From: Dmitry Osipenko <digetx@gmail.com> To: Mikko Perttunen <mperttunen@nvidia.com>, thierry.reding@gmail.com, jonathanh@nvidia.com, airlied@linux.ie, daniel@ffwll.ch Cc: linux-tegra@vger.kernel.org, dri-devel@lists.freedesktop.org Subject: Re: [PATCH v7 02/15] gpu: host1x: Add no-recovery mode Date: Wed, 16 Jun 2021 01:06:02 +0300 [thread overview] Message-ID: <7fc989aa-317a-fedf-b223-fe563da72ff6@gmail.com> (raw) In-Reply-To: <20210610110456.3692391-3-mperttunen@nvidia.com> 10.06.2021 14:04, Mikko Perttunen пишет: > Add a new property for jobs to enable or disable recovery i.e. > CPU increments of syncpoints to max value on job timeout. This > allows for a more solid model for hanged jobs, where userspace > doesn't need to guess if a syncpoint increment happened because > the job completed, or because job timeout was triggered. Userspace should always get proper timeout. Threshold should be wrapped into fence. Fence's error state should be set to -ETIMEDOUT. > On job timeout, we stop the channel, NOP all future jobs on the > channel using the same syncpoint, mark the syncpoint as locked > and resume the channel from the next job, if any. > > The future jobs are NOPed, since because we don't do the CPU > increments, the value of the syncpoint is no longer synchronized, > and any waiters would become confused if a future job incremented > the syncpoint. The syncpoint is marked locked to ensure that any > future jobs cannot increment the syncpoint either, until the > application has recognized the situation and reallocated the > syncpoint. It should be much easier to switch to DRM scheduler, removing lot's of the old code instead of updating it with new quirks that are difficult to follow.
WARNING: multiple messages have this Message-ID (diff)
From: Dmitry Osipenko <digetx@gmail.com> To: Mikko Perttunen <mperttunen@nvidia.com>, thierry.reding@gmail.com, jonathanh@nvidia.com, airlied@linux.ie, daniel@ffwll.ch Cc: dri-devel@lists.freedesktop.org, linux-tegra@vger.kernel.org Subject: Re: [PATCH v7 02/15] gpu: host1x: Add no-recovery mode Date: Wed, 16 Jun 2021 01:06:02 +0300 [thread overview] Message-ID: <7fc989aa-317a-fedf-b223-fe563da72ff6@gmail.com> (raw) In-Reply-To: <20210610110456.3692391-3-mperttunen@nvidia.com> 10.06.2021 14:04, Mikko Perttunen пишет: > Add a new property for jobs to enable or disable recovery i.e. > CPU increments of syncpoints to max value on job timeout. This > allows for a more solid model for hanged jobs, where userspace > doesn't need to guess if a syncpoint increment happened because > the job completed, or because job timeout was triggered. Userspace should always get proper timeout. Threshold should be wrapped into fence. Fence's error state should be set to -ETIMEDOUT. > On job timeout, we stop the channel, NOP all future jobs on the > channel using the same syncpoint, mark the syncpoint as locked > and resume the channel from the next job, if any. > > The future jobs are NOPed, since because we don't do the CPU > increments, the value of the syncpoint is no longer synchronized, > and any waiters would become confused if a future job incremented > the syncpoint. The syncpoint is marked locked to ensure that any > future jobs cannot increment the syncpoint either, until the > application has recognized the situation and reallocated the > syncpoint. It should be much easier to switch to DRM scheduler, removing lot's of the old code instead of updating it with new quirks that are difficult to follow.
next prev parent reply other threads:[~2021-06-15 22:06 UTC|newest] Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-06-10 11:04 [PATCH v7 00/15] TegraDRM UAPI Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 01/15] gpu: host1x: Add DMA fence implementation Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-15 21:47 ` Dmitry Osipenko 2021-06-15 21:47 ` Dmitry Osipenko 2021-06-10 11:04 ` [PATCH v7 02/15] gpu: host1x: Add no-recovery mode Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-15 22:06 ` Dmitry Osipenko [this message] 2021-06-15 22:06 ` Dmitry Osipenko 2021-06-10 11:04 ` [PATCH v7 03/15] gpu: host1x: Add job release callback Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 04/15] gpu: host1x: Add support for syncpoint waits in CDMA pushbuffer Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 05/15] gpu: host1x: Add option to skip firewall for a job Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-22 20:46 ` Michał Mirosław 2021-06-22 20:46 ` Michał Mirosław 2021-06-10 11:04 ` [PATCH v7 06/15] drm/tegra: Extract tegra_gem_lookup Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 07/15] drm/tegra: Add new UAPI to header Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 08/15] drm/tegra: Boot VIC during runtime PM resume Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 09/15] drm/tegra: Allocate per-engine channel in core code Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 10/15] drm/tegra: Implement new UAPI Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 11/15] drm/tegra: Implement syncpoint management UAPI Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 12/15] drm/tegra: Implement syncpoint wait UAPI Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 13/15] drm/tegra: Implement job submission part of new UAPI Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-15 19:00 ` Jon Hunter 2021-06-15 19:00 ` Jon Hunter 2021-06-15 19:03 ` Mikko Perttunen 2021-06-15 19:03 ` Mikko Perttunen 2021-06-15 22:19 ` Dmitry Osipenko 2021-06-15 22:19 ` Dmitry Osipenko 2021-06-15 22:24 ` Dmitry Osipenko 2021-06-15 22:24 ` Dmitry Osipenko 2021-06-16 9:31 ` Jon Hunter 2021-06-16 9:31 ` Jon Hunter 2021-06-10 11:04 ` [PATCH v7 14/15] drm/tegra: Add job firewall Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-10 11:04 ` [PATCH v7 15/15] drm/tegra: Bump driver version Mikko Perttunen 2021-06-10 11:04 ` Mikko Perttunen 2021-06-15 20:32 ` [PATCH v7 00/15] TegraDRM UAPI Dmitry Osipenko 2021-06-15 20:32 ` Dmitry Osipenko
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=7fc989aa-317a-fedf-b223-fe563da72ff6@gmail.com \ --to=digetx@gmail.com \ --cc=airlied@linux.ie \ --cc=daniel@ffwll.ch \ --cc=dri-devel@lists.freedesktop.org \ --cc=jonathanh@nvidia.com \ --cc=linux-tegra@vger.kernel.org \ --cc=mperttunen@nvidia.com \ --cc=thierry.reding@gmail.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.