Hi,
After ~4 years of working on iSER iSCSI, I'm moving on and will be
involved from a different perspective. Therefore, I will be unable to
continue my current maintainership responsibility for iSER related
issues. I want to thank everyone for the great work that I had the
chance to be part
Hi all,
It appears the current Linux iSER initiator does not send the HELLO message
when the connection transits to full feature phase. The stgt target also
ignores this message (if it were to appear). Both of these implementations
use a non-conformant iSER header (they add write_va and
Mike Heinz wrote:
Installation of OFED 1.3.0.0.4 onto a Kusu/OCS cluster does not fully
succeed because of some missing dependencies in the RPM spec files. This
is because Kusu installs nodes over a network by presenting a pool of
RPMs to be installed and letting RPM figure out the order to
usr/iscsi/iscsi_rdma.c | 16 +---
I have no idea what tree this file lives in so I'll just ignore this
patch, right?
As Eli mentioned in PATCH 0/2, the patch set contains a fix for the initiator
side and another fix for the iSER code in stgt. That's why Fujita Tomonori (who
So what was the conclusion on the right way to handle the change that
affects on-the-wire data? Just have a flag day so targets either work
with 2.6.25 and earlier initiators, or work with 2.6.26 and later
initiators, and corrupt data if someone mixes things the wrong way?
See Eli's answer
Erez Zilber wrote:
Pete Wyckoff wrote:
[EMAIL PROTECTED] wrote on Tue, 12 Feb 2008
15:57 -0600:
On Tue, 2008-02-12 at 16:46 -0500, Pete Wyckoff wrote:
[EMAIL PROTECTED] wrote on Tue, 12 Feb 2008
15:10 -0600:
On Tue, 2008-02-12 at 15:54 -0500, Pete Wyckoff wrote
Pete Wyckoff wrote:
[EMAIL PROTECTED] wrote on Tue, 12 Feb 2008 15:57 -0600:
On Tue, 2008-02-12 at 16:46 -0500, Pete Wyckoff wrote:
[EMAIL PROTECTED] wrote on Tue, 12 Feb 2008 15:10 -0600:
On Tue, 2008-02-12 at 15:54 -0500, Pete Wyckoff wrote:
iscsi_iser does not
BTW - I don't mind if other people add the required code to OFED
1.4 for
qla4xxx support.
~sigh~ Yeah.
I wonder how many (if any) of our userbase we are going to upset if we
cease providing the qla4xxx driver in our kernels. On the other hand, I
wonder how many we'd
Roland Dreier wrote:
- itt = ntohl(hdr-itt);
+ itt = hdr-itt;
This still gives the sparse warning
drivers/infiniband/ulp/iser/iser_initiator.c:419:6: warning: incorrect
type in assignment (different base types)
drivers/infiniband/ulp/iser/iser_initiator.c:419:6: expected
The itt field in struct iscsi_data is not defined with any
particular endianness. open-iscsi should use it as-is without
changing its endianness.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iser_initiator.c |4 ++--
1 files changed, 2 insertions(+), 2
When a RDMA_CM_EVENT_DEVICE_REMOVAL event is raised, iSER should release the
connection resources.
This is necessary when the IB HCA module is unloaded while open-iscsi is still
running. Currently, iSER just initiates a BUG() call.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers
The itt field in struct iscsi_data is not defined with any
particular endianness. open-iscsi should use it as-is without
changing its endianness.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iser_initiator.c |2 +-
1 files changed, 1 insertions(+), 1 deletions
Brian J. Murrell wrote:
I have run into a conflict trying to build a matching kernel and
kernel-ib pair for OFED 1.3 and RHEL5's 2.6.18 kernel (although I
suspect this will apply to generally any kernel of the same vintage).
General comment - in the future, I suggest that you send OFED
Is qla4xxx the only problem that you have with open-iscsi in OFED?
Looking through the kernel Kconfig files, it does appear that
SCSI_QLA_ISCSI is the only driver needing SCSI_ISCSI_ATTRS that isn't in
the OFED 1.3 release.
I'm not sure if there's a real demand for this transport
Brian J. Murrell wrote:
On Mon, 2008-04-14 at 16:50 +0300, Erez Zilber wrote:
I'm not sure if there's a real demand for this transport for OFED users,
is there?
Maybe I'm not seeing the bigger picture but it seems pretty orthogonal
to me. Does using OFED 1.3 preclude using
Roland Dreier wrote:
thanks, applied... I assume this much simpler patch replaces the earlier
one completely?
Yes (that's why I added v2 in the subject).
