linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephan Gerhold <stephan.gerhold@linaro.org>
To: Bjorn Andersson <andersson@kernel.org>,
	Konrad Dybcio <konradybcio@kernel.org>,
	Vikash Garodia <quic_vgarodia@quicinc.com>,
	Dikshita Agarwal <quic_dikshita@quicinc.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Mathieu Poirier <mathieu.poirier@linaro.org>,
	Abhinav Kumar <abhinav.kumar@linux.dev>,
	Bryan O'Donoghue <bryan.odonoghue@linaro.org>,
	linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
	linux-media@vger.kernel.org, linux-remoteproc@vger.kernel.org
Subject: Re: [PATCH v2 11/11] media: iris: Enable Secure PAS support with IOMMU managed by Linux
Date: Wed, 20 Aug 2025 15:39:14 +0200	[thread overview]
Message-ID: <aKXQAoXZyR6SRPAA@linaro.org> (raw)
In-Reply-To: <20250820115659.kkngraove46wemxv@hu-mojha-hyd.qualcomm.com>

On Wed, Aug 20, 2025 at 05:26:59PM +0530, Mukesh Ojha wrote:
> On Wed, Aug 20, 2025 at 10:46:31AM +0200, Stephan Gerhold wrote:
> > On Tue, Aug 19, 2025 at 10:24:46PM +0530, Mukesh Ojha wrote:
> > > Most Qualcomm platforms feature a proprietary hypervisor (such as Gunyah
> > > or QHEE), which typically handles IOMMU configuration. This includes
> > > mapping memory regions and device memory resources for remote processors
> > > by intercepting qcom_scm_pas_auth_and_reset() calls. These mappings are
> > > later removed during teardown. Additionally, SHM bridge setup is required
> > > to enable memory protection for both remoteproc metadata and its memory
> > > regions.
> > > 
> > > When the hypervisor is absent, the operating system must perform these
> > > configurations instead.
> > > 
> > > Support for handling IOMMU and SHM setup in the absence of a hypervisor
> > > is now in place. Extend the Iris driver to enable this functionality on
> > > platforms where IOMMU is managed by Linux (i.e., non-Gunyah, non-QHEE).
> > > 
> > > Additionally, the Iris driver must map the firmware and its required
> > > resources to the firmware SID, which is now specified via the device tree.
> > > 
> > > Co-developed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
> > > Signed-off-by: Vikash Garodia <quic_vgarodia@quicinc.com>
> > > Signed-off-by: Mukesh Ojha <mukesh.ojha@oss.qualcomm.com>
> > > ---
> > >  drivers/media/platform/qcom/iris/iris_core.c  |   9 +-
> > >  drivers/media/platform/qcom/iris/iris_core.h  |   6 +
> > >  .../media/platform/qcom/iris/iris_firmware.c  | 156 ++++++++++++++++--
> > >  .../media/platform/qcom/iris/iris_firmware.h  |   2 +
> > >  4 files changed, 155 insertions(+), 18 deletions(-)
> > > 
> > > [...]
> > > diff --git a/drivers/media/platform/qcom/iris/iris_firmware.c b/drivers/media/platform/qcom/iris/iris_firmware.c
> > > index f1b5cd56db32..e3f2fe5c9d7a 100644
> > > --- a/drivers/media/platform/qcom/iris/iris_firmware.c
> > > +++ b/drivers/media/platform/qcom/iris/iris_firmware.c
> > > @@ -3,10 +3,18 @@
> > >   * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved.
> > >   */
> > >  
> > > +#include <linux/device.h>
> > >  #include <linux/firmware.h>
> > > -#include <linux/firmware/qcom/qcom_scm.h>
> > > +#include <linux/kernel.h>
> > > +#include <linux/iommu.h>
> > > +#include <linux/io.h>
