Message-Id: Comments: Authenticated sender is From: "Salvador Eduardo Tropea (SET)" Organization: INTI To: djgpp AT delorie DOT com, shawn AT talula DOT demon DOT co DOT uk Date: Mon, 16 Nov 1998 10:27:01 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: Disable DPMI memory caching? References: <+M9M3kANnJT2EwOV AT talula DOT demon DOT co DOT uk> In-reply-to: X-mailer: Pegasus Mail for Windows (v2.54) Reply-To: djgpp AT delorie DOT com Eli Zaretskii replied to Shawn Hargreaves : > > On Fri, 13 Nov 1998, Shawn Hargreaves wrote: > > > with the use of memory-mapped IO registers that are located in the low > > megabyte of physical memory (0xB8000). At the moment the CPU is caching > > any writes to these registers, which obviously causes the hardware to > > miss a lot of commands! > > Does this happen on several types of motherboards? It surely seems as > a chipset bug to me! The motherboard should have no business caching > memory regions mapped into peripheral devices, ever. I agree with Eli. In my TGUI9440 driver I map the accelerator in the 0xB7F00 range and I never experimented such a thing. > > Is there any way to disable caching for a range of conventional memory > > addresses? I can't find any mention of this in the DPMI spec > > This is not a DPMI issue, so the DPMI spec is IMHO the wrong place to > look for a solution. > > Many motherboards have a BIOS setup program that allows to > enable/disable caching of specific memory address regions. Maybe > somebody has set that machine to cache those areas? If such an option > is available, you could use it to explicitly disable caching. > > Another possibility might be that some memory manager shadows the B800 > region, and the caching actually happens in the remapped addresses. > To see whether this is the cause, boot into plain-vanilla DOS > configuration (empty CONFIG.SYS and AUTOEXEC.BAT) and see if the > problem goes away. If it does, fiddling with the memory manager > command line should do the trick. > > Shadowing is also sometimes controlled from the BIOS setup, so looking > there might also provide some hints. I vote for this theory ;-) SET ------------------------------------ 0 -------------------------------- Visit my home page: http://set-soft.home.ml.org/ or http://www.geocities.com/SiliconValley/Vista/6552/ Salvador Eduardo Tropea (SET). (Electronics Engineer) Alternative e-mail: set-soft AT usa DOT net set AT computer DOT org ICQ: 2951574 Address: Curapaligue 2124, Caseros, 3 de Febrero Buenos Aires, (1678), ARGENTINA TE: +(541) 759 0013