From: "Sven Püschel" <s.pueschel@pengutronix.de>
To: Nicolas Dufresne <nicolas@ndufresne.ca>,
Jacob Chen <jacob-chen@iotwrt.com>,
Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Heiko Stuebner <heiko@sntech.de>, Rob Herring <robh@kernel.org>,
Krzysztof Kozlowski <krzk+dt@kernel.org>,
Conor Dooley <conor+dt@kernel.org>
Cc: linux-media@vger.kernel.org, linux-rockchip@lists.infradead.org,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
kernel@pengutronix.de
Subject: Re: [PATCH v2 14/22] media: rockchip: rga: support external iommus
Date: Wed, 7 Jan 2026 15:19:50 +0100 [thread overview]
Message-ID: <b4b1a113-a8dc-4ed7-b395-eaa0deb6a957@pengutronix.de> (raw)
In-Reply-To: <c5b6707e3e6748bf923aaf34db1ae231c7e121ad.camel@ndufresne.ca>
Hi,
On 12/24/25 4:50 PM, Nicolas Dufresne wrote:
> Le mercredi 03 décembre 2025 à 16:52 +0100, Sven Püschel a écrit :
>> In preparation for the RGA3 add support for external iommus. This is a
>> transition step to just disable the RGA2 specific mmu table setup code.
>>
>> Currently a simple rga_hw struct field is used to set the internal iommu.
>> But to handle the case of more sophisticated detection mechanisms
>> (e.g. check for an iommu property in the device tree), it is abstracted
>> by an inline function.
>>
>> Signed-off-by: Sven Püschel <s.pueschel@pengutronix.de>
>> ---
>> drivers/media/platform/rockchip/rga/rga-buf.c | 46 ++++++++++++++++-----------
>> drivers/media/platform/rockchip/rga/rga-hw.c | 1 +
>> drivers/media/platform/rockchip/rga/rga.c | 11 +++++--
>> drivers/media/platform/rockchip/rga/rga.h | 6 ++++
>> 4 files changed, 43 insertions(+), 21 deletions(-)
>>
>> diff --git a/drivers/media/platform/rockchip/rga/rga-buf.c b/drivers/media/platform/rockchip/rga/rga-buf.c
>> index 8845e1f894f29..e8d1e2e579140 100644
>> --- a/drivers/media/platform/rockchip/rga/rga-buf.c
>> +++ b/drivers/media/platform/rockchip/rga/rga-buf.c
>> @@ -12,6 +12,7 @@
>> #include <media/v4l2-ioctl.h>
>> #include <media/v4l2-mem2mem.h>
>> #include <media/videobuf2-dma-sg.h>
>> +#include <media/videobuf2-dma-contig.h>
>> #include <media/videobuf2-v4l2.h>
>>
>> #include "rga.h"
>> @@ -79,14 +80,16 @@ static int rga_buf_init(struct vb2_buffer *vb)
>> struct rga_frame *f = rga_get_frame(ctx, vb->vb2_queue->type);
>> size_t n_desc = 0;
>>
>> - n_desc = DIV_ROUND_UP(f->size, PAGE_SIZE);
>> + if (rga_has_internal_iommu(rga)) {
>> + n_desc = DIV_ROUND_UP(f->size, PAGE_SIZE);
> n_desc is not used outside of this scope, perhap drop the definition with the 0
> init, and use:
>
> const size_t n_desc = DIV_ROUND_UP(f->size, PAGE_SIZE);
>
>>
>> - rbuf->n_desc = n_desc;
>> - rbuf->dma_desc = dma_alloc_coherent(rga->dev,
>> - rbuf->n_desc * sizeof(*rbuf->dma_desc),
>> - &rbuf->dma_desc_pa, GFP_KERNEL);
>> - if (!rbuf->dma_desc)
>> - return -ENOMEM;
>> + rbuf->n_desc = n_desc;
>> + rbuf->dma_desc = dma_alloc_coherent(rga->dev,
>> + rbuf->n_desc * sizeof(*rbuf->dma_desc),
>> + &rbuf->dma_desc_pa, GFP_KERNEL);
>> + if (!rbuf->dma_desc)
>> + return -ENOMEM;
>> + }
>>
>> return 0;
>> }
>> @@ -133,17 +136,21 @@ static int rga_buf_prepare(struct vb2_buffer *vb)
>> for (i = 0; i < vb->num_planes; i++) {
