public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: Nicolas Dufresne <nicolas.dufresne@collabora.com>
To: Detlev Casanova <detlev.casanova@collabora.com>,
	Mauro Carvalho Chehab	 <mchehab@kernel.org>,
	Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>,
	Heiko Stuebner <heiko@sntech.de>,
	Hans Verkuil <hverkuil+cisco@kernel.org>,
	Jonas Karlman <jonas@kwiboo.se>
Cc: kernel@collabora.com, linux-media@vger.kernel.org,
	 linux-kernel@vger.kernel.org,
	linux-rockchip@lists.infradead.org,
	 linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 7/7] media: rkvdec: Add multicore IOMMU support
Date: Thu, 09 Apr 2026 10:19:10 -0400	[thread overview]
Message-ID: <517dc6e9de0e284b3dc22952d9479616e6c8ec62.camel@collabora.com> (raw)
In-Reply-To: <20260409-rkvdec-multicore-v1-7-62b316abf0f7@collabora.com>

[-- Attachment #1: Type: text/plain, Size: 10653 bytes --]

Le jeudi 09 avril 2026 à 09:50 -0400, Detlev Casanova a écrit :
> As each core has its own IOMMU core, buffers must be mapped in each
> core's IOMMU so that any run() call can use any core without having to
> remap everything.
> 
> To do that, we use rockchip iommu domain's iommu devices list.
> With that, one IOMMU domain can be mapped on multiple devices, meaning
> that each call to iommu_map() will flush the new mapping on all devices
> in the list.
> 
> The IOMMU domain that will have all devices in its list is the first
> core's default domain.
> 
> Another domain cannot be used because VB2 allocates buffers through the
> DMA engine, which uses iommu_get_dma_domain() to find the domain to map
> buffers through.
> 
> The IOMMU restore function can still work as before, but needs to be more
> explicit in what domain to attach the device to.
> That is because detaching the empty domain will reattach the core's default
> domain, which is wrong (except for the first "main" core).
> 
> The RCB temporary buffers are allocated in a dedicated SRAM, each
> core has its own SRAM, so the mapping for each core's SRAM is added in the
> global domain.
> 
> Everything else is mapped through the first core's default domain, making
> the driver write the mappings on both IOMMU cores.

Just raising an issue with the patch ordering here. I'm worried in a git bisect,
the driver will be broken until we apply this last patch. Can we make sure that
the driver bisects ? (or tell me if I'm wrong)

