Andrew Morton wrote:
>
> Unfortunately we have a conflict with the already-queued
> do_mpage_readpage-dont-submit-lots-of-small-bios-on-boundary.patch:
>
How about amending this on top of take #2 ?
Franck
---8<---
diff --git a/fs/mpage.c b/fs/mpage.c
index 74e7f83..ef26eb1 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -157,6 +157,19 @@ map_buffer_to_page(struct page *page, struct buffer_head *bh, int page_block)
}
/*
+ * Since get_block() can map a range of pages, for all these pages the
+ * BH_Boundary flag will be set. But we only need to push what I/O we
+ * have accumulated at the last block of this range.
+*/
+static int last_mapped_page(struct buffer_head *map_bh, struct page *page)
+{
+ unsigned map_index = map_bh->b_size >> PAGE_CACHE_SHIFT;
+
+ return buffer_boundary(map_bh) &&
+ page->index - map_bh->b_page->index >= map_index;
+}
+
+/*
* This is the worker routine which does all the work of mapping the disk
* blocks and constructs largest possible bios, submits them for IO if the
* blocks are not contiguous on the disk.
@@ -312,10 +325,7 @@ alloc_new:
goto alloc_new;
}
- relative_block = block_in_file - *first_logical_block;
- nblocks = map_bh->b_size >> blkbits;
- if ((buffer_boundary(map_bh) && relative_block == nblocks) ||
- (first_hole != blocks_per_page))
+ if (last_mapped_page(map_bh, page) || first_hole != blocks_per_page)
bio = mpage_bio_submit(READ, bio);
else
*last_block_in_bio = blocks[blocks_per_page - 1];
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
