Three comments.
On 12/7/2015 12:43 PM, Chuck Lever wrote:
To support the NFSv4.1 backchannel on RDMA connections, add a
capability for receiving an RPC/RDMA reply on a connection
established by a client.
(snip)
+/* By convention, backchannel calls arrive via rdma_msg type
"By convention"
Two small comments.
On 12/7/2015 12:42 PM, Chuck Lever wrote:
Minor optimization: when dealing with write chunk XDR roundup, do
not post a Write WR for the zero bytes in the pad. Simply update
the write segment in the RPC-over-RDMA header to reflect the extra
pad bytes.
The Reply chunk is also
On 11/30/2015 5:25 PM, Chuck Lever wrote:
To support the server-side of an NFSv4.1 backchannel on RDMA
connections, add a transport class that enables backward
direction messages on an existing forward channel connection.
+static void *
+xprt_rdma_bc_allocate(struct rpc_task *task, size_t
On 11/24/2015 1:52 AM, Christoph Hellwig wrote:
On Mon, Nov 23, 2015 at 07:57:42PM -0500, Tom Talpey wrote:
On 11/23/2015 5:14 PM, Chuck Lever wrote:
FMR's ro_unmap method is already synchronous because ib_unmap_fmr()
is a synchronous verb. However, some improvements can be made here.
I
On 11/24/2015 2:03 AM, Jason Gunthorpe wrote:
On Mon, Nov 23, 2015 at 06:35:28PM -0800, Caitlin Bestler wrote:
Are there actual HCAs that make this mistake?
All IB HCAs have this behavior and require apps to see a send CQ
completion before making any statements about the state of the send Q
On 11/24/2015 5:59 AM, Sagi Grimberg wrote:
As I see it, if we don't wait for local-invalidate to complete before
unmap and IO completion (and no one does)
For the record, that is false. Windows quite strictly performs these
steps, and deliver millions of real 4K IOPS over SMB Direct.
On 11/23/2015 5:20 PM, Chuck Lever wrote:
Extra resources for handling backchannel requests have to be
pre-allocated when a transport instance is created. Set a limit.
Signed-off-by: Chuck Lever
---
include/linux/sunrpc/svc_rdma.h |5 +
On 11/23/2015 8:09 PM, Chuck Lever wrote:
On Nov 23, 2015, at 7:39 PM, Tom Talpey <t...@talpey.com> wrote:
On 11/23/2015 5:20 PM, Chuck Lever wrote:
Extra resources for handling backchannel requests have to be
pre-allocated when a transport instance is created. Set a limit.
Sign
On 11/23/2015 5:21 PM, Chuck Lever wrote:
To support the server-side of an NFSv4.1 backchannel on RDMA
connections, add a transport class for backwards direction
operation.
So, what's special here is that it re-uses an existing forward
channel's connection? If not, it would seem unnecessary to
On 11/23/2015 5:13 PM, Chuck Lever wrote:
Rarely, senders post a Send that is larger than the client's inline
threshold. That can be due to a bug, or the client and server may
not have communicated about their inline limits. RPC-over-RDMA
currently doesn't specify any particular limit on inline
On 11/23/2015 8:16 PM, Chuck Lever wrote:
On Nov 23, 2015, at 7:55 PM, Tom Talpey <t...@talpey.com> wrote:
On 11/23/2015 5:13 PM, Chuck Lever wrote:
Rarely, senders post a Send that is larger than the client's inline
threshold. That can be due to a bug, or the client and server may
no
On 11/23/2015 7:00 PM, Jason Gunthorpe wrote:
On Mon, Nov 23, 2015 at 03:30:42PM -0800, Caitlin Bestler wrote:
The receive completion can be safely assumed to indicate transmit
completion over a reliable connection unless your peer has gone
completely bonkers and is replying to a
On 11/23/2015 5:14 PM, Chuck Lever wrote:
There is a window between the time the RPC reply handler wakes the
waiting RPC task and when xprt_release() invokes ops->buf_free.
