random: clear fast pool, crng, and batches in cpuhp bring up

commit 3191dd5a1179ef0fad5a050a1702ae98b6251e8f upstream.

For the irq randomness fast pool, rather than having to use expensive
atomics, which were visibly the most expensive thing in the entire irq
handler, simply take care of the extreme edge case of resetting count to
zero in the cpuhp online handler, just after workqueues have been
reenabled. This simplifies the code a bit and lets us use vanilla
variables rather than atomics, and performance should be improved.

As well, very early on when the CPU comes up, while interrupts are still
disabled, we clear out the per-cpu crng and its batches, so that it
always starts with fresh randomness.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Sultan Alsawaf <sultan@kerneltoast.com>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Jason A. Donenfeld
2022-02-13 22:48:04 +01:00
committed by Greg Kroah-Hartman
parent 6e1cb84cc6
commit 5064550d42
4 changed files with 65 additions and 15 deletions

View File

@@ -156,4 +156,9 @@ static inline bool __init arch_get_random_long_early(unsigned long *v)
}
#endif
#ifdef CONFIG_SMP
extern int random_prepare_cpu(unsigned int cpu);
extern int random_online_cpu(unsigned int cpu);
#endif
#endif /* _LINUX_RANDOM_H */