Re: [patch] Latency Tracer, voluntary-preempt-2.6.8-rc4-O6

[Ingo Molnar]

* Lee Revell <rlrevell@xxxxxxxxxxx> wrote:

> I have attached a patch that effectively disables extract_entropy.  I
> am adding Theodore T'so to the cc: list as he is the author of the
> code in question.
> 
> For the time being this hack is required to avoid ~0.5 ms
> non-preemptible sections caused by the excessive memcpy's in
> extract_entropy.

i'm not 100% sure it's the memcpy's - but it's extract_entropy()
overhead. Might be the algorithmic slowness of SHATransform().

> +	return nbytes;
> +    

since this effectively disables the random driver i cannot add it to the
patch.

there's another thing you could try: various SHA_CODE_SIZE values in
drivers/char/random.c. Could you try 1, 2 and 3, does it change the
overhead as seen in the trace?

	Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/