On 6/30/2015 12:36 AM, Steve Wise wrote:
The semantics for MR access flags are not consistent across RDMA
protocols. So rather than have applications try and glean what they
need, have them pass in the intended roles and attributes for the MR to
be allocated and let the RDMA core select the
On 30/06/2015 00:36, Steve Wise wrote:
/**
+ * rdma_mr_roles - possible roles an RDMA MR will be used for
+ *
+ * This allows a transport independent RDMA application to
+ * create MRs that are usable for all the desired roles w/o
+ * having to understand which access rights are needed.
+
On 6/8/2015 11:49 PM, Hefty, Sean wrote:
Sean,
IMO, we need to introduce vendor specific header files and interfaces.
It is unmaintainable to drive an API from the bottom up and expose the 'bare
metal'
implementation of a bunch of disjoint pieces of hardware. (Yeah, because we
need yet
On Tue, Jun 30, 2015 at 02:47:00PM +0300, Sagi Grimberg wrote:
Kernel 4.1 introduced the new pmem driver for byte addressable storage
(https://lwn.net/Articles/640115/). It won't be long before we see HA
models where secondary persistent memory devices will sit across an
RDMA fabric.
Currently, IB/cma remove_one flow blocks until all user descriptor managed by
IB/ucma are released. This prevents hot-removal of IB devices. This patch
allows IB/cma to remove devices regardless of user space activity. Upon getting
the RDMA_CM_EVENT_DEVICE_REMOVAL event we close all the underlying
Currently, if there is any user space application using an IB device,
it is impossible to unload the HW device driver for this device.
Similarly, if the device is hot-unplugged or reset, the device driver
hardware removal flow blocks until all user contexts are destroyed.
This patchset removes
On 6/30/2015 3:10 PM, Christoph Hellwig wrote:
On Tue, Jun 30, 2015 at 02:47:00PM +0300, Sagi Grimberg wrote:
Kernel 4.1 introduced the new pmem driver for byte addressable storage
(https://lwn.net/Articles/640115/). It won't be long before we see HA
models where secondary persistent memory
Enables the uverbs_remove_one to succeed despite the fact that there are
running IB applications working with the given ib device. This functionality
enables a HW device to be unbind/reset despite the fact that there are running
user space applications using it.
It exposes a new IB kernel API
From: Kaike Wan kaike@intel.com
This patch routes a SA pathrecord query to netlink first and processes the
response appropriately. If a failure is returned, the request will be sent
through IB. The decision whether to route the request to netlink first is
determined by the presence of a
From: Kaike Wan kaike@intel.com
A SA cache is undeniably critical for fabric scalability and performance.
In user space, the ibacm application provides a good example of pathrecord
cache for address and route resolution. With the recent implementation of
the provider architecture, ibacm
From: Kaike Wan kaike@intel.com
Replace kmalloc with kzalloc so that all uninitialized fields in SA query
will be zero-ed out to avoid unintentional consequence. This prepares the
SA query structure to accept new fields in the future.
Signed-off-by: Kaike Wan kaike@intel.com
From: Kaike Wan kaike@intel.com
This patch enables ibacm to process pathrecord queries through netlink.
Since ibacm can cache pathrecords, this implementation provides an easy
pathrecord cache for kernel components and therefore offers great
performance advantage on large fabric systems.
From: Kaike Wan kaike@intel.com
This patch adds a function to check if listeners for a netlink multicast
group are present.
Signed-off-by: Kaike Wan kaike@intel.com
Signed-off-by: John Fleck john.fl...@intel.com
Signed-off-by: Ira Weiny ira.we...@intel.com
Reviewed-by: Sean Hefty
From: Kaike Wan kaike@intel.com
This patch adds netlink defines for SA client, local service group, local
service operations, and related attributes.
Signed-off-by: Kaike Wan kaike@intel.com
Signed-off-by: John Fleck john.fl...@intel.com
Signed-off-by: Ira Weiny ira.we...@intel.com
-Original Message-
From: Sagi Grimberg [mailto:sa...@dev.mellanox.co.il]
Sent: Tuesday, June 30, 2015 4:26 AM
To: Steve Wise; dledf...@redhat.com
Cc: r...@mellanox.com; sa...@mellanox.com; linux-rdma@vger.kernel.org;
jguntho...@obsidianresearch.com; infinip...@intel.com;
-Original Message-
From: Sagi Grimberg [mailto:sa...@dev.mellanox.co.il]
Sent: Tuesday, June 30, 2015 4:22 AM
To: Steve Wise; dledf...@redhat.com
Cc: r...@mellanox.com; sa...@mellanox.com; linux-rdma@vger.kernel.org;
jguntho...@obsidianresearch.com; infinip...@intel.com;
-Original Message-
From: Sagi Grimberg [mailto:sa...@dev.mellanox.co.il]
Sent: Tuesday, June 30, 2015 4:33 AM
To: Steve Wise; dledf...@redhat.com
Cc: r...@mellanox.com; sa...@mellanox.com; linux-rdma@vger.kernel.org;
jguntho...@obsidianresearch.com; infinip...@intel.com;
On Mon, Jun 29, 2015 at 04:36:18PM -0500, Steve Wise wrote:
+int rdma_device_access_flags(struct ib_pd *pd, int roles, int attrs)
+{
+ int access_flags = attrs;
No RDMA_MRR_SEND ?
