(Sorry for delay, recent absences, but all been in linux-next a week now)
The following changes since commit 2fdc246aaf9a7fa088451ad2a72e9119b5f7f029:
Linus Torvalds (1):
Merge branch 'for-linus' of git://git.kernel.org/.../bp/bp
are available in the git repository at:
ssh://master.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus.git
master
Christian Borntraeger (1):
virtio: let header files include virtio_ids.h
Christoph Hellwig (1):
virtio_blk: revert QUEUE_FLAG_VIRT addition
Rusty Russell (1):
virtio_blk: Revert serial number support
Uwe Kleine-K?nig (2):
move virtballoon_remove to .devexit.text
move virtrng_remove to .devexit.text
Documentation/lguest/lguest.c | 1 -
drivers/block/virtio_blk.c | 39 ++--------------------------------
drivers/char/hw_random/virtio-rng.c | 3 +-
drivers/char/virtio_console.c | 1 -
drivers/net/virtio_net.c | 1 -
drivers/virtio/virtio_balloon.c | 3 +-
include/linux/Kbuild | 1 +
include/linux/virtio_9p.h | 1 +
include/linux/virtio_balloon.h | 1 +
include/linux/virtio_blk.h | 5 +---
include/linux/virtio_console.h | 1 +
include/linux/virtio_net.h | 1 +
include/linux/virtio_rng.h | 1 +
net/9p/trans_virtio.c | 1 -
14 files changed, 12 insertions(+), 48 deletions(-)
commit d5a01175aa25ef30a99545f09e0a2828a226d3cd
Author: Christoph Hellwig <hch at lst.de>
Date: Fri Sep 4 22:44:42 2009 +0200
virtio_blk: revert QUEUE_FLAG_VIRT addition
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>
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
drivers/block/virtio_blk.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)
commit 6ff87d1f188f65a51c28595e9ad16183cbd58d94
Author: Christian Borntraeger <borntraeger at de.ibm.com>
Date: Wed Sep 30 11:17:21 2009 +0200
virtio: let header files include virtio_ids.h
Rusty,
commit 3ca4f5ca73057a617f9444a91022d7127041970a
virtio: add virtio IDs file
moved all device IDs into a single file. While the change itself is
a very good one, it can break userspace applications. For example
if a userspace tool wanted to get the ID of virtio_net it used to
include virtio_net.h. This does no longer work, since virtio_net.h
does not include virtio_ids.h.
This patch moves all "#include <linux/virtio_ids.h>" from
the C
files into the header files, making the header files compatible with
the old ones.
In addition, this patch exports virtio_ids.h to userspace.
CC: Fernando Luis Vazquez Cao <fernando at oss.ntt.co.jp>
Signed-off-by: Christian Borntraeger <borntraeger at de.ibm.com>
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
Documentation/lguest/lguest.c | 1 -
drivers/block/virtio_blk.c | 1 -
drivers/char/hw_random/virtio-rng.c | 1 -
drivers/char/virtio_console.c | 1 -
drivers/net/virtio_net.c | 1 -
drivers/virtio/virtio_balloon.c | 1 -
include/linux/Kbuild | 1 +
include/linux/virtio_9p.h | 1 +
include/linux/virtio_balloon.h | 1 +
include/linux/virtio_blk.h | 1 +
include/linux/virtio_console.h | 1 +
include/linux/virtio_net.h | 1 +
include/linux/virtio_rng.h | 1 +
net/9p/trans_virtio.c | 1 -
14 files changed, 7 insertions(+), 7 deletions(-)
commit ef728ac6296b8ae36791c7644e92b7bc75c0fd88
Author: Rusty Russell <rusty at rustcorp.com.au>
Date: Mon Oct 19 12:58:28 2009 -0600
virtio_blk: Revert serial number support
This reverts "Add serial number support for virtio_blk, V4a".
Turns out that virtio_pci, lguest and s/390 all have an 8 bit limit
on virtio config space, so noone could ever use this.
This is coming back later in a cleaner form.
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
Cc: john cooper <john.cooper at redhat.com>
Cc: Jens Axboe <jens.axboe at oracle.com>
drivers/block/virtio_blk.c | 37 +++----------------------------------
include/linux/virtio_blk.h | 4 ----
2 files changed, 3 insertions(+), 38 deletions(-)
commit 22b4b8fe00398cb66cb9a95d3f0c21b0155e5421
Author: Uwe Kleine-K?nig <u.kleine-koenig at pengutronix.de>
Date: Thu Oct 1 10:28:33 2009 +0200
move virtballoon_remove to .devexit.text
The function virtballoon_remove is used only wrapped by __devexit_p so
define it using __devexit.
Signed-off-by: Uwe Kleine-K?nig <u.kleine-koenig at pengutronix.de>
Acked-by: Sam Ravnborg <sam at ravnborg.org>
Acked-by: Michael S. Tsirkin <mst at redhat.com>
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
drivers/virtio/virtio_balloon.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
commit 360390725dce2961ac35be5f59640f9d44f03d25
Author: Uwe Kleine-K?nig <u.kleine-koenig at pengutronix.de>
Date: Thu Oct 1 10:28:35 2009 +0200
move virtrng_remove to .devexit.text
The function virtrng_remove is used only wrapped by __devexit_p so define
it using __devexit.
Signed-off-by: Uwe Kleine-K?nig <u.kleine-koenig at pengutronix.de>
Acked-by: Sam Ravnborg <sam at ravnborg.org>
Cc: Rusty Russell <rusty at rustcorp.com.au>
Cc: Michael S. Tsirkin <mst at redhat.com>
Acked-by: Christian Borntraeger <borntraeger at de.ibm.com>
Cc: linux-kernel at vger.kernel.org
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
drivers/char/hw_random/virtio-rng.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)