From mboxrd@z Thu Jan 1 00:00:00 1970 From: vinod.koul@intel.com (Vinod Koul) Date: Thu, 15 Oct 2015 09:37:08 +0530 Subject: [PATCH v2 2/4] dmaengine: Add STM32 DMA driver In-Reply-To: References: <1444745127-1105-3-git-send-email-cedric.madianga@gmail.com> <561D1689.90703@linaro.org> <561E58AC.80207@linaro.org> <561E658D.1070900@linaro.org> <561E7497.1050105@linaro.org> Message-ID: <20151015040708.GC27370@localhost> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Wed, Oct 14, 2015 at 05:41:26PM +0200, M'boumba Cedric Madianga wrote: > 2015-10-14 17:28 GMT+02:00 Daniel Thompson : > > On 14/10/15 16:26, M'boumba Cedric Madianga wrote: > >> > >> 2015-10-14 16:24 GMT+02:00 Daniel Thompson : > >>> > >>> > >>> Hmnnn... > >>> > >>> The dmaengine framework will WARN_ONCE() if an dmaengine is removed > >>> whilst > >>> it is active and also works hard to ensure dmaengine modules are not > >>> removed > >>> whilst there are active drivers using the framework. > >>> > >>> How do we get into this function whilst there is still an active DMA > >>> channels? > >> > >> > >> For example, when a user try "rmmod stm32-dma" in uart console. > >> It will enter in stm32_dma_remove while there is potentially still active > >> DMA. > > > > > > Check dmaengine.c for yourself but I think in this case the dmaengine > > framework will hold references to the module and prevent the remove from > > taking place. > > Yes I did it. > As far I understand, the dmaengine framework will print a warning > message but doesn't stop removing operation if there are some active > clients. we hold a ref, so removal wont work, see dma_chan_get() -- ~Vinod