Thanks,
Erez
___
general mailing list
general@lists.openfabrics.org
() call.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iscsi_iser.h |2 ++
drivers/infiniband/ulp/iser/iser_verbs.c | 18 ++
2 files changed, 16 insertions(+), 4 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h
b/drivers
Sean,
I'm trying to add a better implementation to this event in iSER (better
than the current BUG() call that we have). I have 2 questions:
1. Is this event raised for each connection?
2. After the event is raised, I guess that I need to release all IB
resources for that connection,
*OFED 1.4:*
1. Kernel base: since we target 1.4 release to Sep we target the
kernel base to be 2.6.27
This is a good target, but we may need to stay with 2.6.26 if the
kernel progress will not be aligned.
2. Suggestions for new features:
* NFS-RDMA
* Verbs: Reliable
Roland,
I've noticed that your infiniband.git tree contains some branches. I
understand that for-2.6.25 is used for 2.6.25 work etc. What is the
purpose of the master branch?
Thanks,
Erez
___
general mailing list
general@lists.openfabrics.org
list_for_each_entry() instead of the
custom, flawed list iteration code.
Signed-off-by: Arne Redlich [EMAIL PROTECTED]
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iser_verbs.c | 36 +
1 files changed, 16 insertions(+), 20 deletions(-)
diff
failure is handled with a BUG_ON() right
before dereferencing the NULL-pointer - fix this!
Signed-off-by: Arne Redlich [EMAIL PROTECTED]
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iser_verbs.c | 13 +
1 files changed, 9 insertions(+), 4 deletions
Arne Redlich wrote:
Erez Zilber [EMAIL PROTECTED] writes:
Arne Redlich wrote:
The iteration through the list of iser_devices during device
lookup/creation is broken - it might result in an infinite loop if more
than 1 HCA is used with iSER. Use list_for_each_entry() instead
Bart Van Assche wrote:
On Feb 5, 2008 3:41 PM, Erez Zilber [EMAIL PROTECTED] wrote:
stgt (SCSI target) is an open-source framework for storage target
drivers. It supports iSCSI over iSER among other storage target drivers.
Voltaire added a git tree for stgt that will be added to OFED 1.4
Joe Landman wrote:
Erez Zilber wrote:
* READ: 920 MB/sec
* WRITE: 850 MB/sec
Not getting anything even remotely close to this. Are there more
details on configuration somewhere? I followed the web page as
indicated.
Are you running iSCSI over TCP or iSCSI over iSER (over
MB/sec
We hope to have DDR measurements numbers soon.
--
Erez Zilber | 972-9-971-7689
Software Engineer, Storage Solutions
Voltaire – _The Grid Backbone_
www.voltaire.com http://www.voltaire.com
Erez Zilber wrote:
eh_device_reset_handler was already added to scsi_host_template
in iscsi_tcp, and is now added also for iscsi_iser.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
Signed-off-by: Mike Christie [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iscsi_iser.c |1 +
1
eh_device_reset_handler was already added to scsi_host_template
in iscsi_tcp, and is now added also for iscsi_iser.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
Signed-off-by: Mike Christie [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iscsi_iser.c |1 +
1 files changed, 1 insertions
Some RDMA CM events are not supported or not handled in iSER.
This patch adds some info (printk) for the user about them.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iser_verbs.c |6 ++
1 files changed, 2 insertions(+), 4 deletions(-)
diff --git
Add change_queue_depth handler to scsi_host_template in the
iSER driver. This handler was added to iscsi_tcp in order to
solve the problem of queue depth which was too high for some
targets. It is also applicable for iSER.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp
Roland,
The following patch set contains some iSER fixes for 2.6.25.
Thanks,
Erez
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
Bart Van Assche wrote:
On Jan 8, 2008 10:10 PM, Roland Dreier [EMAIL PROTECTED] wrote:
Is the OFED software stack supported on Ubuntu Linux ? While OFED
1.2.5.4 compiles fine on Ubuntu Linux 7.10, I got the following
errors
while installing the RPM's:
Just out of curiosity, what
Erez Zilber wrote:
The url for iSER docs in Kconfig has changed.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/Kconfig |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/Kconfig
b/drivers/infiniband/ulp
Tziporet Koren wrote:
I wish to collect requirements for new features for OFED 1.4
Please reply with any request you have (features of existing modules,
new modules etc.)
Thanks,
Tziporet
Tziporet,
Voltaire will add support for stgt (SCSI target). It includes support
for iSCSI over
The url for iSER docs in Kconfig has changed.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/Kconfig |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/Kconfig
b/drivers/infiniband/ulp/iser/Kconfig
index fe604c8
is incremented in the current for_each_sg
loop. This caused iSER to think that all data is unaligned, and all data
was copied to aligned buffers.
