public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* 2.6.10-rc2-mm2 crashes in early boot
@ 2004-11-20  0:03 Dave Hansen
  2004-11-20  1:02 ` Dave Hansen
  0 siblings, 1 reply; 2+ messages in thread
From: Dave Hansen @ 2004-11-20  0:03 UTC (permalink / raw)
  To: Andrew Morton; +Cc: Linux Kernel Mailing List, Haren Myneni [imap]

I enabled crashdump on 2.6.10-rc2-mm2 with CONFIG_SMP=y and hit the bug
in smp_alloc_memory() because trampoline_base was allocated too high. 
It needs to be below 0x9F000 and crash_reserve_bootmem() reserves
everything below 0xa0000.  I'm not sure how this ever worked with SMP.  

Here's my .config: http://sprucegoose.sr71.net/~dave/lkcd-config

Commenting out the following line makes it boot:

static inline void crash_reserve_bootmem(void)
{
        if (!dump_enabled) {
------->        reserve_bootmem(0, CRASH_RELOCATE_SIZE);
                reserve_bootmem(CRASH_BACKUP_BASE,
                        CRASH_BACKUP_SIZE + CRASH_RELOCATE_SIZE + PAGE_SIZE);
        }
}

The comments don't quite tell *why* LKCD needs any of the memory.  (Oh,
wait, there *aren't* any comments :)

-- Dave


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: 2.6.10-rc2-mm2 crashes in early boot
  2004-11-20  0:03 2.6.10-rc2-mm2 crashes in early boot Dave Hansen
@ 2004-11-20  1:02 ` Dave Hansen
  0 siblings, 0 replies; 2+ messages in thread
From: Dave Hansen @ 2004-11-20  1:02 UTC (permalink / raw)
  To: Andrew Morton; +Cc: Linux Kernel Mailing List, Haren Myneni [imap]

On Fri, 2004-11-19 at 16:03, Dave Hansen wrote:
> I enabled crashdump on 2.6.10-rc2-mm2 with CONFIG_SMP=y and hit the bug
> in smp_alloc_memory() because trampoline_base was allocated too high. 
> It needs to be below 0x9F000 and crash_reserve_bootmem() reserves
> everything below 0xa0000.  I'm not sure how this ever worked with SMP.  
> 
> Here's my .config: http://sprucegoose.sr71.net/~dave/lkcd-config
> 
> Commenting out the following line makes it boot:
> 
> static inline void crash_reserve_bootmem(void)
> {
>         if (!dump_enabled) {
> ------->        reserve_bootmem(0, CRASH_RELOCATE_SIZE);
>                 reserve_bootmem(CRASH_BACKUP_BASE,
>                         CRASH_BACKUP_SIZE + CRASH_RELOCATE_SIZE + PAGE_SIZE);
>         }
> }

OK, found the patch:

http://marc.theaimsgroup.com/?l=linux-kernel&m=110078720123943&w=2

Do we really still need the CRASH_RELOCATE_SIZE #define?  What is it
there for?

-- Dave


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2004-11-20  1:05 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-11-20  0:03 2.6.10-rc2-mm2 crashes in early boot Dave Hansen
2004-11-20  1:02 ` Dave Hansen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox