Andreas Rheinhardt:
> From: Anton Khirnov <an...@khirnov.net>
>
> Many AVFifoBuffer users operate on fixed-size elements (e.g. pointers),
> but the current FIFO API deals exclusively in bytes, requiring extra
> complexity in all these callers.
>
> Add a new AVFifo API creating a FIFO with an element size
> that may be larger than a byte. All operations on such a FIFO then
> operate on complete elements.
>
> This API does not reuse AVFifoBuffer and its API at all, but instead uses
> an opaque struct called AVFifo. The AVFifoBuffer API will be deprecated
> in a future commit once all of its users have been switched to the new
> API.
>
> Not reusing AVFifoBuffer also allowed to use the full range of size_t
> from the beginning.
> ---
> doc/APIchanges | 9 ++
> libavutil/fifo.c | 224 ++++++++++++++++++++++++++++++++++++++++++++
> libavutil/fifo.h | 179 +++++++++++++++++++++++++++++++++++
> libavutil/version.h | 2 +-
> 4 files changed, 413 insertions(+), 1 deletion(-)
>
> diff --git a/doc/APIchanges b/doc/APIchanges
> index 8df0364e4c..57a9df9bef 100644
> --- a/doc/APIchanges
> +++ b/doc/APIchanges
> @@ -14,6 +14,15 @@ libavutil: 2021-04-27
>
> API changes, most recent first:
>
> +2022-01-xx - xxxxxxxxxx - lavu 57.19.100 - fifo.h
> + Add a new FIFO API, which allows setting a FIFO element size.
> + This API operates on these elements rather than on bytes.
> + Add av_fifo_alloc2(), av_fifo_elem_size(), av_fifo_can_read(),
> + av_fifo_can_write(), av_fifo_grow2(), av_fifo_drain2(), av_fifo_write(),
> + av_fifo_write_from_cb(), av_fifo_read(), av_fifo_read_to_cb(),
> + av_fifo_peek(), av_fifo_peek_to_cb(), av_fifo_drain2(), av_fifo_reset2(),
> + av_fifo_freep2().
> +
> 2022-01-04 - 78dc21b123e - lavu 57.16.100 - frame.h
> Add AV_FRAME_DATA_DOVI_METADATA.
>
> diff --git a/libavutil/fifo.c b/libavutil/fifo.c
> index 55621f0dca..0e0d84258f 100644
> --- a/libavutil/fifo.c
> +++ b/libavutil/fifo.c
> @@ -26,6 +26,230 @@
> #include "common.h"
> #include "fifo.h"
>
> +struct AVFifo {
> + uint8_t *buffer;
> +
> + size_t elem_size, nb_elems;
> + size_t offset_r, offset_w;
> + // distinguishes the ambiguous situation offset_r == offset_w
> + int is_empty;
> +};
> +
> +AVFifo *av_fifo_alloc2(size_t nb_elems, size_t elem_size,
> + unsigned int flags)
> +{
> + AVFifo *f;
> + void *buffer;
> +
> + if (!elem_size)
> + return NULL;
> +
> + buffer = av_realloc_array(NULL, nb_elems, elem_size);
After Anton pointed out that it is ill-defined what av_realloc_array()
does in case one requests zero elements (it allocates a single byte,
although it is documented to free the buffer provided to it) this has
been changed to only allocate anything in case a positive number of
elements has been requested:
+ void *buffer = NULL;
+
+ if (!elem_size)
+ return NULL;
+
+ if (nb_elems) {
+ buffer = av_realloc_array(NULL, nb_elems, elem_size);
+ if (!buffer)
+ return NULL;
+ }
+ f = av_mallocz(sizeof(*f));
+ if (!f) {
+ av_free(buffer);
+ return NULL;
+ }
I intend to apply this set with this change and his change requested in
https://ffmpeg.org/pipermail/ffmpeg-devel/2022-February/292349.html
applied (and with another trivial change in qsvdec necessitated by
8ca06a8148) tomorrow unless there are objections.
> + if (!buffer)
> + return NULL;
> + f = av_mallocz(sizeof(*f));
> + if (!f) {
> + av_free(buffer);
> + return NULL;
> + }
> + f->buffer = buffer;
> + f->nb_elems = nb_elems;
> + f->elem_size = elem_size;
> + f->is_empty = 1;
> +
> + return f;
> +}
> +
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
