[LSF/MM TOPIC] NVMe target support

[LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

Hi lsf-pc & Co,

I'd like to propose a NVMe target discussion topic for LSF/MM 2016.

My interest wrt NVMe target drivers is specifically around using shared
infrastructure with existing TCM backend drivers (IBLOCK, FILEIO,
RAMDISK), and how se_device->dev_group configfs symlinks can be
associated with NVMe target driver ports following HCH's WIP code.
Also, there is still a vendor need around some type of NVMe backend
driver (eg: target_core_nvme) for exposing + mapping NVMe-HI queues to
front-end fabric driver RDMA queues.

Beyond the back-end driver discussion, our target_core_fabric_configfs.c
logic for exposing a common user-space ABI has served LIO target drivers
well over the last 10 upstream fabrics, plus other out-of-tree and WIP
drivers.  The design has never once had to break userspace ABI
compatibility, and the only user-visible addition to original design
was allowing network portal attributes to be exposed by fabrics for
iser-target + iscsit_transport friends.

I think one of the more interesting questions is going to be around how
NVMe-over-Fabrics (NVMe_OF) + vhost-nvme will end up using
target_core_fabric_configfs.c code.  What struct config_groups from
target_core_fabric_configfs.c can be common across both SCSI target and
NVMe controller export..?

The folks for such a discussion would include:

   Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi Grimberg,
   Ming Lin, Roland Dreier and Mike Christie.

Thank you,

--nab

[LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

Hi lsf-pc & Co,

I'd like to propose a NVMe target discussion topic for LSF/MM 2016.

My interest wrt NVMe target drivers is specifically around using shared
infrastructure with existing TCM backend drivers (IBLOCK, FILEIO,
RAMDISK), and how se_device->dev_group configfs symlinks can be
associated with NVMe target driver ports following HCH's WIP code.
Also, there is still a vendor need around some type of NVMe backend
driver (eg: target_core_nvme) for exposing + mapping NVMe-HI queues to
front-end fabric driver RDMA queues.

Beyond the back-end driver discussion, our target_core_fabric_configfs.c
logic for exposing a common user-space ABI has served LIO target drivers
well over the last 10 upstream fabrics, plus other out-of-tree and WIP
drivers.  The design has never once had to break userspace ABI
compatibility, and the only user-visible addition to original design
was allowing network portal attributes to be exposed by fabrics for
iser-target + iscsit_transport friends.

I think one of the more interesting questions is going to be around how
NVMe-over-Fabrics (NVMe_OF) + vhost-nvme will end up using
target_core_fabric_configfs.c code.  What struct config_groups from
target_core_fabric_configfs.c can be common across both SCSI target and
NVMe controller export..?

The folks for such a discussion would include:

   Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi Grimberg,
   Ming Lin, Roland Dreier and Mike Christie.

Thank you,

--nab

Re: [LSF/MM TOPIC] NVMe target support

[Christoph Hellwig]

I think the frame is wrong, nevermind that I doubt I'll be able to
discuss anything NVMe Fabrics related before the spec is public.  Which
I doubt is in time for LSF/MM.

[LSF/MM TOPIC] NVMe target support

[Christoph Hellwig]

I think the frame is wrong, nevermind that I doubt I'll be able to
discuss anything NVMe Fabrics related before the spec is public.  Which
I doubt is in time for LSF/MM.

Re: [LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

On Thu, 2016-01-28 at 00:52 -0800, Christoph Hellwig wrote:
> I think the frame is wrong, nevermind that I doubt I'll be able to
> discuss anything NVMe Fabrics related before the spec is public.  Which
> I doubt is in time for LSF/MM.

That fine, but NVMe_OF is certainly not going to be the only fabric
driver out there.

So what I'd like to discuss wrt this topic is what existing common code
relates to vhost-nvme, based on other's fabric driver work already out
there.

[LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

On Thu, 2016-01-28@00:52 -0800, Christoph Hellwig wrote:
> I think the frame is wrong, nevermind that I doubt I'll be able to
> discuss anything NVMe Fabrics related before the spec is public.  Which
> I doubt is in time for LSF/MM.

That fine, but NVMe_OF is certainly not going to be the only fabric
driver out there.

So what I'd like to discuss wrt this topic is what existing common code
relates to vhost-nvme, based on other's fabric driver work already out
there.

