On 04/08/2011 09:37 PM, Marko Ristola wrote:
>
> Here is some statistics without likely and with likely functions.
>
> It seems that using likely() gives better performance with Phenom I too.
I'm not sure whether that's the right conclusion. See below.
> % Plain knl % No likely
On 04/08/2011 05:40 PM, Marko Ristola wrote:
> Avoid unnecessary DVB TS 188 sized packet copying from DMA buffer into stack.
> Backtrack one 188 sized packet just after some garbage bytes when possible.
> This obsoletes patch https://patchwork.kernel.org/patch/118147/
>
> Signed-off-by: Marko Rist
Here is some statistics without likely and with likely functions.
It seems that using likely() gives better performance with Phenom I too.
% Plain knl % No likely % With likely %
dvb_ringbuffer_write5,9 62,88,7 81,35,7 79,2
dvb_dmx_swfilt
Avoid unnecessary DVB TS 188 sized packet copying from DMA buffer into stack.
Backtrack one 188 sized packet just after some garbage bytes when possible.
This obsoletes patch https://patchwork.kernel.org/patch/118147/
Signed-off-by: Marko Ristola marko.rist...@kolumbus.fi
diff --git a/drivers/medi