>> vb2_set_plane_payload(vb, i, f->pix.plane_fmt[i].sizeimage);
>>
>> - /* Create local MMU table for RGA */
>> - n_desc = fill_descriptors(&rbuf->dma_desc[curr_desc],
>> - rbuf->n_desc - curr_desc,
>> - vb2_dma_sg_plane_desc(vb, i));
>> - if (n_desc < 0) {
>> - v4l2_err(&ctx->rga->v4l2_dev,
>> - "Failed to map video buffer to RGA\n");
>> - return n_desc;
>> + if (rga_has_internal_iommu(ctx->rga)) {
>> + /* Create local MMU table for RGA */
>> + n_desc = fill_descriptors(&rbuf->dma_desc[curr_desc],
>> + rbuf->n_desc - curr_desc,
>> + vb2_dma_sg_plane_desc(vb, i));
>> + if (n_desc < 0) {
>> + v4l2_err(&ctx->rga->v4l2_dev,
>> + "Failed to map video buffer to RGA\n");
>> + return n_desc;
>> + }
>> + dma_addrs[i] = curr_desc << PAGE_SHIFT;
>> + curr_desc += n_desc;
>> + } else {
>> + dma_addrs[i] = vb2_dma_contig_plane_dma_addr(vb, i);
>> }
>> - dma_addrs[i] = curr_desc << PAGE_SHIFT;
>> - curr_desc += n_desc;
>> }
>>
>> /* Fill the remaining planes */
>> @@ -173,8 +180,9 @@ static void rga_buf_cleanup(struct vb2_buffer *vb)
>> struct rga_ctx *ctx = vb2_get_drv_priv(vb->vb2_queue);
>> struct rockchip_rga *rga = ctx->rga;
>>
>> - dma_free_coherent(rga->dev, rbuf->n_desc * sizeof(*rbuf->dma_desc),
>> - rbuf->dma_desc, rbuf->dma_desc_pa);
>> + if (rga_has_internal_iommu(rga))
>> + dma_free_coherent(rga->dev, rbuf->n_desc * sizeof(*rbuf->dma_desc),
>> + rbuf->dma_desc, rbuf->dma_desc_pa);
>> }
>>
>> static void rga_buf_return_buffers(struct vb2_queue *q,
>> diff --git a/drivers/media/platform/rockchip/rga/rga-hw.c b/drivers/media/platform/rockchip/rga/rga-hw.c
>> index fe6f518aedd4e..fb0389ae8e132 100644
>> --- a/drivers/media/platform/rockchip/rga/rga-hw.c
>> +++ b/drivers/media/platform/rockchip/rga/rga-hw.c
>> @@ -617,6 +617,7 @@ static struct rga_fmt formats[] = {
>>
>> const struct rga_hw rga2_hw = {
>> .card_type = "rga2",
>> + .has_internal_iommu = true,
>> .formats = formats,
>> .num_formats = ARRAY_SIZE(formats),
>> .cmdbuf_size = RGA_CMDBUF_SIZE,
>> diff --git a/drivers/media/platform/rockchip/rga/rga.c b/drivers/media/platform/rockchip/rga/rga.c
>> index c9719c7bcfc1c..f28ec88c186fa 100644
>> --- a/drivers/media/platform/rockchip/rga/rga.c
>> +++ b/drivers/media/platform/rockchip/rga/rga.c
>> @@ -23,6 +23,7 @@
>> #include <media/v4l2-ioctl.h>
>> #include <media/v4l2-mem2mem.h>
>> #include <media/videobuf2-dma-sg.h>
>> +#include <media/videobuf2-dma-contig.h>
>> #include <media/videobuf2-v4l2.h>
>>
>> #include "rga.h"
>> @@ -95,13 +96,16 @@ queue_init(void *priv, struct vb2_queue *src_vq, struct vb2_queue *dst_vq)
>> src_vq->io_modes = VB2_MMAP | VB2_DMABUF;
>> src_vq->drv_priv = ctx;
>> src_vq->ops = &rga_qops;
>> - src_vq->mem_ops = &vb2_dma_sg_memops;
>> + src_vq->mem_ops = &vb2_dma_contig_memops;
> That's a bit I'm less knowledgeable here, since I never used the sg_memops in
> any of my work, but what is the consequence ? (appart from gaining external
> iommu support)
This is only for the external iommu support, which uses the
vb2_dma_contig_plane_dma_addr function to get a single linear address
(instead of vb2_dma_sg_plane_desc used for the manual iommu mapping).
