Skip to contentSkip to conversation

Maintainer's Summit 2017 Feedback Thread

Ksummit-discuss Theodore Ts'o 10 participants 14 messages Oct 29, 2017 – Oct 30, 2017

lore
Share & tools
lore
#0 lore raw

Hi,

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Thanks!!

- Ted

#1 lore raw
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Hello Ted,

Something that surprised me is although I'm reading Linus' replies to pull requests that appear on the LKML mailing list that during the maintainer summit I learned more about what Linus expects from maintainers than by reading Linus' e-mails. I think this shows that reading Linus' e-mails is not enough to learn what Linus expects from maintainers. This made me wonder how maintainers are informed about what is expected from them, especially new maintainers? Do we perhaps need a document in the kernel tree that explains what is expected from maintainers?

Thanks,

Bart.

#2 lore raw
Quoted 9 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Hello Ted,

Quoted 8 lines

Something that surprised me is although I'm reading Linus' replies to pull requests that appear on the LKML mailing list that during the maintainer summit I learned more about what Linus expects from maintainers than by reading Linus' e-mails. I think this shows that reading Linus' e-mails is not enough to learn what Linus expects from maintainers. This made me wonder how maintainers are informed about what is expected from them, especially new maintainers? Do we perhaps need a document in the kernel tree that explains what is expected from maintainers?

Yes, we do. Especially with the number of people that are sending pull requests directly to Linus. I am taking a look at putting a first draft of such a document together in the next week or so.

#3 lore raw
Quoted 10 lines · Dan Williams
Quoted 9 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Hello Ted,

Quoted 8 lines
Quoted 8 lines

Something that surprised me is although I'm reading Linus' replies to pull requests that appear on the LKML mailing list that during the maintainer summit I learned more about what Linus expects from maintainers than by reading Linus' e-mails. I think this shows that reading Linus' e-mails is not enough to learn what Linus expects from maintainers. This made me wonder how maintainers are informed about what is expected from them, especially new maintainers? Do we perhaps need a document in the kernel tree that explains what is expected from maintainers?

Is there a summary of what Linus said?

Quoted 3 lines

Yes, we do. Especially with the number of people that are sending pull requests directly to Linus. I am taking a look at putting a first draft of such a document together in the next week or so.

Much appreciated! Any chance it will be out before the next open window?

Mimi

#4 lore raw
In reply to Theodore Ts'o
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Something Linus was very clear about is that regressions are not acceptable. The best way I know to reduce the number of regressions is to increase the efforts on automatic testing. Should the number of tests that is run by the zero-day testing infrastructure be increased? Should more tests be added? Should the documentation that is available on
https://01.org/lkp/documentation/0-day-test-service perhaps be moved elsewhere, e.g. inside that repository? Do we need a link on kernel.org or in the kernel documentation to that project?

Thanks,

Bart.

#5 lore raw
Quoted 11 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Something Linus was very clear about is that regressions are not acceptable. The best way I know to reduce the number of regressions is to increase the efforts on automatic testing. Should the number of tests that is run by the zero-day testing infrastructure be increased?

The 0-day system has sent a whole series of reports in response to Linus's v4.14-rc6 announcement. So it sounds like Fengguang is already on it. :-)

At least, for those things where we have tests and where the 0-day hardware tickles the code paths of various arch-specific and drivers.

The good news is that we've picked a lot of the low-hanging fruit. The bad news is that a lot of what is left is hardware-specific driver issues, where fixing a bug for the current generation of devices might break some device from an older generation. I recall one change to the Intel Wireless which only regressed when talking to an Aruba Enterprise-grade Access Point, for which the Intel wireless driver folks did not have in their hardware test library.

I would hope that most maintainers are doing testing on their own subsystems. I certainly do a large amount of exhaustive testing of ext4 before I send a pull request (including most cases doing a trial merge before launching a gce-xfstests run). And I would assume (for example), that Damien does a large amount of test of his ZBC support code against SMR HDD's (bonus points if he tests SMR drives from both WDC and Seagate :-) before he does he submits patches for review, and as they get merged into Linus's tree.

More test in the zero-day testing infrastructure is good, but this can't be a substitute for maintainers testing their own subsystems. That's the only thing that can possibly scale.

- Ted

#6 lore raw
In reply to Bart Van Assche
Quoted 11 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Something Linus was very clear about is that regressions are not acceptable. The best way I know to reduce the number of regressions is to increase the efforts on automatic testing. Should the number of tests that is run by the zero-day testing infrastructure be increased? Should more tests be added?

Are there tests out there that 0-day does not currently run? If there are any out there that 0-day should run that it does not, you can always add new ones, the 0-day client code is on github...

Quoted 4 lines

Should the documentation that is available on
https://01.org/lkp/documentation/0-day-test-service perhaps be moved elsewhere, e.g. inside that repository? Do we need a link on kernel.org or in the kernel documentation to that project?

0-day is just one of the many different test systems for the kernel, should we add links to them all? :)

thanks,

greg k-h

#7 lore raw
Quoted 7 lines · Greg KH
Quoted 4 lines

Should the documentation that is available on
https://01.org/lkp/documentation/0-day-test-service perhaps be moved elsewhere, e.g. inside that repository? Do we need a link on kernel.org or in the kernel documentation to that project?

