PATCH for displaying pages [was: Re: PATCH: [olh at suse.de: kdb and softirqs]]

[Haren Myneni]

mem_map symbol will not be defined for CONFIG_DISCONTIGMEM. 
Probably the following code should use to make it generic:
pg_data_t *pgdat;
for_each_pgdat(pgdat) {
        ulong start = pgdat->node_start_pfn;
        ulong end = pgdat->node_start_pfn + pgdat->node_spanned_pages;
        for (i=start; i < end; i++) {
                struct page *pg = pfn_to_page(i);
                ...
                ...
                ...
        }
}
Sent by:        kdb-bounce at oss.sgi.com
To:     Keith Owens <kaos at sgi.com>
cc:     linas at austin.ibm.com, Olaf Hering <olh at suse.de>, kdb at oss.sgi.com, Anton 
Blanchard <anton at samba.org>, linuxppc64-dev at lists.linuxppc.org 
Subject:        Re: PATCH for displaying pages [was: Re: PATCH: [olh at suse.de: kdb and 
softirqs]]



On Wed, Mar 24, 2004 at 08:39:05AM +1100, Keith Owens wrote:
> >It certainly does not work on DISCONTIG systems.  It should work on all
> >machines without discontig, maybe except for the very strange virtual 
mem
> >map hack that can be optinionall be used on ia64.
> >
> >Was your ia64 an Altix or some whitebox?
>
> From a dim, distant memory, it was an Altix.

Ok.  I think a !defined(CONFIG_DISCONTIGMEM) && 
!defined(CONFIG_VIRTUAL_MEM_MAP)
should put you on the safe side.

---------------------------
Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.



---------------------------
Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.