| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <cwlqzlwpjfthun5lhiaua6e3xt3ta3xzwtltsaxlgqsle2dyw4@d7xfgwvwjk67>
Date: Mon, 31 Mar 2025 17:16:52 +0200
From: Jacopo Mondi <jacopo.mondi@...asonboard.com>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>
Cc: Jacopo Mondi <jacopo.mondi+renesas@...asonboard.com>,
Kieran Bingham <kieran.bingham+renesas@...asonboard.com>,
Niklas Söderlund <niklas.soderlund@...natech.se>, Mauro Carvalho Chehab <mchehab@...nel.org>,
linux-kernel@...r.kernel.org, linux-media@...r.kernel.org, linux-renesas-soc@...r.kernel.org
Subject: Re: [PATCH v2] media: vsp1: Add VSPX support
Hi Laurent
On Mon, Mar 24, 2025 at 07:26:58PM +0200, Laurent Pinchart wrote:
> Hi Jacopo,
>
> Thank you for the patch.
>
> On Wed, Mar 19, 2025 at 04:36:56PM +0100, Jacopo Mondi wrote:
> > Add support for VSPX, a specialized version of the VSP2 that
> > transfer data to the ISP. The VSPX is composed by two RPF units
>
> s/transfer/transfers/
> s/composed by/composed of/
>
> > to read data from external memory and an IIF instance that performs
> > transfer towards the ISP.
> >
> > The VSPX is supported through a newly introduced vsp1_vspx.c file that
> > exposes two interfaces: vsp1_vspx interface, declared in vsp1_vspx.h
> > for the vsp1 core to initialize and cleanup the VSPX, and a vsp1_isp
> > interface, declared in include/media/vsp1.h for the ISP driver to
> > control the VSPX operations.
>
> I have a bit of a hard time reviewing the API exposed to the ISP without
> seeing how it gets used, so I'll start with a partial review.
>
> > Signed-off-by: Jacopo Mondi <jacopo.mondi+renesas@...asonboard.com>
> > ---
> > Add a driver for Rensas VSP2 VSPX function.
> > The VSPX transfer image and config data to the ISP.
> > ---
> > Changes in v2:
> > - Completed documentation of vspx ISP interface in vsp1.h
> > - Link to v1: https://lore.kernel.org/r/20250319-v4h-vspx-v1-1-83761280071b@ideasonboard.com
> > ---
> > drivers/media/platform/renesas/vsp1/Makefile | 1 +
> > drivers/media/platform/renesas/vsp1/vsp1.h | 1 +
> > drivers/media/platform/renesas/vsp1/vsp1_drv.c | 13 +-
> > drivers/media/platform/renesas/vsp1/vsp1_pipe.c | 6 +
> > drivers/media/platform/renesas/vsp1/vsp1_regs.h | 1 +
> > drivers/media/platform/renesas/vsp1/vsp1_vspx.c | 526 ++++++++++++++++++++++++
> > drivers/media/platform/renesas/vsp1/vsp1_vspx.h | 76 ++++
> > include/media/vsp1.h | 138 +++++++
> > 8 files changed, 761 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/media/platform/renesas/vsp1/Makefile b/drivers/media/platform/renesas/vsp1/Makefile
> > index de8c802e1d1a..2057c8f7be47 100644
> > --- a/drivers/media/platform/renesas/vsp1/Makefile
> > +++ b/drivers/media/platform/renesas/vsp1/Makefile
> > @@ -6,5 +6,6 @@ vsp1-y += vsp1_clu.o vsp1_hsit.o vsp1_lut.o
> > vsp1-y += vsp1_brx.o vsp1_sru.o vsp1_uds.o
> > vsp1-y += vsp1_hgo.o vsp1_hgt.o vsp1_histo.o
> > vsp1-y += vsp1_iif.o vsp1_lif.o vsp1_uif.o
> > +vsp1-y += vsp1_vspx.o
> >
> > obj-$(CONFIG_VIDEO_RENESAS_VSP1) += vsp1.o
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1.h b/drivers/media/platform/renesas/vsp1/vsp1.h
> > index 263024639dd2..1872e403f26b 100644
> > --- a/drivers/media/platform/renesas/vsp1/vsp1.h
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1.h
> > @@ -110,6 +110,7 @@ struct vsp1_device {
> > struct media_entity_operations media_ops;
> >
> > struct vsp1_drm *drm;
> > + struct vsp1_vspx *vspx;
> > };
> >
> > int vsp1_device_get(struct vsp1_device *vsp1);
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1_drv.c b/drivers/media/platform/renesas/vsp1/vsp1_drv.c
> > index d13e9b31aa7c..e338ab8af292 100644
> > --- a/drivers/media/platform/renesas/vsp1/vsp1_drv.c
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1_drv.c
> > @@ -38,6 +38,7 @@
> > #include "vsp1_uds.h"
> > #include "vsp1_uif.h"
> > #include "vsp1_video.h"
> > +#include "vsp1_vspx.h"
> >
> > /* -----------------------------------------------------------------------------
