From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vinod Koul Subject: Re: [PATCH v4 2/2] dma: Add Xilinx zynqmp dma engine driver support Date: Sun, 23 Aug 2015 19:38:23 +0530 Message-ID: <20150823140823.GA13546@localhost> References: <1438831173-8761-1-git-send-email-punnaia@xilinx.com> <1438831173-8761-2-git-send-email-punnaia@xilinx.com> <20150820061312.GA13546@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: punnaiah choudary kalluri Cc: Punnaiah Choudary Kalluri , "robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org" , "pawel.moll-5wv7dgnIgG8@public.gmane.org" , "mark.rutland-5wv7dgnIgG8@public.gmane.org" , "ijc+devicetree-KcIKpvwj1kUDXYZnReoRVg@public.gmane.org" , Kumar Gala , "michal.simek-gjFFaj9aHVfQT0dZR+AlfA@public.gmane.org" , dan.j.williams-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org, dmaengine-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, "devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , Punnaiah Choudary List-Id: devicetree@vger.kernel.org On Thu, Aug 20, 2015 at 12:01:27PM +0530, punnaiah choudary kalluri wrote: > On Thu, Aug 20, 2015 at 11:43 AM, Vinod Koul wrote: > > On Thu, Aug 06, 2015 at 08:49:33AM +0530, Punnaiah Choudary Kalluri wrote: > > > >> + list_for_each_entry_safe(desc, next, &chan->done_list, node) { > >> + dma_async_tx_callback callback; > >> + void *callback_param; > >> + > >> + list_del(&desc->node); > >> + > >> + callback = desc->async_tx.callback; > >> + callback_param = desc->async_tx.callback_param; > >> + if (callback) { > >> + if (in_interrupt()) > >> + spin_unlock_bh(&chan->lock); > >> + else > >> + spin_unlock(&chan->lock); > > > > This looks bad! > > Why would callback be called from different context. It should only be > > invoked from your tasklet > > During the terminate call, driver need to clean up the existing BDs so that time > this function will be called from the thread or process context in > addition to the > tasklet context. > > DO you have any suggestion here ? Yes you need to terminate but you don't need to invoke user for terminate flow. It is not expected. Terminate flow expect youu to clean the channels -- ~Vinod -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html