+ if (roles RDMA_MRR_RECV)
+ access_flags |= IB_ACCESS_LOCAL_WRITE;
+
+ if (roles
-Original Message-
From: Or Gerlitz [mailto:ogerl...@mellanox.com]
Sent: Tuesday, June 30, 2015 4:04 AM
To: Steve Wise; Chuck Lever
Cc: dledf...@redhat.com; r...@mellanox.com; sa...@mellanox.com;
linux-rdma@vger.kernel.org; jguntho...@obsidianresearch.com;
infinip...@intel.com;
On Jun 30, 2015, at 10:29 AM, Steve Wise sw...@opengridcomputing.com wrote:
-Original Message-
From: Or Gerlitz [mailto:ogerl...@mellanox.com]
Sent: Tuesday, June 30, 2015 4:04 AM
To: Steve Wise; Chuck Lever
Cc: dledf...@redhat.com; r...@mellanox.com; sa...@mellanox.com;
-Original Message-
From: linux-rdma-ow...@vger.kernel.org
[mailto:linux-rdma-ow...@vger.kernel.org] On Behalf Of Chuck Lever
Sent: Tuesday, June 30, 2015 9:42 AM
To: Steve Wise
Cc: Or Gerlitz; dledf...@redhat.com; r...@mellanox.com; sa...@mellanox.com;
-Original Message-
From: Haggai Eran [mailto:hagg...@mellanox.com]
Sent: Tuesday, June 30, 2015 2:25 AM
To: Steve Wise; dledf...@redhat.com
Cc: r...@mellanox.com; sa...@mellanox.com; linux-rdma@vger.kernel.org;
jguntho...@obsidianresearch.com; infinip...@intel.com;
NFSRDMA currently checks the transport type to decide how to set the
access flags for memory registration. With the new services
exported in this series, we can change/simplify NFSRDMA to not have
to know the transport type.
It would be excellent if this series actually went through and got
Subject: [PATCH V2 2/5] ipath,qib: Expose max_sge_rd correctly
Applications must not assume that max_sge and max_sge_rd are the same,
Hence expose max_sge_rd correctly as well.
Signed-off-by: Steve Wise sw...@opengridcomputing.com
Thanks!
Acked-by: Mike Marciniszyn
On Jun 30, 2015, at 12:54 PM, Jason Gunthorpe jguntho...@obsidianresearch.com
wrote:
On Mon, Jun 29, 2015 at 04:36:13PM -0500, Steve Wise wrote:
Applications must not assume that max_sge and max_sge_rd
are the same, Hence expose max_sge_rd correctly as well.
Chuck,
Now that this works,
include/uapi/rdma/rdma_netlink.h | 87
++
1 files changed, 87 insertions(+), 0 deletions(-)
diff --git a/include/uapi/rdma/rdma_netlink.h
b/include/uapi/rdma/rdma_netlink.h
index 6e4bb42..d2c50e9 100644
--- a/include/uapi/rdma/rdma_netlink.h
+++
-Original Message-
From: Hefty, Sean [mailto:sean.he...@intel.com]
Sent: Tuesday, June 30, 2015 12:04 PM
To: Jason Gunthorpe; Steve Wise
Cc: 'Sagi Grimberg'; dledf...@redhat.com; r...@mellanox.com;
sa...@mellanox.com; linux-rdma@vger.kernel.org; infinipath;
e...@mellanox.com;
On Tue, Jun 30, 2015 at 09:33:21AM -0500, Steve Wise wrote:
I prefer to decouple the iSER changes with this core work.
Jason/Sean... thoughts? I could do the iSER w/o patch 3, and the
follow up with a series that includes our final solution on
transport independent memory registration and
I prefer to decouple the iSER changes with this core work.
Jason/Sean... thoughts? I could do the iSER w/o patch 3, and the
follow up with a series that includes our final solution on
transport independent memory registration and change all the TI
kernel users (iser and nfsrdma) along
On Tue, Jun 30, 2015 at 01:26:03PM +0300, Yishai Hadas wrote:
Signed-off-by: Yishai Hadas yish...@mellanox.com
Signed-off-by: Shachar Raindel rain...@mellanox.com
Reviewed-By: Jason Gunthorpe jguntho...@obsidianresearch.com
err_file:
+ ib_uverbs_free_async_event_file(file);
I suggest to start consolidating to ib_create_mr() that receives an
extensible ib_mr_init_attr and additional attributes can be mr_roles
and mr_attrs.
