Linux ARM-MSM sub-architecture
 help / color / mirror / Atom feed
From: Vikash Garodia <vikash.garodia@oss.qualcomm.com>
To: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Dikshita Agarwal <dikshita.agarwal@oss.qualcomm.com>,
	Abhinav Kumar <abhinav.kumar@linux.dev>,
	Bryan O'Donoghue <bod@kernel.org>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Saravana Kannan <saravanak@kernel.org>,
	Joerg Roedel <joro@8bytes.org>, Will Deacon <will@kernel.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Stefan Schmidt <stefan.schmidt@linaro.org>,
	Hans Verkuil <hverkuil@kernel.org>,
	Krzysztof Kozlowski <krzk@kernel.org>,
	Vishnu Reddy <busanna.reddy@oss.qualcomm.com>,
	Hans Verkuil <hverkuil+cisco@kernel.org>,
	linux-arm-msm@vger.kernel.org, linux-media@vger.kernel.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	iommu@lists.linux.dev,
	Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Subject: Re: [PATCH v2 4/7] media: iris: add context bank devices using iommu-map
Date: Thu, 5 Mar 2026 18:19:52 +0530	[thread overview]
Message-ID: <6553cfcb-9399-4d17-a529-b07b421ed7e8@oss.qualcomm.com> (raw)
In-Reply-To: <vi5v5bczg2wx2adfpr6ppqcad76oecitoyc7zd2i4lahla4buw@mqnppboxcyrs>


On 3/4/2026 3:55 AM, Dmitry Baryshkov wrote:
> On Wed, Mar 04, 2026 at 12:16:50AM +0530, Vikash Garodia wrote:
>>
>> On 2/28/2026 1:50 AM, Dmitry Baryshkov wrote:
>>> On Fri, Feb 27, 2026 at 07:41:20PM +0530, Vikash Garodia wrote:
>>>> Introduce different context banks(CB) and the associated buffer region.
>>>> Different stream IDs from VPU would be associated to one of these CB.
>>>> Multiple CBs are needed to increase the IOVA for the video usecases like
>>>> higher concurrent sessions.
>>>>
>>>> Co-developed-by: Vishnu Reddy <busanna.reddy@oss.qualcomm.com>
>>>> Signed-off-by: Vishnu Reddy <busanna.reddy@oss.qualcomm.com>
>>>> Signed-off-by: Vikash Garodia <vikash.garodia@oss.qualcomm.com>
>>>> ---
>>>>    .../platform/qcom/iris/iris_platform_common.h      | 18 +++++++
>>>>    drivers/media/platform/qcom/iris/iris_probe.c      | 60 ++++++++++++++++++++--
>>>>    drivers/media/platform/qcom/iris/iris_resources.c  | 36 +++++++++++++
>>>>    drivers/media/platform/qcom/iris/iris_resources.h  |  1 +
>>>>    4 files changed, 111 insertions(+), 4 deletions(-)
>>>>
>>>> diff --git a/drivers/media/platform/qcom/iris/iris_platform_common.h b/drivers/media/platform/qcom/iris/iris_platform_common.h
>>>> index 5a489917580eb10022fdcb52f7321a915e8b239d..03c50d6e54853fca34d7d32f65d09eb80945fcdd 100644
>>>> --- a/drivers/media/platform/qcom/iris/iris_platform_common.h
>>>> +++ b/drivers/media/platform/qcom/iris/iris_platform_common.h
>>>> @@ -204,6 +204,22 @@ struct icc_vote_data {
>>>>    	u32 fps;
>>>>    };
>>>> +enum iris_buffer_region {
>>>> +	IRIS_BITSTREAM_REGION		= BIT(0),
>>>> +	IRIS_NON_PIXEL_REGION		= BIT(1),
>>>> +	IRIS_PIXEL_REGION		= BIT(2),
>>>> +	IRIS_SECURE_BITSTREAM_REGION	= BIT(3),
>>>> +	IRIS_SECURE_NON_PIXEL_REGION	= BIT(4),
>>>> +	IRIS_SECURE_PIXEL_REGION	= BIT(5),
>>>
>>> Can a context bank belong to multiple regions at the same time?
>>
>> yes, they would.
> 
> How? Each set of CBs is defined by a separate function in the DT. How
> can CB belong to multiple regions? Could you please provide an example?

SM8550 would have same stream id for VPU hardwares (tensilica and 
vcodec) accessing bitstream and non pixel regions. Thereby non_pixel and 
bitstream regions would map to one CB.
While kaanapali would have different stream id for tensilica accessing 
non pixel region and vcodec accessing bitstream region, thereby having 
different CB.

