Hello, On Thursday, August 25, 2011 1:12 PM Hans Verkuil wrote: > On Thursday, August 25, 2011 12:52:11 Marek Szyprowski wrote: > > This patch changes the order of operations during stream on call. Now the > > buffers are first queued to the driver and then the start_streaming method > > is called. > > > > This resolves the most common case when the driver needs to know buffer > > addresses to enable dma engine and start streaming. Additional parameters > > to start_streaming and buffer_queue methods have been added to simplify > > drivers code. The driver are now obliged to check if the number of queued > > buffers is enough to enable hardware streaming. If not - it should return > > an error. In such case all the buffers that have been pre-queued are > > invalidated. > > > > Drivers that are able to start/stop streaming on-fly, can control dma > > engine directly in buf_queue callback. In this case start_streaming > > callback can be considered as optional. The driver can also assume that > > after a few first buf_queue calls with zero 'streaming' parameter, the core > > will finally call start_streaming callback. > > Looks good! > > > This patch also updates some videobuf2 clients (s5p-fimc, s5p-mfc, s5p-tv, > > mem2mem_testdev and vivi) to work properly with the changed order of > > operations. > > I assume the final patch will update all vb2 clients? Yes, of cource. I just wanted to post the patch ASAP. Updating Samsung and virtual drivers was easy because I already know a bit about them. Updating other drivers requires a bit more work to double check if I didn't break anything. > I have a few very minor comments below: Thanks for your comments :) I will include them in the final version. (snipped) Best regards -- Marek Szyprowski Samsung Poland R&D Center -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
