ALSA: compress_offload: introduce accel operation mode

There is a requirement to expose the audio hardware that accelerates various
tasks for user space such as sample rate converters, compressed
stream decoders, etc.

This is description for the API extension for the compress ALSA API which
is able to handle "tasks" that are not bound to real-time operations
and allows for the serialization of operations.

For details, refer to "compress-accel.rst" document.

Cc: Mark Brown <broonie@kernel.org>
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Cc: Nicolas Dufresne <nicolas@ndufresne.ca>
Cc: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Cc: Pierre-Louis Bossart <pierre-louis.bossart@linux.dev>
Cc: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Tested-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Link: https://patch.msgid.link/20241002093904.1809799-1-perex@perex.cz

2 files changed, 108 insertions, 2 deletions

diff --git a/include/sound/compress_driver.h b/include/sound/compress_driver.h
index bcf872c17dd3..5db4881b0681 100644
--- a/include/sound/compress_driver.h
+++ b/include/sound/compress_driver.h
@@ -20,6 +20,30 @@
 struct snd_compr_ops;
 
 /**
+ * struct snd_compr_task_runtime: task runtime description
+ * @list: list of all managed tasks
+ * @input: input DMA buffer
+ * @output: output DMA buffer
+ * @seqno: sequence number
+ * @input_size: really used data in the input buffer
+ * @output_size: really used data in the output buffer
+ * @flags: see SND_COMPRESS_TFLG_*
+ * @state: actual task state
+ * @private_value: used by the lowlevel driver (opaque)
+ */
+struct snd_compr_task_runtime {
+	struct list_head list;
+	struct dma_buf *input;
+	struct dma_buf *output;
+	u64 seqno;
+	u64 input_size;
+	u64 output_size;
+	u32 flags;
+	u8 state;
+	void *private_value;
+};
+
+/**
  * struct snd_compr_runtime: runtime stream description
  * @state: stream state
  * @ops: pointer to DSP callbacks
@@ -37,6 +61,10 @@ struct snd_compr_ops;
  * @dma_addr: physical buffer address (not accessible from main CPU)
  * @dma_bytes: size of DMA area
  * @dma_buffer_p: runtime dma buffer pointer
+ * @active_tasks: count of active tasks
+ * @total_tasks: count of all tasks
+ * @task_seqno: last task sequence number (!= 0)
+ * @tasks: list of all tasks
  */
 struct snd_compr_runtime {
 	snd_pcm_state_t state;
@@ -54,6 +82,13 @@ struct snd_compr_runtime {
 	dma_addr_t dma_addr;
 	size_t dma_bytes;
 	struct snd_dma_buffer *dma_buffer_p;
+
+#if IS_ENABLED(CONFIG_SND_COMPRESS_ACCEL)
+	u32 active_tasks;
+	u32 total_tasks;
+	u64 task_seqno;
+	struct list_head tasks;
+#endif
 };
 
 /**
@@ -132,6 +167,12 @@ struct snd_compr_ops {
 			struct snd_compr_caps *caps);
 	int (*get_codec_caps) (struct snd_compr_stream *stream,
 			struct snd_compr_codec_caps *codec);
+#if IS_ENABLED(CONFIG_SND_COMPRESS_ACCEL)
+	int (*task_create) (struct snd_compr_stream *stream, struct snd_compr_task_runtime *task);
+	int (*task_start) (struct snd_compr_stream *stream, struct snd_compr_task_runtime *task);
+	int (*task_stop) (struct snd_compr_stream *stream, struct snd_compr_task_runtime *task);
+	int (*task_free) (struct snd_compr_stream *stream, struct snd_compr_task_runtime *task);
+#endif
 };
 
 /**
@@ -242,4 +283,9 @@ int snd_compr_free_pages(struct snd_compr_stream *stream);
 int snd_compr_stop_error(struct snd_compr_stream *stream,
 			 snd_pcm_state_t state);
 
+#if IS_ENABLED(CONFIG_SND_COMPRESS_ACCEL)
+void snd_compr_task_finished(struct snd_compr_stream *stream,
+			     struct snd_compr_task_runtime *task);
+#endif
+
 #endif
diff --git a/include/uapi/sound/compress_offload.h b/include/uapi/sound/compress_offload.h
index d185957f3fe0..98772b0cbcb7 100644
--- a/include/uapi/sound/compress_offload.h
+++ b/include/uapi/sound/compress_offload.h
@@ -14,7 +14,7 @@
 #include <sound/compress_params.h>
 
