Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: tcrypt - do not attempt to write to readonly variable
  random: update interface comments to reflect reality
  crypto: picoxcell - add support for the picoxcell crypto engines
  crypto: sha1 - Add test vector to test partial block processing
  hwrng: omap - Convert release_resource to release_region/release_mem_region
  crypto: aesni-intel - Fix remaining leak in rfc4106_set_hash_key
  crypto: omap-sham - don't treat NULL clk as an error
  crypto: omap-aes - don't treat NULL clk as an error
  crypto: testmgr - mark ghash as fips_allowed
  crypto: testmgr - mark xts(aes) as fips_allowed
  crypto: skcipher - remove redundant NULL check
  hwrng: pixocell - add support for picoxcell TRNG
  crypto: aesni-intel - Don't leak memory in rfc4106_set_hash_subkey

1 files changed, 10 insertions, 3 deletions

diff --git a/drivers/char/random.c b/drivers/char/random.c
index 72a4fcb1774..5e29e8031bb 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -128,6 +128,7 @@
  *      void add_input_randomness(unsigned int type, unsigned int code,
  *                                unsigned int value);
  *      void add_interrupt_randomness(int irq);
+ *      void add_disk_randomness(struct gendisk *disk);
  *
  * add_input_randomness() uses the input layer interrupt timing, as well as
  * the event type information from the hardware.
@@ -136,9 +137,15 @@
  * inputs to the entropy pool.  Note that not all interrupts are good
  * sources of randomness!  For example, the timer interrupts is not a
  * good choice, because the periodicity of the interrupts is too
- * regular, and hence predictable to an attacker.  Disk interrupts are
- * a better measure, since the timing of the disk interrupts are more
- * unpredictable.
+ * regular, and hence predictable to an attacker.  Network Interface
+ * Controller interrupts are a better measure, since the timing of the
+ * NIC interrupts are more unpredictable.
+ *
+ * add_disk_randomness() uses what amounts to the seek time of block
+ * layer request events, on a per-disk_devt basis, as input to the
+ * entropy pool. Note that high-speed solid state drives with very low
+ * seek times do not make for good sources of entropy, as their seek
+ * times are usually fairly consistent.
  *
  * All of these routines try to estimate how many bits of randomness a
  * particular randomness source.  They do this by keeping track of the