[PATCH] tmpfs 4/6 use mark_page_accessed

• Next message: Hugh Dickins: "[PATCH] tmpfs 5/6 use cond_resched"
• Previous message: Hugh Dickins: "[PATCH] tmpfs 3/6 use generic_file_llseek"
• In reply to: Hugh Dickins: "[PATCH] tmpfs 1/6 use generic_write_checks"
• Next in thread: Hugh Dickins: "[PATCH] tmpfs 5/6 use cond_resched"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

tmpfs pages should be surfing the LRUs in the company of their filemap
friends: I was expecting the rules to change, but they've been stable
so long, let's sprinkle mark_page_accessed in the equivalent places
here; but (don't ask me why) SetPageReferenced in shmem_file_write.
Ooh, and shmem_populate was missing a flush_page_to_ram.

--- tmpfs3/mm/shmem.c Tue Apr 1 21:35:10 2003
+++ tmpfs4/mm/shmem.c Tue Apr 1 21:35:21 2003
@@ -951,6 +951,7 @@
         if (error)
                 return (error == -ENOMEM)? NOPAGE_OOM: NOPAGE_SIGBUS;
 
+ mark_page_accessed(page);
         flush_page_to_ram(page);
         return page;
 }
@@ -978,6 +979,8 @@
                 if (err)
                         return err;
                 if (page) {
+ mark_page_accessed(page);
+ flush_page_to_ram(page);
                         err = install_page(mm, vma, addr, page, prot);
                         if (err) {
                                 page_cache_release(page);
@@ -1192,6 +1195,8 @@
                         break;
                 }
 
+ if (!PageReferenced(page))
+ SetPageReferenced(page);
                 set_page_dirty(page);
                 page_cache_release(page);
 
@@ -1264,13 +1269,20 @@
                 }
                 nr -= offset;
 
- /* If users can be writing to this page using arbitrary
- * virtual addresses, take care about potential aliasing
- * before reading the page on the kernel side.
- */
- if (!list_empty(&mapping->i_mmap_shared) &&
- page != ZERO_PAGE(0))
- flush_dcache_page(page);
+ if (page != ZERO_PAGE(0)) {
+ /*
+ * If users can be writing to this page using arbitrary
+ * virtual addresses, take care about potential aliasing
+ * before reading the page on the kernel side.
+ */
+ if (!list_empty(&mapping->i_mmap_shared))
+ flush_dcache_page(page);
+ /*
+ * Mark the page accessed if we read the beginning.
+ */
+ if (!offset)
+ mark_page_accessed(page);
+ }
 
                 /*
                  * Ok, we have the page, and it's up-to-date, so
@@ -1523,6 +1535,7 @@
                 return res;
         res = vfs_readlink(dentry, buffer, buflen, kmap(page));
         kunmap(page);
+ mark_page_accessed(page);
         page_cache_release(page);
         return res;
 }
@@ -1535,6 +1548,7 @@
                 return res;
         res = vfs_follow_link(nd, kmap(page));
         kunmap(page);
+ mark_page_accessed(page);
         page_cache_release(page);
         return res;
 }

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

• Next message: Hugh Dickins: "[PATCH] tmpfs 5/6 use cond_resched"
• Previous message: Hugh Dickins: "[PATCH] tmpfs 3/6 use generic_file_llseek"
• In reply to: Hugh Dickins: "[PATCH] tmpfs 1/6 use generic_write_checks"
• Next in thread: Hugh Dickins: "[PATCH] tmpfs 5/6 use cond_resched"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Mon Apr 07 2003 - 22:00:15 EST