> > > +#include <linux/of.h>
> > >  #include <linux/of_address.h>
> > >  #include <linux/of_reserved_mem.h>
> > > +#include <linux/platform_device.h>
> > > +#include <linux/of_device.h>
> > > +#include <linux/firmware/qcom/qcom_scm.h>
> > > +#include <linux/sizes.h>
> > >  #include <linux/soc/qcom/mdt_loader.h>
> > >  
> > >  #include "iris_core.h"
> > > @@ -17,15 +25,14 @@
> > >  static int iris_load_fw_to_memory(struct iris_core *core, const char *fw_name)
> > >  {
> > >  	u32 pas_id = core->iris_platform_data->pas_id;
> > > +	struct qcom_scm_pas_ctx *ctx;
> > >  	const struct firmware *firmware = NULL;
> > >  	struct device *dev = core->dev;
> > > -	struct reserved_mem *rmem;
> > > -	struct device_node *node;
> > > -	phys_addr_t mem_phys;
> > > -	size_t res_size;
> > > -	ssize_t fw_size;
> > > -	void *mem_virt;
> > > -	int ret;
> > > +	struct reserved_mem *rmem = NULL;
> > > +	struct device_node *node = NULL;
> > > +	ssize_t fw_size = 0;
> > > +	void *mem_virt = NULL;
> > > +	int ret = 0;
> > >  
> > >  	if (strlen(fw_name) >= MAX_FIRMWARE_NAME_SIZE - 4)
> > >  		return -EINVAL;
> > > @@ -39,36 +46,64 @@ static int iris_load_fw_to_memory(struct iris_core *core, const char *fw_name)
> > >  	if (!rmem)
> > >  		return -EINVAL;
> > >  
> > > -	mem_phys = rmem->base;
> > > -	res_size = rmem->size;
> > > +	if (core->has_iommu)
> > > +		dev = core->fw.dev;
> > >  
> > > +	ctx = qcom_scm_pas_ctx_init(dev, pas_id, rmem->base, rmem->size, false);
> > > +	if (!ctx)
> > > +		return -ENOMEM;
> > > +
> > > +	ctx->has_iommu = core->has_iommu;
> > >  	ret = request_firmware(&firmware, fw_name, dev);
> > >  	if (ret)
> > >  		return ret;
> > >  
> > >  	fw_size = qcom_mdt_get_size(firmware);
> > > -	if (fw_size < 0 || res_size < (size_t)fw_size) {
> > > +	if (fw_size < 0 || rmem->size < (size_t)fw_size) {
> > >  		ret = -EINVAL;
> > >  		goto err_release_fw;
> > >  	}
> > >  
> > > -	mem_virt = memremap(mem_phys, res_size, MEMREMAP_WC);
> > > +	mem_virt = memremap(rmem->base, rmem->size, MEMREMAP_WC);
> > >  	if (!mem_virt) {
> > >  		ret = -ENOMEM;
> > >  		goto err_release_fw;
> > >  	}
> > >  
> > > -	ret = qcom_mdt_load(dev, firmware, fw_name,
> > > -			    pas_id, mem_virt, mem_phys, res_size, NULL);
> > > +	ret = qcom_mdt_pas_load(ctx, firmware, fw_name, mem_virt, NULL);
> > >  	if (ret)
> > >  		goto err_mem_unmap;
> > >  
> > > -	ret = qcom_scm_pas_auth_and_reset(pas_id);
> > > +	if (core->has_iommu) {
> > > +		ret = iommu_map(core->fw.iommu_domain, 0, rmem->base, rmem->size,
> > > +				IOMMU_READ | IOMMU_WRITE | IOMMU_PRIV, GFP_KERNEL);
> > 
> > What is the use case for IOMMU_PRIV here? You don't have this flag for
> > the qcom_q6v5_pas change.
> 
> This is there for historic regions, I may not have complete information about why
> is it required but the reference is taken from venus support for chrome.
> 

Setting IOMMU_PRIV results in omitting the ARM_LPAE_PTE_AP_UNPRIV bit in
the IOMMU page tables - have you checked if QHEE sets this? Ideally we
want to do the same QHEE would normally do.

Also, please add a define for the 0 numbere here similar to

#define VENUS_FW_START_ADDR		0x0

It's quite hard to see that this is not an identity-mapping like for
qcom_q6v5_pas.

