As soon as the 2nd stage loader starts, it halts with error message "BTX halted" The full text looks like this BTX loader 1.00 BTX version is 1.01 Console: internal video/keyboard BIOS drive C: is disk0 BIOS drive D: is disk1 BIOS 639kB/2096064kB available memory FreeBSD/i386 bootstrap loader, Revision 1.1 (root@willow.at.sygate.com, Tue Jan 18 04:54:33 UTC 2005) - int=0000000d err=00000000 efl=00030202 eip=00000b1e eax=00002008 ebx=00000004 ecx=00001942 edx=0000fffc esi=00005bae edi=00000004 ebp=000003d7 esp=000003cd cs=e000 ds=0040 es=5bae fs=0000 gs=0000 ss-9e49 cs:eip=ee 58 5a c3 50 52 e8 79-02 8a 46 0c e8 39 00 8a 46 0b e8 33 00 e8 80 02-8a 46 0e e8 2a 00 8a 46 ss:esp=04 20 80 00 00 07 5a 0f-1b 04 f0 01 fc ff 10 00 40 ea 00 f0 00 04 00 7f-4e c0 00 70 00 e1 f0 01 BTX halted Fix: Some folks have reported success disabled DMA in the BIOS. Unfortunately, the AMI box for DL140 does not have an option to disable DMA. Examining the source for loader, it seems that the problem is in acpidetect() or in interact(), however, both of those are in some library and I'm not sure where that code is located. How-To-Repeat: Simply load this build. We load it via PXEboot, completely automated sysinstall. we can provide a CDROM iso with all the bits This same build works on HP DL320, DL380, Dell SC400, SC420. o it appears to be a HP DL140 problem, but we've upgraded the bios to the latest 04502T03, which It appears that Dragonfly has the same issue when dragonfly is tracking 5.3-STABLE. http://leaf.dragonflybsd.org/mailarchive/bugs/2004-12/msg00133.html another instance, http://lists.freebsd.org/pipermail/freebsd-amd64/2005-February/003497.html http://www.freebsdforums.org/forums/showthread.php?s=7b971a2b67cf87ba4b8462e84c4422ce&threadid=28403&goto=nextoldest http://lists.freebsd.org/pipermail/freebsd-i386/2005-January/002037.html
State Changed From-To: open->closed New port added as 'devel/jep'. Thanks for your contribution!
State Changed From-To: closed->open Oops, I closed the wrong PR. Sorry for that.
Can anyone tell me if this problem is the result of a spurious interrupt? We seem to have much better luck booting with the network cables disconnected. I'm wondering if the problem is that a Wake-ON-LAN interrupt or other interrupt arrives before the boot loader has a chance to install interrupt handlers. However, even disabling WOL in the Broadcom config screen and disabling PXEboot does not prevent the problem from happening. we're trying to follow up with HP as well. The problem occurs more frequently under bios 04502T03 than bios 3304T03 2/2/2004. we have it happening on at least 3 DL140s. -chris
State Changed From-To: open->feedback Have asked for feedback
Responsible Changed From-To: freebsd-i386->gavin Track
Hi, and sorry that this PR has sat for so long without anyone looking at it. Can you tell me, have you seen this problem again with newer versions of FreeBSD?
The problem persists. FreeBSD 6.3R, HP ProLiant DL140 G3, BIOS version BIOS O08 (11/16/2007). BTX Halted int=0000000d err=00000000 efl=00010046 eip=000090d8 eax=00000011 ebx=00002820 ecx=00000000 edx=004b08ee esi=00009731 edi=00001f98 ebp=00000000 esp=00001800 cs=0008 ds=0000 es=0000 fs=0020 gs=0020 ss=0010 cs:eip=0f 01 15 00 97 00 00 66-ea e5 90 18 00 b1 20 8e d1 8e d9 8e c1 8e e1 8e-e9 48 0f 22 c0 ea fa 90 ss:esp=0a 69 6e 74 3d 30 30 30-30 30 30 30 64 20 20 65 72 72 sd 30 30 30 30 30-30 30 30 20 20 65 66 6c -- Spartak Radchenko SVR1-RIPE
State Changed From-To: feedback->open Feedback was received
State Changed From-To: open->feedback To submitter: Could you try booting one of the snapshot ISOs located at http://snapshots.us.freebsd.org/snapshots/ and dated after March 12th? Some changes bave recently been made to the BTX loader that may well fix your issues.
State Changed From-To: feedback->closed Close, feedback timeout (3 months). This may well be fixed with the new BTX