Too many holes in the physical address space, giving up?

[Ruslan Ermilov]

On Mon, Nov 13, 2006 at 01:07:33PM -0500, John Baldwin wrote:
> On Monday 13 November 2006 12:48, Sulev-Madis Silber wrote:
> > Strange :) I managed to boot... 6.2-BETA3
> 
> Yeah, beta3 extends the arrays so they can handle more regions.
> 
It's still strange because he was able to boot 4.11 which also
has the same old limit of four chunks ((10 - 2) / 2) of RAM,
he has three non-overlappping RAM chunks reported by SMAP, and
I fail to see how it ends up using five chunks.  Odd...

There's a problem with the PHYS_AVAIL_ARRAY_END in RELENG_4
(type mismatch):

vm_paddr_t phys_avail[10];

/* must be 2 less so 0 0 can signal end of chunks */
#define PHYS_AVAIL_ARRAY_END ((sizeof(phys_avail) / sizeof(vm_offset_t)) - 2)

But it's only applicable to RELENG_4 and PAE.


Cheers,
-- 
Ruslan Ermilov
ru at FreeBSD.org
FreeBSD committer
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-hardware/attachments/20061113/caab92d2/attachment.pgp