During this time, memory regions containing the data payload may
still be accessed by a broken or malicious server, but the
On 11/23/2015 5:20 PM, Chuck Lever wrote:
To support the NFSv4.1 backchannel on RDMA connections, add a
capability for receiving an RPC/RDMA reply on a connection
established by a client.
Signed-off-by: Chuck Lever
---
net/sunrpc/xprtrdma/rpc_rdma.c | 76
On 11/23/2015 5:21 PM, Chuck Lever wrote:
Clean up: The access_flags field is not used outside of
rdma_read_chunk_frmr() and is always set to the same value.
Signed-off-by: Chuck Lever
---
include/linux/sunrpc/svc_rdma.h |1 -
On 11/23/2015 5:14 PM, Chuck Lever wrote:
FMR's ro_unmap method is already synchronous because ib_unmap_fmr()
is a synchronous verb. However, some improvements can be made here.
I thought FMR support was about to be removed in the core.
1. Gather all the MRs for the RPC request onto a
an upper layer question,
not an RPC one.
On 11/23/2015 8:47 PM, Chuck Lever wrote:
On Nov 23, 2015, at 7:44 PM, Tom Talpey <t...@talpey.com> wrote:
On 11/23/2015 5:20 PM, Chuck Lever wrote:
To support the NFSv4.1 backchannel on RDMA connections, add a
capability for receiving an RPC/RDMA
On 11/23/2015 8:36 PM, Chuck Lever wrote:
On Nov 23, 2015, at 8:19 PM, Tom Talpey <t...@talpey.com> wrote:
On 11/23/2015 8:09 PM, Chuck Lever wrote:
On Nov 23, 2015, at 7:39 PM, Tom Talpey <t...@talpey.com> wrote:
On 11/23/2015 5:20 PM, Chuck Lever wrote:
Extra resources
On 11/15/2015 7:30 AM, Matan Barak wrote:
This series adds support for completion timestamp. In order to
support this feature, several extended verbs were implemented
(as instructed in libibverbs).
Can you describe what these timestamps are actually for? It's not
clear at all from the
On 11/10/2015 4:17 PM, Jason Gunthorpe wrote:
On Tue, Nov 10, 2015 at 04:00:43PM -0500, Tom Talpey wrote:
Hmm, agreed, but it must still be acceptable to perform a registration
instead of using the local_dma_lkey. As mentioned earlier, there are
scatter limits and other implications for all
On 11/10/2015 1:25 PM, Jason Gunthorpe wrote:
On Tue, Nov 10, 2015 at 04:04:32AM -0800, Christoph Hellwig wrote:
On Tue, Nov 10, 2015 at 01:46:40PM +0200, Sagi Grimberg wrote:
On 10/11/2015 13:41, Christoph Hellwig wrote:
Oh, and while we're at it. Can someone explain why we're even
using
On 11/10/2015 6:51 AM, Yann Droneaud wrote:
Le mardi 10 novembre 2015 à 12:44 +0200, Sagi Grimberg a écrit :
Also, set rdma_read_access_flags in the relevant device drivers:
mlx4, mlx5, qib, ocrdma, nes: IB_ACCESS_LOCAL_WRITE
cxgb3, cxgb4, hfi: IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE
On 9/6/2015 11:15 AM, Bart Van Assche wrote:
On 09/06/15 00:50, Christoph Hellwig wrote:
Note that the SRP driver already in tree is a good example for this,
although it doesn't use RDMA/CM and thus already operates on a
per-ib_device level.
The challenges with regard to adding RDMA/CM
On 8/25/2015 4:35 AM, Achiad Shochat wrote:
On 8/24/2015 11:40 PM, Tom Talpey wrote:
On 8/20/2015 12:46 PM, Achiad Shochat wrote:
Hi Doug,
This patchset adds RoCE V1 and RoCE V2 support to the mlx5 device
driver.