Nicolas

> 
> Signed-off-by: Detlev Casanova <detlev.casanova@collabora.com>
> ---
>  .../media/platform/rockchip/rkvdec/rkvdec-rcb.c    | 21 ++++++-------
>  .../media/platform/rockchip/rkvdec/rkvdec-rcb.h    |  6 ++--
>  drivers/media/platform/rockchip/rkvdec/rkvdec.c    | 35 +++++++++++++++++-----
>  drivers/media/platform/rockchip/rkvdec/rkvdec.h    |  2 +-
>  4 files changed, 44 insertions(+), 20 deletions(-)
> 
> diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.c b/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.c
> index 190fb7438e8c..977e37cf209b 100644
> --- a/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.c
> +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.c
> @@ -57,7 +57,7 @@ bool rkvdec_rcb_buf_validate_size(struct rkvdec_ctx *ctx)
>  	return ret;
>  }
>  
> -void rkvdec_free_rcb(struct rkvdec_core *core)
> +void rkvdec_free_rcb(struct rkvdec_dev *rkvdec, struct rkvdec_core *core)
>  {
>  	struct rkvdec_rcb_config *cfg = core->rcb_config;
>  	unsigned long virt_addr;
> @@ -76,12 +76,12 @@ void rkvdec_free_rcb(struct rkvdec_core *core)
>  		case RKVDEC_ALLOC_SRAM:
>  			virt_addr = (unsigned long)cfg->rcb_bufs[i].cpu;
>  
> -			if (core->iommu_domain)
> -				iommu_unmap(core->iommu_domain, virt_addr, rcb_size);
> +			if (rkvdec->iommu_global_domain)
> +				iommu_unmap(rkvdec->iommu_global_domain, virt_addr, rcb_size);
>  			gen_pool_free(core->sram_pool, virt_addr, rcb_size);
>  			break;
>  		case RKVDEC_ALLOC_DMA:
> -			dma_free_coherent(core->dev,
> +			dma_free_coherent(rkvdec->main_core->dev,
>  					  rcb_size,
>  					  cfg->rcb_bufs[i].cpu,
>  					  cfg->rcb_bufs[i].dma);
> @@ -97,7 +97,8 @@ void rkvdec_free_rcb(struct rkvdec_core *core)
>  	core->rcb_config = NULL;
>  }
>  
> -int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
> +int rkvdec_allocate_rcb(struct rkvdec_dev *rkvdec, struct rkvdec_core *core,
> +			u32 width, u32 height,
>  			const struct rcb_size_info *size_info,
>  			size_t rcb_count)
>  {
> @@ -132,7 +133,7 @@ int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
>  
>  		/* Try allocating an SRAM buffer */
>  		if (core->sram_pool) {
> -			if (core->iommu_domain)
> +			if (rkvdec->iommu_global_domain)
>  				rcb_size = ALIGN(rcb_size, SZ_4K);
>  
>  			cpu = gen_pool_dma_zalloc_align(core->sram_pool,
> @@ -142,11 +143,11 @@ int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
>  		}
>  
>  		/* If an IOMMU is used, map the SRAM address through it */
> -		if (cpu && core->iommu_domain) {
> +		if (cpu && rkvdec->iommu_global_domain) {
>  			unsigned long virt_addr = (unsigned long)cpu;
>  			phys_addr_t phys_addr = dma;
>  
> -			ret = iommu_map(core->iommu_domain, virt_addr, phys_addr,
> +			ret = iommu_map(rkvdec->iommu_global_domain, virt_addr, phys_addr,
>  					rcb_size, IOMMU_READ | IOMMU_WRITE, 0);
>  			if (ret) {
>  				gen_pool_free(core->sram_pool,
> @@ -166,7 +167,7 @@ int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
>  ram_fallback:
>  		/* Fallback to RAM */
>  		if (!cpu) {
> -			cpu = dma_alloc_coherent(core->dev,
> +			cpu = dma_alloc_coherent(rkvdec->main_core->dev,
>  						 rcb_size,
>  						 &dma,
>  						 GFP_KERNEL);
> @@ -189,7 +190,7 @@ int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
>  	return 0;
>  
>  err_alloc:
> -	rkvdec_free_rcb(core);
> +	rkvdec_free_rcb(rkvdec, core);
>  
>  	return ret;
>  }
> diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.h b/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.h
> index a12af9b7dc2b..d1149afe7fda 100644
> --- a/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.h
> +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec-rcb.h
> @@ -8,6 +8,7 @@
>  
>  #include <linux/types.h>
>  
> +struct rkvdec_dev;
>  struct rkvdec_ctx;
>  struct rkvdec_core;
>  
> @@ -21,11 +22,12 @@ struct rcb_size_info {
>  	enum rcb_axis axis;
>  };
>  
> -int rkvdec_allocate_rcb(struct rkvdec_core *core, u32 width, u32 height,
> +int rkvdec_allocate_rcb(struct rkvdec_dev *rkvdec, struct rkvdec_core *core,
> +			u32 width, u32 height,
>  			const struct rcb_size_info *size_info,
>  			size_t rcb_count);
>  dma_addr_t rkvdec_rcb_buf_dma_addr(struct rkvdec_ctx *ctx, int id);
>  size_t rkvdec_rcb_buf_size(struct rkvdec_ctx *ctx, int id);
>  int rkvdec_rcb_buf_count(struct rkvdec_ctx *ctx);
>  bool rkvdec_rcb_buf_validate_size(struct rkvdec_ctx *ctx);
> -void rkvdec_free_rcb(struct rkvdec_core *core);
> +void rkvdec_free_rcb(struct rkvdec_dev *rkvdec, struct rkvdec_core *core);
> diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec.c b/drivers/media/platform/rockchip/rkvdec/rkvdec.c
> index c2818f1575ef..2930e9b64906 100644
> --- a/drivers/media/platform/rockchip/rkvdec/rkvdec.c
> +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec.c
> @@ -1204,9 +1204,9 @@ static void rkvdec_device_run(void *priv)
>  	}
>  
>  	if (!rkvdec_rcb_buf_validate_size(ctx)) {
> -		rkvdec_free_rcb(ctx->core);
> +		rkvdec_free_rcb(ctx->dev, ctx->core);
>  
> -		ret = rkvdec_allocate_rcb(ctx->core,
> +		ret = rkvdec_allocate_rcb(ctx->dev, ctx->core,
>  					  ctx->decoded_fmt.fmt.pix_mp.width,
>  					  ctx->decoded_fmt.fmt.pix_mp.height,
>  					  ctx->dev->variant->rcb_sizes,
> @@ -1486,6 +1486,7 @@ static void rkvdec_v4l2_cleanup(struct rkvdec_dev *rkvdec)
>  
>  static void rkvdec_iommu_restore(struct rkvdec_core *core)
>  {
> +	int ret;
>  	if (core->empty_domain) {
>  		/*
>  		 * To rewrite mapping into the attached IOMMU core, attach a new empty domain that
> @@ -1494,8 +1495,14 @@ static void rkvdec_iommu_restore(struct rkvdec_core *core)
>  		 * This is safely done in this interrupt handler to make sure no memory get mapped
>  		 * through the IOMMU while the empty domain is attached.
>  		 */
> -		iommu_attach_device(core->empty_domain, core->dev);
> +		iommu_detach_device(core->curr_ctx->dev->iommu_global_domain, core->dev);
> +		ret = iommu_attach_device(core->empty_domain, core->dev);
> +		if (ret)
> +			dev_warn(core->dev, "Cannot attach empty domain: %d\n", ret);
>  		iommu_detach_device(core->empty_domain, core->dev);
> +		ret = iommu_attach_device(core->curr_ctx->dev->iommu_global_domain, core->dev);
> +		if (ret)
> +			dev_warn(core->dev, "Cannot attach global domain: %d\n", ret);
>  	}
>  }
>  
> @@ -1858,6 +1865,8 @@ static int rkvdec_probe(struct platform_device *pdev)
>  
>  	core = &rkvdec->cores[rkvdec->core_count++];
>  
> +	core->id = rkvdec->core_count - 1;
> +
>  	platform_set_drvdata(pdev, rkvdec);
>  	core->dev = &pdev->dev;
>  	INIT_DELAYED_WORK(&core->watchdog_work, rkvdec_watchdog_func);
> @@ -1883,12 +1892,24 @@ static int rkvdec_probe(struct platform_device *pdev)
>  			return PTR_ERR(core->link);
>  	}
>  
> -	core->iommu_domain = iommu_get_domain_for_dev(&pdev->dev);
> -	if (core->iommu_domain) {
> +	if (iommu_get_domain_for_dev(&pdev->dev)) {
>  		core->empty_domain = iommu_paging_domain_alloc(core->dev);
>  
> -		if (!core->empty_domain)
> +		if (IS_ERR(core->empty_domain))
>  			dev_warn(core->dev, "cannot alloc new empty domain\n");
> +
> +		if (!rkvdec->iommu_global_domain) {
> +			rkvdec->iommu_global_domain = iommu_get_domain_for_dev(core->dev);
> +
> +			if (IS_ERR(rkvdec->iommu_global_domain)) {
> +				rkvdec->iommu_global_domain = NULL;
> +				dev_warn_once(core->dev, "cannot alloc new global domain\n");
> +			}
> +		}
> +
> +		ret = iommu_attach_device(rkvdec->iommu_global_domain, core->dev);
> +		if (ret)
> +			dev_warn(core->dev, "cannot attach global domain to core %d\n", core->id);
>  	}
>  
>  	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
> @@ -1961,7 +1982,7 @@ static void rkvdec_remove(struct platform_device *pdev)
>  		if (rkvdec->cores[i].empty_domain)
>  			iommu_domain_free(rkvdec->cores[i].empty_domain);
>  
> -		rkvdec_free_rcb(&rkvdec->cores[i]);
> +		rkvdec_free_rcb(rkvdec, &rkvdec->cores[i]);
>  	}
>  }
>  
> diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec.h b/drivers/media/platform/rockchip/rkvdec/rkvdec.h
> index 4f042a367dc0..ccd766b220c7 100644
> --- a/drivers/media/platform/rockchip/rkvdec/rkvdec.h
> +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec.h
> @@ -135,7 +135,6 @@ struct rkvdec_core {
>  	void __iomem *link;
>  	struct delayed_work watchdog_work;
>  	struct gen_pool *sram_pool;
> -	struct iommu_domain *iommu_domain;
>  	struct iommu_domain *empty_domain;
>  	struct rkvdec_rcb_config *rcb_config;
>  	struct rkvdec_ctx *curr_ctx;
> @@ -155,6 +154,7 @@ struct rkvdec_dev {
>  	unsigned int available_core_count;
>  	spinlock_t cores_lock; /* serializes core list access */
>  	struct rkvdec_core *main_core;
> +	struct iommu_domain *iommu_global_domain;
>  };
>  
>  struct rkvdec_ctx {

