LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Sultan Alsawaf <sultanxda@gmail.com>
To: "Theodore Y. Ts'o" <tytso@mit.edu>,
"Jason A. Donenfeld" <Jason@zx2c4.com>,
Pavel Machek <pavel@ucw.cz>, LKML <linux-kernel@vger.kernel.org>,
Jann Horn <jannh@google.com>
Subject: Re: Linux messages full of `random: get_random_u32 called from`
Date: Sun, 29 Apr 2018 21:34:45 -0700 [thread overview]
Message-ID: <20180430043445.t7wkykxzkhex2isi@sultan-box> (raw)
In-Reply-To: <20180430001106.GS5965@thunk.org>
On Sun, Apr 29, 2018 at 08:11:07PM -0400, Theodore Y. Ts'o wrote:
>
> What your patch does is assume that there is a full bit of uncertainty
> that can be obtained from the information gathered from each
> interrupt. I *might* be willing to assume that to be valid on x86
> systems that have a high resolution cycle counter. But on ARM
> platforms, especially during system bootup when the user isn't typing
> anything and SSD's and flash storage tend to have very predictable
> timing patterns? Not a bet I'd be willing to take. Even with a cycle
> counter, there's a reason why we assumed that we need to mix in timing
> results from 64 interrupts or one second's worth before we would give
> a single bit's worth of entropy credit.
>
> - Ted
What about abusing high-resolution timers to get entropy? Since hrtimers can't
make guarantees down to the nanosecond, there's always a skew between the
requested expiry time and the actual expiry time.
Please see the attached patch and let me know just how horrible it is.
Sultan
>From b0d21c38558c661531d4cb46816fbb36b874a169 Mon Sep 17 00:00:00 2001
From: Sultan Alsawaf <sultanxda@gmail.com>
Date: Sun, 29 Apr 2018 21:28:08 -0700
Subject: [PATCH] random: use high-res timers to generate entropy until crng
init is done
---
drivers/char/random.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/drivers/char/random.c b/drivers/char/random.c
index d9e38523b383..af2d60bbcec3 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -286,6 +286,7 @@
#define OUTPUT_POOL_WORDS (1 << (OUTPUT_POOL_SHIFT-5))
#define SEC_XFER_SIZE 512
#define EXTRACT_SIZE 10
+#define ENTROPY_GEN_INTVL_NS (1 * NSEC_PER_MSEC)
#define LONGS(x) (((x) + sizeof(unsigned long) - 1)/sizeof(unsigned long))
@@ -408,6 +409,8 @@ static struct fasync_struct *fasync;
static DEFINE_SPINLOCK(random_ready_list_lock);
static LIST_HEAD(random_ready_list);
+static struct hrtimer entropy_gen_hrtimer;
+
struct crng_state {
__u32 state[16];
unsigned long init_time;
@@ -2287,3 +2290,47 @@ void add_hwgenerator_randomness(const char *buffer, size_t count,
credit_entropy_bits(poolp, entropy);
}
EXPORT_SYMBOL_GPL(add_hwgenerator_randomness);
+
+/*
+ * Generate entropy on init using high-res timers. Although high-res timers
+ * provide nanosecond precision, they don't actually honor requests to the
+ * nanosecond. The skew between the expected time difference in nanoseconds and
+ * the actual time difference can be used as a way to generate entropy on boot
+ * for machines that lack sufficient boot-time entropy.
