Attachment 295293 Details for Bug 359671 – RHEL4 patch for this issue

[patch] RHEL4 patch for this issue

359671.patch (text/plain), 1.22 KB, created by Prarit Bhargava on 2008-02-19 15:02:37 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Prarit Bhargava

Created: 2008-02-19 15:02:37 UTC

Size: 1.22 KB

patch

obsolete

>Straight backport of
>
>http://marc.info/?l=linux-kernel&m=115467098816513&w=2
>
>We can immediately bale from invalidate_bdev() if the blockdev has no
>pagecache.
>
>This solves the huge IPI storms which hald is causing on the big ia64
>machines when it polls CDROM drives.
>
>Confirmed with jes@sgi.com (acked patch upstream) that no other patches are
>required.
>
>Unfortunately, the hang seen in BZ 359671 is difficult to reproduce.  24 hours
>of booting/rebooting an altix did not cause the hang to occur.
>
>Compile and boot tested by me on 64p/128G Altix, and a 16p x86_64 system.
>
>Resolves BZ 359671.
>
>--- linux-2.6.9.orig/fs/buffer.c	2008-02-19 09:19:48.000000000 -0500
>+++ linux-2.6.9/fs/buffer.c	2008-02-19 09:35:47.000000000 -0500
>@@ -589,13 +589,18 @@ out:
>    pass does the actual I/O. */
> void invalidate_bdev(struct block_device *bdev, int destroy_dirty_buffers)
> {
>+	struct address_space *mapping = bdev->bd_inode->i_mapping;
>+
>+	if (mapping->nrpages == 0)
>+		return;
>+
> 	invalidate_bh_lrus();
> 	/*
> 	 * FIXME: what about destroy_dirty_buffers?
> 	 * We really want to use invalidate_inode_pages2() for
> 	 * that, but not until that's cleaned up.
> 	 */
>-	invalidate_inode_pages(bdev->bd_inode->i_mapping);
>+	invalidate_inode_pages(mapping);
> }
> 
> /*

Actions: View | Diff

Attachments on bug 359671: 295293