Without an external iommu the value is overridden afterwards to
vb2_dma_sg_memops. I probably should move the if up to make clear that
this is only set for the external iommu case.
Or did I misunderstand your question?
>> src_vq->gfp_flags = __GFP_DMA32;
>> src_vq->buf_struct_size = sizeof(struct rga_vb_buffer);
>> src_vq->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_COPY;
>> src_vq->lock = &ctx->rga->mutex;
>> src_vq->dev = ctx->rga->v4l2_dev.dev;
>>
>> + if (rga_has_internal_iommu(ctx->rga))
>> + src_vq->mem_ops = &vb2_dma_sg_memops;
>> +
>> ret = vb2_queue_init(src_vq);
>> if (ret)
>> return ret;
>> @@ -110,13 +114,16 @@ queue_init(void *priv, struct vb2_queue *src_vq, struct vb2_queue *dst_vq)
>> dst_vq->io_modes = VB2_MMAP | VB2_DMABUF;
>> dst_vq->drv_priv = ctx;
>> dst_vq->ops = &rga_qops;
>> - dst_vq->mem_ops = &vb2_dma_sg_memops;
>> + dst_vq->mem_ops = &vb2_dma_contig_memops;
>> dst_vq->gfp_flags = __GFP_DMA32;
>> dst_vq->buf_struct_size = sizeof(struct rga_vb_buffer);
>> dst_vq->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_COPY;
>> dst_vq->lock = &ctx->rga->mutex;
>> dst_vq->dev = ctx->rga->v4l2_dev.dev;
>>
>> + if (rga_has_internal_iommu(ctx->rga))
>> + dst_vq->mem_ops = &vb2_dma_sg_memops;
>> +
>> return vb2_queue_init(dst_vq);
>> }
>>
>> diff --git a/drivers/media/platform/rockchip/rga/rga.h b/drivers/media/platform/rockchip/rga/rga.h
>> index 772f7dff9a8e5..fc81bc5562792 100644
>> --- a/drivers/media/platform/rockchip/rga/rga.h
>> +++ b/drivers/media/platform/rockchip/rga/rga.h
>> @@ -148,6 +148,7 @@ static inline void rga_mod(struct rockchip_rga *rga, u32 reg, u32 val, u32 mask)
>>
>> struct rga_hw {
>> const char *card_type;
>> + bool has_internal_iommu;
>> struct rga_fmt *formats;
>> u32 num_formats;
>> size_t cmdbuf_size;
>> @@ -163,6 +164,11 @@ struct rga_hw {
>> void (*get_version)(struct rockchip_rga *rga);
>> };
>>
>> +static inline bool rga_has_internal_iommu(const struct rockchip_rga *rga)
>> +{
>> + return rga->hw->has_internal_iommu;
>> +}
>> +
>> extern const struct rga_hw rga2_hw;
>>
>> #endif
next prev parent reply other threads:[~2026-01-07 14:20 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-03 15:52 [PATCH v2 00/22] media: platform: rga: Add RGA3 support Sven Püschel
2025-12-03 15:52 ` [PATCH v2 01/22] media: dt-bindings: media: rockchip-rga: add rockchip,rk3588-rga3 Sven Püschel
2025-12-03 20:28 ` Krzysztof Kozlowski
2025-12-03 21:21 ` Sven Püschel
2025-12-03 15:52 ` [PATCH v2 02/22] media: v4l2-common: add has_alpha to v4l2_format_info Sven Püschel
2025-12-24 14:24 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 03/22] media: v4l2-common: add v4l2_fill_pixfmt_mp_aligned helper Sven Püschel
2025-12-24 14:47 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 04/22] media: rockchip: rga: use clk_bulk api Sven Püschel
2025-12-24 14:48 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 05/22] media: rockchip: rga: use stride for offset calculation Sven Püschel
2025-12-24 15:00 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 06/22] media: rockchip: rga: remove redundant rga_frame variables Sven Püschel
