XFS BoF at LSFMM

XFS BoF at LSFMM

[Luis Chamberlain]

How about an XFS BoF at LSFMM? Would it be good to separate the BoF
for XFS and bcachefs so that folks who want to attend both can do so?
How about Wednesday 15:30? That would allow a full hour for bcachefs.

  Luis

Re: [Lsf-pc] XFS BoF at LSFMM

[Amir Goldstein]

On Wed, May 8, 2024 at 10:42 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>
> How about an XFS BoF at LSFMM?

Let me rephrase that.

Darrick,

Would you like to list some items for the XFS BoF so I can put it on
the agenda and link to this thread?

> Would it be good to separate the BoF
> for XFS and bcachefs so that folks who want to attend both can do so?

I have set a side 2.5 hours in the schedule on Wed afternoon for per-FS BoF
3 hours if you include the FS lightning slot afterwards that could be used
as per-FS lightning updates.

> How about Wednesday 15:30? That would allow a full hour for bcachefs.

I have no doubt that people want to be updated about bcaches,
but a full hour for bcachefs? IDK.

FYI, I counted more than 10 attendees that are active contributors or
have contributed to xfs in one way or another.
That's roughly a third of the FS track.

So let's see how much time XFS BoF needs and divide the rest of the
time among other FS that want to do a BoF in the FS room.

If anyone would like to run a *FS BoF in the FS room please let me know
and please try to estimate how much time you will need.
You can also "register" for *FS lightning talk, but we can also arrange this
ad-hoc.

Excited to see you all next week!

Thanks,
Amir.

Re: [Lsf-pc] XFS BoF at LSFMM

[Christoph Hellwig]

On Thu, May 09, 2024 at 08:01:39AM +0300, Amir Goldstein wrote:
> 
> FYI, I counted more than 10 attendees that are active contributors or
> have contributed to xfs in one way or another.
> That's roughly a third of the FS track.

FYI, I'm flying out at 4:15pm on Wednesday, and while I try to keep my
time at the airport short I'd still be gone by 3:30.

But that will only matter if you make the BOF and actual BOF and not the
usual televised crap that happens at LSFMM.

Re: [Lsf-pc] XFS BoF at LSFMM

[Amir Goldstein]

On Thu, May 9, 2024 at 8:06 AM Christoph Hellwig <hch@infradead.org> wrote:
>
> On Thu, May 09, 2024 at 08:01:39AM +0300, Amir Goldstein wrote:
> >
> > FYI, I counted more than 10 attendees that are active contributors or
> > have contributed to xfs in one way or another.
> > That's roughly a third of the FS track.
>
> FYI, I'm flying out at 4:15pm on Wednesday, and while I try to keep my
> time at the airport short I'd still be gone by 3:30.

I've penciled XFS BoF at 2:30

>
> But that will only matter if you make the BOF and actual BOF and not the
> usual televised crap that happens at LSFMM.
>

What happens in XFS BoF is entirely up to the session lead and attendees
to decide.

There is video in the room, if that is what you meant so that remote attendees
that could not make it in person can be included.

We did not hand out free virtual invites to anyone who asked to attend.
Those were sent very selectively.

Any session lead can request to opt-out from publishing the video of the
session publicly or to audit the video before it is published.
This was the same last year and this year this was explicitly mentioned
in the invitation:

"Please note: As with previous years there will be an A/V team on-
site in order to facilitate conferencing and help with virtual
participants. In order to leave room for off-the-record discussions
the storage track completely opts out of recordings. For all other
tracks, please coordinate with your track leads (mentioned below)
whether a session should explicitly opt-out. This can also be
coordinated on-site during or after the workshop. The track leads
then take care that the given session recording will not be
published."

I will take a note to keep XFS BoF off the record if that is what you
want and if the other xfs developers do not object.

Thanks,
Amir.

