Permalink
Switch branches/tags
Nothing to show
Commits on Oct 2, 2015
  1. Add a README.md file

    Some users do not know about the mainline merge and keep proposing updates to
    this old repository.
    
    Make thing clearer in the README file
    ao2 committed Oct 2, 2015
Commits on Dec 27, 2013
  1. The Vaughan device support the 352800 rate and not

    the 352000
    
    Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
    committed Dec 27, 2013
Commits on Jun 21, 2013
  1. Remove pr_debug("%s: called.\n", __func__) statements

    These can be added back locally when debugging, but for the version
    targeting mainline they are not really needed.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 21, 2013
  2. Remove a stale comment

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 21, 2013
  3. Reset panic state to false in hiface_pcm_stream_start()

    A panic in the hiFace driver is a condition due to wrong communication
    over USB, reset it when starting a new USB stream to cover the case of
    a transient failures.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 21, 2013
Commits on Jun 3, 2013
  1. Stop USB stream unconditionally in hiface_pcm_close()

    This fixes a bug which caused the driver to keep sending URBs even after
    the .close() callback was invoked.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  2. Replace pr_debug() with dev_dbg() and pr_err() with dev_err()

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  3. Don't declare multiple variables in the same statement

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  4. Cosmetics: adjust indentation of continuation lines

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  5. Use vmalloc buffers

    For USB devices it's not necessary to allocate physically contiguous
    buffers as there won't be DMA from the device.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  6. Drop the index field of hiface_chip as it's not used anywhere

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  7. Simplify the probing scheme to be per-device instead of per-interface

    The hiFace devices only have one USB interface per device, so there is
    no need to check if the device was already registered with another
    interface.
    
    This allows to simplify the probing scheme. As a consequence the
    "intf_count" and "shutdown" fields of hiface_chip can be dropped.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 3, 2013
  8. hiface_pcm_abort(): don't stop the alsa stream here

    hiface_pcm_abort() is called in the disconnect callback, at this time the
    alsa PCM stream has already been stopped.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 2, 2013
  9. hiface_pcm_close(): close a race window

    Stop the USB stream before deactivating it.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 2, 2013
  10. hiface_pcm_get_substream(): replace pr_debug with dev_err()

    The message really is an error message, and the dev_ variant will be
    more informative about the device producing the message.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 2, 2013
  11. hiface_pcm_trigger(): use spin_lock_irq()

    Reviewers told that spin_lock_irq() can be used here instead of
    spin_lock_irqsave(): nothing else should have disabled interrupts when
    the trigger callback is invoked.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Jun 2, 2013
  12. Drop some useless pr_debug() calls

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed May 28, 2013
  13. Replace strcpy() with strlcpy()

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed May 28, 2013
  14. Replace snd_printk() with the standard dev_err() and friends

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed May 28, 2013
Commits on Feb 13, 2013
  1. Silly Makefile change

    One of those silly details which won't let you sleep.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  2. Fix a couple of checkpatch.pl issues

      ERROR: space required before the open parenthesis '('
      #129: FILE: pcm.c:125:
      +       switch(rate) {
    
      WARNING: braces {} are not necessary for single statement blocks
      #419: FILE: pcm.c:415:
      +               if (!rt->playback.instance) {
      +                       hiface_pcm_stream_stop(rt);
      +               }
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  3. Handle the case of failing URBs in hiface_pcm_out_urb_handler()

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  4. Move the period elapsed check into hiface_pcm_playback()

    This simplifies the unlocking logic in the calling code.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  5. Make hiface_pcm_playback() not returning anything

    In hiface_pcm_playback() there is no need to return an error because the
    format has alredy been checked by ALSA already.
    
    Add a WARN_ON() to spot changes in upper layers if they happen in
    future.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  6. Protect snd_pcm_stop() with snd_pcm_stream_lock_irq()

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 12, 2013
  7. Declare memcpy_swahw32() as static

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  8. Add SNDRV_PCM_RATE_KNOT only when needed

    Add SNDRV_PCM_RATE_KNOT only when extra frequencies are allowed, and in
    this case set up explicit constraints for the supported rates.
    
    Now constraints_rates can also be const as we don't chage its fields at
    runtime.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  9. Cosmetics, align symbolic constant values

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 13, 2013
  10. Rename MAX_BUFSIZE to PCM_BUFFER_SIZE

    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  11. Rename PCM_MAX_PACKET_SIZE to PCM_PACKET_SIZE

    Suggested by Clemens Ladisch.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  12. Make device_table const

    Suggested by Clemens Ladisch.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  13. Downgrade a pr_info to pr_debug in hiface_chip_probe()

    As requested by Clemens Ladisch
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  14. Use again the implicit mechanism to allocate the card private data

    Our hiface_dev_free() has now been simplified so that it just calls
    kfree(), so the implicit mechanism is suitable again for us to use.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  15. Use usb_make_path() when building card->longname

    This was suggested by Clemens Ladisch
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013
  16. Use an explicit switch instead of the rate_value array in hiface_pcm_…

    …set_rate()
    
    This just tells exactly what we are doing, in a more transparent way
    than a for loop.
    
    Also use some symbolic constants to highlight the mapping between the
    sample rate and the value to send via USB.
    
    Signed-off-by: Antonio Ospite <ao2@amarulasolutions.com>
    ao2 committed Feb 11, 2013