[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
- To: alleg-developers@xxxxxxxxxx
- Subject: Re: [AD] malloc
- From: Chris <chris.kcat@xxxxxxxxxx>
- Date: Sun, 20 Nov 2005 20:30:17 -0800
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:from:to:subject:date:user-agent:references:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:message-id; b=p9tvcGOIpl6cmimNXnf24NPl0RQD8ZPwBo5IK/BF9wPYTCfU9AIyc7buu9aTgMvtOJSVqeKiBq1Tqjxp5rHw68ZqTo/OHKDgmxz/gXbA9MffdHpodqP3TXs4p00NgvwDJ+LGoiCWDKPA30iDK7cr0Z5O5KjLX7wGTI0M0JQAgVY=
On Sunday 20 November 2005 08:08 pm, aj wrote:
> i wouldn't be accessing [0][1] SSE is 128bits wide, pixels are 32,
> so i can process 4 pixels at once.
> then the next load would be [0][4] (aligned again).
Allegro processing things per-pixel. If you do bmp_write32(addr+(x*4), color);
it won't be an aligned access if x isn't a multiple of 4.
> >etc). Plus, you can't gaurantee the alignment of video bitmaps since the
> >system allocates/maps them, not your program.
>
> video cards dont have SSE.
> why would you write SSE code for video bitmaps?
You wouldn't (nor shouldn't, directly). The compiler would. And since memory
and video bitmap types are abstracted, if you access a bitmap you'll have no
idea if the pointer the system gives you is aligned or not.
> nor will i386 instructions, so should we not bother with MMX either, as
> i386 is good enough.
I never said don't bother with SSE. I just said MMX isn't going anywhere, and
that it might not be a totally technical reason that Intel is pushing SSE and
downplaying MMX.