Date: 9 Jun 2014 05:23:17 -0400
Message-ID: <20140609092317.17828.qmail@ns.horizon.com>
From: "George Spelvin" <linux@horizon.com>
To: linux@horizon.com, tytso@mit.edu
Subject: Re: [RFC PATCH] drivers/char/random.c: Is reducing locking range like this safe?
Cc: hpa@linux.intel.com, linux-kernel@vger.kernel.org, mingo@kernel.org,
        price@mit.edu
In-Reply-To: <20140609040355.8126.qmail@ns.horizon.com>
Sender: linux-kernel-owner@vger.kernel.org

> Sigh, adventures in "unable to mount root filesystem" currently underway.

PEBKAC resolved.  (I had CONFIG_ATA=m for some testing, but forgot
to change it back to y when compiling a kernel I expected to boot.)

32 MiB write:

	    Without patch		With patch
0 readers: 0.495523 0.494998		0.515995 0.516026
1 reader:  0.842245 0.845308		0.704283 0.704318
2 readers: 1.18658  1.18762		0.904635 0.904844
3 readers: 1.49615  1.49616		1.14311  1.16836

I haven't rebooted back to the without-patch kernel to see if the 20 ms
loss is consistent, but obviously that needs fixing.  (There's more than
just that patch I posted in there.)

Anyway, it's noticeable, but maybe not worth fixing.

The interesting question is whether we should get rid of out[].
That's obscurity that contributes no security, and thus a Bad
Thing by definition.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/