Hi.
The first patch fixes the values of page-flags to solve the problem
that makedumpfile cannot distinguish the cache page correctly.
To distinguish the cache page, makedumpfile uses the values of PG_lru,
PG_swapcache, and PG_private. And there were the mistakes of these
values in the existing makedumpfile. This problem was reported by
Akiyama-san.
| PG_lru | PG_private | PG_swapcache
-----------------+--------+------------+--------------
old makedumpfile| 5 | 12 | 16
2.6.15 | 5 | 11 | 15
2.6.16.27 | 5 | 11 | 15
2.6.17 | 5 | 11 | 15
2.6.18 | 5 | 11 | 15
Thanks
Ken'ichi Ohmichi
diff -puN makedumpfile.org/makedumpfile.h makedumpfile/makedumpfile.h
--- makedumpfile.org/makedumpfile.h 2006-11-15 13:40:57.000000000 +0900
+++ makedumpfile/makedumpfile.h 2006-11-15 13:41:58.000000000 +0900
@@ -51,25 +51,9 @@ enum {
/*
* Page flags
*/
-#define PG_locked (0) /* Page is locked. Don't touch. */
-#define PG_error (1)
-#define PG_referenced (2)
-#define PG_uptodate (3)
-#define PG_dirty (4)
-#define PG_lru (5)
-#define PG_active (6)
-#define PG_slab (7) /* slab debug (Suparna wants this) */
-#define PG_highmem (8)
-#define PG_checked (9) /* kill me in 2.5.<early>. */
-#define PG_arch_1 (10)
-#define PG_reserved (11)
-#define PG_private (12) /* Has something at ->private */
-#define PG_writeback (13) /* Page is under writeback */
-#define PG_nosave (14) /* Used for system suspend/resume */
-#define PG_compound (15) /* Part of a compound page */
-#define PG_swapcache (16) /* Swap page: swp_entry_t in private */
-#define PG_mappedtodisk (17) /* Has blocks allocated on-disk */
-#define PG_reclaim (18) /* To be reclaimed asap */
+#define PG_lru (5)
+#define PG_private (11) /* Has something at ->private */
+#define PG_swapcache (15) /* Swap page: swp_entry_t in private */
#define PAGE_MAPPING_ANON (1)
_
_______________________________________________
fastboot mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/fastboot