> 
>>
>>>
>>>> +};
>>>> +
>>>> +struct iris_context_bank {
>>>> +	struct device *dev;
>>>
>>> Separate data and the actual device. Define a wrapper around struct
>>> device for the actual runtime usage.
>>
>> we still have to store the list of dynamically created device. Name can be
>> used to fetch the device from the list, i think the existing approach is
>> simpler ?
> 
> You don't need a list. You have an array of the size, which is known and
> fixed. You have at most 9 functions, which means less than 9 devices.
> 

as mentioned above, its not the same for all platforms to have one to 
one mapping between CBs and buffer region. Thereby indexing based on 
array would be an issue here
It would end up something like this, considering [dev region] array,

SM8550
non_pixel_device  non_pixel_region
non_pixel_device  bitstream_region
pixel_device      pixel_region

kaanapali
non_pixel_device  non_pixel_region
bitstream_device  bitstream_region
pixel_device      pixel_region


>>
>>>
>>>> +	const char *name;
>>>> +	const u32 f_id;
>>>> +	const enum iris_buffer_region region;
>>>> +};
>>>> +
>>>>    enum platform_pm_domain_type {
>>>>    	IRIS_CTRL_POWER_DOMAIN,
>>>>    	IRIS_HW_POWER_DOMAIN,
>>>> @@ -246,6 +262,8 @@ struct iris_platform_data {
>>>>    	u32 inst_fw_caps_enc_size;
>>>>    	const struct tz_cp_config *tz_cp_config_data;
>>>>    	u32 tz_cp_config_data_size;
>>>> +	struct iris_context_bank *cb_data;
>>>> +	u32 cb_data_size;
>>>
>>> Do they differ from platform to platform?
>> Yes
>>
>>> Mark them as const, it should be data only.
>>
>> cb_data_size can be marked as const
> 
> Why is cb_data non-const?

dev is being updated once created dynamically.