+ */
+static enum hrtimer_restart entropy_timer_cb(struct hrtimer *timer)
+{
+ static u64 prev_ns;
+ u64 curr_ns, delta;
+
+ if (crng_ready())
+ return HRTIMER_NORESTART;
+
+ curr_ns = ktime_get_mono_fast_ns();
+ delta = curr_ns - prev_ns;
+
+ add_interrupt_randomness(delta);
+
+ /* Use the hrtimer skew to make the next interval more unpredictable */
+ if (likely(prev_ns))
+ hrtimer_add_expires_ns(timer, delta);
+ else
+ hrtimer_add_expires_ns(timer, ENTROPY_GEN_INTVL_NS);
+
+ prev_ns = curr_ns;
+ return HRTIMER_RESTART;
+}
+
+static int entropy_gen_hrtimer_init(void)
+{
+ if (!IS_ENABLED(CONFIG_HIGH_RES_TIMERS))
+ return 0;
+
+ hrtimer_init(&entropy_gen_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+
+ entropy_gen_hrtimer.function = entropy_timer_cb;
+ hrtimer_start(&entropy_gen_hrtimer, ns_to_ktime(ENTROPY_GEN_INTVL_NS),
+ HRTIMER_MODE_REL);
+ return 0;
+}
+core_initcall(entropy_gen_hrtimer_init);
--
2.14.1
next prev parent reply other threads:[~2018-04-30 4:34 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-26 4:11 Sultan Alsawaf
2018-04-26 5:00 ` Theodore Y. Ts'o
2018-04-26 5:05 ` Sultan Alsawaf
2018-04-26 7:32 ` Theodore Y. Ts'o
2018-04-26 15:17 ` Sultan Alsawaf
2018-04-26 19:25 ` Theodore Y. Ts'o
2018-04-26 20:22 ` Sultan Alsawaf
2018-04-26 20:47 ` Christian Brauner
2018-04-27 0:00 ` Theodore Y. Ts'o
2018-04-27 15:38 ` Jason A. Donenfeld
2018-04-27 19:14 ` Theodore Y. Ts'o
2018-04-26 23:56 ` Theodore Y. Ts'o
2018-04-27 5:20 ` Sultan Alsawaf
2018-04-27 20:10 ` Theodore Y. Ts'o
2018-04-27 22:59 ` Sultan Alsawaf
2018-04-29 14:32 ` Pavel Machek
2018-04-29 17:05 ` Sultan Alsawaf
2018-04-29 18:41 ` Pavel Machek
2018-04-29 20:20 ` Sultan Alsawaf
2018-04-29 21:18 ` Pavel Machek
2018-04-29 21:34 ` Sultan Alsawaf
2018-04-29 22:05 ` Theodore Y. Ts'o
2018-04-29 22:26 ` Sultan Alsawaf
2018-04-29 22:43 ` Jason A. Donenfeld
2018-04-29 22:49 ` Sultan Alsawaf
2018-04-30 0:11 ` Theodore Y. Ts'o
2018-04-30 4:34 ` Sultan Alsawaf [this message]
2018-04-30 16:11 ` Theodore Y. Ts'o
2018-05-01 19:53 ` Pavel Machek
2018-04-29 22:43 ` Pavel Machek
2018-04-30 0:32 ` Laura Abbott
2018-04-30 21:12 ` Jeremy Cline
2018-05-01 11:52 ` Justin Forbes
2018-05-01 12:55 ` Theodore Y. Ts'o
2018-05-01 22:35 ` Justin Forbes
2018-05-02 0:02 ` Theodore Y. Ts'o
2018-05-02 12:09 ` Justin Forbes
2018-05-02 16:26 ` Theodore Y. Ts'o
2018-05-02 17:49 ` Laura Abbott
2018-05-02 22:25 ` Theodore Y. Ts'o
2018-05-03 6:19 ` Pavel Machek
2018-05-03 12:23 ` Justin Forbes
2018-05-02 0:43 ` Sultan Alsawaf
2018-05-02 0:56 ` Theodore Y. Ts'o
2018-05-02 1:11 ` Sultan Alsawaf
2018-04-29 18:30 ` Sultan Alsawaf
2018-04-29 20:08 ` Theodore Y. Ts'o
2018-05-18 1:27 ` Trent Piepho
2018-05-18 2:32 ` Theodore Y. Ts'o
2018-05-18 22:56 ` Trent Piepho
2018-05-18 23:22 ` Theodore Y. Ts'o
2018-05-21 18:39 ` Trent Piepho
2018-04-29 14:29 ` Pavel Machek
-- strict thread matches above, loose matches on Subject: below --
2018-04-24 11:48 Paul Menzel
2018-04-24 13:56 ` Theodore Y. Ts'o
2018-04-24 14:30 ` Paul Menzel
2018-04-24 15:49 ` Theodore Y. Ts'o
2018-04-24 15:56 ` Paul Menzel
2018-04-25 7:41 ` Theodore Y. Ts'o
2018-04-26 3:48 ` Paul Menzel
2018-04-29 14:22 ` Pavel Machek
2018-04-29 23:02 ` Dave Jones
2018-04-29 23:07 ` Dave Jones
2018-04-30 0:21 ` Theodore Y. Ts'o
2018-04-26 5:51 ` Pavel Machek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180430043445.t7wkykxzkhex2isi@sultan-box \
--to=sultanxda@gmail.com \
--cc=Jason@zx2c4.com \
--cc=jannh@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
--cc=tytso@mit.edu \
--subject='Re: Linux messages full of `random: get_random_u32 called from`' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).