From: Frank Li <Frank.li@oss.nxp.com>
To: James Hilliard <james.hilliard1@gmail.com>
Cc: Vinod Koul <vkoul@kernel.org>, Frank Li <Frank.Li@kernel.org>,
Chen-Yu Tsai <wens@kernel.org>,
Jernej Skrabec <jernej.skrabec@gmail.com>,
Samuel Holland <samuel@sholland.org>,
dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
linux-sunxi@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] dmaengine: sun6i: constrain descriptor addresses through the DMA mask
Date: Thu, 22 Oct 2026 11:10:20 -0500 [thread overview]
Message-ID: <ato1bMSJgTVZnOyd@SMW015318> (raw)
In-Reply-To: <20261021-submit-sun6i-dma-descriptor-mask-v1-v1-1-76e72bd52e57@gmail.com>
On Wed, Oct 21, 2026 at 03:11:29PM -0600, James Hilliard wrote:
> Descriptor allocations pass GFP_DMA32 to dma_pool_alloc(). The DMA API
> chooses the memory zone from the device address mask, rather than accepting
> zone-selection flags from the driver. These flags can also reach the pool
> metadata allocation and trigger a slab allocator warning, as observed on
> H616 when preparing the first UART receive-DMA chain.
>
> Set a 32-bit coherent DMA mask before creating the descriptor pool and
> remove GFP_DMA32 from its allocations. The driver does not encode the
> extra descriptor-address bits supported by some variants, so descriptor
> links must remain below 4 GiB. Keep the streaming mask unchanged for
> variants with wider source and destination addresses.
Spouse other dma address in src/dest in descriptor also limited to 32bit.
Do you means only descripor address have 32bit limiation, other have not
such limiation?
Frank
>
> Fixes: ec31c5c59492 ("dmaengine: sun6i: Add support for 34-bit physical addresses")
> Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
> ---
> drivers/dma/sun6i-dma.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
> index 7704b016aed8..0ef6ed1f936e 100644
> --- a/drivers/dma/sun6i-dma.c
> +++ b/drivers/dma/sun6i-dma.c
> @@ -684,7 +684,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
> if (!txd)
> return NULL;
>
> - v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
> + v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
> if (!v_lli) {
> dev_err(sdev->slave.dev, "Failed to alloc lli memory\n");
> goto err_txd_free;
> @@ -743,7 +743,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_slave_sg(
> return NULL;
>
> for_each_sg(sgl, sg, sg_len, i) {
> - v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
> + v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
> if (!v_lli)
> goto err_lli_free;
>
> @@ -826,7 +826,7 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_cyclic(
> return NULL;
>
> for (i = 0; i < periods; i++) {
> - v_lli = dma_pool_alloc(sdev->pool, GFP_DMA32 | GFP_NOWAIT, &p_lli);
> + v_lli = dma_pool_alloc(sdev->pool, GFP_NOWAIT, &p_lli);
> if (!v_lli) {
> dev_err(sdev->slave.dev, "Failed to alloc lli memory\n");
> goto err_lli_free;
> @@ -1345,6 +1345,11 @@ static int sun6i_dma_probe(struct platform_device *pdev)
> return PTR_ERR(sdc->rstc);
> }
>
> + /* This driver uses unextended 32-bit addresses for descriptor links. */
> + ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
> + if (ret)
> + return ret;
> +
> sdc->pool = dmam_pool_create(dev_name(&pdev->dev), &pdev->dev,
> sizeof(struct sun6i_dma_lli), 4, 0);
> if (!sdc->pool) {
>
> ---
> base-commit: 93f51579e7df248780214094418f205253383cc5
> change-id: 20261021-submit-sun6i-dma-descriptor-mask-v1-35bb381b932a
>
> Best regards,
> --
> James Hilliard <james.hilliard1@gmail.com>
>
next prev parent reply other threads:[~2026-10-22 16:10 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-21 21:11 James Hilliard
2026-10-22 16:10 ` Frank Li [this message]
2026-10-22 19:13 ` James Hilliard
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=ato1bMSJgTVZnOyd@SMW015318 \
--to=frank.li@oss.nxp.com \
--cc=Frank.Li@kernel.org \
--cc=dmaengine@vger.kernel.org \
--cc=james.hilliard1@gmail.com \
--cc=jernej.skrabec@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sunxi@lists.linux.dev \
--cc=samuel@sholland.org \
--cc=vkoul@kernel.org \
--cc=wens@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
all inboxes | Powered by JetHome®