It seemed like a good idea, but it's actually a pain when we get more
than 32 feature bits. Just change it to a u32 for now.
Cc: Ohad Ben-Cohen
Cc: Brian Swetland
Cc: Cornelia Huck
Cc: Christian Borntraeger
Signed-off-by: Rusty Russell
Acked-by: Pawel Moll
---
drivers
Change the u32 to a u64, and make sure to use 1ULL everywhere!
Cc: Ohad Ben-Cohen
Cc: Brian Swetland
Cc: Cornelia Huck
Cc: Christian Borntraeger
Signed-off-by: Rusty Russell
Acked-by: Pawel Moll
---
drivers/char/virtio_console.c |2 +-
drivers/lguest/lguest_device.c
we've already demonstrated
that it's useful to be able to see the ring size, and the new virtio
pci layout wants to know the location of each part of the ring.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_ring.c | 114 +-
include
This lets the transport do endian conversion if necessary, and insulates
the drivers from the difference.
Most drivers can use the simple helpers virtio_cread() and virtio_cwrite().
Signed-off-by: Rusty Russell
---
drivers/block/virtio_blk.c | 77
This lets the transport do endian conversion if necessary, and insulates
the drivers from that change.
Signed-off-by: Rusty Russell
---
include/linux/virtio_config.h | 161 ++---
1 file changed, 134 insertions(+), 27 deletions(-)
diff --git a/include/linux
Simply redirect everything via the byte-at-a-time accessor. Slow, but simple
and this is config reading, which mostly only happens at probe time.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio.c | 69 +
include/linux/virtio_config.h
"Michael S. Tsirkin" writes:
> By the way, Gleb pointed out that on older hosts MMIO will
> always be slower since we need to do a shadow page walk to
> translate virtual to physical address.
> Hopefully not a big concern, and after all we are still
> keeping PIO around for use by BIOS ...
Yeah,
Aiming these for coming merge window.
Nothing should change, but get ready for a non-guest-endian config transports,
and > 32 features bits.
Final one is just an overdue consolidation.
Cheers,
Rusty.
Rusty Russell (7):
virtio_config: introduce size-based accessors.
virtio: use size-ba
"Michael S. Tsirkin" writes:
> On Wed, Apr 03, 2013 at 04:40:29PM +1030, Rusty Russell wrote:
>> Current proposal is a 16 bit 'offset' field in the queue data for each
>> queue, ie.
>> addr = dev->notify_base + vq->notify_off;
>>
>&
"H. Peter Anvin" writes:
> On 03/29/2013 08:19 PM, Rusty Russell wrote:
>>>
>>> Shift count?
>>
>> You can only have 2^16 vqs per device. Is it verboten to write 16-bit
>> values to odd offsets? If so, we've just dropped it to 2^15 befor
Dmitry Fleytman writes:
> From: Dmitry Fleytman
>
> Virtio-net driver currently negotiates network offloads
> on startup via features mechanism and have no ability to
> change offloads state later.
> This patch introduced a new control command that allows
> to configure device network offloads st
Amos Kong writes:
> Some head files were split or moved to uapi/ without
> updating MAINTAINERS.
>
> Signed-off-by: Amos Kong
Thanks, applied.
Cheers,
Rusty.
___
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linux
Paul Bolle writes:
> virtio_balloon.h exports "u16" and "u64" to userspace. Use "__u16" and
> "__u64" instead.
>
> Signed-off-by: Paul Bolle
> ---
> 0) Tested with a kernel build.
>
> 1) Both types are used in this header for quite some time now. So is my
> patch flawed or doesn't userspace actua
"Michael S. Tsirkin" writes:
> Rusty's currently doing some reorgs of -net let's delay
> cleanups there to avoid stepping on each other's toys.
> Let's focus on scsi here.
> E.g. any chance framing assumptions can be fixed in 3.10?
I am waiting for your removal of the dma-compelete ordering stuff
"H. Peter Anvin" writes:
> On 03/27/2013 04:25 AM, Michael S. Tsirkin wrote:
>>
>> Aha. Yes, good idea. As for how large the offsets are,
>> I am guessing we should either just say offset is vqn * X and data is
>> vqn, or give hypervisors full flexibility with 32 bit offset and
>> arbitrary data
Amit Shah writes:
> When multiple ovq operations are being performed (lots of open/close
> operations on virtio_console fds), the __send_control_msg() function can
> get confused without locking.
