writeback: pay attention to wbc->nr_to_write in write_cache_pages
Dave Chinner [Wed, 9 Jun 2010 00:37:18 +0000 (10:37 +1000)]
If a filesystem writes more than one page in ->writepage, write_cache_pages
fails to notice this and continues to attempt writeback when wbc->nr_to_write
has gone negative - this trace was captured from XFS:

    wbc_writeback_start: towrt=1024
    wbc_writepage: towrt=1024
    wbc_writepage: towrt=0
    wbc_writepage: towrt=-1
    wbc_writepage: towrt=-5
    wbc_writepage: towrt=-21
    wbc_writepage: towrt=-85

This has adverse effects on filesystem writeback behaviour. write_cache_pages()
needs to terminate after a certain number of pages are written, not after a
certain number of calls to ->writepage are made.  This is a regression
introduced by 17bc6c30cf6bfffd816bdc53682dd46fc34a2cf4 ("vfs: Add
no_nrwrite_index_update writeback control flag"), but cannot be reverted
directly due to subsequent bug fixes that have gone in on top of it.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/writeback.h
include/trace/events/ext4.h
mm/page-writeback.c

index f641346..d63ef8f 100644 (file)
@@ -56,15 +56,6 @@ struct writeback_control {
        unsigned for_reclaim:1;         /* Invoked from the page allocator */
        unsigned range_cyclic:1;        /* range_start is cyclic */
        unsigned more_io:1;             /* more io to be dispatched */
-       /*
-        * write_cache_pages() won't update wbc->nr_to_write and
-        * mapping->writeback_index if no_nrwrite_index_update
-        * is set.  write_cache_pages() may write more than we
-        * requested and we want to make sure nr_to_write and
-        * writeback_index are updated in a consistent manner
-        * so we use a single control to update them
-        */
-       unsigned no_nrwrite_index_update:1;
 };
 
 /*
index f5b1ba9..f3865c7 100644 (file)
@@ -306,7 +306,6 @@ TRACE_EVENT(ext4_da_writepages_result,
                __field(        int,    pages_written           )
                __field(        long,   pages_skipped           )
                __field(        char,   more_io                 )       
-               __field(        char,   no_nrwrite_index_update )
                __field(       pgoff_t, writeback_index         )
        ),
 
@@ -317,16 +316,14 @@ TRACE_EVENT(ext4_da_writepages_result,
                __entry->pages_written  = pages_written;
                __entry->pages_skipped  = wbc->pages_skipped;
                __entry->more_io        = wbc->more_io;
-               __entry->no_nrwrite_index_update = wbc->no_nrwrite_index_update;
                __entry->writeback_index = inode->i_mapping->writeback_index;
        ),
 
-       TP_printk("dev %s ino %lu ret %d pages_written %d pages_skipped %ld more_io %d no_nrwrite_index_update %d writeback_index %lu",
+       TP_printk("dev %s ino %lu ret %d pages_written %d pages_skipped %ld more_io %d writeback_index %lu",
                  jbd2_dev_to_name(__entry->dev),
                  (unsigned long) __entry->ino, __entry->ret,
                  __entry->pages_written, __entry->pages_skipped,
                  __entry->more_io,
-                 __entry->no_nrwrite_index_update,
                  (unsigned long) __entry->writeback_index)
 );
 
index 5fa63bd..b3dbb80 100644 (file)
@@ -835,7 +835,6 @@ int write_cache_pages(struct address_space *mapping,
        pgoff_t done_index;
        int cycled;
        int range_whole = 0;
-       long nr_to_write = wbc->nr_to_write;
 
        pagevec_init(&pvec, 0);
        if (wbc->range_cyclic) {
@@ -935,11 +934,10 @@ continue_unlock:
                                        done = 1;
                                        break;
                                }
-                       }
+                       }
 
-                       if (nr_to_write > 0) {
-                               nr_to_write--;
-                               if (nr_to_write == 0 &&
+                       if (wbc->nr_to_write > 0) {
+                               if (--wbc->nr_to_write == 0 &&
                                    wbc->sync_mode == WB_SYNC_NONE) {
                                        /*
                                         * We stop writing back only if we are
@@ -970,11 +968,8 @@ continue_unlock:
                end = writeback_index - 1;
                goto retry;
        }
-       if (!wbc->no_nrwrite_index_update) {
-               if (wbc->range_cyclic || (range_whole && nr_to_write > 0))
-                       mapping->writeback_index = done_index;
-               wbc->nr_to_write = nr_to_write;
-       }
+       if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
+               mapping->writeback_index = done_index;
 
        return ret;
 }