Re: [LSF/MM TOPIC] NVMe target support

[Christoph Hellwig]

There is a upcoming 'NVMe over Fabrics' spec, and it seems to be your are
using the word 'fabrics' in a different context, so you might want to
explain it in lay mans term.

If I was dictator for life I'd forbid the use of the word 'fabrics'
for any not related to textiles..

[LSF/MM TOPIC] NVMe target support

[Christoph Hellwig]

There is a upcoming 'NVMe over Fabrics' spec, and it seems to be your are
using the word 'fabrics' in a different context, so you might want to
explain it in lay mans term.

If I was dictator for life I'd forbid the use of the word 'fabrics'
for any not related to textiles..

RE: [LSF/MM TOPIC] NVMe target support

[Judy Brock-SSI]

	>>I'd forbid the use of the word 'fabrics' for any not related to textiles..
lol

Btw, You're not dictator for life?

-----Original Message-----
From: Linux-nvme [mailto:linux-nvme-bounces@lists.infradead.org] On Behalf Of Christoph Hellwig
Sent: Thursday, January 28, 2016 1:12 AM
To: Nicholas A. Bellinger
Cc: Minturn, Dave B; Sagi Grimberg; Roland Dreier; linux-scsi; linux-nvme@lists.infradead.org; Christoph Hellwig; Mike Christie; target-devel; Hannes Reinecke; Ming Lin; lsf-pc; Christoph Hellwig
Subject: Re: [LSF/MM TOPIC] NVMe target support

There is a upcoming 'NVMe over Fabrics' spec, and it seems to be your are using the word 'fabrics' in a different context, so you might want to explain it in lay mans term.

If I was dictator for life I'd forbid the use of the word 'fabrics'
for any not related to textiles..

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

[LSF/MM TOPIC] NVMe target support

[Judy Brock-SSI]

	>>I'd forbid the use of the word 'fabrics' for any not related to textiles..
lol

Btw, You're not dictator for life?

-----Original Message-----
From: Linux-nvme [mailto:linux-nvme-bounces@lists.infradead.org] On Behalf Of Christoph Hellwig
Sent: Thursday, January 28, 2016 1:12 AM
To: Nicholas A. Bellinger
Cc: Minturn, Dave B; Sagi Grimberg; Roland Dreier; linux-scsi; linux-nvme at lists.infradead.org; Christoph Hellwig; Mike Christie; target-devel; Hannes Reinecke; Ming Lin; lsf-pc; Christoph Hellwig
Subject: Re: [LSF/MM TOPIC] NVMe target support

There is a upcoming 'NVMe over Fabrics' spec, and it seems to be your are using the word 'fabrics' in a different context, so you might want to explain it in lay mans term.

If I was dictator for life I'd forbid the use of the word 'fabrics'
for any not related to textiles..

_______________________________________________
Linux-nvme mailing list
Linux-nvme at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

Re: [LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

On Thu, 2016-01-28 at 10:11 +0100, Christoph Hellwig wrote:
> There is a upcoming 'NVMe over Fabrics' spec, 

Yes.

> and it seems to be your are
> using the word 'fabrics' in a different context, so you might want to
> explain it in lay mans term.
> 

I mean in the context of target_core_fabric_ops and front-end driver
association with backend se_device->dev_group configfs symlinks.
Specifically how NVMe target drivers will be mapping these into the
existing TCM configfs layout.

> If I was dictator for life I'd forbid the use of the word 'fabrics'
> for any not related to textiles..

..?

[LSF/MM TOPIC] NVMe target support

[Nicholas A. Bellinger]

On Thu, 2016-01-28@10:11 +0100, Christoph Hellwig wrote:
> There is a upcoming 'NVMe over Fabrics' spec, 

Yes.

> and it seems to be your are
> using the word 'fabrics' in a different context, so you might want to
> explain it in lay mans term.
> 

I mean in the context of target_core_fabric_ops and front-end driver
association with backend se_device->dev_group configfs symlinks.
Specifically how NVMe target drivers will be mapping these into the
existing TCM configfs layout.

> If I was dictator for life I'd forbid the use of the word 'fabrics'
> for any not related to textiles..

..?

Re: [LSF/MM TOPIC] NVMe target support

[Bart Van Assche]

On 01/27/16 20:41, Nicholas A. Bellinger wrote:
> [ ... ]
> The folks for such a discussion would include:
>
>     Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi Grimberg,
>     Ming Lin, Roland Dreier and Mike Christie.

