- Use exact numerator/denominator from spec for CMTime to avoid rounding errors.
- Wrap frame rate setting in @try-catch to prevent crashes on strict drivers.
V4L2 is able to advertise that a video device is able to display any frame
interval within a continuous range. SDL does not allow advertising this and
only exposes discrete frame intervals. To work around this, SDL attempted to
generate a subset of the range with a fixed interval. Unfortunately, the way
this was accomplish is inherently broken and led to attempting to allocate a
very large number of formats per resolution and colorspace. With the Magewell
Pro Capture HDMI, which can expose FRMSIZE_TYPE_CONTINUOUS as well, this can
expose a truly astronomical number of formats, exceeding 1 PB of RAM. This will
lead to an OOM kill for any process that tries to initialize the camera
subsystem.
This patch just tests to see if some common frame rates are within the
contiuous range and expose those. SDL still does not handle
FRMSIZE_TYPE_CONTINUOUS in a graceful way so it still uses over a gigabyte of
RAM for each possible combination of sizes, but with this patch it no longer
leads to an OOM kill. The API will need amending for proper support for both
continuous frame sizes and frame intervals.
This still logs backend choices, but we might use it for other things. For
example, what Android device is being used, or all the devices we enumerated,
etc.
Ideally this eventually logs all the stuff we often have to ask followup
questions about.
The surrounding code in all of these instances expects the Unicode
variants. Previously, this code mixed Unicode and ANSI/ASCII calls if
`UNICODE` was undefined, which caused type and logic errors. Explicitly
spelling out the W removes any reliance on that macro.
The Debian 12 'bookworm' stable release only has 0.3.65, which is earlier
than the introduction of that symbol. Similarly, Steam Runtime 3 'sniper'
and Steam Runtime 2 'soldier' have a backport of 0.3.65 from Debian 12,
so they have the same limitation.
steamrt/tasks#686
Signed-off-by: Simon McVittie <smcv@collabora.com>
This helps the Pipewire camera driver to access cameras
in a sandboxed environment without host Pipewire socket access.
Unlike other platforms, no event is sent when the user rejects
camera access. This is because there is no mechanism to query
cameras through the portal, and we only obtain access to the
Pipewire fd if the user accepts the request. The Pipewire driver
will attempt to open the host socket instead.
This was intended to make the API public, so SDL_hashtable.h got an extreme
documentation makeover, but for now this remains a private header.
This makes several significant interface changes to SDL_HashTable, and
improves code that makes use of it in various ways.
- The ability to make "stackable" tables is removed. Apparently this still
worked with the current implementation, but I could see a future
implementation struggle mightily to support this. It'll be better for
something external to build on top of the table if it needs it, inserting a
linked list of stacked items as the hash values and managing them separately.
There was only one place in SDL using this, unnecessarily, and that has also
been cleaned up to not need it.
- You no longer specify "buckets" when creating a table, but rather an
estimated number of items the table is meant to hold. The bucket count was
crucial to our classic hashtable implementation, but meant less once we
moved to an Open Addressing implementation anyhow, since the bucket count
isn't static (and they aren't really "buckets" anymore either). Now you
can just report how many items you think the hash will hold and SDL will
allocate a reasonable default for you...or 0 to not guess, and SDL will
start small and grow as necessary, which is often the correct thing to do.
- There's no more SDL_IterateHashTableKey because there's no more "stackable"
hash tables.
- SDL_IterateHashTable() now uses a callback, which matches other parts of SDL,
and also lets us hold the read-lock for the entire iteration and get rid of
the goofy iterator state variable.
- SDL_InsertIntoHashTable() now lets you specify whether to replace existing
keys or fail if the key already exists.
- Callbacks now use SDL conventions (userdata as the first param).
- Other naming convention fixes.
I discovered we use a lot of hash tables in SDL3 internally. :) So the bulk
of this work is fixing up that code to use the new interfaces, and simplifying
things (like checking for an item to remove it if it already exists before
inserting a replacement...just do the insert atomically, it'll do all that
for you!).
