Message ID | 1618051221-19962-1-git-send-email-humin29@huawei.com (mailing list archive) |
---|---|
Headers |
Return-Path: <dev-bounces@dpdk.org> X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id F344EA0546; Sat, 10 Apr 2021 12:40:28 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 52A6814122A; Sat, 10 Apr 2021 12:40:13 +0200 (CEST) Received: from szxga04-in.huawei.com (szxga04-in.huawei.com [45.249.212.190]) by mails.dpdk.org (Postfix) with ESMTP id 0091C1411E5 for <dev@dpdk.org>; Sat, 10 Apr 2021 12:40:07 +0200 (CEST) Received: from DGGEMS404-HUB.china.huawei.com (unknown [172.30.72.58]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4FHWf41v68zpWTv; Sat, 10 Apr 2021 18:37:16 +0800 (CST) Received: from localhost.localdomain (10.69.192.56) by DGGEMS404-HUB.china.huawei.com (10.3.19.204) with Microsoft SMTP Server id 14.3.498.0; Sat, 10 Apr 2021 18:40:00 +0800 From: "Min Hu (Connor)" <humin29@huawei.com> To: <dev@dpdk.org> CC: <ferruh.yigit@intel.com>, <shepard.siegel@atomicrules.com>, <qiming.yang@intel.com>, <rosen.xu@intel.com>, <john.mcnamara@intel.com>, <ciara.power@intel.com>, <maxime.coquelin@redhat.com> Date: Sat, 10 Apr 2021 18:40:14 +0800 Message-ID: <1618051221-19962-1-git-send-email-humin29@huawei.com> X-Mailer: git-send-email 2.7.4 MIME-Version: 1.0 Content-Type: text/plain X-Originating-IP: [10.69.192.56] X-CFilter-Loop: Reflected Subject: [dpdk-dev] [PATCH 0/7] support set thread name X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions <dev.dpdk.org> List-Unsubscribe: <https://mails.dpdk.org/options/dev>, <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>, <mailto:dev-request@dpdk.org?subject=subscribe> Errors-To: dev-bounces@dpdk.org Sender: "dev" <dev-bounces@dpdk.org> |
Series |
support set thread name
|
|
Message
humin (Q)
April 10, 2021, 10:40 a.m. UTC
This set of patches support set thread name for debugging. Chengwen Feng (7): net/ark: support set thread name net/ice: support set VSI reset thread name vdpa/ifc: support set notify and vring relay thread name raw/ifpga: support set monitor thread name examples/performance-thread: support set thread name telemetry: support set init threads name examples/vhost_blk: support set ctrl worker thread name drivers/net/ark/ark_ethdev.c | 1 + drivers/net/ice/ice_dcf_parent.c | 5 +++++ drivers/raw/ifpga/ifpga_rawdev.c | 1 + drivers/vdpa/ifc/ifcvf_vdpa.c | 12 ++++++++++++ examples/performance-thread/pthread_shim/main.c | 5 +++++ examples/vhost_blk/vhost_blk.c | 1 + lib/librte_telemetry/telemetry.c | 3 ++- 7 files changed, 27 insertions(+), 1 deletion(-)
Comments
On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> wrote: > > This set of patches support set thread name for debugging. > > Chengwen Feng (7): > net/ark: support set thread name > net/ice: support set VSI reset thread name > vdpa/ifc: support set notify and vring relay thread name > raw/ifpga: support set monitor thread name > examples/performance-thread: support set thread name > telemetry: support set init threads name > examples/vhost_blk: support set ctrl worker thread name Rather than add those calls, can maintainers check if their component can use ctrl threads instead? rte_ctrl_thread_create ensures both that the name is set, and that the ctrl thread won't run on the same cpu as "datapath" threads. I also saw some issues with components creating threads. I'll post a series addressing those later (net/ark, net/ice not detaching/joining created threads + telemetry not checking pthread_create failures).
