On 11/30/2011 11:38 PM, Hefty, Sean wrote:
[...] Also, any thoughts on the direction of AF_IB for adding support for
native IB addresses through the rdma_cm?
yes, see some questions and comments over that thread
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body
On 11/4/2011 9:08 AM, Hefty, Sean wrote:
+static struct idr *cma_select_ib_ps(struct rdma_id_private *id_priv)
+{
+ struct idr *ps = NULL;
+ struct sockaddr_ib *sib;
+ u64 sid_ps, mask, sid;
+
+ sib = (struct sockaddr_ib *)id_priv-id.route.addr.src_addr;
+ mask =
On 11/4/2011 9:08 AM, Hefty, Sean wrote:
+static struct idr *cma_select_ib_ps(struct rdma_id_private *id_priv)
[...]
+ if (ps) {
+ sib-sib_sid = cpu_to_be64(sid_ps | ntohs(cma_port((struct
sockaddr *) sib)));
+ sib-sib_sid_mask = cpu_to_be64(RDMA_IB_IP_PS_MASK
This is a static checker fix, and I'm not super familiar with this code.
My checker complains that user_wr-num_sge + sg_ind can have an integer
overflow and wrap. That's is true, but I don't know the code well
enough to say if it's a problem or not.
Can someone take a look at this?
Hi,
OFED 1.5.4-rc5 is available.
The tarball is available on:
http://www.openfabrics.org/downloads/OFED/ofed-1.5.4/OFED-1.5.4-rc5.tgz
To get BUILD_ID run ofed_info
Please report any issues in bugzilla https://bugs.openfabrics.org/ for OFED
1.5.4
Regards,
Vladimir
Main changes from RC4:
Add additional buffer space for output
so no longer chopped after FECCorrectableBlockCtrLane11
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
diff --git a/src/perfquery.c b/src/perfquery.c
index 0dbb80d..c747d11 100644
--- a/src/perfquery.c
+++ b/src/perfquery.c
@@ -377,7 +377,7 @@ static
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
diff --git a/infiniband-diags/src/perfquery.c b/infiniband-diags/src/perfquery.c
index 859ddbb..3ebf1cc 100644
--- a/infiniband-diags/src/perfquery.c
+++ b/infiniband-diags/src/perfquery.c
@@ -401,12 +401,56 @@ static void
infiniband-diags/perfquery.c: Fix reset for PortExtendedSpeedsCounters
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
Changes since v1:
Patch now against separated/recent infiniband-diags tree
diff --git a/src/perfquery.c b/src/perfquery.c
index 859ddbb..3ebf1cc 100644
--- a/src/perfquery.c
On Wed, Nov 30, 2011 at 1:38 PM, Hefty, Sean sean.he...@intel.com wrote:
Have you had a chance to think about this patch? It would be nice to have
user space xrc support defined to match up with the kernel.
I think this is the right approach -- need to get a big enough chunk
of time to look
On 11/4/2011 9:08 AM, Hefty, Sean wrote:
+static struct idr *cma_select_ib_ps(struct rdma_id_private *id_priv)
+{
+ struct idr *ps = NULL;
+ struct sockaddr_ib *sib;
+ u64 sid_ps, mask, sid;
+
+ sib = (struct sockaddr_ib *)id_priv-id.route.addr.src_addr;
+ mask =
+static struct idr *cma_select_ib_ps(struct rdma_id_private *id_priv)
[...]
+ if (ps) {
+ sib-sib_sid = cpu_to_be64(sid_ps | ntohs(cma_port((struct
sockaddr *) sib)));
+ sib-sib_sid_mask = cpu_to_be64(RDMA_IB_IP_PS_MASK | sib-
sib_sid_mask);
+ }
+ return
This patch series makes the ib_srp driver better suited for use in a H.A.
setup because:
- Switchover without triggering read or write errors become possible. Such
errors are bad because these can make a filesystem switch to read-only
mode.
- A ping mechanism has been added that allows to
Make the logging code more a little more brief by replacing
printk(KERN_WARNING PFX ...) by pr_warn(...) and by replacing
printk(KERN_ERR PFX ...) by pr_err(...). Remove one trailing
space to avoid a checkpatch warning.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: David Dillow
Remove sysfs attributes before removing a target instead of
testing the target state in every sysfs attribute callback
method. Note: it is safe to invoke a sysfs attribute removal
method like device_remove_file() twice on the same attribute.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc:
Currently the ib_srp driver allows to log in multiple times to the
same target via its sysfs interface. This leads to each target LUN
being imported multiple times at the initiator side, something that
should not be allowed.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: David Dillow
Make sure that the default block layer timeout is above the
InfiniBand transport layer timeout. The default block layer
timeout is 30 seconds. Typical values for the QP local ack
timeout and retry count are 19 and 7 respectively, which means
that it can take up to 60.1 seconds before a HCA submits
If a user misconfigures the block layer timeout such that it is below
the InfiniBand RC timeout it can happen that an SRP reply arrives after
the SCSI error handler has already killed the associated SCSI command.
Avoid that late replies cause a kernel crash.