>
> A simple recipe to cause badness is:
> * create a QEMU VM with two virtio-serial ports
> * in the
"Michael S. Tsirkin" writes:
> On Mon, Mar 25, 2013 at 08:30:28PM +1030, Rusty Russell wrote:
>> Let's go back a level. Do we still need I/O bars at all now? Or can we
>> say "if you want hundreds of vqs, use mem bars"?
>>
>> hpa wanted
Andru Gheorghiu writes:
> PTR_RET does return. It's perfectly equivalent to using IS_ERR and the
> returning PTR_ERR. The implementation is here [1].
Um, I read the implementation, thanks.
> The reason for using it is this: if you have a function that does
> something why not call it instead of
Alexandru Gheorghiu writes:
> Used PTR_RET function instead of IS_ERR and PTR_ERR.
> Patch found using coccinelle.
WTF is PTR_RET? PTR_RET doesn't return anything. Why is it called
that? It doesn't even make sense.
ZERO_OR_PTR_ERR() maybe.
But what problem are we solving? Insufficient chur
Ohad Ben-Cohen writes:
> On Fri, Mar 22, 2013 at 4:50 PM, Sjur Brændeland wrote:
>> I guess you would need to update the feature bits in remoteproc as well?
>> e.g. something like:
>>
>> diff --git a/include/linux/remoteproc.h b/include/linux/remoteproc.h
>> index faf3332..148a503 100644
>> --- a
"H. Peter Anvin" writes:
> That might be pushing it, fitting into the 192-byte nonstandard area with
> everything else that might have to go there...
Yeah, and it only ever grows, since we add new fields at the end.
Cheers,
Rusty.
___
Virtualization m
"Michael S. Tsirkin" writes:
> On Fri, Mar 22, 2013 at 01:22:57PM +1030, Rusty Russell wrote:
>> "Michael S. Tsirkin" writes:
>> > I would like an option for hypervisor to simply say "Do IO
>> > to this fixed address for this VQ". Then
Sjur Brændeland writes:
> On Thu, Mar 21, 2013 at 9:29 AM, Rusty Russell wrote:
>> @@ -653,8 +676,14 @@ static int vp_set_vq_affinity(struct virtqueue *vq, int
>> cpu)
>> }
>>
>> static const struct virtio_config_ops virtio_pci_config_ops = {
&g
Sjur Brændeland writes:
> Check that vringh_config is not NULL before using it.
>
> Signed-off-by: Sjur Brændeland
Both applied.
Thanks,
Rusty.
___
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.or
Cornelia Huck writes:
> On Thu, 21 Mar 2013 18:59:24 +1030
> Rusty Russell wrote:
...
>> diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
>> index 6711e65..dcf35b1 100644
>> --- a/drivers/s390/kvm/kvm_virtio.c
>> +++ b/drivers/s390/kvm/kvm_
"Michael S. Tsirkin" writes:
> On Thu, Mar 21, 2013 at 06:59:37PM +1030, Rusty Russell wrote:
>> (MST, is this what you were thinking?)
>
> Almost.
>
> Three points:
>
> 1. this is still an offset in BAR so for KVM we are still forced to use
> an IO BAR.
Cornelia Huck writes:
> On Thu, 21 Mar 2013 18:59:26 +1030
> Rusty Russell wrote:
>
>> Change the u32 to a u64, and make sure to use 1ULL everywhere!
>
> And a not-even-compiled change for virtio_ccw as well:
Thanks, applied that too.
BTW, this will all be in my virtio-pci
Cornelia Huck writes:
> On Thu, 21 Mar 2013 18:59:25 +1030
> Rusty Russell wrote:
>
>> It seemed like a good idea, but it's actually a pain when we get more
>> than 32 feature bits. Just change it to a u32 for now.
>>
...
> I didn't try this patch, but
"Michael S. Tsirkin" writes:
> On Thu, Mar 21, 2013 at 02:44:50PM +0800, Amos Kong wrote:
>> The lengcy guests don't have mac programming command, we don't know when
>> it's safe to use MAC. This patch changed qemu to makes MAC change effect
>> when the last byte of MAC is written to config space.
