|
Bugzilla – Full Text Bug Listing |
| Summary: | Installation on HP dx5150 MT fails in 64bit mode | ||
|---|---|---|---|
| Product: | [openSUSE] SUSE LINUX 10.0 | Reporter: | Veit Laule <admin> |
| Component: | Kernel | Assignee: | Thomas Renninger <trenn> |
| Status: | RESOLVED FIXED | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Major | ||
| Priority: | P5 - None | CC: | mike |
| Version: | Final | ||
| Target Milestone: | --- | ||
| Hardware: | x86-64 | ||
| OS: | SuSE Linux 10.0 | ||
| Whiteboard: | |||
| Found By: | System Test | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Veit Laule
2005-12-04 02:17:31 UTC
Can you try to experiment with vga=normal, nmi_watchdog=1, acpi=on/off/force to see whether some option is capable of reviving the system or at least capturing a trace? Please report how this goes, if a crash or a panic results, please paste this here. No feedback as requested in comment #1. Please re-open if problems persist and if more information is available. These are the last lines of output before the kernel hangs (no errors etc.) when acpi=on & acpi=force: ACPI: Assume root bridge [.\_SB_.PCI0] bus is 0 when acpi=off: PCI: Probing PCI hardware (bus 00) Sounds like an ACPI issue, then. Please try the boot params in the following order until you can boot: pci=noacpi, noapic (if it does not help, try both). Have you already installed the latest YOU update kernel, if not, please do so. If it does boot with pci=noacpi, it's likely that this already got fixed. Please reopen if you still encounter problems. BTW: Please also ensure you have the latest BIOS installed. FYI: Workaround is to specify noapictimer as boot argument Still reproducable with 2.6.13-15.7-default There are know problems on x86_64 with ATI chipsets. Stuff I know is: wrong time, cannot shutdown, ..., not booting is new but seem to have the same cause: a broken (apic?) timer... Andi has added a patch to current SL 10.1, but it does not work well... I didn't have time to dig in deeper. It would be helpful if you could install the latest 10.1 Open SUSE and report back. Also have a look at #145747 for some boot parameters you might want to play with ... Please also reopen the bug if you still have problems with 10.1. Besides hanging on boot i did not actually encounter any problems (single core CPU). The noapictimer boot param works around this for all dx5150 that we use. My commend was mostly intended as FYI to document the workaround i found. BTW: Not booting is not exactly new. This bug is about this issue :-) #145747 is locked If i can find some time i'll try to get a SL10.1 beta and check if its fixed. I will re-open the bug then. Thanks. I changed the product for #145747. Can you now access the bug? Just tested with SL10.1 beta3. dx5150 boots fine no boot param needed. Time keeping seems ok as well. Could not test beyond install boot due to crash with xfs mount. I can now access #145747, thanks. -> no longer a problem Please also try beta4 which comes out soon. There are still changes pending as IIRC not all machines work that fine... |