| Re: [AD] Color convertors |
[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
Eric Botcazou wrote:
[snip]
You're right, there was still an off-by-one-pixel error after I had fixed
the off-by-one-scanline error. As I am a purist, I will probably touch again
the non-MMX code because your fix breaks the instruction pairing on Pentium
(and the non-MMX code was primarily targeted to it).
Heh, allright.
Moreover, I see you replaced the 'movl $0, reg' by 'xorl reg, reg'. Why ?
I'm not sure that's the way to go on Pentium.
Yes it's the way to go. Even Intel recommends it! xorl executes in 1 cycle
and is fully pairable. movl executes in 1 cycle but can only be placed in
the U pipe if it involves another register than %eax, since it's longer than
6 bytes. xorl is also shorter (2 or 3 bytes), which reduces the pressure on
the instruction cache.
Great work !
However, the 16->15 and 15->16 convertors can't be enabled under Windows for
the time being (see the latest docs for the explanation).
Allegro.txt?
The 15->16 one works under Windows (for me at least)
There are also dummy 15->15, 16->16, 24->24, 32->32 "convertors" which are
implemented via a super fast memcpy :)
What are they for ? They won't be used under Windows for example.
So under Windows, if the window has the same depth as the screen, then the
screen -is- the window? or does it still pass via a second buffer like the
other depths?
I'm going to update the Windows windowed driver.
Btw, could you upload somewhere your big diff and let us know the address ?
Already on CVS :)
--
- Robert J Ohannessian
"Microsoft code is probably O(n^20)" (my CS prof)
http://pages.infinit.net/voidstar/