Messages in this thread Patch in this message |  | | | Date | Sun, 2 Aug 1998 15:30:02 +0200 | | From | Thomas Bogendoerfer <> | | Subject | Re: 2.1.113: still an SMP related oops at boot time |
| |
On Sat, Aug 01, 1998 at 05:34:28PM -0700, Bill Broadhurst wrote: > I'm seeing the same oopsen. All values are the same. > > Hardware: > Tyan Tomcat IV with dual P-200MMX. 128M DRAM. All SCSI system on a > BusLogic controller. > > I rolled back to .111 for now.
the bug got introduced by .112 and will only occur on P5 SMP machines. By changing the order of smp_init() and check_bugs(), the idt got locked down read only on P5 systems to workaround the F00F bug. But this gives an oops, when smp_init() initializes the io-apic interrupt table. Below a probably wrongm, but working fix.
Thomas.
Index: main.c =================================================================== RCS file: /usr/src/cvs/linux/init/main.c,v retrieving revision 1.1.1.68 diff -u -r1.1.1.68 main.c --- main.c 1998/08/02 10:01:49 1.1.1.68 +++ main.c 1998/08/02 13:28:54 @@ -1123,11 +1123,11 @@ dquot_init_hash(); #endif printk("POSIX conformance testing by UNIFIX\n"); - check_bugs(); #ifdef __SMP__ smp_init(); #endif + check_bugs(); #if defined(CONFIG_MTRR) /* Do this after SMP initialization */ /*
-- See, you not only have to be a good coder to create a system like Linux, you have to be a sneaky bastard too ;-) [Linus Torvalds in <4rikft$7g5@linux.cs.Helsinki.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.altern.org/andrebalsa/doc/lkml-faq.html
|  |