>-----Original Message----- >From: David Marchand <david.marchand@redhat.com> >Sent: Monday 12 April 2021 08:20 >To: Min Hu (Connor) <humin29@huawei.com>; Shepard Siegel ><shepard.siegel@atomicrules.com>; Yang, Qiming <qiming.yang@intel.com>; >Wang, Xiao W <xiao.w.wang@intel.com>; Xu, Rosen <rosen.xu@intel.com>; >Maxime Coquelin <maxime.coquelin@redhat.com> >Cc: dev <dev@dpdk.org>; Yigit, Ferruh <ferruh.yigit@intel.com>; Mcnamara, >John <john.mcnamara@intel.com>; Power, Ciara <ciara.power@intel.com> >Subject: Re: [dpdk-dev] [PATCH 0/7] support set thread name > >On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> >wrote: >> >> This set of patches support set thread name for debugging. >> >> Chengwen Feng (7): >> net/ark: support set thread name >> net/ice: support set VSI reset thread name >> vdpa/ifc: support set notify and vring relay thread name >> raw/ifpga: support set monitor thread name >> examples/performance-thread: support set thread name >> telemetry: support set init threads name >> examples/vhost_blk: support set ctrl worker thread name > >Rather than add those calls, can maintainers check if their component can use >ctrl threads instead? >rte_ctrl_thread_create ensures both that the name is set, and that the ctrl >thread won't run on the same cpu as "datapath" threads. > >I also saw some issues with components creating threads. >I'll post a series addressing those later (net/ark, net/ice not detaching/joining >created threads + telemetry not checking pthread_create failures). > > Hi David, Telemetry has no dependency on EAL, so using the ctrl thread functions isn't possible for that component. Thanks, Ciara >-- >David Marchand
在 2021/4/12 15:19, David Marchand 写道: > On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> wrote: >> >> This set of patches support set thread name for debugging. >> >> Chengwen Feng (7): >> net/ark: support set thread name >> net/ice: support set VSI reset thread name >> vdpa/ifc: support set notify and vring relay thread name >> raw/ifpga: support set monitor thread name >> examples/performance-thread: support set thread name >> telemetry: support set init threads name >> examples/vhost_blk: support set ctrl worker thread name > > Rather than add those calls, can maintainers check if their component > can use ctrl threads instead? > rte_ctrl_thread_create ensures both that the name is set, and that the > ctrl thread won't run on the same cpu as "datapath" threads. > > I also saw some issues with components creating threads. > I'll post a series addressing those later (net/ark, net/ice not > detaching/joining created threads + telemetry not checking > pthread_create failures). > Totally agree with David. By the way, for "telemetry not checking pthread_create failures", I have sent patches to fix it. Please check it out. Thanks. >
On 2021/4/12 15:19, David Marchand wrote: > On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> wrote: >> >> This set of patches support set thread name for debugging. >> >> Chengwen Feng (7): >> net/ark: support set thread name >> net/ice: support set VSI reset thread name >> vdpa/ifc: support set notify and vring relay thread name >> raw/ifpga: support set monitor thread name >> examples/performance-thread: support set thread name >> telemetry: support set init threads name >> examples/vhost_blk: support set ctrl worker thread name > > Rather than add those calls, can maintainers check if their component > can use ctrl threads instead? > rte_ctrl_thread_create ensures both that the name is set, and that the > ctrl thread won't run on the same cpu as "datapath" threads. > > I also saw some issues with components creating threads. > I'll post a series addressing those later (net/ark, net/ice not > detaching/joining created threads + telemetry not checking > pthread_create failures). > > There already have reviewing patches: ([dpdk-dev] [PATCH v2 0/2] fix missing check for thread creation) to fix telemetry not checking pthread_create failure. Please review it, thanks
13/04/2021 03:12, Min Hu (Connor): > 在 2021/4/12 15:19, David Marchand 写道: > > On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> wrote: > >> > >> This set of patches support set thread name for debugging. > >> > >> Chengwen Feng (7): > >> net/ark: support set thread name > >> net/ice: support set VSI reset thread name > >> vdpa/ifc: support set notify and vring relay thread name > >> raw/ifpga: support set monitor thread name > >> examples/performance-thread: support set thread name > >> telemetry: support set init threads name > >> examples/vhost_blk: support set ctrl worker thread name > > > > Rather than add those calls, can maintainers check if their component > > can use ctrl threads instead? > > rte_ctrl_thread_create ensures both that the name is set, and that the > > ctrl thread won't run on the same cpu as "datapath" threads. > > > > I also saw some issues with components creating threads. > > I'll post a series addressing those later (net/ark, net/ice not > > detaching/joining created threads + telemetry not checking > > pthread_create failures). > > > Totally agree with David. > By the way, for "telemetry not checking pthread_create failures", I have > sent patches to fix it. Please check it out. > Thanks. Do I understand correctly that we prefer switching to rte_ctrl_thread_create() instead of this patch series adding rte_thread_setname()? Any volunteer to add the use of rte_ctrl_thread_create() in these drivers and examples?
在 2021/4/17 2:40, Thomas Monjalon 写道: > 13/04/2021 03:12, Min Hu (Connor): >> 在 2021/4/12 15:19, David Marchand 写道: >>> On Sat, Apr 10, 2021 at 12:40 PM Min Hu (Connor) <humin29@huawei.com> wrote: >>>> >>>> This set of patches support set thread name for debugging. >>>> >>>> Chengwen Feng (7): >>>> net/ark: support set thread name >>>> net/ice: support set VSI reset thread name >>>> vdpa/ifc: support set notify and vring relay thread name >>>> raw/ifpga: support set monitor thread name >>>> examples/performance-thread: support set thread name >>>> telemetry: support set init threads name >>>> examples/vhost_blk: support set ctrl worker thread name >>> >>> Rather than add those calls, can maintainers check if their component >>> can use ctrl threads instead? >>> rte_ctrl_thread_create ensures both that the name is set, and that the >>> ctrl thread won't run on the same cpu as "datapath" threads. >>> >>> I also saw some issues with components creating threads. >>> I'll post a series addressing those later (net/ark, net/ice not >>> detaching/joining created threads + telemetry not checking >>> pthread_create failures). >>> >> Totally agree with David. >> By the way, for "telemetry not checking pthread_create failures", I have >> sent patches to fix it. Please check it out. >> Thanks. > > Do I understand correctly that we prefer switching > to rte_ctrl_thread_create() instead of this patch series > adding rte_thread_setname()? > > Any volunteer to add the use of rte_ctrl_thread_create() > in these drivers and examples? > Hi, Thomos, we will send next patch to fix it, thanks. > > . >