On Wed, Dec 02, 2009 at 12:38:07PM +0100, Takashi Iwai wrote:
At Tue, 1 Dec 2009 13:38:40 -0800, mark gross wrote:
--- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c
(snip)
@@ -506,8 +508,8 @@ static int snd_pcm_hw_free(struct snd_pcm_substream *substream) if (substream->ops->hw_free) result = substream->ops->hw_free(substream); runtime->status->state = SNDRV_PCM_STATE_OPEN;
- pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY,
substream->latency_id);
- pm_qos_remove_request(substream->latency_pm_qos_req);
The NULL check seems needed in the caller side because pm_qos_remove_request() doesn't do it.
ouch. Thanks for noticing this.
Or, would you add a NULL check in pm_qos_remove_request()? It'd be more handy.
I can make the pm_qos_remove_request null pointer safe, its assumed to be a slow path so it would be a good thing to do.
I'll add this to the next patch set that the e1000e guys are making me do against linux-next.
Thanks!
--mgross
thanks,
Takashi