Messages in this thread | | | Date | Mon, 23 Nov 1998 13:45:45 -0600 (EST) | From | Thomas Molina <> | Subject | APM power down problems |
| |
I have a problem with APM powering down my machine using shutdown -h now. My system is a TMC TI5VG+ with MVP3 (82C598AT chipset), AMD K6-2 35oMHz processor. OS is RedHat 5.2, kernel is 2.1.129 with ac3 patch from Alan Cox. The problem also appears with 2.1.125-2.1.129 as well as with 2.0.36. I'm cc'ing both linux kernel and linux-laptop lists, though I'm not subscribed to linux-laptop ( I can do so if needed to follow a discussion on this topic). Stephen Rothwell is listed as responsible for the APM code in the MAINTAINERS file; I hope that's correct.
It's been suggested that it might be caused by a bad bios, having the bios shadowed, or because the bios expected to be in real mode when the kernel was in protected mode. It might be faulty bios code; I've sent an email to TMC asking for tech support on this issue. Of course Winblows handles the power down fine. I do not have the bios shadowed. The only shadowing is the video, but unshadowing that makes no difference. I don't fully grok the APM code, but my reading of the source indicates it is doing all the magic to switch into real mode before invoking the power down. At this point I'd appreciate any suggestions the readers might have.
After doing shutdown -h now I get:
system halted Power down.
After that I get the following: (copied by hand - the gpf is followed by a kernel oops for unable to handle page fault. I didn't copy that down, although I could if necessary)
general protection fault: f000 CPU: 0 EIP: 0050:[<0000898F>] EFLAGS: 00010047 eax: 00005301 ebx: 00000001 ecx: 00000003 edx: 00000000 esi: bfff82e6 edi: 080492e1 ebp: 67890000 esp: c19cbe38 Process halt: (pid: 376, process nr: 15, Stackpage = c19cb000) Stack: 0000bfff be526789 0001c19c 00000000 00030000 53070000 00000000 00000000 82d50058 82aaff04 0000827d 00160000 004881e6 00000003 c0106335 00000010 00000018 00000018 00000000 bfffff04 080492e1 bffffdc4 bfffff04 00000286 Call Trace: [<c0106336>] [<c0106378>] c0107179>] [<c011287f>] [<c01c2292>] [<c0141559>] [<c010e4cd>] [<c010e74c>] [<c010edff>] [<c0117bac>] [<c010fae7>] [<c01087f0>] Code: <1> Unable to handle kernel paging request at virtual address 0000898f current->tss.cr3 = 01953000, %cr3 = 019530000 *pde = 00000000
ksymoops output:
Using `/boot/2.1.129b.map' to map addresses to symbols.
>>EIP: 898f cannot be resolved Trace: c0106336 <apm_set_power_state+3e/70> Trace: c0106378 <apm_power_off+10/14> Trace: c011287f <sys_reboot+18b/20c> Trace: c01c2292 <tvecs+34a/69e8> Trace: c0141559 <ext2_follow_link+65/84> Trace: c010e4cd <send_sig_info+265/274> Trace: c010e74c <kill_something_info+f0/100> Trace: c010edff <sys_kill+4f/5c> Trace: c0117bac <handle_mm_fault+cc/16c> Trace: c010fae7 <schedule+1e3/20c> Trace: c01087f0 <system_call+34/38> Code:
following is a relevant snip from bootup regarding APM:
Nov 23 12:09:11 wr5z kernel: lp0: using parport0 (interrupt-driven). Nov 23 12:09:11 wr5z kernel: APM BIOS version 1.2 Flags 0x07 (Driver version 1.6) Nov 23 12:09:11 wr5z kernel: Entry f000:81c0 cseg16 f000 dseg fdfc cseg len e6ff, cseg16 len 16, dseg len 0 Nov 23 12:09:11 wr5z kernel: Connection version 1.2 Nov 23 12:09:11 wr5z kernel: AC on line, battery status unknown, battery life unknown Nov 23 12:09:11 wr5z kernel: battery flag 0x80, battery life unknown Nov 23 12:09:11 wr5z kernel: <Pro AudioSpectrum 16D rev 255> at 0x388 irq 10 dma 3
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |