All of lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Ajay Kumar Nandam" <ajay.nandam@oss.qualcomm.com>
Cc: conor+dt@kernel.org, devicetree@vger.kernel.org, robh@kernel.org
Subject: Re: [PATCH v5 3/4] dt-bindings: sound: qcom,q6apm-dai: add memory-region and relax iommus
Date: Mon, 21 Sep 2026 19:17:04 +0000	[thread overview]
Message-ID: <20260921191704.E18391F000FF@smtp.kernel.org> (raw)
In-Reply-To: <20260922-vmid-v4-v5-3-e79cfd7af5b3@oss.qualcomm.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dt-bindings: sound: q6apm-dai: schema allows omitting both iommus and memory-region
--

commit 0ba7ab9d36bf919ab318abe00f3ad28279a11313
Author: Ajay Kumar Nandam <ajay.nandam@oss.qualcomm.com>

dt-bindings: sound: qcom,q6apm-dai: add memory-region and relax iommus

This commit adds an optional memory-region property for control-path
and data-path carveouts on modem-DSP platforms. It also removes iommus
from the required list because mDSP platforms lack IOMMU or VMID device
tree properties.

> diff --git a/Documentation/devicetree/bindings/sound/qcom,q6apm-dai.yaml b/Documentation/devicetree/bindings/sound/qcom,q6apm-dai.yaml
> index 9e5b30d9c6e62..7d0415f2463a5 100644
> --- a/Documentation/devicetree/bindings/sound/qcom,q6apm-dai.yaml
> +++ b/Documentation/devicetree/bindings/sound/qcom,q6apm-dai.yaml
> @@ -20,9 +20,19 @@ properties:
>      minItems: 1
>      maxItems: 2
>  
> +  memory-region:
> +    minItems: 1
> +    maxItems: 2
> +    items:
> +      - description:
> +          Control-path buffer (audio_heap_mem) for platforms where the DSP
> +          runs in a stage-2 protected context (e.g. modem-DSP).
> +      - description:
> +          Data-path buffer pool from which per-stream DMA buffers are
> +          allocated.
> +
>  required:
>    - compatible
> -  - iommus

[Severity: Medium]
Does removing iommus here without adding a constraint like oneOf or anyOf
allow the schema to silently pass if both properties are omitted?

If a device tree is written without either iommus or memory-region, it looks
like it will incorrectly pass schema validation. At runtime, the driver would
then allocate normal memory, and when the DSP attempts to access it, it could
trigger an SMMU or stage-2 translation fault and crash the DSP/subsystem.

Should we add a structural constraint to enforce that at least one of the
memory protection mechanisms is specified?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260922-vmid-v4-v5-0-e79cfd7af5b3@oss.qualcomm.com?part=3

  reply	other threads:[~2026-09-21 19:17 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-21 19:08 [PATCH v5 0/4] ASoC: qcom: enable audio on stage-2 protected DSPs (mDSP) Ajay Kumar Nandam
2026-09-21 19:08 ` [PATCH v5 1/4] ASoC: qcom: q6apm: clear g_apm on driver removal Ajay Kumar Nandam
2026-09-21 19:08 ` [PATCH v5 2/4] ASoC: qcom: qdsp6: generalize GPR service domain Ajay Kumar Nandam
2026-09-21 19:08 ` [PATCH v5 3/4] dt-bindings: sound: qcom,q6apm-dai: add memory-region and relax iommus Ajay Kumar Nandam
2026-09-21 19:17   ` sashiko-bot [this message]
2026-09-22 12:21   ` Mark Brown
2026-09-23 11:08     ` Ajay Kumar Nandam
2026-09-22 12:46   ` Rob Herring (Arm)
2026-09-23 11:12     ` Ajay Kumar Nandam
2026-09-23 11:57   ` Krzysztof Kozlowski
2026-09-23 16:22     ` Ajay Kumar Nandam
2026-09-21 19:08 ` [PATCH v5 4/4] ASoC: qcom: q6apm-dai: add SCM buffer assignment for mDSP platforms Ajay Kumar Nandam
2026-09-21 19:25   ` sashiko-bot
2026-09-22 12:30   ` Mark Brown
2026-09-23 11:20     ` Ajay Kumar Nandam

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=20260921191704.E18391F000FF@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=ajay.nandam@oss.qualcomm.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=robh@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.