Re: [PATCH] Fixes for prep_zero_page

[Chris Wright]

* Zwane Mwaikambo (zwane@xxxxxxxxxxxxxxxx) wrote:
> On Sun, 9 Jan 2005, Andrew Morton wrote:
> > Can't we simply move the page zeroing to the very end of __alloc_pages()?
> 
> Ok, i've changed that bit to something like;

I did it the other way around, and moved kernel_map_pages to prep_new_page
so it's called before zeroing to keep that with the other prep bits
in buffered_rmqueue.  Made sense to me that kernel_map_pages is part of
prepping a new page, but this isn't my area, so I could be way off ;-)
It works for me with DEBUG_PAGEALLOC enabled.

===== mm/page_alloc.c 1.251 vs edited =====
--- 1.251/mm/page_alloc.c	2005-01-07 21:44:07 -08:00
+++ edited/mm/page_alloc.c	2005-01-09 14:36:38 -08:00
@@ -413,6 +413,7 @@ static void prep_new_page(struct page *p
 			1 << PG_checked | 1 << PG_mappedtodisk);
 	page->private = 0;
 	set_page_refs(page, order);
+	kernel_map_pages(page, 1 << order, 1);
 }
 
 /* 
@@ -823,7 +824,6 @@ nopage:
 	return NULL;
 got_pg:
 	zone_statistics(zonelist, z);
-	kernel_map_pages(page, 1 << order, 1);
 	return page;
 }
 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/