Question - assuming I read them correctly, these patches add the
RoCE v1
On 8/20/2015 12:46 PM, Achiad Shochat wrote:
Hi Doug,
This patchset adds RoCE V1 and RoCE V2 support to the mlx5 device
driver.
Question - assuming I read them correctly, these patches add the
RoCE v1 and RoCE v2 support on a per-port basis. That is, a port
can be either IB, RoCE v1 or RoCE
On 7/21/2015 7:33 AM, Steve Wise wrote:
-Original Message-
From: linux-rdma-ow...@vger.kernel.org
[mailto:linux-rdma-ow...@vger.kernel.org] On Behalf Of Tom Talpey
Sent: Monday, July 20, 2015 7:15 PM
To: Steve Wise; 'Jason Gunthorpe'
Cc: 'Chuck Lever'; linux-rdma@vger.kernel.org
On 7/20/2015 1:55 PM, Chuck Lever wrote:
On Jul 20, 2015, at 4:34 PM, Tom Talpey t...@talpey.com wrote:
On 7/20/2015 12:03 PM, Chuck Lever wrote:
All HCA providers have an ib_get_dma_mr() verb. Thus
rpcrdma_ia_open() will either grab the device's local_dma_key if one
is available
On 7/20/2015 2:16 PM, Steve Wise wrote:
-Original Message-
From: linux-nfs-ow...@vger.kernel.org [mailto:linux-nfs-ow...@vger.kernel.org]
On Behalf Of Jason Gunthorpe
Sent: Monday, July 20, 2015 4:06 PM
To: Tom Talpey; Steve Wise
Cc: Chuck Lever; linux-rdma@vger.kernel.org; linux
On 7/20/2015 12:03 PM, Chuck Lever wrote:
All HCA providers have an ib_get_dma_mr() verb. Thus
rpcrdma_ia_open() will either grab the device's local_dma_key if one
is available, or it will call ib_get_dma_mr() which is a 100%
guaranteed fallback.
I recall that in the past, some providers did
On 7/20/2015 3:17 PM, Jason Gunthorpe wrote:
On Mon, Jul 20, 2015 at 03:04:21PM -0700, Tom Talpey wrote:
B) why bother to check? Are machines with 4GB interesting, and worth
supporting a special optimization?
mainline drivers shouldn't silently malfunction.
I meant why bother to check, just
On 7/20/2015 5:34 PM, Chuck Lever wrote:
On Jul 20, 2015, at 8:11 PM, Tom Talpey t...@talpey.com wrote:
On 7/20/2015 4:36 PM, Chuck Lever wrote:
On Jul 20, 2015, at 6:41 PM, Jason Gunthorpe jguntho...@obsidianresearch.com
wrote:
On Mon, Jul 20, 2015 at 06:31:11PM -0400, Chuck Lever wrote
On 7/20/2015 3:21 PM, Chuck Lever wrote:
On Jul 20, 2015, at 5:55 PM, Tom Talpey t...@talpey.com wrote:
On 7/20/2015 1:55 PM, Chuck Lever wrote:
On Jul 20, 2015, at 4:34 PM, Tom Talpey t...@talpey.com wrote:
On 7/20/2015 12:03 PM, Chuck Lever wrote:
All HCA providers have an ib_get_dma_mr
know.
Tom Talpey or Tom Tucker? I don't think I used a dma_lkey at all
in the original work, but there were some changes later.
I thought not all rnic's supported a dma_lkey. So requiring it
seems like a bad idea, to me.
