master branch fast-forwarded to v3.7-rc1

master branch fast-forwarded to v3.7-rc1

[Ben Myers]

Hi XFS Folks,

Linux v3.7-rc1 is out.  The XFS master branch
(git://oss.sgi.com/xfs/xfs.git) has been fast-forwarded to v3.7-rc1.

We'd like to take a moment and let you all know that we at SGI are going
to make an effort to communicate more from our end.  Your contribution
is appreciated whether it be features, testing, reviews, docs,
refactoring, cheerleading, or otherwise.  We want you here, and we want
your help!  We'll continue to try to be as responsive as possible.

Please bear in mind that the master branch is a shared resource.  We XFS
developers need a stable and bug free environment to work in and test
our changes.  We simply cannot pull in work with known regressions
because that can stop _all_ development until the problems are resolved.

Much more importantly, our end users have every reason to expect release
quality software when they choose to run tagged releases from
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git.
People entrust XFS with their data, so the stakes are high.  Developers
and testers who run upstream kernels between releases also have every
reason to expect quality.  XFS must not bottleneck their development
efforts through regression.  We must all be committed to a consistently
high level of quality, and this comes before our own convenience as
developers.

When submitting work to the xfs@oss.sgi.com mailing list, please
remember to follow the rules described in your kernel sources under
Documentation/SubmittingPatches, and those in the intro of the
MAINTAINERS file.  In the MAINTAINERS file there is specific emphasis in
#1 on testing patches.  Please take this to heart.  SGI is committed to
testing patches before pulling them in.  And as developers, we each need
to thoroughly test our patches before posting.  Also, please don't get
discouraged if colleagues find bugs in your code by inspection or
through testing.  It happens sometimes.  We shouldn't rake each other
over the coals for it.

Again, thanks to each of you for your participation.  Please continue to
treat each other with respect in your reviews and try to remember that
everyone is doing their best with a complicated project.  Lets try to
keep a friendly and collegial atmosphere on the mailing list.  We should
all be able to have some fun while working on XFS!

Happy hacking!

-The XFS Team @ SGI

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1, and corp-speak mumble

[Eric Sandeen]

On 10/16/12 10:56 AM, Ben Myers wrote:
> Hi XFS Folks,
> 
> Linux v3.7-rc1 is out.  The XFS master branch
> (git://oss.sgi.com/xfs/xfs.git) has been fast-forwarded to v3.7-rc1.
> 
> We'd like to take a moment and let you all know that we at SGI are going
> to make an effort to communicate more from our end.  Your contribution
> is appreciated whether it be features, testing, reviews, docs,
> refactoring, cheerleading, or otherwise.  We want you here, and we want
> your help!  We'll continue to try to be as responsive as possible.
> 
> Please bear in mind that the master branch is a shared resource.  We XFS
> developers need a stable and bug free environment to work in and test
> our changes.  We simply cannot pull in work with known regressions
> because that can stop _all_ development until the problems are resolved.
> 
> Much more importantly, our end users have every reason to expect release
> quality software when they choose to run tagged releases from
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git.
> People entrust XFS with their data, so the stakes are high.  Developers
> and testers who run upstream kernels between releases also have every
> reason to expect quality.  XFS must not bottleneck their development
> efforts through regression.  We must all be committed to a consistently
> high level of quality, and this comes before our own convenience as
> developers.
> 
> When submitting work to the xfs@oss.sgi.com mailing list, please
> remember to follow the rules described in your kernel sources under
> Documentation/SubmittingPatches, and those in the intro of the
> MAINTAINERS file.  In the MAINTAINERS file there is specific emphasis in
> #1 on testing patches.  Please take this to heart.  SGI is committed to
> testing patches before pulling them in.  And as developers, we each need
> to thoroughly test our patches before posting.  Also, please don't get
> discouraged if colleagues find bugs in your code by inspection or
> through testing.  It happens sometimes.  We shouldn't rake each other
> over the coals for it.

