In the end, you will always have to write to non-cached memory or flush
the cache to get it to the device. There should be no performance
penalty in page flipping, but a performance increase. You should also
have a smaller memory footprint.
The typical reason for poor performance of page flipping
Michael,
I tried to implement page flipping in s3c2410fb. No luck. Performance
is worse than in case without flipping.
I suppose SurfaceFlinger uses memcpy (or similar) for double buffering
from/to "offscreen" surface(s).
If offscreen surface resides in Fb memory it's non-cached, therefore
performa
Hi,
Alexey Roslyakov wrote:
> Thanks, Michael,
> great work.
>
> Now bionic code (libc/arch-arm/*.s) is armv4-friendly. You can safely
> use original bionic/libc/Android.mk - optimized versions of memcpy,
> strlen and memcmp.
> Android will run faster a little bit;)
>
I have redone a rebase and
Thanks, Michael,
great work.
Now bionic code (libc/arch-arm/*.s) is armv4-friendly. You can safely
use original bionic/libc/Android.mk - optimized versions of memcpy,
strlen and memcmp.
Android will run faster a little bit;)
On 27 ноя, 03:25, Michael Trimarchi
wrote:
> Hi,
>
> this is my patch t
