Messages in this thread | | | Date | Sat, 27 Mar 2004 09:17:29 -0600 | From | Matt Mackall <> | Subject | Re: [PATCH 21/22] /dev/random: kill batching of entropy mixing |
| |
On Sat, Mar 27, 2004 at 01:52:45PM +0000, Jamie Lokier wrote: > Matt Mackall wrote: > > Rather than batching up entropy samples, resulting in longer lock hold > > times when we actually process the samples, mix in samples > > immediately. The trickle code should eliminate almost all the > > additional interrupt-time overhead this would otherwise incur, with or > > without locking. > > What do you mean by "the trickle code"? I didn't see anything in your > patch set which makes the interrupt-time overhead faster.
This code which is in the existing driver to prevent pathological lock contention on large SMP:
/* if over the trickle threshold, use only 1 in 4096 samples */ if ( input_pool.entropy_count > trickle_thresh && (__get_cpu_var(trickle_count)++ & 0xfff)) return;
-- Matt Mackall : http://www.selenic.com : Linux development and consulting - 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/
| |