It sounds like SGI perceives a problem in the development cycle, based
on all the nice, vague words about testing and collaboration and quality.

Can we lose the management-speak and state clearly what's perceived as
the problem to be solved here?  I'm reading between the lines as best
I can.

Dave had concerns that a regression, which, although quickly fixed, was
cited as the reason for missing a merge window.

This concerns me too, because it's not just SGI's timetables that matter
here; others are also depending on this work getting upstream within certain
deadlines as well.

Reading back through the list, I'm alarmed that SGI wants some unspecified
"soak time," but not upstream, for new work.  There's no better place than
an -rc1 to get soak & exposure for tested patches.  Bugs get found and fixed.
I don't think the XFS developer community needs a lecture on patch submission
processes and quality expectations.

If maintainers get too conservative all it's going to do is slow development,
and slow the discovery & fixes for bugs which will inevitably occur in the
course of active development.

The master branch is a shared *development* resource.  If SGI is treating it
as something which must never be broken under any circumstances, then I humbly
submit that you're doing it wrong.  It may date me, but I'll point out that
it's meant to be a bazaar, not a cathedral, and that release early, release
often is not a new idea.

If you need something solid and slow-moving, that's what forks and branches
and -stable trees are for.


-Eric


> Again, thanks to each of you for your participation.  Please continue to
> treat each other with respect in your reviews and try to remember that
> everyone is doing their best with a complicated project.  Lets try to
> keep a friendly and collegial atmosphere on the mailing list.  We should
> all be able to have some fun while working on XFS!
> 
> Happy hacking!
> 
> -The XFS Team @ SGI
> 
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
> 

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1

[Dave Chinner]

On Tue, Oct 16, 2012 at 10:56:40AM -0500, Ben Myers wrote:
> Hi XFS Folks,
> 
> Linux v3.7-rc1 is out.  The XFS master branch
> (git://oss.sgi.com/xfs/xfs.git) has been fast-forwarded to v3.7-rc1.
> 
> We'd like to take a moment and let you all know that we at SGI are going
> to make an effort to communicate more from our end.  Your contribution
> is appreciated whether it be features, testing, reviews, docs,
> refactoring, cheerleading, or otherwise.  We want you here, and we want
> your help!  We'll continue to try to be as responsive as possible.

"we at SGI".

Curious. Has the XFS maintainer been replaced with a corporate
plurality? :/

FWIW:

http://lwn.net/SubscriberLink/519919/01a4caa1a34465f5/

"Developing in the community has a lot of rewards, including the
fact that credit for the work stays with the developer rather than
accruing to the sponsoring company.

> Please bear in mind that the master branch is a shared resource.  We XFS
> developers need a stable and bug free environment to work in and test
> our changes.  We simply cannot pull in work with known regressions
> because that can stop _all_ development until the problems are resolved.

I disagreed when you first indicated this is what you wanted to do a
couple of weeks ago:

http://oss.sgi.com/archives/xfs/2012-10/msg00124.html

but you didn't respond. I still disagree with it.

FYI, I don't need a stable and bug free environment to work in - I
haven't had one for years. That's because finding and fixing bugs is
what I do every day. I create tens of XFS bugs a day and I fix that
many as well. If the dev tree has bugs in it, then I'll find and fix
those as I trip over them. And I don't care whose bug the code is in
- if it stops me from continuing my line of development then my job
is to immediately find and fix that bug.  This is what *everyone*
use the dev tree should be doing - we are all responsible for finding
and fixing bugs in the XFS tree.

Regardless, instituting new development policies by decree is not
the best way to win friends and influence people in an open
development environment. You cannot force people to agree to new
policies, even as the maintainer. Change is made via discussion and
concensus, but you haven't engaged in disussion of this subject at
all.

Indeed, the development history of Linux and OSS in general points
to this policy change being the precisely the wrong direction to
take. At minimum, the benefits to developer productivity need to be
demonstrated before such a change should be made. Show us why this
is the right direction to take - tell us how you came to this
decision and how it will improve my productivity....