> > * Interrupt Handling
> > @@ -488,7 +489,10 @@ static int vsp1_create_entities(struct vsp1_device *vsp1)
> >
> > ret = media_device_register(mdev);
> > } else {
> > - ret = vsp1_drm_init(vsp1);
> > + if (vsp1->info->version == VI6_IP_VERSION_MODEL_VSPX_GEN4)
> > + ret = vsp1_vspx_init(vsp1);
> > + else
> > + ret = vsp1_drm_init(vsp1);
> > }
> >
> > done:
> > @@ -846,6 +850,13 @@ static const struct vsp1_device_info vsp1_device_infos[] = {
> > .uif_count = 2,
> > .wpf_count = 1,
> > .num_bru_inputs = 5,
> > + }, {
> > + .version = VI6_IP_VERSION_MODEL_VSPX_GEN4,
> > + .model = "VSP2-X",
> > + .gen = 4,
> > + .features = VSP1_HAS_IIF,
> > + .rpf_count = 2,
> > + .wpf_count = 1,
> > },
> > };
> >
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1_pipe.c b/drivers/media/platform/renesas/vsp1/vsp1_pipe.c
> > index a51061738edc..12ede30e0af7 100644
> > --- a/drivers/media/platform/renesas/vsp1/vsp1_pipe.c
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1_pipe.c
> > @@ -24,6 +24,7 @@
> > #include "vsp1_pipe.h"
> > #include "vsp1_rwpf.h"
> > #include "vsp1_uds.h"
> > +#include "vsp1_vspx.h"
> >
> > /* -----------------------------------------------------------------------------
> > * Helper Functions
> > @@ -100,6 +101,11 @@ static const struct vsp1_format_info vsp1_video_formats[] = {
> > VI6_RPF_DSWAP_P_WDS,
> > 1, { 16, 0, 0 }, false, false, 1, 1, false },
> >
> > + /* R-Car V4H Metadata format: Maps on ARGB888 */
>
> s/Maps on/Maps to/
>
> > + { V4L2_META_FMT_RCAR_V4H, MEDIA_BUS_FMT_METADATA_FIXED,
> > + VI6_FMT_ARGB_8888, VI6_RPF_DSWAP_P_LLS | VI6_RPF_DSWAP_P_LWS,
> > + 1, { 32, 0, 0 }, false, false, 1, 1, false },
> > +
> > { V4L2_PIX_FMT_RGB332, MEDIA_BUS_FMT_ARGB8888_1X32,
> > VI6_FMT_RGB_332, VI6_RPF_DSWAP_P_LLS | VI6_RPF_DSWAP_P_LWS |
> > VI6_RPF_DSWAP_P_WDS | VI6_RPF_DSWAP_P_BTS,
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1_regs.h b/drivers/media/platform/renesas/vsp1/vsp1_regs.h
> > index 86e47c2d991f..10cfbcd1b6e0 100644
> > --- a/drivers/media/platform/renesas/vsp1/vsp1_regs.h
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1_regs.h
> > @@ -799,6 +799,7 @@
> > #define VI6_IP_VERSION_MODEL_VSPDL_GEN3 (0x19 << 8)
> > #define VI6_IP_VERSION_MODEL_VSPBS_GEN3 (0x1a << 8)
> > #define VI6_IP_VERSION_MODEL_VSPD_GEN4 (0x1c << 8)
> > +#define VI6_IP_VERSION_MODEL_VSPX_GEN4 (0x1d << 8)
> > /* RZ/G2L SoCs have no version register, So use 0x80 as the model version */
> > #define VI6_IP_VERSION_MODEL_VSPD_RZG2L (0x80 << 8)
> >
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1_vspx.c b/drivers/media/platform/renesas/vsp1/vsp1_vspx.c
> > new file mode 100644
> > index 000000000000..b6dce5f50fed
> > --- /dev/null
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1_vspx.c
> > @@ -0,0 +1,526 @@
> > +// SPDX-License-Identifier: GPL-2.0+
> > +/*
> > + * vsp1_vspx.c -- R-Car Gen 4 VSPX
> > + *
> > + * Copyright (C) 2025 Ideas On Board Oy
> > + * Copyright (C) 2025 Renesas Electronics Corporation
> > + */
> > +
> > +#include "vsp1_vspx.h"
> > +
> > +#include <linux/cleanup.h>
> > +#include <linux/device.h>
> > +#include <linux/dma-mapping.h>
> > +#include <linux/slab.h>
> > +#include <linux/spinlock.h>
> > +
> > +#include <media/media-entity.h>
> > +#include <media/v4l2-subdev.h>
> > +#include <media/vsp1.h>
> > +
> > +#include "vsp1.h"
> > +#include "vsp1_dl.h"
> > +#include "vsp1_iif.h"
> > +#include "vsp1_pipe.h"
> > +#include "vsp1_rwpf.h"
> > +
> > +#define VSPX_RPF_SINK_PAD_IMG 0
> > +#define VSPX_RPF_SINK_PAD_CONFIG 2
> > +
> > +static const struct v4l2_pix_format_mplane vspx_default_fmt = {
> > + .width = 1920,
> > + .height = 1080,
> > + .pixelformat = V4L2_PIX_FMT_SRGGB8,
> > + .field = V4L2_FIELD_NONE,
> > + .num_planes = 1,
> > + .plane_fmt = {
> > + [0] = {
> > + .sizeimage = 1920 * 1080,
> > + .bytesperline = 1920,
> > + },
> > + },
> > +};
> > +
> > +/*
> > + * Apply the given width, height and fourcc to the subdevice inside the
> > + * VSP1 entity.
