Date: Mon, 15 Mar 1999 09:49:13 +0900 From: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: freebsd-hackers@freebsd.org, yokota@zodiac.mech.utsunomiya-u.ac.jp Subject: Re: FreeBSD 3.1-Stable Being Unstable Message-ID: <199903150049.JAA06054@zodiac.mech.utsunomiya-u.ac.jp> In-Reply-To: Your message of "Sun, 14 Mar 1999 11:26:24 PST." <199903141926.LAA93452@apollo.backplane.com> References: <199903132253.QAA09676@isua1.iastate.edu> <199903141926.LAA93452@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> The vga stuff in the call stack is very suspicious. It looks like it > crashed trying to start up the screen saver. > > If you are using any dynamically loaded modules ( such as the splash > module ), make sure you have recompiled and reinstaslled them. > > -Matt > Matthew Dillon > <dillon@backplane.com> To be precise, the system is crashing when syscons is trying to stop the screen saver and restore the video mode. What I know about this crash is as follows: The section of the code which is triggering the panic is in the vga driver in the kernel, not in any KLD modules. The users are seeing the problem on 486DX, P54C/133, Cyrix (I don't know which one it is) CPUs. (I cannot reproduce the problem on my test bed, a dual P-Pro machine.) The vga driver is trying to write to the BIOS data variables in the physical memory region 0x0-0x4ff, the very first page of the physical memory, when this panic occurs The region is supposed to be always mapped to the beginning of the kernel memory space. The vga driver reads from that region when initializing itself and there is no panic at that point. I suspect the page table entry for the above region may be corrupted at some point. Any comment? Kazu 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?199903150049.JAA06054>