Re: [Lsf-pc] XFS BoF at LSFMM

[Ritesh Harjani]

Amir Goldstein <amir73il@gmail.com> writes:

> On Wed, May 8, 2024 at 10:42 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>>
>> How about an XFS BoF at LSFMM?
>
> Let me rephrase that.
>
> Darrick,
>
> Would you like to list some items for the XFS BoF so I can put it on
> the agenda and link to this thread?
>
>> Would it be good to separate the BoF
>> for XFS and bcachefs so that folks who want to attend both can do so?
>
> I have set a side 2.5 hours in the schedule on Wed afternoon for per-FS BoF
> 3 hours if you include the FS lightning slot afterwards that could be used
> as per-FS lightning updates.
>
>> How about Wednesday 15:30? That would allow a full hour for bcachefs.
>
> I have no doubt that people want to be updated about bcaches,
> but a full hour for bcachefs? IDK.
>
> FYI, I counted more than 10 attendees that are active contributors or
> have contributed to xfs in one way or another.
> That's roughly a third of the FS track.
>
> So let's see how much time XFS BoF needs and divide the rest of the
> time among other FS that want to do a BoF in the FS room.

Thanks that would be helpful. I would like to participate in XFS/iomap BoF
session and ext4 too (if that happens). So if we could avoid the
conflict between the two, that would be much appreciated.

>
> If anyone would like to run a *FS BoF in the FS room please let me know
> and please try to estimate how much time you will need.
> You can also "register" for *FS lightning talk, but we can also arrange this
> ad-hoc.
>
> Excited to see you all next week!

Same here, Thanks!

-ritesh

Re: [Lsf-pc] XFS BoF at LSFMM

[Darrick J. Wong]

[adds ritesh to cc]

On Thu, May 09, 2024 at 08:23:25AM +0300, Amir Goldstein wrote:
> On Thu, May 9, 2024 at 8:06 AM Christoph Hellwig <hch@infradead.org> wrote:
> >
> > On Thu, May 09, 2024 at 08:01:39AM +0300, Amir Goldstein wrote:
> > >
> > > FYI, I counted more than 10 attendees that are active contributors or
> > > have contributed to xfs in one way or another.
> > > That's roughly a third of the FS track.
> >
> > FYI, I'm flying out at 4:15pm on Wednesday, and while I try to keep my
> > time at the airport short I'd still be gone by 3:30.
> 
> I've penciled XFS BoF at 2:30

Ritesh and Ted and Jan and I were chatting during the ext4 concall just
now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
bof after that?  That would give us some time to chat with hch about
iomap (and xfs) direction before he has to leave.

Alternately, we could announce a lunchtime discussion group on Monday
following Ritesh's presentation about iomap.  That could fit everyone's
schedule better?  Also everyone's braincaches will likely be warmer.

> >
> > But that will only matter if you make the BOF and actual BOF and not the
> > usual televised crap that happens at LSFMM.
> >
> 
> What happens in XFS BoF is entirely up to the session lead and attendees
> to decide.
> 
> There is video in the room, if that is what you meant so that remote attendees
> that could not make it in person can be included.
> 
> We did not hand out free virtual invites to anyone who asked to attend.
> Those were sent very selectively.
> 
> Any session lead can request to opt-out from publishing the video of the
> session publicly or to audit the video before it is published.
> This was the same last year and this year this was explicitly mentioned
> in the invitation:
> 
> "Please note: As with previous years there will be an A/V team on-
> site in order to facilitate conferencing and help with virtual
> participants. In order to leave room for off-the-record discussions
> the storage track completely opts out of recordings. For all other
> tracks, please coordinate with your track leads (mentioned below)
> whether a session should explicitly opt-out. This can also be
> coordinated on-site during or after the workshop. The track leads
> then take care that the given session recording will not be
> published."
> 
> I will take a note to keep XFS BoF off the record if that is what you
> want and if the other xfs developers do not object.

Survey: How many people want to attend the xfs bof virtually?