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1, and corp-speak mumble

[Ben Myers]

Hey Eric,

On Wed, Oct 17, 2012 at 03:20:59PM -0500, Eric Sandeen wrote:
> On 10/16/12 10:56 AM, Ben Myers wrote:
> > Hi XFS Folks,
> > 
> > Linux v3.7-rc1 is out.  The XFS master branch
> > (git://oss.sgi.com/xfs/xfs.git) has been fast-forwarded to v3.7-rc1.
> > 
> > We'd like to take a moment and let you all know that we at SGI are going
> > to make an effort to communicate more from our end.  Your contribution
> > is appreciated whether it be features, testing, reviews, docs,
> > refactoring, cheerleading, or otherwise.  We want you here, and we want
> > your help!  We'll continue to try to be as responsive as possible.
> > 
> > Please bear in mind that the master branch is a shared resource.  We XFS
> > developers need a stable and bug free environment to work in and test
> > our changes.  We simply cannot pull in work with known regressions
> > because that can stop _all_ development until the problems are resolved.
> > 
> > Much more importantly, our end users have every reason to expect release
> > quality software when they choose to run tagged releases from
> > git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git.
> > People entrust XFS with their data, so the stakes are high.  Developers
> > and testers who run upstream kernels between releases also have every
> > reason to expect quality.  XFS must not bottleneck their development
> > efforts through regression.  We must all be committed to a consistently
> > high level of quality, and this comes before our own convenience as
> > developers.
> > 
> > When submitting work to the xfs@oss.sgi.com mailing list, please
> > remember to follow the rules described in your kernel sources under
> > Documentation/SubmittingPatches, and those in the intro of the
> > MAINTAINERS file.  In the MAINTAINERS file there is specific emphasis in
> > #1 on testing patches.  Please take this to heart.  SGI is committed to
> > testing patches before pulling them in.  And as developers, we each need
> > to thoroughly test our patches before posting.  Also, please don't get
> > discouraged if colleagues find bugs in your code by inspection or
> > through testing.  It happens sometimes.  We shouldn't rake each other
> > over the coals for it.
> 
> It sounds like SGI perceives a problem in the development cycle, based
> on all the nice, vague words about testing and collaboration and quality.
> 
> Can we lose the management-speak and state clearly what's perceived as
> the problem to be solved here?  I'm reading between the lines as best
> I can.

Ew.  Management-speak.  What were we thinking?  ;)

> Dave had concerns that a regression, which, although quickly fixed, was
> cited as the reason for missing a merge window.
>
> This concerns me too, because it's not just SGI's timetables that matter
> here; others are also depending on this work getting upstream within certain
> deadlines as well.
>
> Reading back through the list, I'm alarmed that SGI wants some unspecified
> "soak time," but not upstream, for new work.  There's no better place than an
> -rc1 to get soak & exposure for tested patches.  Bugs get found and fixed.

I think you are referring to my comment in this message:
http://oss.sgi.com/archives/xfs/2012-10/msg00122.html

My soak time comment was with regard to a specific patch set that had a history
of problems.  My comment was not meant to indicate that SGI plans to soak every
patch before we pull it in.  We do run the regression test suite on new patches
before pulling them in though.

> I don't think the XFS developer community needs a lecture on patch submission
> processes and quality expectations.

Sorry it came off that way.  Maybe not everyone else sees it, but we have had a
few situations recently where a patch was posted which clearly hadn't been
adequately tested by the developer.  E.g. something crashes right away, hits an
assert, or whatever.  It's not a huge deal, but that's why we wanted to remind
everyone to test before posting.  I hope that most everyone is not running into
those kinds of problems because they are usually caught before commit time.

> If maintainers get too conservative all it's going to do is slow development,
> and slow the discovery & fixes for bugs which will inevitably occur in the
> course of active development.

