rpms/kernel/F-11 linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch, NONE, 1.1 kernel.spec, 1.1736, 1.1737

[Justin M. Forbes]

Author: jforbes

Update of /cvs/pkgs/rpms/kernel/F-11
In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv17539

Modified Files:
	kernel.spec 
Added Files:
	linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch 
Log Message:
Revert virtio_blk to rotational mode. (#509383)

linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch:
 virtio_blk.c |    1 -
 1 file changed, 1 deletion(-)

--- NEW FILE linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch ---
Subject: [for-2.6.31] virtio_blk: revert QUEUE_FLAG_VIRT addition
Date: Fri, 04 Sep 2009 20:44:42 -0000
From: Christoph Hellwig <hch at lst.de>

It seems like the addition of QUEUE_FLAG_VIRT caueses major performance
regressions for Fedora users:

	https://bugzilla.redhat.com/show_bug.cgi?id=509383
	https://bugzilla.redhat.com/show_bug.cgi?id=505695

while I can't reproduce those extreme regressions myself I think the flag
is wrong.

Rationale:

  QUEUE_FLAG_VIRT expands to QUEUE_FLAG_NONROT which casus the queue
  unplugged immediately.  This is not a good behaviour for at least
  qemu and kvm where we do have significant overhead for every
  I/O operations.  Even with all the latested speeups (native AIO,
  MSI support, zero copy) we can only get native speed for up to 128kb
  I/O requests we already are down to 66% of native performance for 4kb
  requests even on my laptop running the Intel X25-M SSD for which the
  QUEUE_FLAG_NONROT was designed.
  If we ever get virtio-blk overhead low enough that this flag makes
  sense it should only be set based on a feature flag set by the host.
	
Signed-off-by: Christoph Hellwig <hch at lst.de>
Acked-by: Jeff Moyer <jmoyer at redhat.com>

---
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Index: linux-2.6/drivers/block/virtio_blk.c
===================================================================
--- linux-2.6.orig/drivers/block/virtio_blk.c	2009-09-04 17:33:48.802523987 -0300
+++ linux-2.6/drivers/block/virtio_blk.c	2009-09-04 17:33:56.186522158 -0300
@@ -249,7 +249,6 @@ static int __devinit virtblk_probe(struc
        goto out_mempool;
 	}
 
-	queue_flag_set_unlocked(QUEUE_FLAG_VIRT, vblk->disk->queue);
 
 	if (index < 26) {
 		sprintf(vblk->disk->disk_name, "vd%c", 'a' + index % 26);


Index: kernel.spec
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-11/kernel.spec,v
retrieving revision 1.1736
retrieving revision 1.1737
diff -u -p -r1.1736 -r1.1737
--- kernel.spec	15 Sep 2009 19:31:06 -0000	1.1736
+++ kernel.spec	16 Sep 2009 15:38:06 -0000	1.1737
@@ -656,6 +656,7 @@ Patch1000: linux-2.6-neigh_-fix-state-tr
 # Virt Patches
 Patch1200: linux-2.6-xen-fix-brkpoints-hw-watchpoints.patch
 Patch1201: linux-2.6-xen-clean-up-warnings.patch
+Patch1202: linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch
 
 Patch1515: lirc-2.6.31.patch
 Patch1517: hid-ignore-all-recent-imon-devices.patch
@@ -1265,6 +1266,9 @@ ApplyPatch linux-2.6-e1000-ich9.patch
 ApplyPatch linux-2.6-xen-fix-brkpoints-hw-watchpoints.patch
 ApplyPatch linux-2.6-xen-clean-up-warnings.patch
 
+# Misc Virt
+ApplyPatch linux-2.6-virtio_blk-revert-QUEUE_FLAG_VIRT-addition.patch
+
 # Misc fixes
 # The input layer spews crap no-one cares about.
 ApplyPatch linux-2.6-input-kill-stupid-messages.patch
@@ -1967,6 +1971,9 @@ fi
 # and build.
 
 %changelog
+* Wed Sep 16 2009 Justin M. Forbes <jforbes at redhat.com> 2.6.30.7-58
+- Revert virtio_blk to rotational mode. (#509383)
+
 * Tue Sep 15 2009 Chuck Ebbert <cebbert at redhat.com> 2.6.30.7-57
 - Linux 2.6.30.7