Just use
the code you have for fmr with frwr to get the lkey, probably move
On 7/20/2015 3:41 PM, Steve Wise wrote:
-Original Message-
From: Tom Talpey [mailto:t...@talpey.com]
Sent: Monday, July 20, 2015 5:04 PM
To: Steve Wise; 'Jason Gunthorpe'
Cc: 'Chuck Lever'; linux-rdma@vger.kernel.org; linux-...@vger.kernel.org
Subject: Re: [PATCH v3 05/15] xprtrdma
On 7/14/2015 5:22 AM, Sagi Grimberg wrote:
On 7/14/2015 10:37 AM, 'Christoph Hellwig' wrote:
On Mon, Jul 13, 2015 at 03:36:44PM -0400, Tom Talpey wrote:
On 7/11/2015 6:25 AM, 'Christoph Hellwig' wrote:
I think what we need to support for now is FRMR as the primary target,
and FMR
On 7/14/2015 11:36 AM, 'Christoph Hellwig' wrote:
On Tue, Jul 14, 2015 at 12:10:36PM +0300, Sagi Grimberg wrote:
Having an API that does FRMR/FMR/PHYS_MR is even worse from the ULP
PoV. If you expose an API that might schedule (PHYS_MR) it limits the
context that the caller is allowed to call
On 7/14/2015 3:32 PM, Steve Wise wrote:
On Tue, Jul 14, 2015 at 02:25:50PM -0500, Steve Wise wrote:
The benefit is that we don't have to check for iWARP protocol in the
ULP.
IB should have to pay the cost of FRMR for lkey on RDMA READ, I'm
pretty sure you have to check for iWarp at
On 7/13/2015 12:30 PM, Chuck Lever wrote:
RDMA_MSGP type calls insert a zero pad in the middle of the RPC
message to align the RPC request's data payload to the server's
alignment preferences. A server can then page flip the payload
into place to avoid a data copy in certain circumstances.
Assche; Liran Liss;
Hefty,
Sean; Doug Ledford; Tom Talpey
Subject: Re: Kernel fast memory registration API proposal [RFC]
On 7/14/2015 6:33 PM, Christoph Hellwig wrote:
On Tue, Jul 14, 2015 at 11:39:24AM +0300, Sagi Grimberg wrote:
This is exactly what I don't want to do. I don't think
On 7/11/2015 6:25 AM, 'Christoph Hellwig' wrote:
I think what we need to support for now is FRMR as the primary target,
and FMR as a secondar[y].
FMR is a *very* bad choice, for several reasons.
1) It's not supported by very many devices, in fact it might even
be thought to be obsolete.
2)
On 7/13/2015 1:18 PM, Jason Gunthorpe wrote:
On Fri, Jul 10, 2015 at 11:10:23PM -0400, Doug Ledford wrote:
Black hat server is beyond the scope of this discussion.
We cannot assume an all-trusted model here, there are many
configurations to deploy NFS/iSCSI that don't assume that. Even if you
On 7/10/2015 7:29 AM, Devesh Sharma wrote:
we need to honor the max limits of device by checking
dev_attr.max_sge? a vendor may not support 4 sges.
iWARP requires a minimum of 4 send SGEs (draft-hilland-verbs 8.1.3.2)
An RI MUST support at least four Scatter/Gather Elements per
On 7/9/2015 6:53 PM, Jason Gunthorpe wrote:
On Thu, Jul 09, 2015 at 06:18:26PM -0400, Doug Ledford wrote:
A lot of this discussion is interesting and has gone off in an area that
I think is worth pursuing in the long run. However, in the short run,
this patch was a minor
On 7/10/2015 1:56 PM, Doug Ledford wrote:
On 07/10/2015 12:11 PM, Jason Gunthorpe wrote:
On Fri, Jul 10, 2015 at 09:22:24AM -0400, Tom Talpey wrote:
and it is enabled only when the RDMA Read is active.
??? How is that done? ib_get_dma_mr is defined to return a remote
usable rkey
On 7/9/2015 1:01 PM, Jason Gunthorpe wrote:
Laid out like this, I think it even means we can nuke the IB DMA API
for these cases. rdma_post_read and rdma_post_complete_read are the
two points that need dma api calls (cache flushes), and they can just
do them internally.
This also tells me that
On 7/8/2015 3:08 PM, Jason Gunthorpe wrote:
The MR stuff was never really designed, the HW people provided some
capability and the SW side just raw exposed it, thoughtlessly.
Jason, I don't disagree that the API can be improved. I have some
responses to your statements below though.