"H. Peter Anvin" writes:
> On 03/21/2013 10:49 AM, Michael S. Tsirkin wrote:
>>
>> Just to clarify, I expect BIOS to use it *for config access*.
>> Notification will support IO anyway because it's
>> faster on KVM, so BIOS can use it directly.
>>
>
> Ah, yes, of course.
>
> Quite frankly, I don'
"Michael S. Tsirkin" writes:
> On Thu, Mar 21, 2013 at 06:59:34PM +1030, Rusty Russell wrote:
>> Differences:
>> 1) Uses 4 pci capabilities to demark common, irq, notify and dev-specific
>> areas.
>> 2) Guest sets queue size, using host-provided maximum.
>
Because we have potentially-infinite feature bits, it's hard for the device
to know when features are finalized.
This adds a new status bit, VIRTIO_CONFIG_S_FEATURES_DONE, which is only
set by the modern virtio_pci driver at the moment.
Signed-off-by: Rusty Russell
---
drivers/v
Our helpers can take a virtio_pci_device, rather than converting from
a virtio_device all the time. They couldn't do this when they were
called from the common virtio code, but now we wrap them anyway, it
simplifies things.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci-com
There's a great deal of work in setting up and disabling interrupts,
particularly with MSI-X, which is generic. So we move most of the
work out to helpers which take the location of the msix_config
register, and setup_vq and del_vq functions.
Signed-off-by: Rusty Russell
---
drivers/v
If we make the legacy driver set up the ->isr pointer in
the struct virtio_pci_device structure, we can use that in common code.
(the positions have changed, but the semantics haven't).
Signed-off-by: Rusty Russell
---
drivers/virtio/Makefile|4 +-
drivers/virtio/vi
They're almost identical: we add a "legacy" ioregion (what was
"ioaddr" in the legacy driver), and move it out to virtio_pci-common.h.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci-common.h | 76 +
drivers/virtio/v
By removing the redundant num field and including a pointer to where we
should notify, it exactly matches the virtio_pci one now.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci_legacy.c | 21 +++--
1 file changed, 11 insertions(+), 10 deletions(-)
diff --git a
en: it can still distinguish different notifications).
Cc: "Michael S. Tsirkin"
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci.c | 61 +++
include/uapi/linux/virtio_pci.h |2 +-
2 files changed, 44 insertions(+), 19 deletions(-)
Previously, it was defined as "guest-endian". This was always confusing
for PCI, for which everything else is defined a little endian.
The ring itself is unchanged, this is just the per-device config info.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pc
queue_address was at offset 28 (0x1C),
which is unusual, so add more padding to take the first 64-bit value
to offset 32 (0x20).
Cc: H. Peter Anvin
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci.c | 48 +++
include/uapi/linux/virtio_pci.h | 11
Differences:
1) Uses 4 pci capabilities to demark common, irq, notify and dev-specific areas.
2) Guest sets queue size, using host-provided maximum.
3) Guest sets queue alignment, rather than ABI-defined 4096.
4) More than 32 feature bits (a lot more!).
Signed-off-by: Rusty Russell
Signed-off-by
Another HPA suggestion: that the device be allowed to offer duplicate
capabilities, particularly so it can offer a mem and an I/O bar and let
the guest decide (Linux guest probably doesn't care?).
Cc: H. Peter Anvin
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci_legacy.c |
With module option to override.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci_legacy.c | 20 +++-
include/linux/virtio_pci.h | 16
2 files changed, 35 insertions(+), 1 deletion(-)
diff --git a/drivers/virtio/virtio_pci_legacy.c
b
include/linux/virtio_pci.h turns off the compat defines, and we use it
rather than including include/uapi/linux/virtio_pci.h directly.
This makes it obvious if we use legacy defines elsewhere.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_pci_legacy.c | 62
We don't *remove* the old ones, unless VIRTIO_PCI_NO_LEGACY is defined,
but they get a friendly #warning about the change.
Note that config option is not prompted; we always enable it for now.