> > + */
> > +static int vsp1_vspx_rwpf_set_subdev_fmt(struct vsp1_device *vsp1,
> > + struct vsp1_rwpf *rwpf, u32 fourcc,
> > + unsigned int width,
> > + unsigned int height)
> > +{
> > + struct vsp1_entity *ent = &rwpf->entity;
> > + const struct vsp1_format_info *fmtinfo;
> > + struct v4l2_subdev_format format = {};
> > + int ret;
> > +
> > + fmtinfo = vsp1_get_format_info(vsp1, fourcc);
> > + if (!fmtinfo) {
> > + dev_dbg(vsp1->dev, "Unsupported pixel format %08x\n", fourcc);
> > + return -EINVAL;
> > + }
> > +
> > + rwpf->fmtinfo = fmtinfo;
> > +
> > + format.which = V4L2_SUBDEV_FORMAT_ACTIVE;
> > + format.pad = RWPF_PAD_SINK;
> > + format.format.width = width;
> > + format.format.height = height;
> > + format.format.field = V4L2_FIELD_NONE;
> > + format.format.code = fmtinfo->mbus;
> > +
> > + ret = v4l2_subdev_call(&ent->subdev, pad, set_fmt, NULL, &format);
> > + if (ret < 0)
> > + return ret;
> > +
> > + return 0;
> > +}
> > +
> > +/*
> > + * Configure RPF0 for ConfigDMA or RAW image transfer.
> > + */
> > +static int vsp1_vspx_rpf0_configure(struct vsp1_device *vsp1,
> > + dma_addr_t addr, u32 fourcc,
> > + unsigned int width, unsigned int height,
> > + unsigned int rpf_sink_pad,
>
> The variable name is a bit misleading, this is the sink pad of the
> entity connected to the RPF output, i.e. the IIF. I'd name it
> iif_sink_pad, and rename the VSPX_RPF_SINK_PAD_* macros. Those macros
> should be moved to vsp1_iif.h (probably in patch "[PATCH v6 1/7] media:
> vsp1: Add support IIF ISP Interface").
>
> > + struct vsp1_dl_list *dl,
> > + struct vsp1_dl_body *dlb)
> > +{
> > + const struct vsp1_format_info *fmtinfo;
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > + struct vsp1_pipeline *pipe;
> > + struct vsp1_rwpf *rpf0;
>
> You can write
>
> struct vsp1_pipeline *pipe = &vsp1->vspx->pipe.pipe;
> struct vsp1_rwpf *rpf0 = pipe->inputs[0];
> const struct vsp1_format_info *fmtinfo;
>
> > + int ret;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > + pipe = &vspx_pipe->pipe;
> > + rpf0 = pipe->inputs[0];
>
> And drop this.
>
> > +
> > + ret = vsp1_vspx_rwpf_set_subdev_fmt(vsp1, rpf0, fourcc, width, height);
> > + if (ret)
> > + return ret;
> > +
> > + /*
> > + * The RPF0 sink pad is used to configure the entity routing table
> > + * and compute stride.
>
> This should also be updated, maybe to
>
> * Connect RPF0 to the IIF sink pad corresponding to the config or image
> * path.
>
> > + */
> > + rpf0->entity.sink_pad = rpf_sink_pad;
> > +
> > + if (rpf_sink_pad == VSPX_RPF_SINK_PAD_CONFIG) {
> > + /* ConfigDMA is a 1-D buffer with 4 bytes entries. */
> > + rpf0->format.plane_fmt[0].bytesperline = 4;
>
> This doesn't seem right.
>
Stride doesn't seem to make a difference for ConfigDMA.
As the ConfigDMA buffer is logically a sequence of u32 it felt 4 bytes
was a natural choice.
Otherwise I should set this to pairs * 2 + 2 (the whole size of the
ConfigDMA buffer), but this feels a tad weird ?
> > + } else {
> > + /* fmtinfo is valid as per vsp1_vspx_rwpf_set_subdev_fmt(). */
> > + fmtinfo = vsp1_get_format_info(vsp1, fourcc);
> > + rpf0->format.plane_fmt[0].bytesperline = width *
> > + (7 + fmtinfo->bpp[0]) / 8;
>
> The stride should be passed by the caller, likely using
> job->img.fmt.fmt.pix_mp.bytesperline. For symmetry, I would also pass
> the config stride from the caller to this function, as you pass the
> width.
Doesn't this have enough parameters ?
ret = vsp1_vspx_rpf0_configure(vsp1, desc->img.mem,
desc->img.fmt.fmt.pix_mp.pixelformat,
desc->img.fmt.fmt.pix_mp.width,
desc->img.fmt.fmt.pix_mp.height,
VSPX_RPF_SINK_PAD_IMG, dl, dlb);
I would keep the trivial calculation of stride here, we have
everything we need to perform it in this function.
>
> > + }
> > +
> > + pipe->part_table[0].rpf[0].width = width;
> > + pipe->part_table[0].rpf[0].height = height;
> > +
> > + rpf0->mem.addr[0] = addr;
> > + rpf0->mem.addr[1] = 0;
> > + rpf0->mem.addr[2] = 0;
> > +
> > + vsp1_entity_route_setup(&rpf0->entity, pipe, dlb);
> > + vsp1_entity_configure_stream(&rpf0->entity, rpf0->entity.state, pipe,
> > + dl, dlb);
> > + vsp1_entity_configure_partition(&rpf0->entity, pipe,
> > + &pipe->part_table[0], dl, dlb);
> > +
> > + return 0;
> > +}
> > +
> > +static int vsp1_vspx_job_schedule(struct vsp1_device *vsp1)
> > +{
> > + struct vsp1_dl_list *second_dl = NULL;
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > + struct vsp1_dl_list *first_dl;
> > + struct vsp1_isp_job_desc *job;
> > + struct vsp1_pipeline *pipe;
> > + struct vsp1_dl_body *dlb;
> > + struct vsp1_dl_list *dl;
> > + unsigned long flags;
> > + int ret = 0;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > + pipe = &vspx_pipe->pipe;
> > +
> > + mutex_lock(&vspx_pipe->vspx_lock);
>
> I think you can use
>
> guard(mutex, &vspx_pipe->vspx_lock);
>
> and drop the error_unlock_and_return label. Similary, with
>
> struct vsp1_isp_job_desc *job __free(kfree) = NULL;
>
> you could drop the error_free_and_unlock label. The ret variable
> wouldn't need to be initialized in that case.