Why is
On 4/27/2015 2:52 PM, ira.weiny wrote:
On Mon, Apr 27, 2015 at 09:39:05AM +0200, Michael Wang wrote:
On 04/24/2015 05:12 PM, Liran Liss wrote:
From: linux-rdma-ow...@vger.kernel.org [mailto:linux-rdma-
[snip]
a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index
65994a1..d54f91e
On 4/27/2015 5:36 PM, Doug Ledford wrote:
On Mon, 2015-04-27 at 17:16 -0700, Tom Talpey wrote:
On 4/27/2015 2:52 PM, ira.weiny wrote:
On Mon, Apr 27, 2015 at 09:39:05AM +0200, Michael Wang wrote:
On 04/24/2015 05:12 PM, Liran Liss wrote:
[snip]
Like:
enum rdma_protocol
On 4/27/2015 6:24 PM, Doug Ledford wrote:
On Mon, 2015-04-27 at 17:53 -0700, Tom Talpey wrote:
On 4/27/2015 5:36 PM, Doug Ledford wrote:
On Mon, 2015-04-27 at 17:16 -0700, Tom Talpey wrote:
On 4/27/2015 2:52 PM, ira.weiny wrote:
On Mon, Apr 27, 2015 at 09:39:05AM +0200, Michael Wang wrote
On 4/24/2015 8:23 AM, Michael Wang wrote:
Add new callback query_transport() and implement for each HW.
Mapping List:
node-type link-layer old-transport new-transport
...
mlx4IB_CA IB/ETH IB IB/IBOE
mlx5IB_CA
On 4/24/2015 10:35 AM, Michael Wang wrote:
On 04/24/2015 04:29 PM, Tom Talpey wrote:
On 4/24/2015 8:23 AM, Michael Wang wrote:
[snip]
+static enum rdma_transport_type
+mlx5_ib_query_transport(struct ib_device *device, u8 port_num)
+{
+return RDMA_TRANSPORT_IB;
+}
+
Just noticed
On 4/16/2015 11:22 AM, Michael Wang wrote:
On 04/16/2015 04:31 PM, Hefty, Sean wrote:
This is equivalent to today where the checks are per node rather than
per port.
Should all checks here be port 1 based or only certain ones like listen
? For example, in connect/reject/disconnect, don't we
On 4/10/2015 5:06 PM, ira.weiny wrote:
On Fri, Apr 10, 2015 at 01:17:23PM -0600, Jason Gunthorpe wrote:
...
So trying to sum up.
Have we settled on the following capabilities? Helper function names aside.
/* legacy to communicate to userspace */
RDMA_LINK_LAYER_IB = 0x0001,
On 4/7/2015 8:25 AM, Michael Wang wrote:
Mapping List:
node-type link-layer old-transport new-transport
nes RNICETH IWARP IWARP
amso1100RNICETH IWARP IWARP
cxgb3 RNIC
On 4/8/2015 4:10 PM, Jason Gunthorpe wrote:
On Wed, Apr 08, 2015 at 02:29:46PM -0400, Doug Ledford wrote:
...
rdma_port_get_read_sge(dev, port)
{
if (rdma_transport_is_iwarp)
return 1;
return dev-port[port]-max_sge;
}
Hum, that is nice, but it doesn't quite
On 4/7/2015 8:34 AM, Michael Wang wrote:
/**
+ * cap_read_multi_sge - Check if the port of device has the capability
+ * RDMA Read Multiple Scatter-Gather Entries.
+ *
+ * @device: Device to be checked
+ * @port_num: Port number of the device
+ *
+ * Return 0 when port of the device don't
On 3/31/2015 3:39 AM, Michael Wang wrote:
On 03/31/2015 12:35 AM, Jason Gunthorpe wrote:
On Mon, Mar 30, 2015 at 05:10:12PM +0200, Michael Wang wrote:
I found that actually we don't have to touch this one which
only used by HW driver currently.