Signed-off-by: Rusty Russell
---
drivers/virtio/Kconfig | 12 +
drivers/virtio/Mak
From: Michael S Tsirkin
Virtio drivers should map the part of the range they need, not necessarily
all of it. They also need non-cacheable mapping even for
prefetchable BARs.
Signed-off-by: Michael S. Tsirkin
Signed-off-by: Rusty Russell
---
include/asm-generic/pci_iomap.h |5 +
lib
Based on patch by Michael S. Tsirkin , but I found it
hard to follow so changed to use structures which are more
self-documenting.
Signed-off-by: Rusty Russell
---
include/uapi/linux/virtio_pci.h | 41 +++
1 file changed, 41 insertions(+)
diff --git a
Change the u32 to a u64, and make sure to use 1ULL everywhere!
Cc: Ohad Ben-Cohen
Cc: Brian Swetland
Cc: Cornelia Huck
Cc: Pawel Moll
Cc: Christian Borntraeger
Signed-off-by: Rusty Russell
---
drivers/char/virtio_console.c |2 +-
drivers/lguest/lguest_device.c | 10
we've already demonstrated
that it's useful to be able to see the ring size, and the new virtio
pci layout wants to know the location of each part of the ring.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_ring.c | 114 +-
include
All transports just pass through at the moment.
Cc: Ohad Ben-Cohen
Cc: Brian Swetland
Cc: Cornelia Huck
Cc: Pawel Moll
Cc: Christian Borntraeger
Signed-off-by: Rusty Russell
---
drivers/lguest/lguest_device.c | 79 ++--
drivers/net/caif/caif_virtio.c
It seemed like a good idea, but it's actually a pain when we get more
than 32 feature bits. Just change it to a u32 for now.
Cc: Ohad Ben-Cohen
Cc: Brian Swetland
Cc: Cornelia Huck
Cc: Pawel Moll
Cc: Christian Borntraeger
Signed-off-by: Rusty Russell
---
drivers/char/virtio_cons
This lets the transport do endian conversion if necessary, and insulates
the drivers from that change.
Most drivers can use the simple helpers virtio_cread() and virtio_cwrite().
Signed-off-by: Rusty Russell
---
drivers/block/virtio_blk.c | 77
This lets the transport do endian conversion if necessary, and insulates
the drivers from that change.
Signed-off-by: Rusty Russell
---
include/linux/virtio_config.h | 161 ++---
1 file changed, 134 insertions(+), 27 deletions(-)
diff --git a/include/linux
etely untested.
Comments gratefully received!
Rusty.
Michael S Tsirkin (1):
pci: add pci_iomap_range
Rusty Russell (21):
virtio_config: introduce size-based accessors.
virtio_config: use size-based accessors.
virtio_config: make transports implement accessors.
virtio: use u
Ohad Ben-Cohen writes:
> On Mon, Mar 18, 2013 at 11:34 AM, Rusty Russell wrote:
>> We never add buffers with input and output parts, so use the new accessors.
>>
>> Cc: Ohad Ben-Cohen
>> Signed-off-by: Rusty Russell
>
> Acked-by: Ohad Ben-Cohen
Thanks. U
Rusty Russell writes:
> virtio_add_buf() is going away, replaced with virtio_add_sgs() which
> takes multiple terminated scatterlists.
>
> Cc: Eric Van Hensbergen
> Signed-off-by: Rusty Russell
This one was buggy. Testing and reading the patch do wonderful things.
Che
Rusty Russell writes:
> Add virtqueue_add_sgs which is more general than virtqueue_add_buf,
> which makes virtio-scsi and virtio-blk nicer, then add virtqueue_add_inbuf
> and virtqueue_add_outbuf which handle the more general case, and finally
> delete virtqueue_add_buf().
>
&g
Reviewed-by: Asias He
> Signed-off-by: Rusty Russell
Thanks... and I agree with Paulo: this is nice.
I've replaced the patch I posted with this one.
Cheers,
Rusty.
___
Virtualization mailing list
Virtualization@lists.linux-foundation.org
http
"Michael S. Tsirkin" writes:
> On Thu, Mar 14, 2013 at 01:10:20PM +0200, Michael S. Tsirkin wrote:
>> Add field for guest to specify command size for virtio-blk.