This patch increments the missing counter inside the for_each_sg loop
whenever necessary.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers
Sufficool, Stanley wrote:
Does anyone know a source for Windows initiators for iSER?
Currently, there's no open-source iSER initiator for Windows. However,
we (Voltaire) are working on it.
--
Erez Zilber | 972-9-971-7689
Tziporet Koren wrote:
OFED October 8 meeting summary on OFED 1.3 alpha status and beta tasks:
Meeting summary:
1. Alpha release status:
* Cisco - tested only RHEL 4 5 on x86_64 systems
* Qlogic - in general looks good
* Intel - test Intel MPI on 16 nodes cluster on RHEL 5.1
*
5. iSER
o ???
open-iscsi is based on r865.12. There are no changes in iSER itself.
Erez
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
Michael S. Tsirkin wrote:
Hello!
I have updated the OFED 1.3 kernel tree at
git://git.openfabrics.org/ofed_1_3/linux-2.6.git ofed_kernel
to upstream 2.6.23-rc8.
I have resolved minor conflicts in libiscsi backports for RHEL4,
and everything seems to build fine now. iSER maintainers, please
Roland Dreier wrote:
err, subject should have been [PATCH] IB/iser: Remove unnecessary includes
It looks ok. If it complies on your machines, I'm ok with it.
Erez
___
general mailing list
general@lists.openfabrics.org
Quoting Kilian Cavalotti [EMAIL PROTECTED]:
Subject: Re: OFED 1.2.5 - GA release
Hi all,
Quoting Tziporet Koren [EMAIL PROTECTED]:
I am happy to announce on OFED 1.2.5 GA release.
Thanks for the release. I just wanted to let you know that I got a
compilation error (using RHEL4) in
,
--
Erez Zilber | 972-9-971-7689
Software Engineer, Storage Team
Voltaire – _The Grid Backbone_
__
www.voltaire.com http://www.voltaire.com/
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin
That seems to be what happens if you don't load the IB driver
modules. We use a startup script that does the moral equivalent
of:
modprobe ib_mthca
modprobe rdma_ucm
If you're using RHEL5, a good approach is just to install the
openib, libmthca, libibverbs{,-devel},
Tziporet,
Does 1.2.c-9 also include FMR support? Else, should we wait for 1.2.c-10?
Thanks,
Erez
Tziporet Koren wrote:
Hi All,
OFED 1.2.c-9 is available now on the OFA server under:
_http://www.openfabrics.org/builds/connectx/release/_
Note: this release was tested with FW 2.1.000
Will you send patches for iSER soon? I'd like to test it, and make sure
that iSER over IB is not damaged.
Our patches may not interest you since we are using an older version of
the iSER code. However, we will also be exploring the use of IB with our
implementations. Will this
Michael S. Tsirkin wrote:
Quoting Erez Zilber [EMAIL PROTECTED]:
Subject: Re: add_open_iscsi_h.patch
Michael S. Tsirkin wrote:
Erez, add_open_iscsi_h currently does:
-#include scsi/iscsi_if.h
+#include iscsi_if.h
why is ths bit needed?
Strange. I remember that I
Michael S. Tsirkin wrote:
Quoting Erez Zilber [EMAIL PROTECTED]:
Subject: Re: add_open_iscsi_h.patch
Michael S. Tsirkin wrote:
Quoting Erez Zilber [EMAIL PROTECTED]:
Subject: Re: add_open_iscsi_h.patch
Michael S. Tsirkin wrote:
Erez, add_open_iscsi_h currently
Ok, so this isn't somthing that I will need to worry a lot about if I am
planning on using iWARP?
You will need to use the iSER header as defined in the iSER spec.
I hope this helps.
It does, thank you.
BTW - do you plan to use the current iSER initiator
code for iWARP?
Yes, we
Roland Dreier wrote:
[adding [EMAIL PROTECTED] and general@lists.openfabrics.org -- Roland]
I would like to test that on iSER. Where can I download all 33 patches from?
Thanks,
Erez
___
general mailing list
general@lists.openfabrics.org
Jens Axboe wrote:
On Thu, Jul 19 2007, Erez Zilber wrote:
Roland Dreier wrote:
[adding [EMAIL PROTECTED] and general@lists.openfabrics.org -- Roland]
I would like to test that on iSER. Where can I download all 33 patches from?
I can provide a rolled up patch
Roland Dreier wrote:
Make iser_conn_release() and iser_start_rdma_unaligned_sg() static,
since they are only used in the .c file where they are defined. In
addition to being a cleanup, this even shrinks the generated code by
allowing the single call of iser_start_rdma_unaligned_sg() to be
?