>
> This being said, as Niklas pointed out, this function is called from
> interrupt context, so you need a spinlock instead. There's lots of work
> being done here to be run in an interrupt handler. It would be better to
> create the display list in user context (from vsp1_isp_queue_job()) and
> only commit it in this function.
I've split display list creation and commit to two different
functions.
>
> > + if (vspx_pipe->processing)
> > + goto error_unlock_and_return;
> > +
> > + /* Extract one job, if available, from the jobs list. */
> > + scoped_guard(mutex, &vspx_pipe->jobs_lock) {
> > + job = list_first_entry_or_null(&vspx_pipe->jobs,
> > + struct vsp1_isp_job_desc,
> > + job_queue);
> > + if (!job)
> > + goto error_unlock_and_return;
> > +
> > + list_del(&job->job_queue);
> > + }
> > +
> > + /*
> > + * Transfer the buffers described in the job: (optional) ConfigDMA and
> > + * RAW image.
> > + */
> > +
> > + first_dl = vsp1_dl_list_get(pipe->output->dlm);
> > +
> > + dl = first_dl;
> > + dlb = vsp1_dl_list_get_body0(dl);
> > +
> > + /* Disable RPF1. */
> > + vsp1_dl_body_write(dlb, vsp1->rpf[1]->entity.route->reg,
> > + VI6_DPR_NODE_UNUSED);
> > +
> > + /* Configure IIF routing and enable IIF function */
> > + vsp1_entity_route_setup(pipe->iif, pipe, dlb);
> > + vsp1_entity_configure_stream(pipe->iif, pipe->iif->state, pipe,
> > + first_dl, dlb);
> > +
> > + /* Configure WPF0 to enable RPF0 as source*/
> > + vsp1_entity_route_setup(&pipe->output->entity, pipe, dlb);
> > + vsp1_entity_configure_stream(&pipe->output->entity,
> > + pipe->output->entity.state, pipe,
> > + first_dl, dlb);
> > +
> > + if (job->config.pairs) {
> > + /*
> > + * Configure RPF0 for config data. Transfer the number of
> > + * configuration pairs plus 2 words for the header.
> > + */
> > + ret = vsp1_vspx_rpf0_configure(vsp1, job->config.mem,
> > + V4L2_META_FMT_RCAR_V4H,
> > + job->config.pairs * 2 + 2, 1,
> > + VSPX_RPF_SINK_PAD_CONFIG,
> > + dl, dlb);
> > + if (ret)
> > + goto error_free_and_unlock;
> > +
> > + second_dl = vsp1_dl_list_get(pipe->output->dlm);
> > +
> > + dl = second_dl;
> > + dlb = vsp1_dl_list_get_body0(dl);
> > + }
> > +
> > + /* Configure RPF0 for RAW image transfer. */
> > + ret = vsp1_vspx_rpf0_configure(vsp1, job->img.mem,
> > + job->img.fmt.fmt.pix_mp.pixelformat,
> > + job->img.fmt.fmt.pix_mp.width,
> > + job->img.fmt.fmt.pix_mp.height,
> > + VSPX_RPF_SINK_PAD_IMG, dl, dlb);
> > + if (ret)
> > + goto error_free_and_unlock;
> > +
> > + if (second_dl)
> > + vsp1_dl_list_add_chain(first_dl, second_dl);
> > +
> > + vsp1_dl_list_commit(first_dl, 0);
> > +
> > + vspx_pipe->processing = true;
> > + mutex_unlock(&vspx_pipe->vspx_lock);
> > +
> > + kfree(job);
> > +
> > + /* Trigger VSPX processing by setting VI6_CMD[STRCMD]. */
> > + spin_lock_irqsave(&pipe->irqlock, flags);
> > + vsp1_pipeline_run(pipe);
> > + spin_unlock_irqrestore(&pipe->irqlock, flags);
>
> You can use a scoped guard. Same below.
>
> > +
> > + return 0;
> > +
> > +error_free_and_unlock:
> > + kfree(job);
> > +error_unlock_and_return:
> > + mutex_unlock(&vspx_pipe->vspx_lock);
> > +
> > + return ret;
> > +}
> > +
> > +/*******************************************************************************
> > + * Interrupt handling
> > + */
>
> Following the driver coding style,
>
> /* -----------------------------------------------------------------------------
> * Interrupt handling
> */
>
> Same below.
>
> > +static void vsp1_vspx_pipeline_frame_end(struct vsp1_pipeline *pipe,
> > + unsigned int completion)
> > +{
> > + struct vsp1_vspx_pipeline *vspx_pipe = to_vsp1_vspx_pipeline(pipe);
> > + struct vsp1_vspx *vspx = to_vsp1_vspx(vspx_pipe);
> > + struct vsp1_device *vsp1 = vspx->vsp1;
>
> This is used in a single place below, where you can use vspx->vsp1
> without declaring a local variable.
>
> > + unsigned long flags;
> > +
> > + if (vspx_pipe->vspx_frame_end)
> > + vspx_pipe->vspx_frame_end(vspx_pipe->frame_end_data);
> > +
> > + /*
> > + * Set the pipeline state to stopped to ensure the next call to
> > + * vsp1_pipeline_run() actually starts the VSPX.