Yep, I understand.  I don't want to slow development either.  I do want to try
to get the relevent content in before -rc7 and play by the rules with the
upstream tree.  And, I understand that bugs slip through.
 
> The master branch is a shared *development* resource.  If SGI is treating it
> as something which must never be broken under any circumstances, then I
> humbly submit that you're doing it wrong.  It may date me, but I'll point out
> that it's meant to be a bazaar, not a cathedral, and that release early,
> release often is not a new idea.

I think we're treating the master branch correctly.  If a patch is 1)
adequately reviewed and 2) doesn't regress something, we're good to go.  I'll
grant you the statement "We simply cannot pull in work with known regressions"
was a bit strong.  There are probably some good reasons to pull in known broken
code, e.g. the new breakage is preferable to the old, but usually it should be
fixed first.

> If you need something solid and slow-moving, that's what forks and branches
> and -stable trees are for.

I take your point but I think we also have to be careful not to break stuff
that will affect a lot of people.  When we break the master branch it affects
maybe tens of people.  I really don't know how many people use the -rc releases
upstream.  So here's Ben at v3.6, considering pulling in a broken patchset with
a challenging history, along with a one-day-old fix, without much testing, for
something that is already fixed in 3.6, in the middle of the merge window,
shortly after being told not to do that, when a mistake will affect a lot of
people.  So I made the call.  But it's only for a specific patch set.  It does
not represent a problem or a change in the development cycle, except to say
that we'd like to try and get stuff in by -rc7 next time so we can play by the
rules, to remind devs to test before posting, and to explain why we have to
push back sometimes.

Does that make things any clearer?  Make it worse?

Regards,
Ben

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1, and corp-speak mumble

[Dave Chinner]

On Thu, Oct 18, 2012 at 05:28:38PM -0500, Ben Myers wrote:
> On Wed, Oct 17, 2012 at 03:20:59PM -0500, Eric Sandeen wrote:
> > Dave had concerns that a regression, which, although quickly fixed, was
> > cited as the reason for missing a merge window.
> >
> > This concerns me too, because it's not just SGI's timetables that matter
> > here; others are also depending on this work getting upstream within certain
> > deadlines as well.
> >
> > Reading back through the list, I'm alarmed that SGI wants some unspecified
> > "soak time," but not upstream, for new work.  There's no better place than an
> > -rc1 to get soak & exposure for tested patches.  Bugs get found and fixed.
> 
> I think you are referring to my comment in this message:
> http://oss.sgi.com/archives/xfs/2012-10/msg00122.html
> 
> My soak time comment was with regard to a specific patch set that had a history
> of problems.  My comment was not meant to indicate that SGI plans to soak every
> patch before we pull it in.  We do run the regression test suite on new patches
> before pulling them in though.

I think the misunderstanding stems from the approach taken to
resolve this particular case combined it with abstract references to
stabilty and avoiding all regressions in the dev tree.  We need to
keep communication clear and direct...

> > I don't think the XFS developer community needs a lecture on patch submission
> > processes and quality expectations.
> 
> Sorry it came off that way.  Maybe not everyone else sees it, but we have had a
> few situations recently where a patch was posted which clearly hadn't been
> adequately tested by the developer.  E.g. something crashes right away, hits an
> assert, or whatever.

This is something that the review process is supposed to cover. If
you don't think a patch set is adequately tested, then that should
be part of your review feedback.

i.e. timely communication is very helpful here, so we all should be
trying to review code within a couple of days of it being posted and
not leaving it for a week or two to rot before looking at it....

> It's not a huge deal, but that's why we wanted to remind
> everyone to test before posting.  I hope that most everyone is not running into
> those kinds of problems because they are usually caught before commit time.

Developers do test their code - they test them to the best of their
ability and available hardware. That doesn't mean the code is
perfect or bug free - we know from years of experience that a change
that passes xfstests on several developers machines might cause a
repeated ASSERT failure on one specific machine with a specific
configuration somewhere else.