2025-12-24 15:03 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 07/22] media: rockchip: rga: move hw specific parts to a dedicated struct Sven Püschel
2025-12-24 15:05 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 08/22] media: rockchip: rga: move cmdbuf to rga_ctx Sven Püschel
2025-12-24 15:21 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 09/22] media: rockchip: rga: align stride to 4 bytes Sven Püschel
2025-12-24 15:26 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 10/22] media: rockchip: rga: prepare cmdbuf on streamon Sven Püschel
2025-12-24 15:29 ` Nicolas Dufresne
2026-01-07 12:00 ` Sven Püschel
2025-12-03 15:52 ` [PATCH v2 11/22] media: rockchip: rga: check scaling factor Sven Püschel
2025-12-24 15:39 ` Nicolas Dufresne
2026-01-09 12:18 ` Sven Püschel
2025-12-03 15:52 ` [PATCH v2 12/22] media: rockchip: rga: use card type to specify rga type Sven Püschel
2025-12-24 15:40 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 13/22] media: rockchip: rga: change offset to dma_addresses Sven Püschel
2025-12-24 15:41 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 14/22] media: rockchip: rga: support external iommus Sven Püschel
2025-12-24 15:50 ` Nicolas Dufresne
2026-01-07 14:19 ` Sven Püschel [this message]
2026-01-07 14:24 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 15/22] media: rockchip: rga: share the interrupt when an external iommu is used Sven Püschel
2025-12-24 15:50 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 16/22] media: rockchip: rga: remove size from rga_frame Sven Püschel
2025-12-24 15:54 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 17/22] media: rockchip: rga: remove stride " Sven Püschel
2025-12-24 15:56 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 18/22] media: rockchip: rga: move rga_fmt to rga-hw.h Sven Püschel
2025-12-24 15:59 ` Nicolas Dufresne
2026-01-07 14:52 ` Sven Püschel
2025-12-03 15:52 ` [PATCH v2 19/22] media: rockchip: rga: add feature flags Sven Püschel
2025-12-24 16:00 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 20/22] media: rockchip: rga: disable multi-core support Sven Püschel
2025-12-24 16:02 ` Nicolas Dufresne
2025-12-03 15:52 ` [PATCH v2 21/22] media: rockchip: rga: add rga3 support Sven Püschel
2025-12-24 16:34 ` Nicolas Dufresne
2026-01-21 14:40 ` Sven Püschel
2025-12-03 15:52 ` [PATCH v2 22/22] arm64: dts: rockchip: add rga3 dt nodes Sven Püschel
2025-12-05 22:36 ` [PATCH v2 00/22] media: platform: rga: Add RGA3 support Rob Herring
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=b4b1a113-a8dc-4ed7-b395-eaa0deb6a957@pengutronix.de \
--to=s.pueschel@pengutronix.de \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=ezequiel@vanguardiasur.com.ar \
--cc=heiko@sntech.de \
--cc=jacob-chen@iotwrt.com \
--cc=kernel@pengutronix.de \
--cc=krzk+dt@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-rockchip@lists.infradead.org \
--cc=mchehab@kernel.org \
--cc=nicolas@ndufresne.ca \
--cc=robh@kernel.org \
/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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®