[PATCH 1/1] em28xx: Fix for Slow Memory Leak

[PATCH 1/1] em28xx: Fix for Slow Memory Leak

[Robert Krakora]

em28xx: Fix for Slow Memory Leak

From: Robert Krakora <rob.krakora@messagenetsystems.com>

Test Code:  (Provided by Douglas)

v4l-dvb/v4l2-apps/test/stress-buffer.c

The audio DMA area was never being freed and would slowly leak over
time as the v4l device was opened and closed by an application.

Thanks again to Douglas for generating the test code to help locate
memory leaks!!!

Priority: normal

Signed-off-by: Robert Krakora <rob.krakora@messagenetsystems.com>

diff -r 5567e82c34a0 linux/drivers/media/video/em28xx/em28xx-audio.c
--- a/linux/drivers/media/video/em28xx/em28xx-audio.c   Tue Mar 31
07:24:14 2009 -0300
+++ b/linux/drivers/media/video/em28xx/em28xx-audio.c   Tue Apr 14
10:16:45 2009 -0400
@@ -278,6 +278,7 @@
 #endif

        dprintk("Allocating vbuffer\n");
+
        if (runtime->dma_area) {
                if (runtime->dma_bytes > size)
                        return 0;
@@ -385,6 +386,18 @@
        mutex_lock(&dev->lock);
        dev->adev.users--;
        em28xx_audio_analog_set(dev);
+       if (substream == dev->adev.capture_pcm_substream)
+       {
+               if (substream && substream->runtime &&
substream->runtime->dma_area) {
+                       dprintk("freeing\n");
+                       vfree(substream->runtime->dma_area);
+                       substream->runtime->dma_area = NULL;
+               }
+       }
+       else
+       {
+               em28xx_errdev("substream(%p) !=
dev->adev.capture_pcm_substream(%p)\n", substream,
dev->adev.capture_pcm_substream);
+       }
        mutex_unlock(&dev->lock);

        return 0;

Re: [PATCH 1/1] em28xx: Fix for Slow Memory Leak

[Douglas Schilling Landgraf]

Hello Robert,

On Tue, 14 Apr 2009 15:17:16 -0400
Robert Krakora <rob.krakora@messagenetsystems.com> wrote:

> em28xx: Fix for Slow Memory Leak

Thanks, I am going to test and commit your patch.

Cheers,
Douglas
> From: Robert Krakora <rob.krakora@messagenetsystems.com>
> 
> Test Code:  (Provided by Douglas)
> 
> v4l-dvb/v4l2-apps/test/stress-buffer.c
> 
> The audio DMA area was never being freed and would slowly leak over
> time as the v4l device was opened and closed by an application.
> 
> Thanks again to Douglas for generating the test code to help locate
> memory leaks!!!
> 
> Priority: normal
> 
> Signed-off-by: Robert Krakora <rob.krakora@messagenetsystems.com>
> 
> diff -r 5567e82c34a0 linux/drivers/media/video/em28xx/em28xx-audio.c
> --- a/linux/drivers/media/video/em28xx/em28xx-audio.c   Tue Mar 31
> 07:24:14 2009 -0300
> +++ b/linux/drivers/media/video/em28xx/em28xx-audio.c   Tue Apr 14
> 10:16:45 2009 -0400
> @@ -278,6 +278,7 @@
>  #endif
> 
>         dprintk("Allocating vbuffer\n");
> +
>         if (runtime->dma_area) {
>                 if (runtime->dma_bytes > size)
>                         return 0;
> @@ -385,6 +386,18 @@
>         mutex_lock(&dev->lock);
>         dev->adev.users--;
>         em28xx_audio_analog_set(dev);
> +       if (substream == dev->adev.capture_pcm_substream)
> +       {
> +               if (substream && substream->runtime &&
> substream->runtime->dma_area) {
> +                       dprintk("freeing\n");
> +                       vfree(substream->runtime->dma_area);
> +                       substream->runtime->dma_area = NULL;
> +               }
> +       }
> +       else
> +       {
> +               em28xx_errdev("substream(%p) !=
> dev->adev.capture_pcm_substream(%p)\n", substream,
> dev->adev.capture_pcm_substream);
> +       }
>         mutex_unlock(&dev->lock);
> 
>         return 0;
> --
> To unsubscribe from this list: send the line "unsubscribe
> linux-media" in the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html