[PATCH 4.9 035/159] random: avoid reading two cache lines on irq randomness

From: Greg Kroah-Hartman
Date: Mon Oct 24 2022 - 07:45:44 EST


From: Jason A. Donenfeld <Jason@xxxxxxxxx>

commit 9ee0507e896b45af6d65408c77815800bce30008 upstream.

In order to avoid reading and dirtying two cache lines on every IRQ,
move the work_struct to the bottom of the fast_pool struct. add_
interrupt_randomness() always touches .pool and .count, which are
currently split, because .mix pushes everything down. Instead, move .mix
to the bottom, so that .pool and .count are always in the first cache
line, since .mix is only accessed when the pool is full.

Fixes: 58340f8e952b ("random: defer fast pool mixing to worker")
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Signed-off-by: Jason A. Donenfeld <Jason@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/char/random.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -891,10 +891,10 @@ void __init add_bootloader_randomness(co
}

struct fast_pool {
- struct work_struct mix;
unsigned long pool[4];
unsigned long last;
unsigned int count;
+ struct work_struct mix;
};

static DEFINE_PER_CPU(struct fast_pool, irq_randomness) = {