0-day is just one of the many different test systems for the kernel, should we add links to them all? :)

Worth pointing out that there's https://bottest.wiki.kernel.org/ that is probably a good place for this.

-K

#8 lore raw
Quoted 6 lines · Konstantin Ryabitsev
Quoted 2 lines · Greg KH
Quoted 2 lines

Should the documentation that is available on
https://01.org/lkp/documentation/0-day-test-service perhaps be

moved

Quoted 1 line
Quoted 1 line

elsewhere, e.g. inside that repository? Do we need a link on

kernel.org or

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

in the kernel documentation to that project?

0-day is just one of the many different test systems for the kernel, should we add links to them all? :)

Worth pointing out that there's https://bottest.wiki.kernel.org/ that is
probably a good place for this.

Actually,adding what runs automatically, what you have to request coverage for and what you have to run yourself would be useful info for the wiki.

James

Quoted 5 lines

-K
_______________________________________________
Ksummit-discuss mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/ksummit-discuss

Signature · 1 line
Sent from my Android device with K-9 Mail. Please excuse my brevity.
#9 lore raw
In reply to Greg KH
Quoted 8 lines · Greg KH
Quoted 7 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Quoted 8 lines
Quoted 4 lines

Something Linus was very clear about is that regressions are not acceptable. The best way I know to reduce the number of regressions is to increase the efforts on automatic testing. Should the number of tests that is run by the zero-day testing infrastructure be increased? Should more tests be added?

Are there tests out there that 0-day does not currently run? If there are any out there that 0-day should run that it does not, you can always add new ones, the 0-day client code is on github...

0-day is the only test infrastructure of which I know that it is run automatically for the Linux kernel. How should a developer decide which project to add new tests to: 0-day, ltp (Linux Test Project), ktest, xfstests, blktests or any other Linux kernel testing project that I am not aware of?

Thanks,

Bart.

#10 lore raw
Quoted 14 lines · Bart Van Assche
Quoted 13 lines · Greg KH
Quoted 11 lines · Bart Van Assche
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

Something Linus was very clear about is that regressions are not acceptable. The best way I know to reduce the number of regressions is to increase the efforts on automatic testing. Should the number of tests that is run by the zero-day testing infrastructure be increased? Should more tests be added?

Quoted 6 lines
Quoted 3 lines

Are there tests out there that 0-day does not currently run? If there are any out there that 0-day should run that it does not, you can always add new ones, the 0-day client code is on github...

0-day is the only test infrastructure of which I know that it is run automatically for the Linux kernel.

There are many others, kernel.ci is one of them, the link Konstantin pointed to has a larger list.

Quoted 3 lines

How should a developer decide which project
to add new tests to: 0-day, ltp (Linux Test Project), ktest, xfstests, blktests or any other Linux kernel testing project that I am not aware of?

It depends on what type of test you are wanting to add. And you forgot kselftest :)

thanks,

greg k-h

#11 lore raw
Quoted 6 lines
Quoted 3 lines

How should a developer decide which project
to add new tests to: 0-day, ltp (Linux Test Project), ktest, xfstests, blktests or any other Linux kernel testing project that I am not aware of?

It depends on what type of test you are wanting to add. And you forgot kselftest :)

I think that 0-day takes tests from elsewhere. You don't add tests to it.

julia

#12 lore raw
In reply to Theodore Ts'o
Quoted 8 lines · Theodore Ts'o

Hi,

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

I find such a small group is suitable for discussions, too, as the atmosphere there was a kind of extended hallway sessions. Especially I felt it after the tables were gathered and the mic got off. A slightly more attendees would still work, too, though.

But the room assignment needs more consideration: not only pillars in the middle, the room was too large for a small group, and the constant noise for preparing the lunch was a bit annoying.

With my distro maintainer hat on, I believe that it would have been better if we had an open session for things like "bash distro kernels (or vice versa)" before the maintainers summit. The "regression tracking" and "bash kernel maintainers" topics went well in the maintainers summit since we had sorted out the focused points beforehand. Otherwise the discussion will be blurred.

thanks,

Takashi

#13 lore raw
In reply to Theodore Ts'o
Quoted 5 lines · Theodore Ts'o

Please reply to this thread if you have any comments about how we can organize the Maintainer's Summit for next year. Given that Linus seemed fairly happy with how things went, it's likely we will stick with the same format for next year, but if there are any details about how we could do things better, I'd greatly appreciate them.

I think the format generally proved to make sense; it really helped the conversation to flow in an informal but still productive fashion.

The two downsides from my personal POV:

- room choice; I guess that's pretty obvious and not really worth too much discussion (especially the pillars and the constant noise being generated by the staff in the background)

- I can't help it, but the overall number of attendees seemed to be a little bit *too* small. I'm not asking for a revival of a 90+ people on the kernel summit (that absolutely doesn't fit the nature of discussions and the format), but perhaps 40 would make sense? There definitely were a few moments when I felt that the discussion came to a point where having one particular missing person would be helpful.

I guess that given this was the first time we've had this limited setup, we are still converging to the "best" group size?

Thanks,

Signature · 2 lines
Jiri Kosina
SUSE Labs