> > + */
> > + spin_lock_irqsave(&pipe->irqlock, flags);
> > + pipe->state = VSP1_PIPELINE_STOPPED;
> > + spin_unlock_irqrestore(&pipe->irqlock, flags);
> > +
> > + scoped_guard(mutex, &vspx_pipe->vspx_lock)
> > + vspx_pipe->processing = false;
>
> As this is a relatively new construct we'll need to figure out the
> coding style. The absence of curly braces looks weird to me, I would
> have written
>
> scoped_guard(mutex, &vspx_pipe->vspx_lock) {
> vspx_pipe->processing = false;
> }
>
> > +
> > + /* Try schedule a new job from the queue. */
> > + vsp1_vspx_job_schedule(vsp1);
> > +}
> > +
> > +/*******************************************************************************
> > + * vsp1_isp interface (include/media/vsp1.h)
>
> I'd write "ISP Driver API" to match vsp1_drm.c.
>
> > + */
> > +
> > +int vsp1_isp_init(struct device *dev)
> > +{
> > + struct vsp1_device *vsp1 = dev_get_drvdata(dev);
> > +
> > + if (!vsp1)
> > + return -EPROBE_DEFER;
> > +
> > + return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(vsp1_isp_init);
> > +
> > +struct device *vsp1_isp_get_bus_master(struct device *dev)
> > +{
> > + struct vsp1_device *vsp1 = dev_get_drvdata(dev);
> > +
> > + if (!vsp1)
> > + return ERR_PTR(-ENODEV);
> > +
> > + return vsp1->bus_master;
> > +}
> > +EXPORT_SYMBOL_GPL(vsp1_isp_get_bus_master);
> > +
> > +int vsp1_isp_alloc_buffers(struct device *dev, size_t size,
> > + struct vsp1_isp_buffer_desc *buffer_desc)
> > +{
> > + struct device *bus_master = vsp1_isp_get_bus_master(dev);
> > +
> > + if (IS_ERR_OR_NULL(bus_master))
> > + return -ENODEV;
> > +
> > + buffer_desc->cpu_addr = dma_alloc_coherent(bus_master, size,
> > + &buffer_desc->dma_addr,
> > + GFP_KERNEL);
> > + if (IS_ERR_OR_NULL(buffer_desc->cpu_addr))
> > + return -EINVAL;
> > +
> > + return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(vsp1_isp_alloc_buffers);
> > +
> > +int vsp1_isp_configure(struct device *dev,
> > + const struct v4l2_pix_format_mplane *fmt)
> > +{
> > + struct vsp1_device *vsp1 = dev_get_drvdata(dev);
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > + struct vsp1_pipeline *pipe;
> > + int ret;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > + pipe = &vspx_pipe->pipe;
> > +
> > + guard(mutex)(&vspx_pipe->vspx_lock);
> > +
> > + /*
> > + * Apply the same format to the RPF0 and WPF0 so that the partition
> > + * calculation results in a single partition.
> > + */
> > + ret = vsp1_vspx_rwpf_set_subdev_fmt(vsp1, pipe->inputs[0],
> > + fmt->pixelformat, fmt->width,
> > + fmt->height);
> > + if (ret)
> > + return ret;
> > +
> > + ret = vsp1_vspx_rwpf_set_subdev_fmt(vsp1, pipe->output, fmt->pixelformat,
> > + fmt->width, fmt->height);
> > + if (ret)
> > + return ret;
> > +
> > + vsp1_pipeline_calculate_partition(pipe, &pipe->part_table[0],
> > + fmt->width, 0);
> > +
> > + return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(vsp1_isp_configure);
> > +
> > +static void vsp1_vspx_release_jobs(struct vsp1_device *vsp1)
> > +{
> > + struct vsp1_isp_job_desc *job, *tmp;
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > +
> > + guard(mutex)(&vspx_pipe->jobs_lock);
> > +
> > + list_for_each_entry_safe(job, tmp, &vspx_pipe->jobs, job_queue) {
> > + list_del(&job->job_queue);
> > + kfree(job);
> > + }
> > +}
> > +
> > +static void vsp1_isp_stop_streaming(struct vsp1_device *vsp1)
> > +{
> > + struct vsp1_vspx_pipeline *vspx_pipe = &vsp1->vspx->pipe;
> > + struct vsp1_pipeline *pipe = &vspx_pipe->pipe;
> > +
> > + pipe->state = VSP1_PIPELINE_STOPPED;
> > + vsp1_pipeline_stop(pipe);
> > + vsp1_vspx_release_jobs(vsp1);
> > + vspx_pipe->processing = false;
> > + vspx_pipe->vspx_frame_end = NULL;
> > + vsp1_device_put(vsp1);
> > +}
> > +
> > +int vsp1_isp_start_streaming(struct device *dev,
> > + struct vsp1_vspx_frame_end *frame_end,
> > + bool enable)
> > +{
> > + struct vsp1_device *vsp1 = dev_get_drvdata(dev);
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > + struct vsp1_pipeline *pipe;
> > + u32 value;
> > + int ret;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > + pipe = &vspx_pipe->pipe;
> > +
> > + guard(mutex)(&vspx_pipe->vspx_lock);
> > +
> > + if (!enable) {
> > + vsp1_isp_stop_streaming(vsp1);
> > + return 0;
> > + }
> > +
> > + if (!frame_end)
> > + return -EINVAL;
> > +
> > + vspx_pipe->vspx_frame_end = frame_end->vspx_frame_end;
> > + vspx_pipe->frame_end_data = frame_end->frame_end_data;
> > +
> > + /* Make sure VSPX is not active. */
> > + value = vsp1_read(vsp1, VI6_CMD(0));
> > + if (value & VI6_CMD_STRCMD) {
> > + dev_err(vsp1->dev,
> > + "%s: Starting of WPF0 already reserved\n", __func__);
> > + return -EBUSY;
> > + }
> > +
> > + value = vsp1_read(vsp1, VI6_STATUS);
> > + if (value & VI6_STATUS_SYS_ACT(0)) {
> > + dev_err(vsp1->dev,
> > + "%s: WPF0 has not entered idle state\n", __func__);
> > + return -EBUSY;
> > + }
> > +
> > + /* Enable the VSP1 and prepare for streaming. */
> > + vsp1_pipeline_dump(pipe, "VSPX job");
> > +
> > + ret = vsp1_device_get(vsp1);
> > + if (ret < 0)
> > + return ret;
> > +
> > + return 0;
> > +}
> > +EXPORT_SYMBOL_GPL(vsp1_isp_start_streaming);
> > +
> > +int vsp1_isp_queue_job(struct device *dev, const struct vsp1_isp_job_desc *job)
> > +{
> > + struct vsp1_device *vsp1 = dev_get_drvdata(dev);
> > + struct vsp1_vspx_pipeline *vspx_pipe = &vsp1->vspx->pipe;
> > + struct vsp1_isp_job_desc *newjob;
> > +
> > + /*
> > + * Copy the job description and append it to the jobs queue.
