Re: [PATCH 1 of 3] Introduce __raw_memcpy_toio32

[Roland Dreier]

    Bryan> I'll take a look at whether the extra call/ret indirection
    Bryan> affects performance in a measurable fashion.

Your current implementation is out-of-line, right?

I would be surprised if calling a function has any overhead on x86_64,
since the function call is a jump that can be predicted perfectly.
The only issue is the code to shuffle values into the right registers.

 - R.
-
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/