Re: Optimizing bitblit.c / fb_pad_*

[Knut Petersen <Knut_Petersen@t-online.de>]

Hi Antonino,

>> +                if (fast_8x16)
>> +                    for (i = 16; i--; ) {
>> +                        *dstp = src[15-i];
>> +                        dstp += pitch;
>> +                    }
>
>
> why not make it to something like this?
>
>         for (i = cellsize; i--;) {
>             *dstp = src[cellsize-1-i];
>             dstp += pitch;
>         }
>
> This way, it will work with any fontsize, as long as the width
> is a 8-bit size-aligned.  You can do the same with
> fb_pad_aligned_buffer(), so the rest can benefit too.
>
fb_pad_unaligned() could be optimized localy, there are other cases of
memcpy(), e.g. in softcursor, etc.

Well, this was nothing but a quick ´n dirty hack, I tried to find out if 
there
is still potential for optimizations. If people agree that the gain in speed
justifies it, I will try to look at those parts of the kernel _after_ I have
submitted my cyberblade/i1 driver. I have to admit that I ´m not familar
(enough) with the layers above the hardware driver level right now.

> PS: Next time you submit a patch, add a Signed-off-line.

ack. btw, should a new driver first be presented in
linux-fbdev-devel@lists.sourceforge.net or lkml?

cu,
 Knut


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click