From: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
To: Jason Gunthorpe <jgg@nvidia.com>, "Wan, Kaike" <kaike.wan@intel.com>
Cc: "dledford@redhat.com" <dledford@redhat.com>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
"Rimmer, Todd" <todd.rimmer@intel.com>
Subject: Re: [PATCH RFC 0/9] A rendezvous module
Date: Fri, 19 Mar 2021 15:22:45 -0400 [thread overview]
Message-ID: <29061edb-b40c-67a9-c329-3c9446f0f434@cornelisnetworks.com> (raw)
In-Reply-To: <20210319154805.GV2356281@nvidia.com>
On 3/19/2021 11:48 AM, Jason Gunthorpe wrote:
> On Fri, Mar 19, 2021 at 02:49:29PM +0000, Wan, Kaike wrote:
>>> From: Jason Gunthorpe <jgg@nvidia.com>
>>> Sent: Friday, March 19, 2021 9:53 AM
>>> To: Wan, Kaike <kaike.wan@intel.com>
>>> Cc: dledford@redhat.com; linux-rdma@vger.kernel.org; Rimmer, Todd
>>> <todd.rimmer@intel.com>
>>> Subject: Re: [PATCH RFC 0/9] A rendezvous module
>>>
>>> On Fri, Mar 19, 2021 at 08:56:26AM -0400, kaike.wan@intel.com wrote:
>>>
>>>> - Basic mode of operations (PSM3 is used as an example for user
>>>> applications):
>>>> - A middleware (like MPI) has out-of-band communication channels
>>>> between any two nodes, which are used to establish high performance
>>>> communications for providers such as PSM3.
>>>
>>> Huh? Doesn't PSM3 already use it's own special non-verbs char devices that
>>> already have memory caches and other stuff? Now you want to throw that
>>> all away and do yet another char dev just for HFI? Why?
>
>> [Wan, Kaike] I think that you are referring to PSM2, which uses the
>> OPA hfi1 driver that is specific to the OPA hardware. PSM3 uses
>> standard verbs drivers and supports standard RoCE.
>
> Uhhh.. "PSM" has always been about the ipath special char device, and
> if I recall properly the library was semi-discontinued and merged into
> libfabric.
This driver is intended to work with a fork of the PSM2 library. The
PSM2 library which is for Omni-Path is now maintained by Cornelis
Networks on our GitHub. PSM3 is something from Intel for Ethernet. I
know it's a bit confusing.
> So here you are talking about a libfabric verbs provider that doesn't
> use the ipath style char interface but uses verbs and this rv thing so
> we call it a libfabric PSM3 provider because thats not confusing to
> anyone at all..
>
>> A focus is the Intel RDMA Ethernet NICs. As such it cannot use the
>> hfi1 driver through the special PSM2 interface.
>
> These are the drivers that aren't merged yet, I see. So why are you
> sending this now? I'm not interested to look at even more Intel code
> when their driver saga is still ongoing for years.
>
>> Rather it works with the hfi1 driver through standard verbs
>> interface.
>
> But nobody would do that right? You'd get better results using the
> hif1 native interfaces instead of their slow fake verbs stuff.
I can't imagine why. I'm not sure what you mean by our slow fake verbs
stuff? We support verbs just fine. It's certainly not fake.
>>> I also don't know why you picked the name rv, this looks like it has little to do
>>> with the usual MPI rendezvous protocol. This is all about bulk transfers. It is
>>> actually a lot like RDS. Maybe you should be using RDS?
.
.
>> The name "rv" is chosen simply because this module is designed to
>> enable the rendezvous protocol of the MPI/OFI/PSM3 application stack
>> for large messages. Short messages are handled by eager transfer
>> through UDP in PSM3.
>
> A bad name seems like it will further limit potential re-use of this
> code.
As to the name, I started this driver when I was still at Intel before
handing the reins to Kaike when we spun out. I plucked the name rv out
of the air without much consideration. While the code seems to have
changed a lot, the name seems to have stuck. It was as Kaike mentions
because it helps enable the rendezvous side of PSM3.
