On Wed, Aug 20, 2025 at 10:40:25AM +0200, Stephan Gerhold wrote:
> On Tue, Aug 19, 2025 at 10:24:45PM +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 aforementioned hypervisor is absent, the operating system must
> > perform these configurations instead.
> >
> > When Linux runs as the hypervisor (at EL2) on a SoC, it will have its
> > own device tree overlay file that specifies the firmware stream ID now
> > managed by Linux for a particular remote processor. If the iommus
> > property is specified in the remoteproc device tree node, it indicates
> > that IOMMU configuration must be handled by Linux. In this case, the
> > has_iommu flag is set for the remote processor, which ensures that the
> > resource table, carveouts, and SHM bridge are properly configured before
> > memory is passed to TrustZone for authentication. Otherwise, the
> > has_iommu flag remains unset, which is the default behavior.
> >
> > Enables Secure PAS support for remote processors when IOMMU configuration
> > is managed by Linux.
> >
> > Signed-off-by: Mukesh Ojha <mukesh.o...@oss.qualcomm.com>
> > ---
> > drivers/remoteproc/qcom_q6v5_pas.c | 63 +++++++++++++++++++++++++++---
> > 1 file changed, 57 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/remoteproc/qcom_q6v5_pas.c
> > b/drivers/remoteproc/qcom_q6v5_pas.c
> > index 1e0f09bf1ef2..180528bcd57c 100644
> > --- a/drivers/remoteproc/qcom_q6v5_pas.c
> > +++ b/drivers/remoteproc/qcom_q6v5_pas.c
> > [...]
> > @@ -424,7 +459,8 @@ static int qcom_pas_parse_firmware(struct rproc *rproc,
> > const struct firmware *f
> > if (!rproc->has_iommu)
> > return ret;
> >
> > - ret = qcom_scm_pas_get_rsc_table(pas->pas_id, NULL, 0, &output_rt,
> > &output_rt_size);
> > + ret = qcom_scm_pas_get_rsc_table(pas->pas_ctx, NULL, 0,
> > + &output_rt, &output_rt_size);
>
> Unrelated formatting change, should be in previous commit.
Ack, will fix.
>
> > if (ret) {
> > dev_err(pas->dev, "error %d getting resource_table\n", ret);
> > return ret;
> > @@ -726,6 +762,20 @@ static int qcom_pas_probe(struct platform_device *pdev)
> > return -ENOMEM;
> > }
> >
> > + if (of_property_present(pdev->dev.of_node, "iommus")) {
>
> I think you need a dt-bindings change for this? You had one in v1, but
> dropped it entirely for some reason.
I missed to mentioned it in changelog.
As I said in another patch, I intended to send it separately however,
will send it along with this series in the next version.
>
> > + struct of_phandle_args args;
> > +
> > + ret = of_parse_phandle_with_args(pdev->dev.of_node, "iommus",
> > + "#iommu-cells", 0, &args);
> > + if (ret < 0)
> > + return ret;
> > +
> > + rproc->has_iommu = true;
> > + of_node_put(args.np);
> > + } else {
> > + rproc->has_iommu = false;
> > + }
> > +
> > rproc->auto_boot = desc->auto_boot;
> > rproc_coredump_set_elf_info(rproc, ELFCLASS32, EM_NONE);
> >
> > @@ -800,6 +850,7 @@ static int qcom_pas_probe(struct platform_device *pdev)
> > if (!pas->dtb_pas_ctx)
> > goto remove_ssr_sysmon;
> >
> > + pas->pas_ctx->has_iommu = pas->dtb_pas_ctx->has_iommu =
> > rproc->has_iommu;
>
> Nitpick: I think this would look cleaner if you separate it into two
> lines (only one assignment on each line).
>
Sure.
Thanks for the review.
> Thanks,
> Stephan
--
-Mukesh Ojha
