On Wed, 23 Oct 2019 14:01:14 +0200, Srinivas Kandagatla wrote:

From: Preetam Singh Ranawat apranawat@codeaurora.org

When period interrupts are disabled, while loop in snd_pcm_update_hw_ptr0() results in the machine locking up if runtime->hw_ptr_buffer_jiffies is 0.

Validate runtime->hw_ptr_buffer_jiffies value before while loop to avoid delta check.

Although the fix itself is fine, I still wonder how this is triggered. Could you check in what kind of PCM state is it?

Basically snd_pcm_update_hw_ptr*() is called only in the running PCM state, and for that, we should have set hw_ptr_buffer_jiffies at snd_pcm_post_start(). It's

runtime->hw_ptr_buffer_jiffies = (runtime->buffer_size * HZ) / runtime->rate;

So maybe the given buffer size is so small and hw_ptr_buffer_jiffies is really 0?

thanks,

Takashi

Signed-off-by: Preetam Singh Ranawat apranawat@codeaurora.org Signed-off-by: Srinivas Kandagatla srinivas.kandagatla@linaro.org

---

This issue was noticed in Android setup from fastcapture path.

sound/core/pcm_lib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c index d80041ea4e01..f407f6895f8a 100644 --- a/sound/core/pcm_lib.c +++ b/sound/core/pcm_lib.c @@ -341,7 +341,8 @@ static int snd_pcm_update_hw_ptr0(struct snd_pcm_substream *substream, * the elapsed time to detect xruns. */ jdelta = curr_jiffies - runtime->hw_ptr_jiffies;

• if (jdelta < runtime->hw_ptr_buffer_jiffies / 2)
  

• if ((jdelta < runtime->hw_ptr_buffer_jiffies / 2) ||
  

•     !runtime->hw_ptr_buffer_jiffies)
  goto no_delta_check;
  

  hdelta = jdelta - delta * HZ / runtime->rate; xrun_threshold = runtime->hw_ptr_buffer_jiffies / 2 + 1;

-- 2.21.0