Lockdep complains about conflicts between isight->mutex,
ALSA's register_mutex, mm->mmap_sem, and pcm->open_mutex.
This can be fixed by moving the calls to isight_pcm_abort(),
snd_card_disconnect(), and fw_iso_resources_update() out of
isight->mutex. These functions are designed to be called
asynchronously; the mutex needs to protect only the device
streaming state modified by isight_start/stop_streaming().
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Experiments have shown this driver to work now.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Tested-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When aborting a PCM stream, the xrun is signaled only if the stream is
running. When disconnecting a PCM stream, calling snd_card_disconnect()
too early would change the stream into a non-running state and thus
prevent the xrun from being noticed by user space.
To prevent this, move the snd_card_disconnect() call after the xrun.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
which is needed to get the iSight to talk.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Set the .header_size field when queueing packets to avoid a division by
zero.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
After handling a received packet, we want to resubmit the same packet,
so do not increase the packet index too early.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Fix crashes in isight_pcm_abort() that happen when the driver tries to
access isight->pcm->runtime which does not exist when the device is not
open. Introduce a new field pcm_active to track this state.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This adds an experimental driver for the front and rear microphones of
the Apple iSight web camera.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When queueing iso packets, the run time is dominated by the two
MMIO accesses that set the DMA context's wake bit. Because most
drivers submit packets in batches, we can save much time by
removing all but the last wakeup.
The internal kernel API is changed to require a call to
fw_iso_context_queue_flush() after a batch of queued packets.
The user space API does not change, so one call to
FW_CDEV_IOC_QUEUE_ISO must specify multiple packets to take
advantage of this optimization.
In my measurements, this patch reduces the time needed to queue
fifty skip packets from userspace to one sixth on a 2.5 GHz CPU,
or to one third at 800 MHz.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
We do not need slab allocations anymore in order to satisfy
streaming DMA mapping constraints, thanks to commit da28947e7e
"firewire: ohci: avoid separate DMA mapping for small AT payloads".
(Besides, the slab-allocated buffers that firewire-core, firewire-sbp2,
and firedtv used to provide for 8-byte write and lock requests were
still not fully portable since they crossed cacheline boundaries or
shared a cacheline with unrelated CPU-accessed data. snd-firewire-lib
got this aspect right by using an extra kmalloc/ kfree just for the
8-byte transaction buffer.)
This change replaces kmalloc'ed lock transaction scratch buffers in
firewire-core, firedtv, and snd-firewire-lib by local stack allocations.
Perhaps the most notable result of the change is simpler locking because
there is no need to serialize usages of preallocated per-device buffers
anymore. Also, allocations and deallocations are simpler.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
When aborting a PCM stream, the xrun is signaled only if the stream is
running. When disconnecting a PCM stream, calling snd_card_disconnect()
too early would change the stream into a non-running state and thus
prevent the xrun from being noticed by user space.
To prevent this, move the snd_card_disconnect() call after the xrun.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
fixes this error:
sound/firewire/fcp.c: In function 'fcp_avc_transaction':
sound/firewire/fcp.c:103: error: implicit declaration of function 'msleep'
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add an AMDTP stream error state that occurs when we fail to queue
another packet. In this case, the stream is stopped, and the error can
be reported when the application tries to restart the PCM stream.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
For correct cache coherency on some architectures, DMA buffers must be
allocated in a different cache line than data that is concurrently used
by the CPU.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
In non-blocking mode, the SYT_INTERVAL is larger than the number of
audio frames in each packet, so there are packets that do not contain
any frame to which the SYT could be applied. For these packets, the
SYT must not be the timestamp of the next valid SYT frame, but the
special no-info SYT value.
This fixes broken playback on the FireWave at 44.1 kHz.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add a driver for two playback-only FireWire devices based on the OXFW970
chip.
v2: better AMDTP API abstraction; fix fw_unit leak; small fixes
v3: cache the iPCR value
v4: FireWave constraints; fix fw_device reference counting;
fix PCR caching; small changes and fixes
v5: volume/mute support; fix crashing due to pcm stop races
v6: fix build; one-channel volume for LaCie
v7: use signed values to make volume (range checks) work; fix function
block IDs for volume/mute; always use channel 0 for LaCie volume
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Tested-by: Jay Fenlason <fenlason@redhat.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