--
Erez Zilber | 972-9-971-7689
Software Engineer, Storage Team
Voltaire – _The Grid Backbone_
__
www.voltaire.com http://www.voltaire.com/
___
general mailing list
general@lists.openfabrics.org
http
,
--
Erez Zilber | 972-9-971-7689
Software Engineer, Storage Team
Voltaire – _The Grid Backbone_
__
www.voltaire.com http://www.voltaire.com/
___
general mailing list
general@lists.openfabrics.org
http
Erez Zilber wrote:
Add support for open-iscsi over iSER in RHAS4 in OFED's scripts.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
build.sh |2 +-
build_env.sh |4 ++--
install.sh |2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/build.sh b
and the
parameters.
TODO: use scsi_for_each_sg().
Signed-off-by: FUJITA Tomonori [EMAIL PROTECTED]
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iscsi_iser.c |4 ++--
drivers/infiniband/ulp/iser/iser_initiator.c | 14 --
2 files changed, 6 insertions
FUJITA Tomonori wrote:
From: Erez Zilber [EMAIL PROTECTED]
Subject: [Fwd: [PATCH 2/2] iscsi_iser: convert to use the data buffer
accessors]
Date: Tue, 19 Jun 2007 09:20:33 +0300
Roland,
Can you add the patch below to 2.6.23?
Thanks, but the patch was already added to James' scsi
Michael S. Tsirkin wrote:
Since we'll have RC6, I'd like to add the following cosmetic change:
move iscsi_scsi_makefile from kernel_patches/fixes to ofed_scripts,
and use a softlink to put in in place.
The solves the following problem: if I do git clone on ofed
tree without -n, and try to
I'm trying to build srpt according to the instructions, but it does
not get built at all. Here's what I did:
tar xzf OFED-1.2-rc3.tgz
cd OFED-1.2-rc3/SRPMS
rpm2cpio ofa_kernel-1.2-rc3.src.rpm |cpio -i
tar xzf ofa_kernel-1.2.tgz
cd ofa_kernel-1.2
patch -p1
I am doing that. However, attribute_container.c includes base.h which is in
the kernel_addons dir. Since attribute_container.c is no longer there, I need
to add the following line:
-I$(PWD)/kernel_addons/backport/2.6.9_U3/include/src/
It is not very very ugly, so I think that we can
I have the following files in backport/2.6.9_UX/include/src/:
attribute_container.c - almost identical to the file on 2.6.20. I had to
change one line in it.
could be a patch ...
which line?
Now, attribute_container.c, klist.c transport_class.c are copied from
the kernel tree.
However, you are copying a ton of files from upstream kernel.
Sticking extra files in include might interfere with newer
kernels, so I don't have better ideas for this for 1.2
(for 1.3 I am hoping we'll use the submodule support in git,
so we'll be able to re-use headers as well).
Michael S. Tsirkin wrote:
Quoting Erez Zilber [EMAIL PROTECTED]:
Subject: [PATCH 2/2] IB/iser: add backport kernel addons for open-iscsi
over iSER support for RHAS4 up3 and up4
Add the required backport patches kernel addons for open-iscsi
over iSER in RHAS4 up3 and up4.
Signed-off
The following patches add the required backports kernel addons in order to
support open-iscsi over iSER in RHAS4 up3 up4 in OFED (currently SLES 10,
SLES 10 sp1 RHEL 5 are supported).
--
Erez Zilber | 972-9-971-7689
Software
Roland Dreier wrote:
The following patch replaces the bad patch (iser_conn should not be
released while its workqueue is active) that I sent a few days
ago. Again, if it's possible, I'd like to have it merged into 2.6.21
(it is a bug fix).
We can still merge bug fixes, but I need
Roland,
Or I found a bug in this patch. I hope to send a fix for it in the next few
days. Meanwhile, please don't merge it.
Erez
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
Roland,
Here is a series of patches for iSER. All of them are bug fixes. I hope
that they can be added to 2.6.21.
Thanks
Erez
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To
pointer will be referenced.
This fix waits until the workqueue function is called and only
then it releases the connection object.
Signed-off-by: Erez Zilber [EMAIL PROTECTED]
---
drivers/infiniband/ulp/iser/iscsi_iser.h |1 +
drivers/infiniband/ulp/iser/iser_verbs.c |9 +++--
2 files
/mailman/listinfo/ewg
--
Erez Zilber | 972-9-971-7689
Software Engineer, Storage Team
Voltaire – _The Grid Backbone_
__
www.voltaire.com http://www.voltaire.com/
___
general mailing list
69 matches
Mail list logo