Patch from Dmitry Monakhov:
Previous fix for retries in ext3_prepare_write() violation
was a bit errorneuos:
- it missed return of error code from ext3_journal_stop()
- it missed do_journal_get_write_access() before commit_write

a few comments added also.

Signed-Off-By: Dmitry Monakhov <[EMAIL PROTECTED]>
Signed-Off-By: Kirill Korotaev <[EMAIL PROTECTED]>


diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index a48ada9..1acb528 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -1162,13 +1162,14 @@ static int ext3_prepare_failure(struct f
        struct buffer_head *bh, *head, *next;
        unsigned block_start, block_end;
        unsigned blocksize;
+       int ret;
+       handle_t *handle = ext3_journal_current_handle();
 
        mapping = page->mapping;
        if (ext3_should_writeback_data(mapping->host)) {
                /* optimization: no constraints about data */
 skip:
-               ext3_journal_stop(ext3_journal_current_handle());
-               return 0;
+               return ext3_journal_stop(handle);
        }
 
        head = page_buffers(page);
@@ -1186,7 +1187,19 @@ skip:
                        break;
                }
                if (!buffer_mapped(bh))
+               /* prepare_write failed on this bh */
                        break;
+               if (ext3_should_journal_data(mapping->host)) {
+                       ret = do_journal_get_write_access(handle, bh);
+                       if (ret) {
+                               ext3_journal_stop(handle);
+                               return ret;
+                       }
+               }
+       /*
+        * block_start here becomes the first block where the current iteration
+        * of prepare_write failed.
+        */
        }
        if (block_start <= from)
                goto skip;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to