> > + * Memory is released when the job is consumed.
> > + */
> > + newjob = kmalloc(sizeof(*newjob), GFP_KERNEL);
> > + if (!newjob)
> > + return -ENOMEM;
> > +
> > + *newjob = *job;
> > +
> > + scoped_guard(mutex, &vspx_pipe->jobs_lock)
> > + list_add_tail(&newjob->job_queue, &vspx_pipe->jobs);
> > +
> > + vsp1_vspx_job_schedule(vsp1);
> > +
> > + return 0;
> > +}
> > +EXPORT_SYMBOL(vsp1_isp_queue_job);
>
> Any reason for this one not being GPL ?
>
Ofc course not.
I'll fix the rest and send the driver on top of IIF support patches
listed as pre-requisites here.
> > +
> > +/*******************************************************************************
> > + * vsp1_vspx interface (drivers/media/platform/renesas/vsp1/vsp1_vspx.h)
>
> I'd just write "Initialization & cleanup".
>
> > + */
> > +
> > +int vsp1_vspx_init(struct vsp1_device *vsp1)
> > +{
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > + struct vsp1_pipeline *pipe;
> > +
> > + vsp1->vspx = devm_kzalloc(vsp1->dev, sizeof(*vsp1->vspx), GFP_KERNEL);
> > + if (!vsp1->vspx)
> > + return -ENOMEM;
> > +
> > + vsp1->vspx->vsp1 = vsp1;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
> > + vspx_pipe->processing = false;
> > +
> > + pipe = &vspx_pipe->pipe;
> > +
> > + vsp1_pipeline_init(pipe);
> > +
> > + pipe->partitions = 1;
> > + pipe->part_table = &vspx_pipe->partition;
> > + pipe->interlaced = 0;
>
> This is a boolean field, you can use false.
>
> > + pipe->frame_end = vsp1_vspx_pipeline_frame_end;
> > +
> > + INIT_LIST_HEAD(&vspx_pipe->jobs);
> > + mutex_init(&vspx_pipe->vspx_lock);
> > + mutex_init(&vspx_pipe->jobs_lock);
> > +
> > + pipe->output = vsp1->wpf[0];
> > + pipe->output->entity.pipe = pipe;
> > + vsp1_vspx_rwpf_set_subdev_fmt(vsp1, pipe->output,
> > + vspx_default_fmt.pixelformat,
> > + vspx_default_fmt.width,
> > + vspx_default_fmt.height);
> > + list_add_tail(&pipe->output->entity.list_pipe, &pipe->entities);
> > +
> > + pipe->iif = &vsp1->iif->entity;
> > + pipe->iif->pipe = pipe;
> > + pipe->iif->sink = &pipe->output->entity;
>
> pipe->iif->sink_pad = RWPF_PAD_SINK;
>
> for clarity.
>
> > + list_add(&pipe->iif->list_pipe, &pipe->entities);
> > +
> > + /*
> > + * Initialize RPF0 as inputs for VSPX and use it unconditionally for
> > + * now.
> > + */
> > + pipe->inputs[0] = vsp1->rpf[0];
> > + pipe->inputs[0]->entity.pipe = pipe;
> > + pipe->inputs[0]->entity.sink = pipe->iif;
> > + vsp1_vspx_rwpf_set_subdev_fmt(vsp1, pipe->inputs[0],
> > + vspx_default_fmt.pixelformat,
> > + vspx_default_fmt.width,
> > + vspx_default_fmt.height);
> > + list_add_tail(&pipe->inputs[0]->entity.list_pipe, &pipe->entities);
>
> I think you can use list_add() for all three entities, to have the
> pipe->entities list sorted in source to sink order.
>
> > +
> > + return 0;
> > +}
> > +
> > +void vsp1_vspx_cleanup(struct vsp1_device *vsp1)
> > +{
> > + struct vsp1_vspx_pipeline *vspx_pipe;
> > +
> > + vspx_pipe = &vsp1->vspx->pipe;
>
> struct vsp1_vspx_pipeline *vspx_pipe = &vsp1->vspx->pipe;
>
> > +
> > + vsp1_vspx_release_jobs(vsp1);
> > +
> > + mutex_destroy(&vspx_pipe->jobs_lock);
>
> You should also destroy vspx_lock.
