Date: Wed, 17 Mar 1999 08:50:01 -0600 (CST) From: Jonathan Lemon <jlemon@americantv.com> To: tony@dell.com, freebsd-hackers@freebsd.org Subject: Re: Proposal: Define MAXMEM in GENERIC Message-ID: <199903171450.IAA20156@free.pcs> In-Reply-To: <local.mail.freebsd-hackers/3.0.6.32.19990314235738.03ba2a10@bugs.us.dell.com> References: <local.mail.freebsd-hackers/3.0.6.32.19990314123925.039187c0@bugs.us.dell.com> <local.mail.freebsd-hackers/Your message of "Mon, 15 Mar 199901:54:46 %2B0900."<local.mail.freebsd-hackers/36EBE9D6.7E6320FE@newsguy.com> <local.mail.freebsd-hackers/63209.921431112@axl.noc.iafrica.com> <local.mail.freebsd-hackers/199903141920.NAA16212@free.pcs>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <local.mail.freebsd-hackers/3.0.6.32.19990314235738.03ba2a10@bugs.us.dell.com> you write: >At 01:20 PM 3/14/99 -0600, Jonathan Lemon wrote: >>>p.s. FreeBSD seems to prefer E801h over E820h. I'd like to see it >>> the other way around, since you could gain almost 64 KB of >>> extra memory in some cases. >> >>Not true, if you're referring to the VM86 memory probe. It tries >>INT 15h, AX=E820, then INT 15h, AX=E801, then INT 15h, AX=88, in >>that order. > >Sorry about that. I have all three of the BIOS functions in my >box, but I still see this: > >BIOS basemem: 639K, extmem: 64512K (from 0xe801 call) > >so I made an assumption. I did say it *seems* to prefer E801h. Actually, it appears that you are right, the E820 call was broken by an earlier commit (circa September), and hasn't worked since. I just sync-ed up to current yesterday and found this out. Once this is fixed, I think that the VM86 code should become mandatory, and the entire "speculative memory probe" should just disappear. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199903171450.IAA20156>