From: Andy Shevchenko <andy.shevchenko@gmail.com> To: Lars-Peter Clausen <lars@metafoo.de> Cc: Vinod Koul <vinod.koul@intel.com>, Takashi Iwai <tiwai@suse.com>, Dan Williams <dan.j.williams@intel.com>, Russell King <rmk+kernel@arm.linux.org.uk>, Laurent Pinchart <renesas@ideasonboard.com>, Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>, Qiao Zhou <zhouqiao@marvell.com>, Shengjiu Wang <shengjiu.wang@freescale.com>, Matt Campbell <mcampbell@izotope.com>, Jonah Petri <jpetri@izotope.com>, dmaengine <dmaengine@vger.kernel.org>, alsa-devel@alsa-project.org, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org> Subject: Re: [PATCH 1/4] dmaengine: Add transfer termination synchronization support Date: Thu, 29 Oct 2015 23:59:40 +0200 [thread overview] Message-ID: <CAHp75Vey9zQMCWB63BDsOqEGKQh3SvANWCbH1har-vmXoAquXQ@mail.gmail.com> (raw) In-Reply-To: <1445334391-12272-2-git-send-email-lars@metafoo.de> On Tue, Oct 20, 2015 at 12:46 PM, Lars-Peter Clausen <lars@metafoo.de> wrote: > The DMAengine API has a long standing race condition that is inherent to > the API itself. Calling dmaengine_terminate_all() is supposed to stop and > abort any pending or active transfers that have previously been submitted. > Unfortunately it is possible that this operation races against a currently > running (or with some drivers also scheduled) completion callback. [] > +/** > + * dmaengine_terminate_sync() - Terminate all active DMA transfers > + * @chan: The channel for which to terminate the transfers > + * > + * Calling this function will terminate all active and pending transfers > + * that have previously been submitted to the channel. It is similar to > + * dmaengine_terminate_async() but guarantees that the DMA transfer has actually > + * stopped and that all complete callbacks have finished running when the > + * function returns. > + * > + * This function must only be called from non-atomic context and must not be > + * called from within a complete callback of a descriptor submitted on the same > + * channel. > + */ > +static inline int dmaengine_terminate_sync(struct dma_chan *chan) > +{ > + int ret; Might be a good idea to add might_sleep(); here. > + > + ret = dmaengine_terminate_async(chan); > + if (ret) > + return ret; > + > + dmaengine_synchronize(chan); > + > + return 0; > +} -- With Best Regards, Andy Shevchenko
WARNING: multiple messages have this Message-ID (diff)
From: Andy Shevchenko <andy.shevchenko@gmail.com> To: Lars-Peter Clausen <lars@metafoo.de> Cc: alsa-devel@alsa-project.org, Shengjiu Wang <shengjiu.wang@freescale.com>, Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>, Vinod Koul <vinod.koul@intel.com>, Takashi Iwai <tiwai@suse.com>, "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, Qiao Zhou <zhouqiao@marvell.com>, Laurent Pinchart <renesas@ideasonboard.com>, Matt Campbell <mcampbell@izotope.com>, dmaengine <dmaengine@vger.kernel.org>, Russell King <rmk+kernel@arm.linux.org.uk>, Dan Williams <dan.j.williams@intel.com>, Jonah Petri <jpetri@izotope.com> Subject: Re: [PATCH 1/4] dmaengine: Add transfer termination synchronization support Date: Thu, 29 Oct 2015 23:59:40 +0200 [thread overview] Message-ID: <CAHp75Vey9zQMCWB63BDsOqEGKQh3SvANWCbH1har-vmXoAquXQ@mail.gmail.com> (raw) In-Reply-To: <1445334391-12272-2-git-send-email-lars@metafoo.de> On Tue, Oct 20, 2015 at 12:46 PM, Lars-Peter Clausen <lars@metafoo.de> wrote: > The DMAengine API has a long standing race condition that is inherent to > the API itself. Calling dmaengine_terminate_all() is supposed to stop and > abort any pending or active transfers that have previously been submitted. > Unfortunately it is possible that this operation races against a currently > running (or with some drivers also scheduled) completion callback. [] > +/** > + * dmaengine_terminate_sync() - Terminate all active DMA transfers > + * @chan: The channel for which to terminate the transfers > + * > + * Calling this function will terminate all active and pending transfers > + * that have previously been submitted to the channel. It is similar to > + * dmaengine_terminate_async() but guarantees that the DMA transfer has actually > + * stopped and that all complete callbacks have finished running when the > + * function returns. > + * > + * This function must only be called from non-atomic context and must not be > + * called from within a complete callback of a descriptor submitted on the same > + * channel. > + */ > +static inline int dmaengine_terminate_sync(struct dma_chan *chan) > +{ > + int ret; Might be a good idea to add might_sleep(); here. > + > + ret = dmaengine_terminate_async(chan); > + if (ret) > + return ret; > + > + dmaengine_synchronize(chan); > + > + return 0; > +} -- With Best Regards, Andy Shevchenko
next prev parent reply other threads:[~2015-10-29 21:59 UTC|newest] Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-10-20 9:46 [PATCH 0/4] dmaengine: Add transfer termination synchronization support Lars-Peter Clausen 2015-10-20 9:46 ` [PATCH 1/4] " Lars-Peter Clausen 2015-10-20 9:46 ` Lars-Peter Clausen 2015-10-29 21:59 ` Andy Shevchenko [this message] 2015-10-29 21:59 ` Andy Shevchenko 2015-10-30 14:16 ` Lars-Peter Clausen 2015-10-30 14:16 ` Lars-Peter Clausen 2015-10-20 9:46 ` [PATCH 2/4] dmaengine: virt-dma: Add synchronization helper function Lars-Peter Clausen 2015-10-20 9:46 ` [PATCH 3/4] dmaengine: axi_dmac: Add synchronization support Lars-Peter Clausen 2015-10-20 9:46 ` [PATCH 4/4] ALSA: pcm_dmaengine: Properly synchronize DMA on shutdown Lars-Peter Clausen 2015-10-20 11:17 ` Takashi Iwai 2015-10-20 11:17 ` Takashi Iwai 2015-10-20 11:40 ` Lars-Peter Clausen 2015-10-20 11:40 ` Lars-Peter Clausen 2015-10-20 12:36 ` Takashi Iwai 2015-10-20 13:01 ` Lars-Peter Clausen 2015-10-29 1:26 ` [PATCH 0/4] dmaengine: Add transfer termination synchronization support Vinod Koul
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=CAHp75Vey9zQMCWB63BDsOqEGKQh3SvANWCbH1har-vmXoAquXQ@mail.gmail.com \ --to=andy.shevchenko@gmail.com \ --cc=alsa-devel@alsa-project.org \ --cc=dan.j.williams@intel.com \ --cc=dmaengine@vger.kernel.org \ --cc=jpetri@izotope.com \ --cc=kuninori.morimoto.gx@renesas.com \ --cc=lars@metafoo.de \ --cc=linux-kernel@vger.kernel.org \ --cc=mcampbell@izotope.com \ --cc=renesas@ideasonboard.com \ --cc=rmk+kernel@arm.linux.org.uk \ --cc=shengjiu.wang@freescale.com \ --cc=tiwai@suse.com \ --cc=vinod.koul@intel.com \ --cc=zhouqiao@marvell.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.