 
-#define SNDRV_COMPRESS_VERSION SNDRV_PROTOCOL_VERSION(0, 2, 0)
+#define SNDRV_COMPRESS_VERSION SNDRV_PROTOCOL_VERSION(0, 3, 0)
 /**
  * struct snd_compressed_buffer - compressed buffer
  * @fragment_size: size of buffer fragment in bytes
@@ -68,7 +68,8 @@ struct snd_compr_avail {
 
 enum snd_compr_direction {
 	SND_COMPRESS_PLAYBACK = 0,
-	SND_COMPRESS_CAPTURE
+	SND_COMPRESS_CAPTURE,
+	SND_COMPRESS_ACCEL
 };
 
 /**
@@ -127,6 +128,57 @@ struct snd_compr_metadata {
 	 __u32 value[8];
 } __attribute__((packed, aligned(4)));
 
+/* flags for struct snd_compr_task */
+#define SND_COMPRESS_TFLG_NEW_STREAM		(1<<0)	/* mark for the new stream data */
+
+/**
+ * struct snd_compr_task - task primitive for non-realtime operation
+ * @seqno: sequence number (task identifier)
+ * @origin_seqno: previous sequence number (task identifier) - for reuse
+ * @input_fd: data input file descriptor (dma-buf)
+ * @output_fd: data output file descriptor (dma-buf)
+ * @input_size: filled data in bytes (from caller, must not exceed fragment size)
+ * @flags: see SND_COMPRESS_TFLG_* defines
+ */
+struct snd_compr_task {
+	__u64 seqno;
+	__u64 origin_seqno;
+	int input_fd;
+	int output_fd;
+	__u64 input_size;
+	__u32 flags;
+	__u8 reserved[16];
+} __attribute__((packed, aligned(4)));
+
+/**
+ * enum snd_compr_state - task state
+ * @SND_COMPRESS_TASK_STATE_IDLE: task is not queued
+ * @SND_COMPRESS_TASK_STATE_ACTIVE: task is in the queue
+ * @SND_COMPRESS_TASK_STATE_FINISHED: task was processed, output is available
+ */
+enum snd_compr_state {
+	SND_COMPRESS_TASK_STATE_IDLE = 0,
+	SND_COMPRESS_TASK_STATE_ACTIVE,
+	SND_COMPRESS_TASK_STATE_FINISHED
+};
+
+/**
+ * struct snd_compr_task_status - task status
+ * @seqno: sequence number (task identifier)
+ * @input_size: filled data in bytes (from user space)
+ * @output_size: filled data in bytes (from driver)
+ * @output_flags: reserved for future (all zeros - from driver)
+ * @state: actual task state (SND_COMPRESS_TASK_STATE_*)
+ */
+struct snd_compr_task_status {
+	__u64 seqno;
+	__u64 input_size;
+	__u64 output_size;
+	__u32 output_flags;
+	__u8 state;
+	__u8 reserved[15];
+} __attribute__((packed, aligned(4)));
+
 /*
  * compress path ioctl definitions
  * SNDRV_COMPRESS_GET_CAPS: Query capability of DSP
@@ -164,6 +216,14 @@ struct snd_compr_metadata {
 #define SNDRV_COMPRESS_DRAIN		_IO('C', 0x34)
 #define SNDRV_COMPRESS_NEXT_TRACK	_IO('C', 0x35)
 #define SNDRV_COMPRESS_PARTIAL_DRAIN	_IO('C', 0x36)
+
+
+#define SNDRV_COMPRESS_TASK_CREATE	_IOWR('C', 0x60, struct snd_compr_task)
+#define SNDRV_COMPRESS_TASK_FREE	_IOW('C', 0x61, __u64)
+#define SNDRV_COMPRESS_TASK_START	_IOWR('C', 0x62, struct snd_compr_task)
+#define SNDRV_COMPRESS_TASK_STOP	_IOW('C', 0x63, __u64)
+#define SNDRV_COMPRESS_TASK_STATUS	_IOWR('C', 0x68, struct snd_compr_task_status)
+
 /*
  * TODO
  * 1. add mmap support