Pull random number generator updates from Jason Donenfeld:

 - Dynamically allocate cpumasks off of the stack if the kernel is
   configured for a lot of CPUs, to handle a -Wframe-larger-than case

 - The removal of next_pseudo_random32() after the last user was
   switched over to the prandom interface

 - The removal of get_random_u{8,16,32,64}_wait() functions, as there
   were no users of those at all

 - Some house keeping changes - a few grammar cleanups in the
   comments, system_unbound_wq was renamed to system_dfl_wq, and
   static_key_initialized no longer needs to be checked

* tag 'random-6.19-rc1-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:
  random: complete sentence of comment
  random: drop check for static_key_initialized
  random: remove unused get_random_var_wait functions
  random: replace use of system_unbound_wq with system_dfl_wq
  random: use offstack cpumask when necessary
  prandom: remove next_pseudo_random32
  media: vivid: use prandom
  random: add missing words in function comments

Complete the sentence by adding "is set", rather than having it dangle
as a sentence fragment.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

Commit e871abcda3b6 ("random: handle creditable entropy from atomic
process context") added the use of workqueues, which meant testing
whether the workqueue is valid, but it did not remove the existing check
of whether static keys have been initialized. This static key check is
unnecessary because workqueues are initialized long after it. And
semantically it doesn't make much sense either, because it's not really
directly calling a static key function in the condition.

Remove the now unnecessary check.

Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
[Jason: rewrite commit message with different explanation, rebase on
        random.git, and update code comment.]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

system_unbound_wq has been renamed to system_dfl_wq in 128ea9f6ccfb
("workqueue: Add system_percpu_wq and system_dfl_wq"), so update
random.c's usage of it system_unbound_wq to reflect the new change. The
old system_unbound_wq is slated for removal in the next few cycles.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

The entropy generation function keeps a local cpu mask on the stack,
which can trigger warnings in configurations with a large number of
CPUs:

    drivers/char/random.c:1292:20: error: stack frame size (1288)
    exceeds limit (1280) in 'try_to_generate_entropy' [-Werror,-Wframe-larger-than]

Use the cpumask interface to dynamically allocate it in those
configurations.

Fixes: 1c21fe00eda7 ("random: spread out jitter callback to different CPUs")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

s/good as/as good as/

Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

For consistency with the SHA-1, SHA-2, SHA-3 (in development), and MD5
library APIs, rename blake2s_state to blake2s_ctx.

As a refresher, the ctx name:

- Is a bit shorter.
- Avoids confusion with the compression function state, which is also
  often called the state (but is just part of the full context).
- Is consistent with OpenSSL.

Not a big deal, of course.  But consistency is nice.  With a BLAKE2b
library API about to be added, this is a convenient time to update this.

Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20251018043106.375964-3-ebiggers@kernel.org
Signed-off-by: Eric Biggers <ebiggers@kernel.org>

Reorder the parameters of blake2s() from (out, in, key, outlen, inlen,
keylen) to (key, keylen, in, inlen, out, outlen).

This aligns BLAKE2s with the common conventions of pairing buffers and
their lengths, and having outputs follow inputs.  This is widely used
elsewhere in lib/crypto/ and crypto/, and even elsewhere in the BLAKE2s
code itself such as blake2s_init_key() and blake2s_final().  So
blake2s() was a bit of an exception.

Notably, this results in the same order as hmac_*_usingrawkey().

Note that since the type signature changed, it's not possible for a
blake2s() call site to be silently missed.

Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20251018043106.375964-2-ebiggers@kernel.org
Signed-off-by: Eric Biggers <ebiggers@kernel.org>

Pull rate-limit updates from Paul McKenney:
 "lib/ratelimit: Reduce false-positive and silent misses:

   - Reduce open-coded use of ratelimit_state structure fields.

   - Convert the ->missed field to atomic_t.

   - Count misses that are due to lock contention.

   - Eliminate jiffies=0 special case.

   - Reduce ___ratelimit() false-positive rate limiting (Petr Mladek).

   - Allow zero ->burst to hard-disable rate limiting.

   - Optimize away atomic operations when a miss is guaranteed.

   - Warn if ->interval or ->burst are negative (Petr Mladek).

   - Simplify the resulting code.

  A smoke test and stress test have been created, but they are not yet
  ready for mainline. With luck, we will offer them for the v6.17 merge
  window"

* tag 'ratelimit.2025.05.25a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
  ratelimit: Drop redundant accesses to burst
  ratelimit: Use nolock_ret restructuring to collapse common case code
  ratelimit: Use nolock_ret label to collapse lock-failure code
  ratelimit: Use nolock_ret label to save a couple of lines of code
  ratelimit: Simplify common-case exit path
  ratelimit: Warn if ->interval or ->burst are negative
  ratelimit: Avoid atomic decrement under lock if already rate-limited
  ratelimit: Avoid atomic decrement if already rate-limited
  ratelimit: Don't flush misses counter if RATELIMIT_MSG_ON_RELEASE
  ratelimit: Force re-initialization when rate-limiting re-enabled
  ratelimit: Allow zero ->burst to disable ratelimiting
  ratelimit: Reduce ___ratelimit() false-positive rate limiting
  ratelimit: Avoid jiffies=0 special case
  ratelimit: Count misses due to lock contention
  ratelimit: Convert the ->missed field to atomic_t
  drm/amd/pm: Avoid open-coded use of ratelimit_state structure's internals
  drm/i915: Avoid open-coded use of ratelimit_state structure's ->missed field
  random: Avoid open-coded use of ratelimit_state structure's ->missed field
  ratelimit: Create functions to handle ratelimit_state internals

Pull timer cleanups from Thomas Gleixner:
 "Another set of timer API cleanups:

    - Convert init_timer*(), try_to_del_timer_sync() and
      destroy_timer_on_stack() over to the canonical timer_*()
      namespace convention.

  There is another large conversion pending, which has not been included
  because it would have caused a gazillion of merge conflicts in next.
  The conversion scripts will be run towards the end of the merge window
  and a pull request sent once all conflict dependencies have been
  merged"

* tag 'timers-cleanups-2025-05-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  treewide, timers: Rename destroy_timer_on_stack() as timer_destroy_on_stack()
  treewide, timers: Rename try_to_del_timer_sync() as timer_delete_sync_try()
  timers: Rename init_timers() as timers_init()
  timers: Rename NEXT_TIMER_MAX_DELTA as TIMER_NEXT_MAX_DELTA
  timers: Rename __init_timer_on_stack() as __timer_init_on_stack()
  timers: Rename __init_timer() as __timer_init()
  timers: Rename init_timer_on_stack_key() as timer_init_key_on_stack()
  timers: Rename init_timer_key() as timer_init_key()