linux-kernel - [PATCH 4.19 081/234] random: remove batched entropy locking

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220623164345.350543448@linuxfoundation.org>
Date:   Thu, 23 Jun 2022 18:42:28 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Dominik Brodowski <linux@...inikbrodowski.net>,
        Eric Biggers <ebiggers@...gle.com>,
        Andy Lutomirski <luto@...nel.org>,
        Jonathan Neuschäfer <j.neuschaefer@....net>,
        "Jason A. Donenfeld" <Jason@...c4.com>
Subject: [PATCH 4.19 081/234] random: remove batched entropy locking

From: "Jason A. Donenfeld" <Jason@...c4.com>

commit 77760fd7f7ae3dfd03668204e708d1568d75447d upstream.

Rather than use spinlocks to protect batched entropy, we can instead
disable interrupts locally, since we're dealing with per-cpu data, and
manage resets with a basic generation counter. At the same time, we
can't quite do this on PREEMPT_RT, where we still want spinlocks-as-
mutexes semantics. So we use a local_lock_t, which provides the right
behavior for each. Because this is a per-cpu lock, that generation
counter is still doing the necessary CPU-to-CPU communication.

This should improve performance a bit. It will also fix the linked splat
that Jonathan received with a PROVE_RAW_LOCK_NESTING=y.

Reviewed-by: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Reviewed-by: Dominik Brodowski <linux@...inikbrodowski.net>
Reviewed-by: Eric Biggers <ebiggers@...gle.com>
Suggested-by: Andy Lutomirski <luto@...nel.org>
Reported-by: Jonathan Neuschäfer <j.neuschaefer@....net>
Tested-by: Jonathan Neuschäfer <j.neuschaefer@....net>
Link: https://lore.kernel.org/lkml/YfMa0QgsjCVdRAvJ@latitude/
Signed-off-by: Jason A. Donenfeld <Jason@...c4.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
 drivers/char/random.c |   58 +++++++++++++++++++++++---------------------------
 1 file changed, 27 insertions(+), 31 deletions(-)

--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -1719,13 +1719,15 @@ struct ctl_table random_table[] = {
 };
 #endif	/* CONFIG_SYSCTL */
 
+static atomic_t batch_generation = ATOMIC_INIT(0);
+
 struct batched_entropy {
 	union {
 		u64 entropy_u64[CHACHA20_BLOCK_SIZE / sizeof(u64)];
 		u32 entropy_u32[CHACHA20_BLOCK_SIZE / sizeof(u32)];
 	};
 	unsigned int position;
-	spinlock_t batch_lock;
+	int generation;
 };
 
 /*
@@ -1736,9 +1738,7 @@ struct batched_entropy {
  * wait_for_random_bytes() should be called and return 0 at least once at any
  * point prior.
  */
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64) = {
-	.batch_lock = __SPIN_LOCK_UNLOCKED(batched_entropy_u64.lock),
-};
+static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64);
 
 u64 get_random_u64(void)
 {
@@ -1746,67 +1746,63 @@ u64 get_random_u64(void)
 	unsigned long flags;
 	struct batched_entropy *batch;
 	static void *previous;
+	int next_gen;
 
 	warn_unseeded_randomness(&previous);
 
+	local_irq_save(flags);
 	batch = raw_cpu_ptr(&batched_entropy_u64);
-	spin_lock_irqsave(&batch->batch_lock, flags);
-	if (batch->position % ARRAY_SIZE(batch->entropy_u64) == 0) {
+
+	next_gen = atomic_read(&batch_generation);
+	if (batch->position % ARRAY_SIZE(batch->entropy_u64) == 0 ||
+	    next_gen != batch->generation) {
 		extract_crng((u8 *)batch->entropy_u64);
 		batch->position = 0;
+		batch->generation = next_gen;
 	}
+
 	ret = batch->entropy_u64[batch->position++];
-	spin_unlock_irqrestore(&batch->batch_lock, flags);
+	local_irq_restore(flags);
 	return ret;
 }
 EXPORT_SYMBOL(get_random_u64);
 
-static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32) = {
-	.batch_lock = __SPIN_LOCK_UNLOCKED(batched_entropy_u32.lock),
-};
+static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32);
+
 u32 get_random_u32(void)
 {
 	u32 ret;
 	unsigned long flags;
 	struct batched_entropy *batch;
 	static void *previous;
+	int next_gen;
 
 	warn_unseeded_randomness(&previous);
 
+	local_irq_save(flags);
 	batch = raw_cpu_ptr(&batched_entropy_u32);
-	spin_lock_irqsave(&batch->batch_lock, flags);
-	if (batch->position % ARRAY_SIZE(batch->entropy_u32) == 0) {
+
+	next_gen = atomic_read(&batch_generation);
+	if (batch->position % ARRAY_SIZE(batch->entropy_u32) == 0 ||
+	    next_gen != batch->generation) {
 		extract_crng((u8 *)batch->entropy_u32);
 		batch->position = 0;
+		batch->generation = next_gen;
 	}
+
 	ret = batch->entropy_u32[batch->position++];
-	spin_unlock_irqrestore(&batch->batch_lock, flags);
+	local_irq_restore(flags);
 	return ret;
 }
 EXPORT_SYMBOL(get_random_u32);
 
 /* It's important to invalidate all potential batched entropy that might
  * be stored before the crng is initialized, which we can do lazily by
- * simply resetting the counter to zero so that it's re-extracted on the
- * next usage. */
+ * bumping the generation counter.
+ */
 static void invalidate_batched_entropy(void)
 {
-	int cpu;
-	unsigned long flags;
-
-	for_each_possible_cpu(cpu) {
-		struct batched_entropy *batched_entropy;
-
-		batched_entropy = per_cpu_ptr(&batched_entropy_u32, cpu);
-		spin_lock_irqsave(&batched_entropy->batch_lock, flags);
-		batched_entropy->position = 0;
-		spin_unlock(&batched_entropy->batch_lock);
-
-		batched_entropy = per_cpu_ptr(&batched_entropy_u64, cpu);
-		spin_lock(&batched_entropy->batch_lock);
-		batched_entropy->position = 0;
-		spin_unlock_irqrestore(&batched_entropy->batch_lock, flags);
-	}
+	atomic_inc(&batch_generation);
 }
 
 /**