I'm having a hard time understanding this, the
On 3/31/2015 7:41 AM, Michael Wang wrote:
Hi, Tom
Thanks for the comments :-)
On 03/31/2015 01:19 PM, Tom Talpey wrote:
[oops - repeating my reply with full cc's]
[snip]
Actually I'm thinking about Doug's idea to use rdma_transport_is_XX()
instead of the current basic helper, thus may
On 2/26/2014 8:44 AM, Chuck Lever wrote:
Hi-
Shirley Ma and I are reviving work on the NFS/RDMA client code base in the
Linux kernel. So far we’ve built and run functional tests to determine what is
working and what is broken.
One complication is the number of memory registration modes
On 8/21/2013 11:55 AM, Wendy Cheng wrote:
On Thu, Aug 15, 2013 at 11:08 AM, Wendy Cheng s.wendy.ch...@gmail.com wrote:
On Thu, Aug 15, 2013 at 5:46 AM, Tom Talpey t...@talpey.com wrote:
On 8/14/2013 8:14 PM, Wendy Cheng wrote:
Longer version of the question:
I'm trying to enable NFS-RDMA
On 8/14/2013 8:14 PM, Wendy Cheng wrote:
Longer version of the question:
I'm trying to enable NFS-RDMA on an embedded system (based on 2.6.38
kernel) as a client. The IB stacks are taken from OFED 1.5.4. NFS
server is a RHEL 6.3 Xeon box. The connection uses mellox-4 driver.
Memory registration
On 4/30/2013 1:09 AM, Yan Burman wrote:
-Original Message-
From: J. Bruce Fields [mailto:bfie...@fieldses.org]
Sent: Sunday, April 28, 2013 17:43
To: Yan Burman
Cc: Wendy Cheng; Atchley, Scott; Tom Tucker; linux-rdma@vger.kernel.org;
linux-...@vger.kernel.org; Or Gerlitz
Subject: Re:
On 4/30/2013 1:09 AM, Yan Burman wrote:
I now get up to ~95K IOPS and 4.1GB/sec bandwidth.
...
ib_send_bw with intel iommu enabled did get up to 4.5GB/sec
BTW, you may want to verify that these are the same GB. Many
benchmarks say KB/MB/GB when they really mean KiB/MiB/GiB.
At GB/GiB, the
On 4/30/2013 10:23 AM, Yan Burman wrote:
-Original Message-
From: Tom Talpey [mailto:t...@talpey.com]
On Sun, Apr 28, 2013 at 06:28:16AM +, Yan Burman wrote:
I finally got up to 4.1GB/sec bandwidth with RDMA (ipoib-CM bandwidth is
also way higher now).
For some reason when I had
On 4/25/2013 1:18 PM, Wendy Cheng wrote:
On Wed, Apr 24, 2013 at 11:26 AM, Tom Talpey t...@talpey.com wrote:
On Wed, Apr 24, 2013 at 9:27 AM, Wendy Cheng s.wendy.ch...@gmail.com
wrote:
So I did a quick read on sunrpc/xprtrdma source (based on OFA 1.5.4.1
tar ball) ... Here is a random
On 4/25/2013 3:01 PM, Phil Pishioneri wrote:
On 4/25/13 1:18 PM, Wendy Cheng wrote:
On Wed, Apr 24, 2013 at 11:26 AM, Tom Talpey t...@talpey.com wrote:
1)
The client slot count is not hard-coded, it can easily be changed by
writing a value to /proc and initiating a new mount. But I doubt
On 4/24/2013 2:04 PM, Wendy Cheng wrote:
On Wed, Apr 24, 2013 at 9:27 AM, Wendy Cheng s.wendy.ch...@gmail.com wrote:
On Wed, Apr 24, 2013 at 8:26 AM, J. Bruce Fields bfie...@fieldses.org wrote:
On Wed, Apr 24, 2013 at 11:05:40AM -0400, J. Bruce Fields wrote:
On Wed, Apr 24, 2013 at 12:35:03PM
67 matches
Mail list logo