Hello,
(Machine details at the end)
With 2.3.99 pre6-4 and pre6-5 I get following message, plus
oops following it: (Another problem possibly derives from this)
bad magic fffffc0003b60aed (should be fffffc0000dda690,
creator fffffc00004b35c0), wq bug, forcing oops.
This is indicative of wait-queue head corruption.
The creator address seems to be in mm/mm.o object module .text.init
section, but even '-Map filename --cref' flags added to the top-level
Makefile help to locate the troublesome initializer into function
free_area_init_core() (debugger is unable to find it, perhaps because
so much of debug data is purged at several linking phases..)
...
.text.init 0xfffffc00004b2480 0x18a0 mm/mm.o
0xfffffc00004b2480 vma_init
...
0xfffffc00004b3320 swap_setup
0xfffffc00004b33e0 free_area_init_core
0xfffffc00004b3ba0 free_area_init
0xfffffc00004b3cc0 free_area_init_node
.text.init 0xfffffc00004b3d20 0x1060 fs/fs.o
Another problem is that even if by some luck no oops happens during
bootup, starting X is instant death. Nothing comes out to serial console,
and no reaction to magic sysreq (I explicitely enable it at my rc.local).
The machine:
- DEC PWS 433a running varying kernel versions
- Kernel is configured for MIATA
- There is 0.5 GB RAM
What might have been changed recently so that ia32 and Alpha are
again out of sync ?
/Matti Aarnio <matti.aarnio@sonera.fi>
-
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/
This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:22 EST