Changes

vty: fix osmo_cpu_sched_vty_init() prototype

Change return type of osmo_cpu_sched_vty_init() from void to int
to match its implementation.

This fixes a prototype mismatch rejected by stricter compilers
(e.g. Emscripten).

Change-Id: Ia86e8063178ad82de098c2ff693cfbcd7e9f2192

vty: make CPU scheduling optional based on platform support

Add configure checks for sched_* APIs and guard their usage
in cpu_sched_vty with HAVE_SCHED_* macros.

If scheduler functions are not available, return -ENOTSUP
instead of failing at build time.

This improves portability on platforms lacking full sched
support (e.g. non-Linux or restricted environments).

Change-Id: Ic5b7e39fac16531d370cb81f769ba87fef18cb81

vty: make CPU scheduling optional based on platform support

Add configure checks for sched_* APIs and guard their usage
in cpu_sched_vty with HAVE_SCHED_* macros.

If scheduler functions are not available, return -ENOTSUP
instead of failing at build time.

Also change osmo_cpu_sched_vty_init() to return int.

This improves portability on platforms lacking full sched
support (e.g. non-Linux or restricted environments).

Change-Id: Ic5b7e39fac16531d370cb81f769ba87fef18cb81

osmo_io: iofd_txqueue_dequeue(): fix double next->io_len decrement

The fill-up logic in iofd_txqueue_dequeue() was incorrectly
decrementing next->io_len twice: once when removing a buffer that
didn't fit, and again when assigning to msg_iovlen.  This may cause
next->io_len to reach 0 and trigger an assert at osmo_io.c:274:

  Assert failed next->io_len > 0

Add separate osmo_io_backpressure_test to verify queue management
under network backpressure conditions.  Use separate .ok files for
poll vs io_uring backends since they produce different I/O patterns.
This unit test reproduces the assert when the fix is not applied.

Change-Id: Ia96e25d11d8d683fa9dd0c56bdac257992edd782
Fixes: 3c2a02db4 ("osmo-io: Put together message buffers when dequeued from tx queue")
Related: OS#6997

osmo_io: iofd_txqueue_dequeue(): fix double next->io_len decrement

The fill-up logic in iofd_txqueue_dequeue() was incorrectly
decrementing next->io_len twice: once when removing a buffer that
didn't fit, and again when assigning to msg_iovlen.  This may cause
next->io_len to reach 0 and trigger an assert at osmo_io.c:274:

  Assert failed next->io_len > 0

Add separate osmo_io_backpressure_test to verify queue management
under network backpressure conditions.  Use separate .ok files for
poll vs io_uring backends since they produce different I/O patterns.
This unit test reproduces the assert when the fix is not applied.

Change-Id: Ia96e25d11d8d683fa9dd0c56bdac257992edd782
Fixes: 3c2a02db4 ("osmo-io: Put together message buffers when dequeued from tx queue")
Related: OS#6997

Bump version: 1.13.1.12-8e6ea → 1.14.0

Change-Id: Ifa079c76148fd25f543b7875d3d89b35c19ffe6b