Re: [PATCH v14 32/53] ALSA: usb-audio: Check for support for requested audio format

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Takashi,

On 2/9/2024 1:34 PM, Wesley Cheng wrote:
Hi Takashi,

On 2/9/2024 2:42 AM, Takashi Iwai wrote:
On Fri, 09 Feb 2024 00:13:45 +0100,
Wesley Cheng wrote:

Allow for checks on a specific USB audio device to see if a requested PCM
format is supported.  This is needed for support when playback is
initiated by the ASoC USB backend path.

Signed-off-by: Wesley Cheng <quic_wcheng@xxxxxxxxxxx>
---
  sound/usb/card.c | 31 +++++++++++++++++++++++++++++++
  sound/usb/card.h | 11 +++++++++++
  2 files changed, 42 insertions(+)

diff --git a/sound/usb/card.c b/sound/usb/card.c
index 7dc8007ba839..1ad99a462038 100644
--- a/sound/usb/card.c
+++ b/sound/usb/card.c
@@ -155,6 +155,37 @@ int snd_usb_unregister_platform_ops(void)
  }
  EXPORT_SYMBOL_GPL(snd_usb_unregister_platform_ops);
+/*
+ * Checks to see if requested audio profile, i.e sample rate, # of
+ * channels, etc... is supported by the substream associated to the
+ * USB audio device.
+ */
+struct snd_usb_stream *snd_usb_find_suppported_substream(int card_idx,
+            struct snd_pcm_hw_params *params, int direction)
+{
+    struct snd_usb_audio *chip;
+    struct snd_usb_substream *subs;
+    struct snd_usb_stream *as;
+
+    /*
+     * Register mutex is held when populating and clearing usb_chip
+     * array.
+     */
+    guard(mutex)(&register_mutex);
+    chip = usb_chip[card_idx];
+
+    if (chip && enable[card_idx]) {
+        list_for_each_entry(as, &chip->pcm_list, list) {
+            subs = &as->substream[direction];
+            if (snd_usb_find_substream_format(subs, params))
+                return as;
+        }
+    }
+
+    return NULL;
+}
+EXPORT_SYMBOL_GPL(snd_usb_find_suppported_substream);
+
  /*
   * disconnect streams
   * called from usb_audio_disconnect()
diff --git a/sound/usb/card.h b/sound/usb/card.h
index 02e4ea898db5..ed4a664e24e5 100644
--- a/sound/usb/card.h
+++ b/sound/usb/card.h
@@ -217,4 +217,15 @@ struct snd_usb_platform_ops {
  int snd_usb_register_platform_ops(struct snd_usb_platform_ops *ops);
  int snd_usb_unregister_platform_ops(void);
+
+#if IS_ENABLED(CONFIG_SND_USB_AUDIO)
+struct snd_usb_stream *snd_usb_find_suppported_substream(int card_idx,
+            struct snd_pcm_hw_params *params, int direction);
+#else
+static struct snd_usb_stream *snd_usb_find_suppported_substream(int card_idx,
+            struct snd_pcm_hw_params *params, int direction)
+{
+    return NULL;
+}
+#endif /* IS_ENABLED(CONFIG_SND_USB_AUDIO) */

The usefulness of ifdef guard here is doubtful, IMO.  This header is
only for USB-audio driver enablement, and not seen as generic
helpers.  So, just add the new function declarations without dummy
definitions.


Got it, will remove it.  We also have a dependency in place for the qc_audio_offload driver and SND USB AUDIO in the Kconfig.


Looking at this again after trying some mixed Kconfig settings. These declarations aren't specific for USB-audio. They are helpers that are exposed to soc usb, so that it can do some basic verification with soc usb before allowing the enable stream to continue. Since the ASoC layer doesn't have insight on what audio profiles are supported by the usb device, this API will ensure that the request profile is supported.

Issues are seen when we disable SND USB audio config and enable the ASoC parts.

Thanks
Wesley Cheng



[Index of Archives]     [ALSA User]     [Linux Audio Users]     [Pulse Audio]     [Kernel Archive]     [Asterisk PBX]     [Photo Sharing]     [Linux Sound]     [Video 4 Linux]     [Gimp]     [Yosemite News]

  Powered by Linux