[I said]:
 * > I think loading the alternate drivers would be preferable since if
 * > you have to check a flag every time you attempt a DMAX, you're going
 * > to get some performance problems.
 * 

der Mouse sez:
/*
 * Taking the DMA-completion interrupt is far, far more expensive than
 * making a single test of a flag.  At least on any architecture I've ever
 * looked at, which means VAX, 68k, and SPARC off the top of my head.  And
 * if you're polling for DMA completion, you obviously don't care about
 * the performance hit of testing a single flag bit.  Not unless you've
 * got one king hell of a DMA engine, with memory speeds to match.

Yes, but each test of

	if (somestruct.someflag & HAS_DMA) {...	}
	else {...}

Or even

	(((somestruct.someflag & HAS_DMA) &&
		retval = do_dma(argv)) ||
		retval = do_bounce(argv));

is going to add SOMETHING to a driver.  I can't help but think that some-
how it would be a nontrivial expense.

"Of course, I _could_ be wrong..." -- B. Bunny

 *
 * 					der Mouse
 * 
 * 			    mouse@collatz.mcrcim.mcgill.edu
 */


				--*greywolf;
--
Friends don't let friends use Windows NT.