> 
>>
>>>
>>>>    	u32 core_arch;
>>>>    	u32 hw_response_timeout;
>>>>    	struct ubwc_config_data *ubwc_config;
>>>> diff --git a/drivers/media/platform/qcom/iris/iris_probe.c b/drivers/media/platform/qcom/iris/iris_probe.c
>>>> index ddaacda523ecb9990af0dd0640196223fbcc2cab..557adb038328a75510591d91569819abc0b7b1c9 100644
>>>> --- a/drivers/media/platform/qcom/iris/iris_probe.c
>>>> +++ b/drivers/media/platform/qcom/iris/iris_probe.c
>>>> @@ -123,6 +123,49 @@ static int iris_init_resets(struct iris_core *core)
>>>>    				     core->iris_platform_data->controller_rst_tbl_size);
>>>>    }
>>>> +static void iris_destroy_child_device(struct iris_context_bank *cb)
>>>> +{
>>>> +	struct device *dev = cb->dev;
>>>> +
>>>> +	if (dev)
>>>> +		device_unregister(dev);
>>>> +
>>>> +	cb->dev = NULL;
>>>> +}
>>>> +
>>>> +static void iris_deinit_context_bank_devices(struct iris_core *core)
>>>> +{
>>>> +	struct iris_context_bank *cb;
>>>> +	int i;
>>>> +
>>>> +	for (i = 0; i < core->iris_platform_data->cb_data_size; i++) {
>>>> +		cb = &core->iris_platform_data->cb_data[i];
>>>> +		iris_destroy_child_device(cb);
>>>> +	}
>>>> +}
>>>> +
>>>> +static int iris_init_context_bank_devices(struct iris_core *core)
>>>> +{
>>>> +	struct iris_context_bank *cb;
>>>> +	int ret, i;
>>>> +
>>>> +	for (i = 0; i < core->iris_platform_data->cb_data_size; i++) {
>>>> +		cb = &core->iris_platform_data->cb_data[i];
>>>> +
>>>> +		ret = iris_create_child_device_and_map(core, cb);
>>>> +		if (ret)
>>>> +			goto err_deinit_cb;
>>>> +	}
>>>> +
>>>> +	return 0;
>>>> +
>>>> +err_deinit_cb:
>>>> +	while (i-- > 0)
>>>> +		iris_destroy_child_device(&core->iris_platform_data->cb_data[i]);
>>>> +
>>>> +	return ret;
>>>> +}
>>>> +
>>>>    static int iris_init_resources(struct iris_core *core)
>>>>    {
>>>>    	int ret;
>>>> @@ -193,6 +236,7 @@ static void iris_remove(struct platform_device *pdev)
>>>>    		return;
>>>>    	iris_core_deinit(core);
>>>> +	iris_deinit_context_bank_devices(core);
>>>>    	video_unregister_device(core->vdev_dec);
>>>>    	video_unregister_device(core->vdev_enc);
>>>> @@ -275,12 +319,18 @@ static int iris_probe(struct platform_device *pdev)
>>>>    	platform_set_drvdata(pdev, core);
>>>> -	dma_mask = core->iris_platform_data->dma_mask;
>>>> -
>>>> -	ret = dma_set_mask_and_coherent(dev, dma_mask);
>>>> +	ret = iris_init_context_bank_devices(core);
>>>>    	if (ret)
>>>>    		goto err_vdev_unreg_enc;
>>>> +	dma_mask = core->iris_platform_data->dma_mask;
>>>> +
>>>> +	if (device_iommu_mapped(core->dev)) {
>>>> +		ret = dma_set_mask_and_coherent(core->dev, dma_mask);
>>>> +		if (ret)
>>>> +			goto err_deinit_cb;
>>>> +	}
>>>> +
>>>>    	dma_set_max_seg_size(&pdev->dev, DMA_BIT_MASK(32));
>>>>    	dma_set_seg_boundary(&pdev->dev, DMA_BIT_MASK(32));
>>>> @@ -288,10 +338,12 @@ static int iris_probe(struct platform_device *pdev)
>>>>    	pm_runtime_use_autosuspend(core->dev);
>>>>    	ret = devm_pm_runtime_enable(core->dev);
>>>>    	if (ret)
>>>> -		goto err_vdev_unreg_enc;
>>>> +		goto err_deinit_cb;
>>>>    	return 0;
>>>> +err_deinit_cb:
>>>> +	iris_deinit_context_bank_devices(core);
>>>>    err_vdev_unreg_enc:
>>>>    	video_unregister_device(core->vdev_enc);
>>>>    err_vdev_unreg_dec:
>>>> diff --git a/drivers/media/platform/qcom/iris/iris_resources.c b/drivers/media/platform/qcom/iris/iris_resources.c
>>>> index 773f6548370a257b8ae7332242544266cbbd61a9..be58e8620086d0f82c2c2bda29247483f5c56d79 100644
>>>> --- a/drivers/media/platform/qcom/iris/iris_resources.c
>>>> +++ b/drivers/media/platform/qcom/iris/iris_resources.c
>>>> @@ -6,6 +6,7 @@
>>>>    #include <linux/clk.h>
>>>>    #include <linux/devfreq.h>
>>>>    #include <linux/interconnect.h>
>>>> +#include <linux/iris_vpu_bus.h>
>>>>    #include <linux/pm_domain.h>
>>>>    #include <linux/pm_opp.h>
>>>>    #include <linux/pm_runtime.h>
>>>> @@ -141,3 +142,38 @@ int iris_disable_unprepare_clock(struct iris_core *core, enum platform_clk_type
>>>>    	return 0;
>>>>    }
>>>> +
>>>> +static void iris_device_release(struct device *dev)
>>>> +{
>>>> +	dev_set_drvdata(dev, NULL);
>>>> +	kfree(dev);
>>>> +}
>>>> +
>>>> +int iris_create_child_device_and_map(struct iris_core *core, struct iris_context_bank *cb)
>>>> +{
>>>> +	struct device *dev;
>>>> +	int ret;
>>>> +
>>>> +	dev = kzalloc_obj(*dev);
>>>> +	if (!dev)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	dev->release = iris_device_release;
>>>> +	dev->bus = &iris_vpu_bus_type;
>>>> +	dev->parent = core->dev;
>>>> +	dev->coherent_dma_mask = core->iris_platform_data->dma_mask;
>>>> +	dev->dma_mask = &dev->coherent_dma_mask;
>>>> +
>>>> +	dev_set_name(dev, "%s", cb->name);
>>>> +	dev_set_drvdata(dev, cb);
>>>> +
>>>> +	ret = device_register(dev);
>>>> +	if (ret) {
>>>> +		put_device(dev);
>>>> +		return ret;
>>>> +	}
>>>> +
>>>> +	cb->dev = dev;
>>>> +
>>>> +	return 0;
>>>> +}
>>>> diff --git a/drivers/media/platform/qcom/iris/iris_resources.h b/drivers/media/platform/qcom/iris/iris_resources.h
>>>> index 6bfbd2dc6db095ec05e53c894e048285f82446c6..b7efe15facb203eea9ae13d5f0abdcc2ea718b4d 100644
>>>> --- a/drivers/media/platform/qcom/iris/iris_resources.h
>>>> +++ b/drivers/media/platform/qcom/iris/iris_resources.h
>>>> @@ -15,5 +15,6 @@ int iris_unset_icc_bw(struct iris_core *core);
>>>>    int iris_set_icc_bw(struct iris_core *core, unsigned long icc_bw);
>>>>    int iris_disable_unprepare_clock(struct iris_core *core, enum platform_clk_type clk_type);
>>>>    int iris_prepare_enable_clock(struct iris_core *core, enum platform_clk_type clk_type);
>>>> +int iris_create_child_device_and_map(struct iris_core *core, struct iris_context_bank *cb);
>>>>    #endif
>>>>
>>>> -- 
>>>> 2.34.1
>>>>
>>>
>>
> 


  reply	other threads:[~2026-03-05 12:50 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-27 14:11 [PATCH v2 0/7] media: iris: add support for kaanapali platform Vikash Garodia
2026-02-27 14:11 ` [PATCH v2 1/7] media: dt-bindings: qcom-kaanapali-iris: Add kaanapali video codec binding Vikash Garodia
2026-02-27 15:42   ` Rob Herring (Arm)
2026-02-27 15:50   ` Krzysztof Kozlowski
2026-02-27 15:55     ` Krzysztof Kozlowski
2026-03-03 18:10     ` Vikash Garodia
2026-03-04 15:07       ` Krzysztof Kozlowski
2026-03-05 12:23         ` Vikash Garodia
2026-02-27 19:52   ` Dmitry Baryshkov
2026-03-03 18:13     ` Vikash Garodia
2026-03-03 23:53       ` Dmitry Baryshkov
2026-03-04 15:02         ` Vikash Garodia
2026-03-05  3:33           ` Dmitry Baryshkov
2026-02-27 14:11 ` [PATCH v2 2/7] media: iris: switch to hardware mode after firmware boot Vikash Garodia
2026-02-27 16:49   ` Konrad Dybcio
2026-03-03 18:15     ` Vikash Garodia
2026-03-04  8:57       ` Konrad Dybcio
2026-02-27 14:11 ` [PATCH v2 3/7] media: iris: add iris vpu bus support and register it with iommu_buses Vikash Garodia
2026-02-27 15:49   ` Krzysztof Kozlowski
2026-03-03 18:26     ` Vikash Garodia
2026-02-27 20:14   ` Dmitry Baryshkov
2026-02-27 14:11 ` [PATCH v2 4/7] media: iris: add context bank devices using iommu-map Vikash Garodia
2026-02-27 20:20   ` Dmitry Baryshkov
2026-03-03 18:46     ` Vikash Garodia
2026-03-03 22:25       ` Dmitry Baryshkov
2026-03-05 12:49         ` Vikash Garodia [this message]
2026-03-05 14:21           ` Dmitry Baryshkov
2026-03-05 17:26             ` Vikash Garodia
2026-03-05 18:32               ` Dmitry Baryshkov
2026-03-05 21:34                 ` Vikash Garodia
2026-03-13 19:26                   ` Dmitry Baryshkov
2026-02-27 14:11 ` [PATCH v2 5/7] media: iris: add helper to select context bank device Vikash Garodia
2026-02-27 20:27   ` Dmitry Baryshkov
2026-03-03 19:16     ` Vikash Garodia
2026-03-03 22:27       ` Dmitry Baryshkov
2026-03-04 15:29         ` Vikash Garodia
2026-03-05  3:40           ` Dmitry Baryshkov
2026-02-27 14:11 ` [PATCH v2 6/7] media: iris: add iris4 specific H265 line buffer calculation Vikash Garodia
2026-02-27 20:28   ` Dmitry Baryshkov
2026-03-03 18:48     ` Vikash Garodia
2026-03-03 22:28       ` Dmitry Baryshkov
2026-03-04 15:40         ` Vikash Garodia
2026-03-05  3:41           ` Dmitry Baryshkov
2026-02-27 14:11 ` [PATCH v2 7/7] media: iris: add platform data for kaanapali Vikash Garodia
2026-02-27 15:48   ` Krzysztof Kozlowski
2026-02-28  7:31     ` Dmitry Baryshkov
2026-02-27 15:42 ` [PATCH v2 0/7] media: iris: add support for kaanapali platform Bryan O'Donoghue

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=6553cfcb-9399-4d17-a529-b07b421ed7e8@oss.qualcomm.com \
    --to=vikash.garodia@oss.qualcomm.com \
    --cc=abhinav.kumar@linux.dev \
    --cc=bod@kernel.org \
    --cc=bryan.odonoghue@linaro.org \
    --cc=busanna.reddy@oss.qualcomm.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=dikshita.agarwal@oss.qualcomm.com \
    --cc=dmitry.baryshkov@oss.qualcomm.com \
    --cc=hverkuil+cisco@kernel.org \
    --cc=hverkuil@kernel.org \
    --cc=iommu@lists.linux.dev \
    --cc=joro@8bytes.org \
    --cc=krzk+dt@kernel.org \
    --cc=krzk@kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=mchehab@kernel.org \
    --cc=robh@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=saravanak@kernel.org \
    --cc=stefan.schmidt@linaro.org \
    --cc=will@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