From: Jason A. Donenfeld <Jason@zx2c4.com>
Date: Thu, 22 Sep 2022 16:46:04 +0000 (+0200)
Subject: random: avoid reading two cache lines on irq randomness
X-Git-Url: http://git.maquefel.me/?a=commitdiff_plain;h=9ee0507e896b45af6d65408c77815800bce30008;p=linux.git

random: avoid reading two cache lines on irq randomness

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

diff --git a/drivers/char/random.c b/drivers/char/random.c
index 2f370aa248b22..a90d96f4b3bb4 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -918,10 +918,10 @@ EXPORT_SYMBOL_GPL(unregister_random_vmfork_notifier);
 #endif
 
 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) = {