I think this makes sense, but does it really help? If the end result is that
the app and providers basically end up switching on
From: Hefty, Sean
Sent: Tuesday, June 30, 2015 1:17 PM
To: Wan, Kaike; linux-rdma@vger.kernel.org
Cc: Wan, Kaike; Fleck, John; Weiny, Ira
Subject: RE: [PATCH v7 1/4] IB/netlink: Add defines for local service requests
through netlink
include/uapi/rdma/rdma_netlink.h | 87
On Tue, Jun 30, 2015 at 01:26:05PM +0300, Yishai Hadas wrote:
struct ib_uverbs_device {
- struct kref ref;
+ struct kref comp_ref;
+ struct kref free_ref;
So.. I was looking at this, and there is
On Mon, Jun 29, 2015 at 04:36:13PM -0500, Steve Wise wrote:
Applications must not assume that max_sge and max_sge_rd
are the same, Hence expose max_sge_rd correctly as well.
Chuck,
Now that this works, can we change NFS RDMA and get rid of the
rdma_cap_read_multi_sge stuff?
Thanks,
Jason
--
On Tue, Jun 30, 2015 at 01:26:04PM +0300, Yishai Hadas wrote:
Done in preparation for deploying RCU for the device removal
flow. Allows isolating the RCU handling to the uverb_main layer and
keeping the uverbs_cmd code as is.
Signed-off-by: Yishai Hadas yish...@mellanox.com
Signed-off-by:
-Original Message-
From: Jason Gunthorpe [mailto:jguntho...@obsidianresearch.com]
Sent: Tuesday, June 30, 2015 11:21 AM
To: Steve Wise
Cc: dledf...@redhat.com; r...@mellanox.com; sa...@mellanox.com;
linux-rdma@vger.kernel.org; infinip...@intel.com;
e...@mellanox.com;
On 06/30/2015 02:42 PM, Steve Wise wrote:
-Original Message-
From: Hefty, Sean [mailto:sean.he...@intel.com]
Sent: Tuesday, June 30, 2015 12:04 PM
To: Jason Gunthorpe; Steve Wise
Cc: 'Sagi Grimberg'; dledf...@redhat.com; r...@mellanox.com;
sa...@mellanox.com;
Hi Doug,
a gentle reminder to pull this simple patch-set to your tree.
-Regards
Devesh
On Fri, Jun 12, 2015 at 10:15 PM, Devesh Sharma
devesh.sha...@avagotech.com wrote:
A series to update the license from GPL to GPL/Dual-BSD licensing
for ocrdma source.
Devesh Sharma (2):
RDMA/ocrdma:
On Fri, Jun 12, 2015 at 10:15:03PM +0530, Devesh Sharma wrote:
A series to update the license from GPL to GPL/Dual-BSD licensing
for ocrdma source.
Do you have a written consent from everyone who contributed to the
driver to do this?
--
To unsubscribe from this list: send the line unsubscribe
On 6/30/2015 12:36 AM, Steve Wise wrote:
Limit the sg tablesize based on the device fast reg depth.
Use rdma_get_dma_mr() to allocate the DMA MR.
Use rdma_fast_reg_access_flags() to set the access_flags for fast
register work requests.
Steve,
I wander if it would make more sense to get the
On 6/28/2015 1:33 PM, Or Gerlitz wrote:
On Thu, Jun 25, 2015 at 5:45 PM, Or Gerlitz ogerl...@mellanox.com wrote:
For IB links, reading HCA flow counters through iboe_process_mad() should
be used when mlx4_ib_process_mad() is invoked only for VFs PMA queries and
exactly nothing else.
Fixes:
On 6/30/2015 12:36 AM, Steve Wise wrote:
The following series implements support for iWARP transports in the iSER
initiator and target. This is based on Doug's k.o/for-4.2 branch.
I've tested this on cxgb4 and mlx4 hardware.
Changes since V1:
Introduce and use transport-independent RDMA core
Hi Christoph, I really do not have it.
However, this change is initiated with consent of Emulex management.
-Regards
Devesh
On Tue, Jun 30, 2015 at 11:36 AM, Christoph Hellwig h...@infradead.org wrote:
On Fri, Jun 12, 2015 at 10:15:03PM +0530, Devesh Sharma wrote:
A series to update the
Hello. I found the slides from presentation (by mellanox) about
multipath rdma, does source code somewhere available? When mellanox
provide more info and docs?
Thanks!
--
Vasiliy Tolstov,
e-mail: v.tols...@selfip.ru
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the
On Tue, Jun 30, 2015 at 09:45:55AM -0400, kaike@intel.com wrote:
+static inline int ib_nl_is_good_resolve_resp(const struct nlmsghdr *nlh)
+{
+ const struct nlattr *head, *curr;
+ int len, rem;
+
+ if (nlh-nlmsg_flags RDMA_NL_LS_F_ERR)
+ return 0;
+
+ if
45 matches
Mail list logo