>>
>> Signed-off-by: Michael S. Tsirkin
>
> There's one concern here: are we going to add a
> feature flag for flexible layout? If yes
"Michael S. Tsirkin" writes:
> On Mon, Mar 18, 2013 at 08:04:44PM +1030, Rusty Russell wrote:
>> It's a bit cleaner to hand multiple sgs, rather than one big one.
>>
>> Cc: "Michael S. Tsirkin"
>> Signed-off-by: Rusty
All users changed to virtqueue_add_sg() or virtqueue_add_outbuf/inbuf.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_ring.c | 31 ---
include/linux/virtio.h |7 ---
2 files changed, 38 deletions(-)
diff --git a/drivers/virtio/virtio_ring.c b
Signed-off-by: Rusty Russell
---
tools/virtio/linux/virtio.h |7 ---
tools/virtio/virtio_test.c |6 +++---
tools/virtio/vringh_test.c | 22 --
3 files changed, 15 insertions(+), 20 deletions(-)
diff --git a/tools/virtio/linux/virtio.h b/tools/virtio/linux
virtio_add_buf() is going away, replaced with virtio_add_sgs() which
takes multiple terminated scatterlists.
Cc: Eric Van Hensbergen
Signed-off-by: Rusty Russell
---
net/9p/trans_virtio.c | 48 +++-
1 file changed, 39 insertions(+), 9 deletions
It's a bit cleaner to hand multiple sgs, rather than one big one.
Cc: "Michael S. Tsirkin"
Signed-off-by: Rusty Russell
---
drivers/net/virtio_net.c | 50 +++---
1 file changed, 25 insertions(+), 25 deletions(-)
diff --git a/drivers/net
We never add buffers with input and output parts, so use the new accessors.
Cc: Ohad Ben-Cohen
Signed-off-by: Rusty Russell
---
drivers/rpmsg/virtio_rpmsg_bus.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg
We never add buffers with input and output parts, so use the new accessors.
Cc: Sjur Brendeland
Signed-off-by: Rusty Russell
---
drivers/net/caif/caif_virtio.c |3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/caif/caif_virtio.c b/drivers/net/caif
We never add buffers with input and output parts, so use the new accessors.
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_balloon.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index
We never add buffers with input and output parts, so use the new accessors.
Cc: Amit Shah
Signed-off-by: Rusty Russell
---
drivers/char/virtio_console.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
We never add buffers with input and output parts, so use the new accessors.
Signed-off-by: Rusty Russell
---
drivers/char/hw_random/virtio-rng.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/char/hw_random/virtio-rng.c
b/drivers/char/hw_random/virtio-rng.c
We never add buffers with input and output parts, so use the new accessors.
Cc: "Michael S. Tsirkin"
Signed-off-by: Rusty Russell
---
drivers/net/virtio_net.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/vi
It's simply a flag as to whether we have data now, so make it an
explicit function parameter rather than a member of struct
virtblk_req.
Signed-off-by: Rusty Russell
Reviewed-by: Asias He
---
drivers/block/virtio_blk.c | 22 +++---
1 file changed, 11 insertions(+
: Rusty Russell
---
drivers/scsi/virtio_scsi.c | 91 +---
1 file changed, 34 insertions(+), 57 deletions(-)
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index 0f5dd28..bb67576 100644
--- a/drivers/scsi/virtio_scsi.c
+++ b/drivers
It's a bit clearer, and add_buf is going away.
Signed-off-by: Rusty Russell
---
drivers/scsi/virtio_scsi.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index bb67576..f679b8c 100644
--- a/drivers
#x27;t use vbr->sg, however, we
need to add a couple of arguments to __virtblk_add_req.
We also need to teach __virtblk_add_req how to build SCSI command
requests.
Signed-off-by: Paolo Bonzini
Signed-off-by: Rusty Russell
Reviewed-by: Asias He
---
drivers/block/virtio
parately using virtqueue_add_sgs().
With this change, virtio-blk (with use_bio) is not relying anymore on
the virtio functions ignoring the end markers in a scatterlist.
The next patch will do the same for the other path.
Signed-off-by: Paolo Bonzini
Signed-off-by: Rusty Russell
Reviewed-by:
These are specialized versions of virtqueue_add_buf(), which cover
over 80% of cases and are far clearer.
