Date: Sat, 02 Jun 2001 08:52:48 +0300 From: "Eli Zaretskii" Sender: halo1 AT zahav DOT net DOT il To: yurick AT oblik DOT dp DOT ua Message-Id: <2110-Sat02Jun2001085247+0300-eliz@is.elta.co.il> X-Mailer: Emacs 20.6 (via feedmail 8.3.emacs20_6 I) and Blat ver 1.8.9 CC: djgpp AT delorie DOT com In-reply-to: (yurick@oblik.dp.ua) Subject: Re: __djgpp_map_physical_memory - some questions References: <7458-Thu31May2001062819+0300-eliz AT is DOT elta DOT co DOT il> Reply-To: djgpp AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk > From: yurick AT oblik DOT dp DOT ua > Newsgroups: comp.os.msdos.djgpp > Date: Fri, 1 Jun 2001 18:21:23 +0300 > > Thu, 31 May 2001 06:28:19 +0300, Eli Zaretskii wrote: > > >> 2. I've tried this with djgpp 2.03 under Win95OSR2. Is it supports > >> this call (DPMI 0508)? > EZ> The docs clearly says that this is DPMI 1.0 function, and that most > EZ> DPMI hosts don't support it. Windows doesn't. > I has been read docs but I hoped against hope :) Abandon hope, anyone who uses Windows ;-) > Well, what can I do with it? Can I use _farpoke*() with _dos_ds > for this? I need to access physical address 0xeb000000 (which is not > in 1st meg of course). And will DPMI host allows such access or I must > make some other calls for this? What is at that address? If it's some memory-mapped device, you should be able to use the method described in section 18.7 of the FAQ. > And if _farpoke*() is answer, I can't understand, how this works? > I hasn't allocated memory at this address so DPMI host must blame me, > hasn't it? Without mapping that address into some segment for which there's a valid descriptor, you will indeed get a Page Fault. _farpeek* and _farpoke* only work for valid selectors. One of the things the code presented in section 18.7 of the FAQ does is indeed create a descriptor for the memory you are mapping in.