrs8WU01/POVsqDcfvz7vu2St8hqqTiSIdQGS2zyTKB2/DvPSM3jws3udkIYSuhn+X4\n QBiV6lkVZ7DSE6a065gnAauAql+b32Eymy+xnG5jCt1tR+0Cp2VZYCR9OU2gmomUKBDoX/He\n pSgED01CqYPNjN+TddirwmQX7ep4DtXc8FWvv2g/pq9WZFQk2QARAQABzSNLZXZpbiBUcmF5\n bm9yIDxrdHJheW5vckByZWRoYXQuY29tPsLBjgQTAQgAOBYhBAoiOaH51tHF7VYtEI9CINER\n a+yJBQJdidmsAhsDBQsJCAcCBhUKCQgLAgQWAgMBAh4BAheAAAoJEI9CINERa+yJoxIP/3VF\n 2TIgW4ckxhRFCvFu/606bnvCPie88ake4uWVWMAWwcMc4fKEltRWRCpkSVOwgqoMHnyHxK5r\n kOKzx2CLJMX5TgTMfKzPuaBDHngHLUzl2DStpBzrod0cVg5TShdmmfjY61uxRJKz+DlSkwgJ\n riADdVF5PPosQXTkKSGf2ombpTGpx/pue9ocjnr3x4SDpRLlnooM6Jf/3Y3Ib4jX6HPEyWuY\n b+owIIk9y2nRRGPQ6jbqAhsrXd9V+77UL0QuGWloMuKMZFbNg8hbu7X5aFijAbfxj4YUgojS\n ba7gfGZQan8h32A9KGQWrmsCBc3j2GqEPsX0r05X7cn7WL6IOPgQJ5EiQ7PlazQYVLrvZg9B\n n0GKK0k6895mLG0ZZ5v/qajOPF52etSmvFD1WUPb4OqaHqGA9ZtMpaKFRt7Y6rpXqKNU1xzW\n F5KjbTPtTb9WF3An8dciVv+AYUI7totkZYkWvQtgss8lfaX3NKUvXLVxqK0z3dQyr7rF/tYz\n PneTKypSksjCgaEBLSrsRmM5zKfe7tSNF/fDntfIq/029Jtcw29TcWEP57peNu6TtejewQD9\n sTI+oqiXvW2D5l7LNUDYG8eMJp2oT7I0ZSBRvwcbmjH0DtN/bXCCFfCvk8Yic68F3tV1ctix\n wQARVKDBhT30uCxycRWojCYqTgNJJS71zsFNBF2J2awBEADP57PR2IpSYBeNSrsAjeIcsahE\n N4SQP2C4s50S8QEWAUhqMRI7WNv5cfeef0nDvcl1IUA6oz5SokbcsbMa+mRgaNF4N5KikWTO\n LPYxq2YVJoXwJ+tKmNzyOLFUIfFJ4NBJZple5dTfWzD00Dbb19Mri1hy1mWMqNTPGBee1+hw\n Qcp6n3mmGECvajs8G5A7NyXbwL8ihN7HX9D01ucD62b4G03yKe2g/hvKgcdUVmhCldJlF27I\n 2fSR9tDxH9pZqRODY4rjbFZEey/vWKXqjE+DQ8AtMSEaDfFe5D+i4Aw6erWQ3Wr+DwZt1/7G\n dIAElGA/q90T1ENVwJX9y7fsQssawKYYdDqURHCl5JuDXI+VXUypExipUUT5SPycMmbLsx0D\n iKEqPPDQWKxkIDVKqj2+EhamSuJznZUwBLJKn0h4zrIWiXWUy07lRwtVuhaDXhF3GfW+5W/x\n wAg7Qg3w00ASsb/XTHBIhMnenKDfS7ihtQA8SacwX8ySdxb+15XPyiplM979qBQ0mhnilulm\n MIJzEf/JxoYR5huuj4f1PFqqrsP06Dl+YGB7dQZp3IKggS5c3/TAynARRg9N89UsDXNtp7X0\n tgIPFF5k6fnHE0J5O64GYHeTqN/1aE6dAEOV9WrGzQAJxU9ipikb8jKAWXzLewRIKGmoPcRZ\n WdB0NmIjmQARAQABwsF2BBgBCAAgFiEECiI5ofnW0cXtVi0Qj0Ig0RFr7IkFAl2J2awCGwwA\n CgkQj0Ig0RFr7IkkORAAl/NbX93WK5MEoRw7/DaPTo/Lo6Pj1XMeSqGyACigHK/452UDvlEH\n NjNJMzYYrNIjMtEmN9VVCfjT38CSca7mpGQVwchc0mC7QSPAETLCS+UacVf/Kwxz5FfkEUUw\n UT7A+uyVOIgW3d9ldlRzkHA2czonSSgTQU+i2g6DM4ha+BuQb4byAXH6HQHt/Zh1J64z0ohH\n v6iGsCzCY/sMWF8+LEGSnzMGRCLiiwSF0vJBHbzWK68fANaF4gBV0Z/+6tQRFN7YMhj/INmk\n qgvHj1ZzHFNtirjMGPRxoZs51YoLQM/aBPxKrnmXThx1ufH+0L6sGmFTugiDt0XSEkC5reH7\n a+VhQ1VTFFQrClA8NmDSPzFeuhru4ryaaDHO+uEB16cNHxHrQtlP/2hts2JM5lwkZRWJ5A57\n h8eDEIK5be47T85NVHfuTaboNRmgg1HygVejhGUtt69u/0MVRg/roUTa0FyEbNsvz4qAecyW\n yWzMcVrcGJDQLC9JLKEpoyUF6gdTKaiDL2Vao4+XRIA3Y57b6MO35a3HuzAv7+i5Z0mnDEJO\n XxXqTOmKYpMIGexzM/PtuA0712sT1abG9tAJ17ao/B7cqMW5IkKkalemFbWfI2unns4Papvo\n tk9igVqyp6EJDU98z5TJioCVojwK2laDaoIjTJk9YYv3iwCsqPd5feU=","To":"Ilya Maximets , ovs-dev@openvswitch.org","In-Reply-To":"<20260408171644.1404735-1-i.maximets@ovn.org>","X-Mimecast-Spam-Score":"0","X-Mimecast-MFC-PROC-ID":"NJfE6IF3nYYCdJ3LZ0RlvqJC0jNYzkMfuU261Mbc_io_1776336064","X-Mimecast-Originator":"redhat.com","Content-Language":"en-US","Subject":"Re: [ovs-dev] [PATCH] dpif-netdev: Remove pmd-stats-show in favor\n of pmd-perf-show.","X-BeenThere":"ovs-dev@openvswitch.org","X-Mailman-Version":"2.1.30","Precedence":"list","List-Id":"","List-Unsubscribe":",\n ","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n ","From":"Kevin Traynor via dev ","Reply-To":"Kevin Traynor ","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"ovs-dev-bounces@openvswitch.org","Sender":"\"dev\" "}}]