>  
> > > +		if (ret)
> > > +			goto err_mem_unmap;
> > > +
> > > +		/*
> > > +		 * Firmware has no support for resource table for now, so, lets
> > > +		 * pass NULL and zero for input resource table and input resource
> > > +		 * table respectively.
> > > +		 */
> > > +		ret = qcom_mdt_pas_map_devmem_rscs(ctx, core->fw.iommu_domain, NULL, 0);
> > > +		if (ret)
> > > +			goto err_unmap_carveout;
> > > +	}
> > > +
> > > +	ret = qcom_scm_pas_prepare_and_auth_reset(ctx);
> > >  	if (ret)
> > > -		goto err_mem_unmap;
> > > +		goto err_unmap_devmem_rscs;
> > > +
> > > +	core->fw.ctx = ctx;
> > >  
> > >  	return ret;
> > >  
> > > +err_unmap_devmem_rscs:
> > > +	if (core->has_iommu)
> > > +		qcom_mdt_pas_unmap_devmem_rscs(ctx, core->fw.iommu_domain);
> > > +err_unmap_carveout:
> > > +	if (core->has_iommu)
> > > +		iommu_unmap(core->fw.iommu_domain, 0, rmem->size);
> > >  err_mem_unmap:
> > >  	memunmap(mem_virt);
> > >  err_release_fw:
> > > @@ -109,10 +144,97 @@ int iris_fw_load(struct iris_core *core)
> > >  
> > >  int iris_fw_unload(struct iris_core *core)
> > >  {
> > > -	return qcom_scm_pas_shutdown(core->iris_platform_data->pas_id);
> > > +	struct qcom_scm_pas_ctx *ctx;
> > > +	int ret;
> > > +
> > > +	ctx = core->fw.ctx;
> > > +	ret = qcom_scm_pas_shutdown(ctx->peripheral);
> > > +	if (core->has_iommu) {
> > > +		iommu_unmap(core->fw.iommu_domain, 0, ctx->mem_size);
> > > +		qcom_mdt_pas_unmap_devmem_rscs(ctx, core->fw.iommu_domain);
> > > +	}
> > > +
> > > +	return ret;
> > >  }
> > >  
> > >  int iris_set_hw_state(struct iris_core *core, bool resume)
> > >  {
> > >  	return qcom_scm_set_remote_state(resume, 0);
> > >  }
> > > +
> > > +int iris_fw_init(struct iris_core *core)
> > > +{
> > > +	struct platform_device_info info;
> > > +	struct iommu_domain *iommu_dom;
> > > +	struct platform_device *pdev;
> > > +	struct device_node *np;
> > > +	int ret;
> > > +
> > > +	np = of_get_child_by_name(core->dev->of_node, "video-firmware");
> > > +	if (!np)
> > > +		return 0;
> > 
> > You need a dt-bindings change for this as well. This is documented only
> > for Venus.
> 
> You are right, wanted to send device tree and binding support separately.
> But if required, will add with the series in the next version.
> 

You can send device tree changes separately, but dt-binding changes
always need to come before the driver changes.