Hello Nic,

Since the LSF/MM summit is organized by the Linux Foundation and since 
one of the goals of the Linux Foundation is to promote Linux I think 
every company that is active in both the NVMeOF committee and in the 
Linux kernel community should be invited. That includes but is not 
limited to companies like Oracle, Micron, EMC, NetApp, HGST and also my 
own employer SanDisk (see e.g. http://www.nvmexpress.org/).

Thanks,

Bart.

[LSF/MM TOPIC] NVMe target support

[Bart Van Assche]

On 01/27/16 20:41, Nicholas A. Bellinger wrote:
> [ ... ]
> The folks for such a discussion would include:
>
>     Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi Grimberg,
>     Ming Lin, Roland Dreier and Mike Christie.

Hello Nic,

Since the LSF/MM summit is organized by the Linux Foundation and since 
one of the goals of the Linux Foundation is to promote Linux I think 
every company that is active in both the NVMeOF committee and in the 
Linux kernel community should be invited. That includes but is not 
limited to companies like Oracle, Micron, EMC, NetApp, HGST and also my 
own employer SanDisk (see e.g. http://www.nvmexpress.org/).

Thanks,

Bart.

Re: [Lsf-pc] [LSF/MM TOPIC] NVMe target support

[James Bottomley]

On Thu, 2016-01-28 at 07:16 -0800, Bart Van Assche wrote:
> On 01/27/16 20:41, Nicholas A. Bellinger wrote:
> > [ ... ]
> > The folks for such a discussion would include:
> > 
> >     Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi
> > Grimberg,
> >     Ming Lin, Roland Dreier and Mike Christie.
> 
> Hello Nic,
> 
> Since the LSF/MM summit is organized by the Linux Foundation

This is a misperception.  LSF/MM is organised by a group of volunteers
as a Linux Kernel mini summit to discuss current topics around those
subsystems.

>  and since one of the goals of the Linux Foundation is to promote
> Linux I think every company that is active in both the NVMeOF
> committee and in the Linux kernel community should be invited.

  We use the LF to do the logistics, but it's in no way controlled by
them or aligned with their member companies.  The first LSF/MM was
aided logistically by google, the next by Intel, then USENIX for a
while and now the LF.

I certainly wouldn't be in favour of turning LSF/MM into a corporate
discussion forum ... that's what standards bodies are for.

James

>  That includes but is not limited to companies like Oracle, Micron,
> EMC, NetApp, HGST and also my own employer SanDisk (see e.g. 
> http://www.nvmexpress.org/).
> 
> Thanks,
> 
> Bart.
> _______________________________________________
> Lsf-pc mailing list
> Lsf-pc@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/lsf-pc
> 

[Lsf-pc] [LSF/MM TOPIC] NVMe target support

[James Bottomley]

On Thu, 2016-01-28@07:16 -0800, Bart Van Assche wrote:
> On 01/27/16 20:41, Nicholas A. Bellinger wrote:
> > [ ... ]
> > The folks for such a discussion would include:
> > 
> >     Christoph Hellwig, Hannes Reinecke, Dave Minturn, Sagi
> > Grimberg,
> >     Ming Lin, Roland Dreier and Mike Christie.
> 
> Hello Nic,
> 
> Since the LSF/MM summit is organized by the Linux Foundation

This is a misperception.  LSF/MM is organised by a group of volunteers
as a Linux Kernel mini summit to discuss current topics around those
subsystems.

>  and since one of the goals of the Linux Foundation is to promote
> Linux I think every company that is active in both the NVMeOF
> committee and in the Linux kernel community should be invited.

  We use the LF to do the logistics, but it's in no way controlled by
them or aligned with their member companies.  The first LSF/MM was
aided logistically by google, the next by Intel, then USENIX for a
while and now the LF.

I certainly wouldn't be in favour of turning LSF/MM into a corporate
discussion forum ... that's what standards bodies are for.

James

>  That includes but is not limited to companies like Oracle, Micron,
> EMC, NetApp, HGST and also my own employer SanDisk (see e.g. 
> http://www.nvmexpress.org/).
> 
> Thanks,
> 
> Bart.
> _______________________________________________
> Lsf-pc mailing list
> Lsf-pc at lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/lsf-pc
>