-Denny
next prev parent reply other threads:[~2021-03-19 19:24 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-19 12:56 [PATCH RFC 0/9] A rendezvous module kaike.wan
2021-03-19 12:56 ` [PATCH RFC 1/9] RDMA/rv: Public interferce for the RDMA Rendezvous module kaike.wan
2021-03-19 16:00 ` Jason Gunthorpe
2021-03-19 12:56 ` [PATCH RFC 2/9] RDMA/rv: Add the internal header files kaike.wan
2021-03-19 16:02 ` Jason Gunthorpe
2021-03-19 12:56 ` [PATCH RFC 3/9] RDMA/rv: Add the rv module kaike.wan
2021-03-19 12:56 ` [PATCH RFC 4/9] RDMA/rv: Add functions for memory region cache kaike.wan
2021-03-19 12:56 ` [PATCH RFC 5/9] RDMA/rv: Add function to register/deregister memory region kaike.wan
2021-03-19 12:56 ` [PATCH RFC 6/9] RDMA/rv: Add connection management functions kaike.wan
2021-03-19 12:56 ` [PATCH RFC 7/9] RDMA/rv: Add functions for RDMA transactions kaike.wan
2021-03-19 12:56 ` [PATCH RFC 8/9] RDMA/rv: Add functions for file operations kaike.wan
2021-03-19 12:56 ` [PATCH RFC 9/9] RDMA/rv: Integrate the file operations into the rv module kaike.wan
2021-03-19 13:53 ` [PATCH RFC 0/9] A rendezvous module Jason Gunthorpe
2021-03-19 14:49 ` Wan, Kaike
2021-03-19 15:48 ` Jason Gunthorpe
2021-03-19 19:22 ` Dennis Dalessandro [this message]
2021-03-19 19:44 ` Jason Gunthorpe
2021-03-19 20:12 ` Rimmer, Todd
2021-03-19 20:26 ` Jason Gunthorpe
2021-03-19 20:46 ` Rimmer, Todd
2021-03-19 20:54 ` Jason Gunthorpe
2021-03-19 20:59 ` Wan, Kaike
2021-03-19 21:28 ` Dennis Dalessandro
2021-03-19 21:58 ` Wan, Kaike
2021-03-19 22:35 ` Jason Gunthorpe
2021-03-19 22:57 ` Rimmer, Todd
2021-03-19 23:06 ` Jason Gunthorpe
2021-03-20 16:39 ` Dennis Dalessandro
2021-03-21 8:56 ` Leon Romanovsky
2021-03-21 16:24 ` Dennis Dalessandro
2021-03-21 16:45 ` Jason Gunthorpe
2021-03-21 17:21 ` Dennis Dalessandro
2021-03-21 18:08 ` Jason Gunthorpe
2021-03-22 15:17 ` Rimmer, Todd
2021-03-22 16:47 ` Jason Gunthorpe
2021-03-22 17:31 ` Hefty, Sean
2021-03-23 22:56 ` Jason Gunthorpe
2021-03-23 23:29 ` Rimmer, Todd
2021-03-21 19:19 ` Wan, Kaike
2021-03-23 15:36 ` Christoph Hellwig
2021-03-23 15:35 ` Christoph Hellwig
2021-03-23 15:33 ` Christoph Hellwig
2021-03-23 15:30 ` Christoph Hellwig
2021-03-23 15:46 ` Jason Gunthorpe
2021-03-23 16:07 ` Christoph Hellwig
2021-03-23 17:25 ` Rimmer, Todd
2021-03-23 17:44 ` Jason Gunthorpe
2021-03-19 20:18 ` Dennis Dalessandro
2021-03-19 20:30 ` Jason Gunthorpe
2021-03-19 20:34 ` Hefty, Sean
2021-03-21 12:08 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=29061edb-b40c-67a9-c329-3c9446f0f434@cornelisnetworks.com \
--to=dennis.dalessandro@cornelisnetworks.com \
--cc=dledford@redhat.com \
--cc=jgg@nvidia.com \
--cc=kaike.wan@intel.com \
--cc=linux-rdma@vger.kernel.org \
--cc=todd.rimmer@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).