In particular, the scatterlists passed to these functions don't have
to be clean (ie. we ignore end markers).
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_ring.c |
-39063(39062)
Host: notified 39037-39063(39062), pinged 0
Wall time:1.64-1.81(1.676875)
Signed-off-by: Rusty Russell
---
tools/virtio/linux/virtio.h | 10 ++
tools/virtio/vringh_test.c |8
2 files changed, 14 insertions(+), 4 deletions(-)
diff
Bonzini
Reviewed-by: Asias He
Signed-off-by: Rusty Russell
---
drivers/block/virtio_blk.c | 55
1 file changed, 20 insertions(+), 35 deletions(-)
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 922bcb9..b271650 100644
--- a
From: Paolo Bonzini
This is useful in places that recycle the same scatterlist multiple
times, and do not want to incur the cost of sg_init_table every
time in hot paths.
Acked-by: Jens Axboe
Signed-off-by: Paolo Bonzini
Signed-off-by: Rusty Russell
---
block/blk-integrity.c |2
1.70-1.95(1.723542)
After:
Using CPUS 0 and 3
Guest: notified 0, pinged 39019-39063(39061)
Host: notified 39019-39063(39061), pinged 0
Wall time:2.09-2.52(2.188542)
Signed-off-by: Rusty Russell
Reviewed-by: Wanlong Gao
Reviewed-by: Asias He
--
whole series, and it can
move from my pending-rebases tree into virtio-next.
Thanks!
Rusty.
Paolo Bonzini (4):
scatterlist: introduce sg_unmark_end
virtio-blk: reorganize virtblk_add_req
virtio-blk: use virtqueue_add_sgs on bio path
virtio-blk: use virtqueue_add_sgs on req path
Rusty Russel
63)
Host: notified 39062-39063(39063), pinged 0
Wall time:1.76-2.22(1.789167)
Signed-off-by: Rusty Russell
---
drivers/virtio/virtio_ring.c | 91 ++
1 file changed, 56 insertions(+), 35 deletions(-)
diff --git a/drivers/virti
(39062), pinged 0
Wall time:1.90-2.35(1.921875)
Signed-off-by: Rusty Russell
Reviewed-by: Wanlong Gao
---
drivers/virtio/virtio_ring.c | 147 +++---
1 file changed, 80 insertions(+), 67 deletions(-)
diff --git a/drivers/virtio/virtio_ring.c
evice
>>>> struct. This enables virtio drivers to manage virtio
>>>> host rings without detailed knowledge of how the
>>>> vrings are created and deleted.
>>>>
>>>> The function vringh_notify() is added so vringh clients
>>>>
Amit Shah writes:
> On (Wed) 13 Mar 2013 [12:10:48], sjur.brandel...@stericsson.com wrote:
>> From: Sjur Brændeland
>>
>>
>> This reverts commit 8078db789a92b10ff6e2d713231b5367e014c53b, and
>> adds a lengthy comment explaining the problem area.
>>
>> The reverted patch caused opening of port
anks,
Rusty.
caif_virtio: remove unneeded remoteproc.h
Reported-by: Ohad Ben-Cohen
Signed-off-by: Rusty Russell
diff --git a/drivers/net/caif/caif_virtio.c b/drivers/net/caif/caif_virtio.c
index 6a745dd..b1e1205 100644
--- a/drivers/net/caif/caif_virtio.c
+++ b/drivers/net/caif/caif_virtio.c
@@ -1
David Miller writes:
> From: Erwan Yvin
> Date: Fri, 15 Mar 2013 10:42:17 +0100
>
>> caif-virtio is going to replace caif-shm.
>> This patch should be merged in rusty's tree. (vringh)
>> because there is a dependency with vringh wrapper.
>
> Feel free to add my:
>
> Acked-by: David S. Miller
T
n_pool_create" [drivers/net/caif/caif_virtio.ko] undefined!
ERROR: "gen_pool_destroy" [drivers/net/caif/caif_virtio.ko] undefined!
ERROR: "gen_pool_free" [drivers/net/caif/caif_virtio.ko] undefined!