>
> > +}
> > diff --git a/drivers/media/platform/renesas/vsp1/vsp1_vspx.h b/drivers/media/platform/renesas/vsp1/vsp1_vspx.h
> > new file mode 100644
> > index 000000000000..69f73e1addba
> > --- /dev/null
> > +++ b/drivers/media/platform/renesas/vsp1/vsp1_vspx.h
> > @@ -0,0 +1,76 @@
> > +/* SPDX-License-Identifier: GPL-2.0+ */
> > +/*
> > + * vsp1_vspx.h -- R-Car Gen 4 VSPX
> > + *
> > + * Copyright (C) 2025 Ideas On Board Oy
> > + * Copyright (C) 2025 Renesas Electronics Corporation
> > + */
> > +#ifndef __VSP1_VSPX_H__
> > +#define __VSP1_VSPX_H__
> > +
> > +#include <linux/container_of.h>
> > +#include <linux/device.h>
>
> Is device.h needed ?
>
> > +#include <linux/list.h>
> > +#include <linux/mutex.h>
> > +
> > +#include <media/vsp1.h>
>
> This header doesn't seem needed either.
>
> > +
> > +#include "vsp1.h"
> > +#include "vsp1_pipe.h"
> > +
> > +/* Pixel format for ConfigDMA buffers. */
> > +#define V4L2_META_FMT_RCAR_V4H v4l2_fourcc('R', 'C', 'A', '4')
>
> V4L2 formats belong to UAPI headers. I understand the format is used
> internally only though. The commit message, or the comment above, should
> explain why this is fine.
>
> You should include the header that defines v4l2_fourcc.
>
> Thinking more about this, we could simply avoid adding this new formats,
> and instead use V4L2_PIX_FMT_XBGR32 in vsp1_vspx_job_schedule() with an
> appropriate comment.
>
> > +
> > +/*
> > + * struct vsp1_vspx_pipeline - VSPX pipeline
> > + * @pipe: the VSP1 pipeline
> > + * @partition: the pre-calculated partition used by the pipeline
> > + * @vspx_lock: protect access to the VSPX configuration
> > + * @processing: VSPX busy flag
> > + * @jobs_lock: protect the jobs queue
> > + * @jobs: jobs queue
> > + * @vspx_frame_end: frame end callback
> > + * @frame_end_data: data for the frame end callback
> > + */
> > +struct vsp1_vspx_pipeline {
> > + struct vsp1_pipeline pipe;
> > + struct vsp1_partition partition;
> > +
> > + /* Protects the pipeline configuration */
> > + struct mutex vspx_lock;
> > + bool processing;
> > +
> > + /* Protects the jobs list */
> > + struct mutex jobs_lock;
> > + struct list_head jobs;
> > +
> > + void (*vspx_frame_end)(void *frame_end_data);
> > + void *frame_end_data;
> > +};
> > +
> > +static inline struct vsp1_vspx_pipeline *
> > +to_vsp1_vspx_pipeline(struct vsp1_pipeline *pipe)
> > +{
> > + return container_of(pipe, struct vsp1_vspx_pipeline, pipe);
> > +}
> > +
> > +/*
> > + * struct vsp1_vspx - VSPX device
> > + * @vsp1: the VSP1 device
> > + * @pipe: the VSPX pipeline
> > + */
> > +struct vsp1_vspx {
> > + struct vsp1_device *vsp1;
> > + struct vsp1_vspx_pipeline pipe;
> > +};
> > +
> > +static inline struct vsp1_vspx *
> > +to_vsp1_vspx(struct vsp1_vspx_pipeline *vspx_pipe)
> > +{
> > + return container_of(vspx_pipe, struct vsp1_vspx, pipe);
> > +}
>
> I think all of the above could be moved to vsp1_vspx.c to provide better
> encapsulation and avoid exposing internals. I know vsp1_drm.h has a
> similar structure, which should probably be fixed too (not as part of
> this patch).
>
> > +
> > +int vsp1_vspx_init(struct vsp1_device *vsp1);
> > +void vsp1_vspx_cleanup(struct vsp1_device *vsp1);
> > +
> > +#endif /* __VSP1_VSPX_H__ */
> > diff --git a/include/media/vsp1.h b/include/media/vsp1.h
> > index 48f4a5023d81..668f5896503b 100644
> > --- a/include/media/vsp1.h
> > +++ b/include/media/vsp1.h
> > @@ -9,12 +9,17 @@
> > #ifndef __MEDIA_VSP1_H__
> > #define __MEDIA_VSP1_H__
> >
> > +#include <linux/list.h>
> > #include <linux/scatterlist.h>
> > #include <linux/types.h>
> > #include <linux/videodev2.h>
> >
> > struct device;
> >
> > +/*******************************************************************************
> > + * VSP1 DU interface
> > + */
> > +
> > int vsp1_du_init(struct device *dev);
> >
> > #define VSP1_DU_STATUS_COMPLETE BIT(0)
> > @@ -117,4 +122,137 @@ void vsp1_du_atomic_flush(struct device *dev, unsigned int pipe_index,
> > int vsp1_du_map_sg(struct device *dev, struct sg_table *sgt);
> > void vsp1_du_unmap_sg(struct device *dev, struct sg_table *sgt);
> >
> > +/*******************************************************************************
> > + * VSP1 ISP interface
> > + */
> > +
> > +/**
> > + * struct vsp1_isp_buffer_desc - Describe a buffer allocated by VSPX
> > + *
> > + * @cpu_addr: CPU-mapped address of a buffer allocated by VSPX
> > + * @dma_addr: bus address of a buffer allocated by VSPX
> > + */
> > +struct vsp1_isp_buffer_desc {
> > + void *cpu_addr;
> > + dma_addr_t dma_addr;
> > +};
> > +
> > +/**
> > + * struct vsp1_isp_job_desc - Describe a VSPX buffer transfer request
> > + *
> > + * Describe a transfer request for the VSPX to perform on behalf of the ISP.