Thanks,
Stephan

  reply	other threads:[~2025-08-20 13:39 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-19 16:54 [PATCH v2 00/11] Peripheral Image Loader support for Qualcomm SoCs running Linux host at EL2 Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 01/11] firmware: qcom_scm: Introduce PAS context initialization helper Mukesh Ojha
2025-08-19 17:17   ` Pavan Kondeti
2025-08-20  6:19     ` Mukesh Ojha
2025-08-20 11:40   ` Bryan O'Donoghue
2025-08-20 12:28     ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 02/11] soc: qcom: mdtloader: Add context aware qcom_mdt_pas_load() helper Mukesh Ojha
2025-08-20 11:48   ` Bryan O'Donoghue
2025-08-20 12:25     ` Mukesh Ojha
2025-09-03 15:03   ` Bryan O'Donoghue
2025-09-04  9:52     ` Mukesh Ojha
2025-09-04 10:15       ` Bryan O'Donoghue
2025-09-04 11:43         ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 03/11] firmware: qcom_scm: Add a prep version of auth_and_reset function Mukesh Ojha
2025-08-20 12:03   ` Bryan O'Donoghue
2025-08-20 12:24     ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 04/11] firmware: qcom_scm: Simplify qcom_scm_pas_init_image() Mukesh Ojha
2025-08-21 14:36   ` Bryan O'Donoghue
2025-08-21 16:29     ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 05/11] firmware: qcom_scm: Add shmbridge support to pas_init/release function Mukesh Ojha
2025-08-21 15:23   ` Bryan O'Donoghue
2025-08-21 17:03     ` Mukesh Ojha
2025-08-22 16:52       ` Mukesh Ojha
2025-08-22 23:13       ` Bryan O'Donoghue
2025-08-19 16:54 ` [PATCH v2 06/11] remoteproc: Move resource table data structure to its own header Mukesh Ojha
2025-08-20  8:12   ` Stephan Gerhold
2025-08-20 15:18     ` Mukesh Ojha
2025-08-20 15:31       ` Stephan Gerhold
2025-08-22  7:56         ` Mukesh Ojha
2025-08-20 16:32       ` Mukesh Ojha
2025-08-20 16:53         ` Stephan Gerhold
2025-08-22  9:21           ` Mukesh Ojha
2025-08-22  8:35         ` Krzysztof Kozlowski
2025-08-22  9:30           ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 07/11] firmware: qcom_scm: Add qcom_scm_pas_get_rsc_table() to get resource table Mukesh Ojha
2025-08-21 15:05   ` Krzysztof Kozlowski
2025-08-21 17:20     ` Mukesh Ojha
2025-08-22  6:22       ` Krzysztof Kozlowski
2025-08-22  7:21         ` Mukesh Ojha
2025-08-22  8:30         ` Krzysztof Kozlowski
2025-08-19 16:54 ` [PATCH v2 08/11] soc: qcom: mdt_loader: Add helper functions to map and unmap resources Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 09/11] remoteproc: pas: Extend parse_fw callback to parse resource table Mukesh Ojha
2025-08-20  8:36   ` Stephan Gerhold
2025-08-20 11:14     ` Mukesh Ojha
2025-08-20 13:07       ` Stephan Gerhold
2025-08-21 14:49     ` Krzysztof Kozlowski
2025-08-21 17:41       ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 10/11] remoteproc: qcom: pas: Enable Secure PAS support with IOMMU managed by Linux Mukesh Ojha
2025-08-20  8:40   ` Stephan Gerhold
2025-08-20 12:03     ` Mukesh Ojha
2025-08-19 16:54 ` [PATCH v2 11/11] media: iris: " Mukesh Ojha
2025-08-20  8:46   ` Stephan Gerhold
2025-08-20 11:56     ` Mukesh Ojha
2025-08-20 13:39       ` Stephan Gerhold [this message]
2025-08-22  4:26         ` Vikash Garodia
2025-08-22  8:46           ` Stephan Gerhold
2025-08-22 15:06             ` Mukesh Ojha
2025-08-22 16:26               ` Stephan Gerhold
2025-08-22 16:40                 ` Mukesh Ojha
2025-08-23 20:43                   ` Stephan Gerhold
2025-08-25 11:19                     ` Mukesh Ojha
2025-08-20 11:33   ` Bryan O'Donoghue
2025-08-20 12:00     ` Mukesh Ojha
2025-08-22  8:45   ` Krzysztof Kozlowski
2025-08-22 15:13     ` Mukesh Ojha
2025-08-23 15:41       ` Krzysztof Kozlowski
2025-08-23 15:46         ` Krzysztof Kozlowski
2025-08-23 15:52           ` Krzysztof Kozlowski
2025-08-20 11:03 ` [PATCH v2 00/11] Peripheral Image Loader support for Qualcomm SoCs running Linux host at EL2 Bryan O'Donoghue
2025-08-20 11:22   ` Mukesh Ojha
2025-09-03 11:56     ` Konrad Dybcio
2025-09-03 13:31       ` Bryan O'Donoghue
2025-09-03 14:02         ` Dmitry Baryshkov
2025-09-03 14:05           ` Bryan O'Donoghue
2025-09-03 14:13           ` Bryan O'Donoghue
2025-09-03 14:21             ` Bryan O'Donoghue
2025-09-03 14:28             ` Dmitry Baryshkov

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=aKXQAoXZyR6SRPAA@linaro.org \
    --to=stephan.gerhold@linaro.org \
    --cc=abhinav.kumar@linux.dev \
    --cc=andersson@kernel.org \
    --cc=bryan.odonoghue@linaro.org \
    --cc=konradybcio@kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-remoteproc@vger.kernel.org \
    --cc=mathieu.poirier@linaro.org \
    --cc=mchehab@kernel.org \
    --cc=quic_dikshita@quicinc.com \
    --cc=quic_vgarodia@quicinc.com \
    /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;
as well as URLs for NNTP newsgroup(s).