ERROR: "gen_pool_alloc" [drivers/net/caif/caif_virtio.ko] undefined!
Sig
"Michael S. Tsirkin" writes:
> On Thu, Mar 14, 2013 at 05:08:22PM +1030, Rusty Russell wrote:
>> "Michael S. Tsirkin" writes:
>> > On Wed, Mar 13, 2013 at 05:19:51PM +1030, Rusty Russell wrote:
>> >> OK, I've been trying to read the vhost
sjur.brandel...@stericsson.com writes:
> From: Sjur Brændeland
>
>
> This reverts commit 8078db789a92b10ff6e2d713231b5367e014c53b, and
> adds a lengthy comment explaining the problem area.
>
> The reverted patch caused opening of ports to fail for rproc_serial.
> In probe guest_connected was set
"Michael S. Tsirkin" writes:
> On Wed, Mar 13, 2013 at 05:19:51PM +1030, Rusty Russell wrote:
>> OK, I've been trying to read the vhost and vhost net code, and I'm
>> struggling with basic questions.
>>
>> 1) Why do we allow userspace to change an
OK, I've been trying to read the vhost and vhost net code, and I'm
struggling with basic questions.
1) Why do we allow userspace to change an already-set-up vhost device?
Why not have:
open()
ioctl(VHOST_GET_FEATURES)
ioctl(VHOST_SET_VRING) x n (sets num, addresses, kick
Sjur Brændeland writes:
> On Tue, Mar 12, 2013 at 8:34 AM, Amit Shah wrote:
>> On (Mon) 11 Mar 2013 [16:15:00], sjur.brandel...@stericsson.com wrote:
>>> From: Sjur Brćndeland
>>>
>>> This reverts commit 8078db789a92b10ff6e2d713231b5367e014c53b.
>>>
>>> The reverted patch caused opening of ports
Paolo Bonzini writes:
> Il 07/03/2013 04:29, Wanlong Gao ha scritto:
>> Convert the virtio-scsi driver to use pr_err() instead of printk().
>>
>> Signed-off-by: Wanlong Gao
>
> Acked-by: Paolo Bonzini
Thanks, applied.
Cheers,
Rusty.
___
Virtualizat
sjur.brandel...@stericsson.com writes:
> From: Sjur Brændeland
>
> This reverts commit 8078db789a92b10ff6e2d713231b5367e014c53b.
>
> The reverted patch caused opening of ports to fail for rproc_serial.
> In probe guest_connected was set to true, but port_fops_open()
> fails with -EMFILE if guest_
Sjur Brændeland writes:
> Hi Rusty,
>
> The two similar functions in vringh and virtqueue for turning on
> interrupts has opposite return values if there are buffers available
> in the ring. I think it would be better if these two functions aligned
> the use of return values. Maybe it's just me,
Cornelia Huck writes:
> On Thu, 7 Mar 2013 20:02:21 +0200
> "Michael S. Tsirkin" wrote:
>
>> virtio-s390 on kvm can use a cookie value passed to guest
>
> s/virtio-s390/virtio-ccw/ (to avoid confusion with s390-virtio, which
> was never specced)
>
>> to optimize channel/VQ lookups.
>> Document th
Hi Michael,
I'm a bit confused about why ubuf_info and heads are UIO_MAXIOV
length arrays, rather than being the size of the ring? In particular,
this is suspicious:
linux/drivers/vhost/net.c:342: struct ubuf_info *ubuf = &vq->ubuf_info[head];
And it seems to assume we trust head: a ma
"Michael S. Tsirkin" writes:
> Here's a patchset to update tools/virtio with
> changes in 3.8, as well as the proposed API extension.
>
> Useful to see whether the extra level of indirection
> adds measureable overhead.
>
> I will upstream at least patch 1 soon.
What happened to this, BTW?
> Mic
"Michael S. Tsirkin" writes:
> On Thu, Mar 07, 2013 at 02:48:24PM +1100, Rusty Russell wrote:
>> "Michael S. Tsirkin" writes:
>> > On Wed, Mar 06, 2013 at 03:54:42PM +1100, Rusty Russell wrote:
>> >> In the coming vringh_test, we share an mma
301 - 400 of 1795 matches
Mail list logo
