All of lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Bartosz Golaszewski" <bartosz.golaszewski@oss.qualcomm.com>
Cc: vkoul@kernel.org, Frank.Li@kernel.org, dmaengine@vger.kernel.org
Subject: Re: [PATCH v21 14/14] crypto: qce - Communicate the base physical address to the dmaengine
Date: Mon, 13 Jul 2026 13:26:02 +0000	[thread overview]
Message-ID: <20260713132602.DA5BA1F00A3A@smtp.kernel.org> (raw)
In-Reply-To: <20260713-qcom-qce-cmd-descr-v21-14-bc2583e18475@oss.qualcomm.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The error path in `qce_submit_cmd_desc()` attempts to free a non-reusable DMA descriptor using `dmaengine_desc_free()`, which will fail and leak the descriptor.
--

--- Patch [14]: [PATCH v21 14/14] crypto: qce - Communicate the base physical address to the dmaengine ---
commit abb736f89a48907c45dba4ccf37ffa56752bda22
Author: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>

crypto: qce - Communicate the base physical address to the dmaengine

This commit attaches metadata containing the scratchpad address and direction
to the BAM DMA descriptor. It also adds an error handling path to free the
descriptor if metadata attachment or submission fails.

> diff --git a/drivers/crypto/qce/dma.c b/drivers/crypto/qce/dma.c
> index 1b43c56503334..6410f8dc5bcf5 100644
> --- a/drivers/crypto/qce/dma.c
> +++ b/drivers/crypto/qce/dma.c

[ ... ]

> @@ -60,15 +65,21 @@ int qce_submit_cmd_desc(struct qce_device *qce)
>  		goto err_unmap_sg;
>  	}
>  
> +	ret = dmaengine_desc_attach_metadata(dma_desc, &meta, sizeof(meta));
> +	if (ret)
> +		goto err_free_desc;
> +
>  	qce_desc->dma_desc = dma_desc;
>  	cookie = dmaengine_submit(qce_desc->dma_desc);
>  
>  	ret = dma_submit_error(cookie);
>  	if (ret)
> -		goto err_unmap_sg;
> +		goto err_free_desc;
>  
>  	return 0;
>  
> +err_free_desc:
> +	dmaengine_desc_free(dma_desc);

[Severity: Medium]
Does this error path leak the DMA descriptor?

Looking at dmaengine_desc_free() in include/linux/dmaengine.h, it checks if
the descriptor is reusable and returns -EPERM if it is not:

static inline int dmaengine_desc_free(struct dma_async_tx_descriptor *desc)
{
        if (!dmaengine_desc_test_reuse(desc))
                return -EPERM;

        return desc->desc_free(desc);
}

Since the descriptor allocated earlier in qce_submit_cmd_desc() by
dmaengine_prep_slave_sg() is a standard descriptor and not marked as
reusable, it appears dmaengine_desc_free() will fail without actually
freeing the memory if either dmaengine_desc_attach_metadata() or
dma_submit_error() triggers the goto.

Is there a different method intended for cleaning up non-reusable
descriptors in this error path, or should the descriptor be allowed to
complete with a failure instead?

>  err_unmap_sg:
>  	dma_unmap_sg(qce->dev, bam_txn->wr_sgl, bam_txn->wr_sgl_cnt, DMA_TO_DEVICE);
>  	return ret;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260713-qcom-qce-cmd-descr-v21-0-bc2583e18475@oss.qualcomm.com?part=14

      reply	other threads:[~2026-07-13 13:26 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-13 13:01 [PATCH v21 00/14] crypto/dmaengine: qce: introduce BAM locking and use DMA for register I/O Bartosz Golaszewski
2026-07-13 13:01 ` [PATCH v21 01/14] dmaengine: constify struct dma_descriptor_metadata_ops Bartosz Golaszewski
2026-07-13 13:01 ` [PATCH v21 02/14] dmaengine: qcom: bam_dma: free interrupt before the clock in error path Bartosz Golaszewski
2026-07-13 13:14   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 03/14] dmaengine: qcom: bam_dma: convert tasklet to a BH workqueue Bartosz Golaszewski
2026-07-13 13:17   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 04/14] dmaengine: qcom: bam_dma: Extend the driver's device match data Bartosz Golaszewski
2026-07-13 13:01 ` [PATCH v21 05/14] dmaengine: qcom: bam_dma: Add pipe_lock_supported flag support Bartosz Golaszewski
2026-07-13 13:01 ` [PATCH v21 06/14] dmaengine: qcom: bam_dma: add support for BAM locking Bartosz Golaszewski
2026-07-13 13:17   ` sashiko-bot
2026-07-14  9:49   ` Stephan Gerhold
2026-07-13 13:01 ` [PATCH v21 07/14] crypto: qce - Cancel work on device detach Bartosz Golaszewski
2026-07-13 13:15   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 08/14] crypto: qce - Include algapi.h in the core.h header Bartosz Golaszewski
2026-07-13 13:18   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 09/14] crypto: qce - Remove unused ignore_buf Bartosz Golaszewski
2026-07-13 13:14   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 10/14] crypto: qce - Simplify arguments of devm_qce_dma_request() Bartosz Golaszewski
2026-07-13 13:01 ` [PATCH v21 11/14] crypto: qce - Use existing devres APIs in devm_qce_dma_request() Bartosz Golaszewski
2026-07-13 13:16   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 12/14] crypto: qce - Map crypto memory for DMA Bartosz Golaszewski
2026-07-13 13:22   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 13/14] crypto: qce - Add BAM DMA support for crypto register I/O Bartosz Golaszewski
2026-07-13 13:24   ` sashiko-bot
2026-07-13 13:01 ` [PATCH v21 14/14] crypto: qce - Communicate the base physical address to the dmaengine Bartosz Golaszewski
2026-07-13 13:26   ` sashiko-bot [this message]

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=20260713132602.DA5BA1F00A3A@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=Frank.Li@kernel.org \
    --cc=bartosz.golaszewski@oss.qualcomm.com \
    --cc=dmaengine@vger.kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=vkoul@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 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.