On Fri, Feb 11, 2005 at 09:58:27PM -0800, Grant Grundler wrote:
> I'm guessing _sdp_iocb_page_save() is just broken.
> The ia64 kernel builds fine.
> And every other use of pmd_offset() looks like this:
> mm/rmap.c: pmd = pmd_offset(pud, address);
The following patch makes _sdp_iocb_page_save() look like the
code in mm/rmap.c. I have no clue if it's right or not.
But it now builds on ia64. Not tested yet.
And rmap.c uses pXX_present() macro.
I don't know if that should be used as well.
Signed-off-by: Grant Grundler <[EMAIL PROTECTED]>
Index: ulp/sdp/sdp_iocb.c
===================================================================
--- ulp/sdp/sdp_iocb.c (revision 1781)
+++ ulp/sdp/sdp_iocb.c (working copy)
@@ -139,6 +139,7 @@ static int _sdp_iocb_page_save(struct sd
struct page *page;
unsigned long pfn;
pgd_t *pgd;
+ pud_t *pud;
pmd_t *pmd;
pte_t *ptep;
pte_t pte;
@@ -186,12 +187,17 @@ static int _sdp_iocb_page_save(struct sd
0 < size;
counter++, addr += PAGE_SIZE, size -= PAGE_SIZE) {
- pgd = pgd_offset_gate(iocb->mm, addr);
+ pgd = pgd_offset(iocb->mm, addr);
if (!pgd || pgd_none(*pgd)) {
break;
}
-
- pmd = pmd_offset(pgd, addr);
+
+ pud = pud_offset(pgd, addr);
+ if (!pud || pud_none(*pud)) {
+ break;
+ }
+
+ pmd = pmd_offset(pud, addr);
if (!pmd || pmd_none(*pmd)) {
break;
}
_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