> > + * The transfer job descriptor contains an optional ConfigDMA buffer and
> > + * one RAW image buffer. Set config.pairs to 0 if no ConfigDMA buffer should
> > + * be transferred.
> > + *
> > + * @config: ConfigDMA buffer
> > + * @config.pairs: number of reg-value pairs in the ConfigDMA buffer
> > + * @config.mem: bus address of the ConfigDMA buffer
> > + * @img: RAW image buffer
> > + * @img.fmt: RAW image format
> > + * @img.mem: bus address of the RAW image buffer
> > + * @job_queue: list handle
> > + */
> > +struct vsp1_isp_job_desc {
> > + struct {
> > + unsigned int pairs;
> > + dma_addr_t mem;
> > + } config;
> > + struct {
> > + struct v4l2_format fmt;
> > + dma_addr_t mem;
> > + } img;
> > + struct list_head job_queue;
> > +};
> > +
> > +/**
> > + * struct vsp1_vspx_frame_end - VSPX frame end callback data
> > + *
> > + * @vspx_frame_end: Frame end callback. Called after a transfer job has been
> > + * completed. If the job includes both a ConfigDMA and a
> > + * RAW image, the callback is called after both have been
> > + * transferred
> > + * @frame_end_data: Frame end callback data, passed to vspx_frame_end
> > + */
> > +struct vsp1_vspx_frame_end {
> > + void (*vspx_frame_end)(void *data);
> > + void *frame_end_data;
> > +};
> > +
> > +/**
> > + * vsp1_isp_init() - Initialize the VSPX
> > + *
> > + * @dev: The VSP1 struct device
> > + *
> > + * Return: %0 on success or a negative error code on failure
> > + */
>
> Functions for the DU API are documented in vsp1_drm.c, I would do the
> same for the ISP API and move the documentation to vsp1_vspx.c.
>
> > +int vsp1_isp_init(struct device *dev);
> > +
> > +/**
> > + * vsp1_isp_get_bus_master - Get VSPX bus master
> > + *
> > + * The VSPX access memory through an FCPX instance. When allocating memory
> > + * buffers that will have to be accessed by the VSPX the 'struct device' of
> > + * the FCPX should be used. Use this function to get a reference to it.
> > + *
> > + * @dev: The VSP1 struct device
> > + *
> > + * Return: a pointer to the bus master's device
> > + */
> > +struct device *vsp1_isp_get_bus_master(struct device *dev);
> > +
> > +/**
> > + * vsp1_isp_alloc_buffers - Allocate buffers in the VSPX address space
> > + *
> > + * Allocate buffers that will be later accessed by the VSPX.
> > + *
> > + * @dev: The VSP1 struct device
> > + * @size: The size of the buffer to be allocated by the VSPX
> > + * @buffer_desc: The allocated buffer description, will be filled with the
> > + * buffer CPU-mapped address and the bus address
> > + *
> > + * Return: %0 on success or a negative error code on failure
> > + */
> > +int vsp1_isp_alloc_buffers(struct device *dev, size_t size,
> > + struct vsp1_isp_buffer_desc *buffer_desc);
> > +
> > +/**
> > + * vsp1_isp_configure - Configure the VSPX with the RAW image format
> > + *
> > + * Apply to the VSPX RPF/WPF the size of the RAW image that will be transferred
> > + * to the ISP.
> > + *
> > + * @dev: The VSP1 struct device
> > + * @fmt: The RAW image format description
> > + *
> > + * Return: %0 on success or a negative error code on failure
> > + */
> > +int vsp1_isp_configure(struct device *dev,
> > + const struct v4l2_pix_format_mplane *fmt);
> > +
> > +/**
> > + * vsp1_isp_start_streaming - Start processing VSPX jobs
> > + *
> > + * Start the VSPX and prepare for accepting buffer transfer job requests.
> > + *
> > + * @dev: The VSP1 struct device
> > + * @frame_end: The frame end callback description
> > + * @enable: The enable/disable streaming flag
> > + *
> > + * Return: %0 on success or a negative error code on failure
> > + */
> > +int vsp1_isp_start_streaming(struct device *dev,
> > + struct vsp1_vspx_frame_end *frame_end,
> > + bool enable);
> > +
> > +/**
> > + * vsp1_isp_queue_job - Queue a buffer transfer job request
> > + *
> > + * @dev: The VSP1 struct device
> > + * @job: The job description
> > + *
> > + * Return: %0 on success or a negative error code on failure
> > + */
> > +int vsp1_isp_queue_job(struct device *dev, const struct vsp1_isp_job_desc *job);
> > +
> > #endif /* __MEDIA_VSP1_H__ */
> >
> > ---
> > base-commit: f2151613e040973c868d28c8b00885dfab69eb75
> > change-id: 20250228-v4h-vspx-6a537fc83806
> > prerequisite-change-id: 20250123-v4h-iif-a1dda640c95d:v5
> > prerequisite-patch-id: 700c4bf62e4d8fbd61bf2d7f0289ad6a62ff37d1
> > prerequisite-patch-id: 4af28f0fe9d553b6834c9340969dc9b1c69ceeb9
> > prerequisite-patch-id: ce124996dcd23a437d4d6e1d8e8511a2381ef7b2
> > prerequisite-patch-id: 1aef52a94a4b67ac092e1d182110cfaad573fec2
> > prerequisite-patch-id: 51b2930186407e748f27042bdc314d3ecadd5b2c
> > prerequisite-patch-id: bbdd47075847923c82dbf7a547d0cf8e6b145677
> > prerequisite-patch-id: 91ae1a6a845e756ca98e35e1902fd223f1c5ec3d
>
> --
> Regards,
>
> Laurent Pinchart
Powered by blists - more mailing lists