Signed-off-by: Bart Van Assche
Help the CPU branch predictor by rearranging two if-statements such
that the most likely code is in the if-part instead of the else-part.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: David Dillow dillo...@ornl.gov
Cc: Roland Dreier rol...@purestorage.com
---
Factor out common code into a new function.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: David Dillow dillo...@ornl.gov
Cc: Roland Dreier rol...@purestorage.com
---
drivers/infiniband/ulp/srp/ib_srp.c | 19 +++
1 files changed, 11 insertions(+), 8 deletions(-)
diff
Document the srp_transport sysfs attributes according to the rules specified
in Documentation/ABI/README.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: FUJITA Tomonori fujita.tomon...@lab.ntt.co.jp
Cc: Brian King brk...@linux.vnet.ibm.com
---
Documentation/ABI/stable/sysfs-transport-srp
Register transport attributes after the attribute array has been
set up instead of before.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: FUJITA Tomonori fujita.tomon...@lab.ntt.co.jp
Cc: Brian King brk...@linux.vnet.ibm.com
Cc: David Dillow dillo...@ornl.gov
Cc: Roland Dreier
Eliminate the private_rport_attrs[] array and the SETUP_*() macros
used to set up that array since the information in that array
duplicates the information in the static device attributes. Also,
trigger a kernel warning if SRP_RPORT_ATTRS is not large enough
since it is easy to forget to update
On Mon, 28 Nov 2011 06:01:28 -0800
Hal Rosenstock h...@dev.mellanox.co.il wrote:
Signed-off-by: Hal Rosenstock h...@mellanox.com
---
diff --git a/include/infiniband/mad.h b/include/infiniband/mad.h
index 199b05f..09b334f 100644
--- a/include/infiniband/mad.h
+++ b/include/infiniband/mad.h
Add a time-based transport layer test such that fail-over in a multipath
setup can happen quickly.
Add the necessary functions in the SRP transport module to allow an SRP
initiator driver to implement this.
Add a slave_delete callback in the SCSI host template such that SCSI
hosts that hold a
Make it possible to disconnect via sysfs the IB RC connection used by the
SRP protocol to communicate with a target.
Let the SRP transport layer create a sysfs delete attribute for
initiator drivers that support this functionality.
Signed-off-by: Bart Van Assche bvanass...@acm.org
Cc: David
On Thu, 1 Dec 2011 08:21:12 -0800
Hal Rosenstock h...@dev.mellanox.co.il wrote:
Add additional buffer space for output
so no longer chopped after FECCorrectableBlockCtrLane11
Signed-off-by: Hal Rosenstock h...@mellanox.com
Thanks applied,
Ira
---
diff --git a/src/perfquery.c
On Thu, 1 Dec 2011 08:49:09 -0800
Hal Rosenstock h...@dev.mellanox.co.il wrote:
infiniband-diags/perfquery.c: Fix reset for PortExtendedSpeedsCounters
Signed-off-by: Hal Rosenstock h...@mellanox.com
Thanks applied,
Ira
---
Changes since v1:
Patch now against separated/recent
We propose a new process for the OFED releases starting from next OFED release:
- OFED content will be the relevant kernel.org modules and user space released
packages
- OFED will offer only backports to the distros (no fixes)
- OFED package will be used for easy installation of all packages in
perhaps this should use the unlikely() macro instead:
if (unlikely(wc.status)) {
...
}
...
On Thu, 1 Dec 2011 20:02:17 +0100
Bart Van Assche bvanass...@acm.org wrote:
Help the CPU branch predictor by rearranging two if-statements such
that the most
On Thu, 2011-12-01 at 13:55 -0500, Bart Van Assche wrote:
Make the logging code more a little more brief by replacing
printk(KERN_WARNING PFX ...) by pr_warn(...) and by replacing
printk(KERN_ERR PFX ...) by pr_err(...). Remove one trailing
space to avoid a checkpatch warning.
This is a good
On Thu, 2011-12-01 at 13:54 -0500, Bart Van Assche wrote:
This patch series makes the ib_srp driver better suited for use in a H.A.
setup because:
- Switchover without triggering read or write errors become possible. Such
errors are bad because these can make a filesystem switch to read-only
We propose a new process for the OFED releases starting from next OFED
release:
- OFED content will be the relevant kernel.org modules and user space released
packages
- OFED will offer only backports to the distros (no fixes)
I think this point needs to be clarified - at least to me
Sean wrote,
- OFED will offer only backports to the distros (no fixes)
I think this point needs to be clarified - at least to me anyway. :)
What this means is that the OFED code base will be identical to what
is included in the upstream kernel and libs.
OFED will provide only backports to
On Thu, Dec 01, 2011 at 04:13:41PM -0800, Woodruff, Robert J wrote:
Sean wrote,
- OFED will offer only backports to the distros (no fixes)
I think this point needs to be clarified - at least to me anyway. :)
What this means is that the OFED code base will be identical to what
is included
On Fri, Dec 2, 2011 at 1:04 AM, Hefty, Sean sean.he...@intel.com wrote:
- What should we do with modules like SDP that are not in kernel?
Either remove them or carry them forward as experimental features.
Wat I expect is that reworking the SDP implementation such that it can
be included
35 matches
Mail list logo