Messages in this thread | | | Date | Wed, 17 Jun 1998 09:34:07 -0500 (CDT) | From | Jeffrey Hundstad <> | Subject | Re: MMX emulator ? |
| |
On 17 Jun, Johan Myréen wrote: > On Tue, 16 Jun 1998, Jeffrey Hundstad wrote: > >> > i = detect_MMMX(); >> > if !i then usenormalstuff=1; >> > else usemmx=1; > >> If you can assume the MMX functions are there then EACH application >> doesn't have to emulate all of the MMX stuff itself if it doesn't have >> the REAL hardware. > >> This should be treated EXACTLY like the Floating Point hardware... at >> least in my opinion. > > The 'normal stuff' in a real application that cared about > decent speed would NOT be emulating the MMX instructions, but > would do things differently. MMX is all about speed. I'd > expect emulated MMX to be far slower than writing the code > optimally using the instruction set available on a non-MMX > CPU. And this is not counting the overhead from trapping the > MMX opcodes. > > If writing an MMX emulator leads to application programmers > being lazy and just doing MMX versions of their apps, because > "MMX is always available on Linux", then you have made a great > disservice to the users. > >
I don't think the MMX emulator would exclude the programmer from checking the capabilities of the CPU and deciding NOT to use the emulator. This way people who want to write MMX enabled code get their way, people who want performance on an MMX and non-MMX machine get their way and "lazy programmers" get their way, it doesn't force people to do anything they don't want to do. ...and you NEVER get a program that says it works only on an MMX machine which will decrease the ``wine'' factor from users.
Jeffrey E. Hundstad
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |