From: sutar.mounesh@gmail.com
To: patch@alsa-project.org
Cc: Timo Wischer <twischer@de.adit-jv.com>,
alsa-devel@alsa-project.org,
Ravikiran Polepalli <ravikiran_polepalli@mentor.com>,
Mounesh Sutar <sutar.mounesh@gmail.com>,
Mikhail Durnev <mikhail_durnev@mentor.com>,
mounesh_sutar@mentor.com
Subject: [PATCH 1/6] pcm: dmix_rewind corrupts application pointer fix
Date: Fri, 17 Feb 2017 12:45:36 +0530 [thread overview]
Message-ID: <1487315736-8950-1-git-send-email-sutar.mounesh@gmail.com> (raw)
From: Timo Wischer <twischer@de.adit-jv.com>
sometimes pulseaudio stops with the following assertion in libasound.so:
alsa-lib-1.0.29/src/pcm/pcm.c:2761:
snd_pcm_area_copy: Assertion `dst < src || dst >= src + bytes' failed.
Application pointer is handled properly, in cases of rewind operations.
Signed-off-by: Timo Wischer <twischer@de.adit-jv.com>
Signed-off-by: Ravikiran Polepalli <ravikiran_polepalli@mentor.com>
Signed-off-by: Mikhail Durnev <mikhail_durnev@mentor.com>
Signed-off-by: Mounesh Sutar <sutar.mounesh@gmail.com>
diff --git a/src/pcm/pcm_dmix.c b/src/pcm/pcm_dmix.c
index dd0356e..4b763e2 100644
--- a/src/pcm/pcm_dmix.c
+++ b/src/pcm/pcm_dmix.c
@@ -706,7 +706,7 @@ static snd_pcm_sframes_t snd_pcm_dmix_rewind(snd_pcm_t *pcm, snd_pcm_uframes_t f
{
snd_pcm_direct_t *dmix = pcm->private_data;
snd_pcm_uframes_t slave_appl_ptr, slave_size;
- snd_pcm_uframes_t appl_ptr, size, transfer, result;
+ snd_pcm_uframes_t appl_ptr, size, transfer, result, frames_to_remix;
int err;
const snd_pcm_channel_area_t *src_areas, *dst_areas;
@@ -717,6 +717,11 @@ static snd_pcm_sframes_t snd_pcm_dmix_rewind(snd_pcm_t *pcm, snd_pcm_uframes_t f
return err;
}
+ /* (appl_ptr - last_appl_ptr) indicates the frames which are not already mixed
+ (last_appl_ptr - hw_ptr) indicates the frames which are already mixed but not played yet.
+ So they can be remixed.
+ */
+
if (dmix->last_appl_ptr < dmix->appl_ptr)
size = dmix->appl_ptr - dmix->last_appl_ptr;
else
@@ -729,6 +734,9 @@ static snd_pcm_sframes_t snd_pcm_dmix_rewind(snd_pcm_t *pcm, snd_pcm_uframes_t f
return size;
result = size;
+ /* Always at this point last_appl_ptr == appl_ptr
+ So (appl_ptr - hw_ptr) indicates the frames which can be remixed
+ */
if (dmix->hw_ptr < dmix->appl_ptr)
size = dmix->appl_ptr - dmix->hw_ptr;
else
@@ -741,9 +749,12 @@ static snd_pcm_sframes_t snd_pcm_dmix_rewind(snd_pcm_t *pcm, snd_pcm_uframes_t f
slave_size = dmix->slave_appl_ptr + (pcm->boundary - dmix->slave_hw_ptr);
if (slave_size < size)
size = slave_size;
- frames -= size;
- result += size;
-
+
+ /* frames which should be remixed will be saved
+ to also backward the appl pointer on success
+ */
+ frames_to_remix = size;
+
/* add sample areas here */
src_areas = snd_pcm_mmap_areas(pcm);
dst_areas = snd_pcm_mmap_areas(dmix->spcm);
@@ -769,15 +780,15 @@ static snd_pcm_sframes_t snd_pcm_dmix_rewind(snd_pcm_t *pcm, snd_pcm_uframes_t f
appl_ptr += transfer;
appl_ptr %= pcm->buffer_size;
}
- dmix->last_appl_ptr -= frames;
- dmix->last_appl_ptr %= pcm->boundary;
- dmix->slave_appl_ptr -= frames;
- dmix->slave_appl_ptr %= dmix->slave_boundary;
dmix_up_sem(dmix);
- snd_pcm_mmap_appl_backward(pcm, frames);
+ snd_pcm_mmap_appl_backward(pcm, frames_to_remix);
+ result += frames_to_remix;
+ /* At this point last_appl_ptr and appl_ptr has to indicate the
+ * position of the first not mixed frame
+ */
- return result + frames;
+ return result;
}
static snd_pcm_sframes_t snd_pcm_dmix_forwardable(snd_pcm_t *pcm)
--
2.7.4
next reply other threads:[~2017-02-17 7:15 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-17 7:15 sutar.mounesh [this message]
2017-02-17 17:45 ` [PATCH 1/6] pcm: dmix_rewind corrupts application pointer fix Takashi Iwai
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=1487315736-8950-1-git-send-email-sutar.mounesh@gmail.com \
--to=sutar.mounesh@gmail.com \
--cc=alsa-devel@alsa-project.org \
--cc=mikhail_durnev@mentor.com \
--cc=mounesh_sutar@mentor.com \
--cc=patch@alsa-project.org \
--cc=ravikiran_polepalli@mentor.com \
--cc=twischer@de.adit-jv.com \
/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;
as well as URLs for NNTP newsgroup(s).