diff options
author | Sage Weil <sage@newdream.net> | 2011-08-24 14:07:01 -0700 |
---|---|---|
committer | Sage Weil <sage@newdream.net> | 2011-10-25 16:10:16 -0700 |
commit | 83eaea22bdfc9e1cec88f81be5b64f30f6c37e8b (patch) | |
tree | f63f415ce4241c12639ef78f4920ed6b99915760 /fs | |
parent | 80db8bea6a0f4fd047eafd8329a44d5a110f462b (diff) | |
download | op-kernel-dev-83eaea22bdfc9e1cec88f81be5b64f30f6c37e8b.zip op-kernel-dev-83eaea22bdfc9e1cec88f81be5b64f30f6c37e8b.tar.gz |
Revert "ceph: don't truncate dirty pages in invalidate work thread"
This reverts commit c9af9fb68e01eb2c2165e1bc45cfeeed510c64e6.
We need to block and truncate all pages in order to reliably invalidate
them. Otherwise, we could:
- have some uptodate pages in the cache
- queue an invalidate
- write(2) locks some pages
- invalidate_work skips them
- write(2) only overwrites part of the page
- page now dirty and uptodate
-> partial leakage of invalidated data
It's not entirely clear why we started skipping locked pages in the first
place. I just ran this through fsx and didn't see any problems.
Signed-off-by: Sage Weil <sage@newdream.net>
Diffstat (limited to 'fs')
-rw-r--r-- | fs/ceph/inode.c | 46 |
1 files changed, 1 insertions, 45 deletions
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 095799b..5dde7d51 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -9,7 +9,6 @@ #include <linux/namei.h> #include <linux/writeback.h> #include <linux/vmalloc.h> -#include <linux/pagevec.h> #include "super.h" #include "mds_client.h" @@ -1364,49 +1363,6 @@ void ceph_queue_invalidate(struct inode *inode) } /* - * invalidate any pages that are not dirty or under writeback. this - * includes pages that are clean and mapped. - */ -static void ceph_invalidate_nondirty_pages(struct address_space *mapping) -{ - struct pagevec pvec; - pgoff_t next = 0; - int i; - - pagevec_init(&pvec, 0); - while (pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) { - for (i = 0; i < pagevec_count(&pvec); i++) { - struct page *page = pvec.pages[i]; - pgoff_t index; - int skip_page = - (PageDirty(page) || PageWriteback(page)); - - if (!skip_page) - skip_page = !trylock_page(page); - - /* - * We really shouldn't be looking at the ->index of an - * unlocked page. But we're not allowed to lock these - * pages. So we rely upon nobody altering the ->index - * of this (pinned-by-us) page. - */ - index = page->index; - if (index > next) - next = index; - next++; - - if (skip_page) - continue; - - generic_error_remove_page(mapping, page); - unlock_page(page); - } - pagevec_release(&pvec); - cond_resched(); - } -} - -/* * Invalidate inode pages in a worker thread. (This can't be done * in the message handler context.) */ @@ -1429,7 +1385,7 @@ static void ceph_invalidate_work(struct work_struct *work) orig_gen = ci->i_rdcache_gen; spin_unlock(&inode->i_lock); - ceph_invalidate_nondirty_pages(inode->i_mapping); + truncate_inode_pages(&inode->i_data, 0); spin_lock(&inode->i_lock); if (orig_gen == ci->i_rdcache_gen && |