Hi all,
I'm trying to get VM started with libiscsi (iSCSI initiator) device. Every
thing seems good from iSCSI point of view, but I'm stuck in VM boot,
"Booting from Hard Disk".
Any idea what can get wrong? how I can assure that the boot itself is from
the wanted disk.
Also will be helpful if
are pre-allocated and pre-registered,
hence in order to work with iSER from QEMU, one need to enable
MEMLOCK attribute in the VM to be large enough for all iSER buffers and RDMA
resources.
Signed-off-by: Roy Shterman <ro...@mellanox.com>
---
block/iscsi.c
to configure iSER in URI options
Roy Shterman (2):
block/iscsi: Introducing new zero-copy API
block/iscsi: Adding new iSER transport layet option
block/iscsi.c | 83 +++-
1 files changed, 81 insertions(+), 2 deletions(-)
--
1.7.8.2
A new API to deploy zero-copy command submission. The new API takes I/O
vectors list and number of I/O vectors to submit as input parameters
when initiating the command. New API must be used if working with
iSER transport option.
Signed-off-by: Roy Shterman <ro...@mellanox.com>
---
block/i
On 9/27/2016 2:52 PM, Paolo Bonzini wrote:
On 27/09/2016 13:37, Roy Shterman wrote:
+iscsi_url = iscsi_parse_full_url(iscsi,
uri_string_unescape(filename, -1, NULL));
if (iscsi_url == NULL) {
-error_setg(errp, "Failed to parse URL : %s", filename);
+error
On 9/27/2016 1:28 PM, Stefan Hajnoczi wrote:
On Wed, Sep 21, 2016 at 09:19:44PM +0300, Roy Shterman wrote:
iSER is a new transport layer supported in Libiscsi,
iSER provides a zero-copy RDMA capable interface that can
improve performance.
New API is introduced in abstracion of the Libiscsi
zero-copy command submission. iSER is differing from TCP in
data-path, hence IO vectors must be transferred already when queueing
the PDU.
changes from v1:
- Adding iser as an additional block driver
Signed-off-by: Roy Shterman <ro...@mellanox.com>
---
block/iscsi.c
.
Thanks,
Roy
-Original Message-
From: Peter Lieven [mailto:p...@kamp.de]
Sent: Thursday, July 28, 2016 11:46 AM
To: Roy Shterman <ro...@mellanox.com>; qemu-devel@nongnu.org
Cc: ronniesahlb...@gmail.com; pbonz...@redhat.com
Subject: Re: [PATCH] block/iscsi: Adding iser support in Li
On 8/1/2016 4:52 PM, Paolo Bonzini wrote:
On 28/07/2016 10:45, Peter Lieven wrote:
Furthermore I would suggest using LIBISCS_FEATURE_ISER and not the API
version in the
preprocessor commands.
Actually I disagree with this suggestion. The new API could be used
also if iSER is not used, can
zero-copy command submission. iSER is differing from TCP in
data-path, hence IO vectors must be transferred already when queueing
the PDU.
Signed-off-by: Roy Shterman <ro...@mellanox.com>
---
block/iscsi.c | 45 +
1 files changed, 41 insertions
o Bonzini wrote:
> >
> >
> > On 28/01/2016 08:28, Roy Shterman wrote:
> > >
> > > Important to understand that after modifying and saving configuration
> of
> > > xml with virsh edit $name_of_guest
> > >
> > > when i reenter the xml
to just upraise the maximum available pinnable-memory of
guest? without locking all 2G (in my guest)
because default maximum pinnable memory in QEMU is 60k.
Thank you all for your help,
Roy
On Sat, Jan 30, 2016 at 10:29 AM, Roy Shterman <roy.shter...@gmail.com>
wrote:
> You were right,
scsi-block, id "scsi1-0-0-0"
drive = "drive-scsi1-0-0-0"
channel = 0 (0x0)
scsi-id = 0 (0x0)
lun = 0 (0x0)
Thnaks,
Roy
On Wed, Jan 27, 2016 at 9:03 PM, Roy Shterman <roy.shter...@gmail.com>
wrote:
> Hi,
>
, in RH 7.2 data-plane is default if one is choosing to work with virtio?
Thank you very much,
Roy
On Thu, Jan 21, 2016 at 11:01 AM, Paolo Bonzini <pbonz...@redhat.com> wrote:
>
>
> On 20/01/2016 21:12, Roy Shterman wrote:
> > Hi,
> >
> > I have
Hi,
I have two questions,
First, I'm developing for Libiscsi and trying to work with virtio-scsi
dataplane or even virtio-blk dataplane and it doesn't works well.
I'm working with latest qemu and latest Libiscsi in RedHat 7 libvirt
package.
my iscsi xml part is :
virtio-blk -
15 matches
Mail list logo