From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030237AbcJRQJI (ORCPT ); Tue, 18 Oct 2016 12:09:08 -0400 Received: from bhuna.collabora.co.uk ([46.235.227.227]:57704 "EHLO bhuna.collabora.co.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1031000AbcJRQI7 (ORCPT ); Tue, 18 Oct 2016 12:08:59 -0400 Subject: Re: [PATCH 2/2] [media] vb2: Add support for use_dma_bidirectional queue flag To: Sakari Ailus References: <1476446894-4220-1-git-send-email-thierry.escande@collabora.com> <1476446894-4220-3-git-send-email-thierry.escande@collabora.com> <20161017100632.GM9460@valkosipuli.retiisi.org.uk> Cc: Mauro Carvalho Chehab , linux-media@vger.kernel.org, linux-kernel@vger.kernel.org, Pawel Osciak , Marek Szyprowski , Kyungmin Park From: Thierry Escande Message-ID: <664b9541-e7ab-f1b3-b12c-677db0e63fc0@collabora.com> Date: Tue, 18 Oct 2016 18:08:53 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0 MIME-Version: 1.0 In-Reply-To: <20161017100632.GM9460@valkosipuli.retiisi.org.uk> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Sakari, On 17/10/2016 12:06, Sakari Ailus wrote: > Hi Thierry, > > Thanks for the set. A few comments below. > > On Fri, Oct 14, 2016 at 02:08:14PM +0200, Thierry Escande wrote: >> From: Pawel Osciak >> >> When this flag is set for CAPTURE queues by the driver on calling >> vb2_queue_init(), it forces the buffers on the queue to be >> allocated/mapped with DMA_BIDIRECTIONAL direction flag, instead of >> DMA_FROM_DEVICE. This allows the device not only to write to the >> buffers, but also read out from them. This may be useful e.g. for codec >> hardware, which may be using CAPTURE buffers as reference to decode >> other buffers. > > Just out of curiosity --- when do you return these buffers back to the user? > Once they're no longer needed as reference frames? Tbh, I don't now. This is used by a rockchip vpu driver not yet upstreamed in the chromeos v4.4 kernel tree. Pawel might answer this question I guess. > >> >> This flag is ignored for OUTPUT queues, as we don't want to allow HW to >> be able to write to OUTPUT buffers. >> >> Signed-off-by: Pawel Osciak >> Tested-by: Pawel Osciak >> Reviewed-by: Tomasz Figa >> Signed-off-by: Thierry Escande >> --- >> drivers/media/v4l2-core/videobuf2-v4l2.c | 8 ++++++-- >> include/media/videobuf2-core.h | 4 ++++ >> 2 files changed, 10 insertions(+), 2 deletions(-) >> >> diff --git a/drivers/media/v4l2-core/videobuf2-v4l2.c b/drivers/media/v4l2-core/videobuf2-v4l2.c >> index fde1e2d..9255291 100644 >> --- a/drivers/media/v4l2-core/videobuf2-v4l2.c >> +++ b/drivers/media/v4l2-core/videobuf2-v4l2.c >> @@ -659,8 +659,12 @@ int vb2_queue_init(struct vb2_queue *q) >> * queues will always initialize waiting_for_buffers to false. >> */ >> q->quirk_poll_must_check_waiting_for_buffers = true; >> - q->dma_dir = V4L2_TYPE_IS_OUTPUT(q->type) >> - ? DMA_TO_DEVICE : DMA_FROM_DEVICE; >> + >> + if (V4L2_TYPE_IS_OUTPUT(q->type)) >> + q->dma_dir = DMA_TO_DEVICE; >> + else >> + q->dma_dir = q->use_dma_bidirectional >> + ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; >> >> return vb2_core_queue_init(q); >> } >> diff --git a/include/media/videobuf2-core.h b/include/media/videobuf2-core.h >> index 38410dd..e613c74 100644 >> --- a/include/media/videobuf2-core.h >> +++ b/include/media/videobuf2-core.h >> @@ -433,6 +433,9 @@ struct vb2_buf_ops { >> * @quirk_poll_must_check_waiting_for_buffers: Return POLLERR at poll when QBUF >> * has not been called. This is a vb1 idiom that has been adopted >> * also by vb2. >> + * @use_dma_bidirectional: use DMA_BIDIRECTIONAL for CAPTURE buffers; this >> + * allows HW to read from the CAPTURE buffers in >> + * addition to writing; ignored for OUTPUT queues >> * @lock: pointer to a mutex that protects the vb2_queue struct. The >> * driver can set this to a mutex to let the v4l2 core serialize >> * the queuing ioctls. If the driver wants to handle locking >> @@ -500,6 +503,7 @@ struct vb2_queue { >> unsigned fileio_write_immediately:1; >> unsigned allow_zero_bytesused:1; >> unsigned quirk_poll_must_check_waiting_for_buffers:1; >> + unsigned use_dma_bidirectional:1; > > This field is in the same struct as dma_dir which it directly affects. > > How about adding a macro instead to give you the queue DMA direction > instead? > > E.g. > > #define vb2_dma_dir(q) \ > (V4L2_TYPE_IS_OUTPUT((q)->type) ? DMA_TO_DEVICE : \ > (q)->use_dma_bidirectional ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE) > > I would call this capture_dma_bidirectional as it only affects capture. Or > simply choose DMA_BIDIRECTIONAL whenever the flag is set. Sure. Will do. Regards, Thierry