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

  reply	other threads:[~2026-04-09 14:19 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-09 13:50 [PATCH 0/7] media: rkvdec: Enable multi-core support Detlev Casanova
2026-04-09 13:50 ` [PATCH 1/7] media: v4l2-mem2mem: Add v4l2_m2m_buf_done_manual() Detlev Casanova
2026-04-09 13:50 ` [PATCH 2/7] media: v4l2-mem2mem: Remove WARN_ON() in v4l2_m2m_job_finish() Detlev Casanova
2026-04-09 13:50 ` [PATCH 3/7] media: rkvdec: Keep RCB to the correct size Detlev Casanova
2026-04-09 13:50 ` [PATCH 4/7] media: rkvdec: Remove unused need_reset Detlev Casanova
2026-04-09 13:50 ` [PATCH 5/7] media: rkvdec: Add multicore support Detlev Casanova
2026-04-09 13:50 ` [PATCH 6/7] media: rkvdec: Wait for all buffers before stop_streaming Detlev Casanova
2026-04-09 13:50 ` [PATCH 7/7] media: rkvdec: Add multicore IOMMU support Detlev Casanova
2026-04-09 14:19   ` Nicolas Dufresne [this message]
2026-04-09 14:49     ` Detlev Casanova
2026-04-09 15:24       ` Nicolas Dufresne

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=517dc6e9de0e284b3dc22952d9479616e6c8ec62.camel@collabora.com \
    --to=nicolas.dufresne@collabora.com \
    --cc=detlev.casanova@collabora.com \
    --cc=ezequiel@vanguardiasur.com.ar \
    --cc=heiko@sntech.de \
    --cc=hverkuil+cisco@kernel.org \
    --cc=jonas@kwiboo.se \
    --cc=kernel@collabora.com \
    --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 \
    /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