SUPERH platform development
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: linux-sh@vger.kernel.org
Subject: Re: [PATCH/RFC 2/8] serial: sh-sci: Fix scatterlist mapping leak
Date: Sat, 23 May 2015 19:04:49 +0000	[thread overview]
Message-ID: <3470551.8sqo3UHZvD@avalon> (raw)
In-Reply-To: <1432145174-11534-3-git-send-email-geert+renesas@glider.be>

Hi Geert,

Thank you for the patch.

On Wednesday 20 May 2015 20:06:08 Geert Uytterhoeven wrote:
> The mapped scatterlist is never unmapped. This leaks quite some
> mappings, as the mapping is done in uart_ops.startup(), i.e. every time
> the device is opened. Unmap the scatterlist on device close.
> 
> Note that we have to restore the original DMA entry length and DMA
> address, as it has been modified in work_fn_tx(). Else warnings are
> printed if CONFIG_DMA_API_DEBUG=y:
> 
>     WARNING: CPU: 0 PID: 20 at lib/dma-debug.c:1103
> check_unmap+0x24c/0x85c() rcar-dmac e6700000.dma-controller: DMA-API:
> device driver frees DMA memory with different size [device
> address=0x000000006e15f000] [map size@96 bytes] [unmap size=3 bytes]
> 
> and:
> 
>     WARNING: CPU: 0 PID: 1364 at lib/dma-debug.c:1093
> check_unmap+0x178/0x85c() rcar-dmac e6700000.dma-controller: DMA-API:
> device driver tries to free DMA memory it has not allocated [device
> address=0x000000006e16b006] [size@96 bytes]
> 
> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> ---
> As the transmit scatterlist contains one single entry, which is reused
> and modified, I think it's better to use dma_{,un}map_single() instead
> of dma_{,un}map_sg(). But that requires more invasive changes.

It would indeed be more invasive, but I think it would be cleaner. Can we go 
that way directly ? Modifying the address and length in work_fn_tx() is a big 
hack.

> ---
>  drivers/tty/serial/sh-sci.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
> index 8756d186e86891ac..0aec66cf68615972 100644
> --- a/drivers/tty/serial/sh-sci.c
> +++ b/drivers/tty/serial/sh-sci.c
> @@ -1390,6 +1390,14 @@ static void sci_tx_dma_release(struct sci_port *s,
> bool enable_pio)
> 
>  	s->chan_tx = NULL;
>  	s->cookie_tx = -EINVAL;
> +	if (s->sg_len_tx) {
> +		/* Restore sg_dma_len() and sg_dma_address() */
> +		sg_dma_len(&s->sg_tx) = UART_XMIT_SIZE;
> +		sg_dma_address(&s->sg_tx) = sg_dma_address(&s->sg_tx) &
> +					    ~(UART_XMIT_SIZE - 1);
> +		dma_unmap_sg(chan->device->dev, &s->sg_tx, 1, DMA_TO_DEVICE);
> +		s->sg_len_tx = 0;
> +	}
>  	dma_release_channel(chan);
>  	if (enable_pio)
>  		sci_start_tx(port);

-- 
Regards,

Laurent Pinchart


      reply	other threads:[~2015-05-23 19:04 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-20 18:06 [PATCH/RFC 2/8] serial: sh-sci: Fix scatterlist mapping leak Geert Uytterhoeven
2015-05-23 19:04 ` Laurent Pinchart [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=3470551.8sqo3UHZvD@avalon \
    --to=laurent.pinchart@ideasonboard.com \
    --cc=linux-sh@vger.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