I'd be ok with an OTR discussion with no video, though I reserve the
right to change my mind if the score becomes chair: 2 djwong: 0. :P

--D

> Thanks,
> Amir.
> 

Re: [Lsf-pc] XFS BoF at LSFMM

[Christoph Hellwig]

On Thu, May 09, 2024 at 08:55:28AM -0700, Darrick J. Wong wrote:
> Ritesh and Ted and Jan and I were chatting during the ext4 concall just
> now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
> bof after that?  That would give us some time to chat with hch about
> iomap (and xfs) direction before he has to leave.

Well, I probably need to leave at about 3pm if I don't want to push it
too much.

> Alternately, we could announce a lunchtime discussion group on Monday
> following Ritesh's presentation about iomap.  That could fit everyone's
> schedule better?  Also everyone's braincaches will likely be warmer.

Sounds way more useful.

Re: [Lsf-pc] XFS BoF at LSFMM

[Amir Goldstein]

On Thu, May 9, 2024 at 6:55 PM Darrick J. Wong <djwong@kernel.org> wrote:
>
> [adds ritesh to cc]
>
> On Thu, May 09, 2024 at 08:23:25AM +0300, Amir Goldstein wrote:
> > On Thu, May 9, 2024 at 8:06 AM Christoph Hellwig <hch@infradead.org> wrote:
> > >
> > > On Thu, May 09, 2024 at 08:01:39AM +0300, Amir Goldstein wrote:
> > > >
> > > > FYI, I counted more than 10 attendees that are active contributors or
> > > > have contributed to xfs in one way or another.
> > > > That's roughly a third of the FS track.
> > >
> > > FYI, I'm flying out at 4:15pm on Wednesday, and while I try to keep my
> > > time at the airport short I'd still be gone by 3:30.
> >
> > I've penciled XFS BoF at 2:30
>
> Ritesh and Ted and Jan and I were chatting during the ext4 concall just
> now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
> bof after that?  That would give us some time to chat with hch about
> iomap (and xfs) direction before he has to leave.
>
> Alternately, we could announce a lunchtime discussion group on Monday
> following Ritesh's presentation about iomap.  That could fit everyone's
> schedule better?  Also everyone's braincaches will likely be warmer.
>

Seems to me that there will be a wider interest in iomap BoF
Not sure what you mean by lunchtime discussion.
We can move Willy's GFP_NOFS talk to 15:30 and have the iomap BoF
after Ritesh's session.

> > >
> > > But that will only matter if you make the BOF and actual BOF and not the
> > > usual televised crap that happens at LSFMM.
> > >
> >
> > What happens in XFS BoF is entirely up to the session lead and attendees
> > to decide.
> >
> > There is video in the room, if that is what you meant so that remote attendees
> > that could not make it in person can be included.
> >
> > We did not hand out free virtual invites to anyone who asked to attend.
> > Those were sent very selectively.
> >
> > Any session lead can request to opt-out from publishing the video of the
> > session publicly or to audit the video before it is published.
> > This was the same last year and this year this was explicitly mentioned
> > in the invitation:
> >
> > "Please note: As with previous years there will be an A/V team on-
> > site in order to facilitate conferencing and help with virtual
> > participants. In order to leave room for off-the-record discussions
> > the storage track completely opts out of recordings. For all other
> > tracks, please coordinate with your track leads (mentioned below)
> > whether a session should explicitly opt-out. This can also be
> > coordinated on-site during or after the workshop. The track leads
> > then take care that the given session recording will not be
> > published."
> >
> > I will take a note to keep XFS BoF off the record if that is what you
> > want and if the other xfs developers do not object.
>
> Survey: How many people want to attend the xfs bof virtually?
>
> I'd be ok with an OTR discussion with no video, though I reserve the
> right to change my mind if the score becomes chair: 2 djwong: 0. :P

The middle ground is to allow video for the selective virtual attendees
(and trust them not to record and publish) and not publish the video
on LSMFF site.

Thanks,
Amir.

Re: [Lsf-pc] XFS BoF at LSFMM

[Darrick J. Wong]

On Thu, May 09, 2024 at 08:29:19PM +0300, Amir Goldstein wrote:
> On Thu, May 9, 2024 at 6:55 PM Darrick J. Wong <djwong@kernel.org> wrote:
> >
> > [adds ritesh to cc]
> >
> > On Thu, May 09, 2024 at 08:23:25AM +0300, Amir Goldstein wrote:
> > > On Thu, May 9, 2024 at 8:06 AM Christoph Hellwig <hch@infradead.org> wrote:
> > > >
> > > > On Thu, May 09, 2024 at 08:01:39AM +0300, Amir Goldstein wrote:
> > > > >
> > > > > FYI, I counted more than 10 attendees that are active contributors or
> > > > > have contributed to xfs in one way or another.
> > > > > That's roughly a third of the FS track.
> > > >
> > > > FYI, I'm flying out at 4:15pm on Wednesday, and while I try to keep my
> > > > time at the airport short I'd still be gone by 3:30.
> > >
> > > I've penciled XFS BoF at 2:30
> >
> > Ritesh and Ted and Jan and I were chatting during the ext4 concall just
> > now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
> > bof after that?  That would give us some time to chat with hch about
> > iomap (and xfs) direction before he has to leave.
> >
> > Alternately, we could announce a lunchtime discussion group on Monday
> > following Ritesh's presentation about iomap.  That could fit everyone's
> > schedule better?  Also everyone's braincaches will likely be warmer.
> >
> 
> Seems to me that there will be a wider interest in iomap BoF
> Not sure what you mean by lunchtime discussion.

Monday 90-minute lunch is posted as being in "Grand ballroom C", so I
would tell everyone to come find the table(s) I'm sitting at for a
discussion over lunch.  We can move out to a hallway after everyone's
done eating.

> We can move Willy's GFP_NOFS talk to 15:30 and have the iomap BoF
> after Ritesh's session.

<shrug> If you like, though I don't think it's totally necessary.  But
you might have a better idea of what the venue is like than I do, so
I'll let you make that call. :)

> > > >
> > > > But that will only matter if you make the BOF and actual BOF and not the
> > > > usual televised crap that happens at LSFMM.
> > > >
> > >
> > > What happens in XFS BoF is entirely up to the session lead and attendees
> > > to decide.
> > >
> > > There is video in the room, if that is what you meant so that remote attendees
> > > that could not make it in person can be included.
> > >
> > > We did not hand out free virtual invites to anyone who asked to attend.
> > > Those were sent very selectively.
> > >
> > > Any session lead can request to opt-out from publishing the video of the
> > > session publicly or to audit the video before it is published.
> > > This was the same last year and this year this was explicitly mentioned
> > > in the invitation:
> > >
> > > "Please note: As with previous years there will be an A/V team on-
> > > site in order to facilitate conferencing and help with virtual
> > > participants. In order to leave room for off-the-record discussions
> > > the storage track completely opts out of recordings. For all other
> > > tracks, please coordinate with your track leads (mentioned below)
> > > whether a session should explicitly opt-out. This can also be
> > > coordinated on-site during or after the workshop. The track leads
> > > then take care that the given session recording will not be
> > > published."
> > >
> > > I will take a note to keep XFS BoF off the record if that is what you
> > > want and if the other xfs developers do not object.
> >
> > Survey: How many people want to attend the xfs bof virtually?
> >
> > I'd be ok with an OTR discussion with no video, though I reserve the
> > right to change my mind if the score becomes chair: 2 djwong: 0. :P
> 
> The middle ground is to allow video for the selective virtual attendees
> (and trust them not to record and publish) and not publish the video
> on LSMFF site.

I didn't know that was also an option; I'll keep that in mind.

--D

> Thanks,
> Amir.
> 

Re: [Lsf-pc] XFS BoF at LSFMM

[Theodore Ts'o]

On Thu, May 09, 2024 at 10:47:45AM -0700, Darrick J. Wong wrote:
> > > Ritesh and Ted and Jan and I were chatting during the ext4 concall just
> > > now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
> > > bof after that?  That would give us some time to chat with hch about
> > > iomap (and xfs) direction before he has to leave.
> > >
> > > Alternately, we could announce a lunchtime discussion group on Monday
> > > following Ritesh's presentation about iomap.  That could fit everyone's
> > > schedule better?  Also everyone's braincaches will likely be warmer.
> > 
> > Seems to me that there will be a wider interest in iomap BoF
> > Not sure what you mean by lunchtime discussion.
> 
> Monday 90-minute lunch is posted as being in "Grand ballroom C", so I
> would tell everyone to come find the table(s) I'm sitting at for a
> discussion over lunch.  We can move out to a hallway after everyone's
> done eating.
> 
> > We can move Willy's GFP_NOFS talk to 15:30 and have the iomap BoF
> > after Ritesh's session.
> 
> <shrug> If you like, though I don't think it's totally necessary.  But
> you might have a better idea of what the venue is like than I do, so
> I'll let you make that call. :)

How did we decide to resolve this?  If we have the iomap BOF at
2:30pm, with a hard stop at 3:00pm, that would work for everyone
including Cristoph.

Or we can try to get all of the people interested in talking about
iomap at lunch on Wednesday.

Or we could do something that requires a lot more juggling of slots,
which I'll leave to the track leads.  :-)

All of these ideas work for me.


As far as scheduling per-file system BOF's, I'd like to claim the slot
immediately after XFS for an ext4 BOF, it that would be OK.

Thanks,

						- Ted

Re: [Lsf-pc] XFS BoF at LSFMM

[Darrick J. Wong]

On Mon, May 13, 2024 at 07:48:16AM -0600, Theodore Ts'o wrote:
> On Thu, May 09, 2024 at 10:47:45AM -0700, Darrick J. Wong wrote:
> > > > Ritesh and Ted and Jan and I were chatting during the ext4 concall just
> > > > now.  Could we have a 30 minute iomap bof at 2:30pm followed by the XFS
> > > > bof after that?  That would give us some time to chat with hch about
> > > > iomap (and xfs) direction before he has to leave.
> > > >
> > > > Alternately, we could announce a lunchtime discussion group on Monday
> > > > following Ritesh's presentation about iomap.  That could fit everyone's
> > > > schedule better?  Also everyone's braincaches will likely be warmer.
> > > 
> > > Seems to me that there will be a wider interest in iomap BoF
> > > Not sure what you mean by lunchtime discussion.
> > 
> > Monday 90-minute lunch is posted as being in "Grand ballroom C", so I
> > would tell everyone to come find the table(s) I'm sitting at for a
> > discussion over lunch.  We can move out to a hallway after everyone's
> > done eating.
> > 
> > > We can move Willy's GFP_NOFS talk to 15:30 and have the iomap BoF
> > > after Ritesh's session.
> > 
> > <shrug> If you like, though I don't think it's totally necessary.  But
> > you might have a better idea of what the venue is like than I do, so
> > I'll let you make that call. :)
> 
> How did we decide to resolve this?  If we have the iomap BOF at
> 2:30pm, with a hard stop at 3:00pm, that would work for everyone
> including Cristoph.
> 
> Or we can try to get all of the people interested in talking about
> iomap at lunch on Wednesday.

It's currently scheduled for 12:30 (Mountain Time, or GMT -0600) tomorrow the
14th.  Schedule still subject to change tho

https://docs.google.com/spreadsheets/d/176LXLys9Uh6A-Eal2flrzcbUSJMUXGkGwyihr9jAAaQ/edit#gid=0

--D

> Or we could do something that requires a lot more juggling of slots,
> which I'll leave to the track leads.  :-)
> 
> All of these ideas work for me.
> 
> 
> As far as scheduling per-file system BOF's, I'd like to claim the slot
> immediately after XFS for an ext4 BOF, it that would be OK.
> 
> Thanks,
> 
> 						- Ted