Such a problem is not the devloper's fault, though, and keeping the
code out of the dev tree because of such a problem is a harsh
penalty. And it's apenalty that just places more burden on
developers and reviewers.

i.e. Instead of having to just review/track a single patch for the
regression, the entire patch set has to be kept under management and
reposted and updated over and over again while the regression is
triaged and fixes are tested.  Reviewers have to look at the whole
patch series to see what they need to re-review, it needs to be
retested, etc. It's messy and time consuming and reduces developer
productivity because of the additional overhead it imposes on them.
Dealing with such problems post-integration is much more efficient
for everyone.

> > If maintainers get too conservative all it's going to do is slow development,
> > and slow the discovery & fixes for bugs which will inevitably occur in the
> > course of active development.
> 
> Yep, I understand.  I don't want to slow development either.  I do want to try
> to get the relevent content in before -rc7 and play by the rules with the
> upstream tree.  And, I understand that bugs slip through.

Then, please, say exactly that.  Tell us what your merge cycle plan
is so we can organise our work flows around that.  It'll make
everyone's life easier if we know what targets we need to work
towards.  I know I sound like a broken record, but it's that
communication thing again...  :/

> > The master branch is a shared *development* resource.  If SGI is treating it
> > as something which must never be broken under any circumstances, then I
> > humbly submit that you're doing it wrong.  It may date me, but I'll point out
> > that it's meant to be a bazaar, not a cathedral, and that release early,
> > release often is not a new idea.
> 
> I think we're treating the master branch correctly.  If a patch is 1)
> adequately reviewed and 2) doesn't regress something, we're good to go.  I'll
> grant you the statement "We simply cannot pull in work with known regressions"
> was a bit strong.  There are probably some good reasons to pull in known broken
> code, e.g. the new breakage is preferable to the old, but usually it should be
> fixed first.

Sure. But the crux of the concern is that "2) doesn't regress
something" is new merge criteria because "something" has been
redefined to have a much wider scope.

What it comes down to is that minor or isolated regressions in the
dev tree are an acceptible trade-off for getting code into the dev
tree quickly and efficiently. Getting the code into the tree fast
has much wider benefits to the dev community than ensuring that
merged code does not have any regressions at all.

> > If you need something solid and slow-moving, that's what forks and branches
> > and -stable trees are for.
> 
> I take your point but I think we also have to be careful not to break stuff
> that will affect a lot of people.  When we break the master branch it affects
> maybe tens of people.

Development trees are for developers. Development trees, by
definition, are unstable. Developers understand this, and so expect
to find regressions and breakages in the developement tree.

> I really don't know how many people use the -rc releases
> upstream.

Tens of thousands. Maybe even hundreds of thousands of developer and
test machines are running -rc kernels. Machines that expect -rc
kernels to have problems and are running -rc kernels to find those
problems and get them fixed.  Again, regressions in -rc1 are not the
end of the world.

> So here's Ben at v3.6, considering pulling in a broken patchset with
> a challenging history, along with a one-day-old fix, without much testing, for
> something that is already fixed in 3.6, in the middle of the merge window,
> shortly after being told not to do that, when a mistake will affect a lot of
> people.

I'm not going get into a he-said, she-said argument over this,
because....

> Does that make things any clearer?  Make it worse?

... the problem was very clear: a lack of communication.

(<smash> Oh, look, another broken record. :)

In the above case, an early "I don't think this is ready for merge"
statement would have resulted in discussing the issue immediately
and deciding the way to progress. No surprises, and everyone ends up
happy with the outcome because they know exactly what is going on.

The maintainer role is not just being a code monkey - that's the
easy part. You also have to be a release manager, a front line
support engineer, a QA manager and - last but not least - a very
good developer.  All of these tasks require close communication with
affected parties, so the maintainer must also be able to communicate
quickly and effectively.

Put simply, the #1 job of the maintainer is communication and
co-ordination. If the maintainer is not communicating and/or
co-ordinating well then everything else breaks down.

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1, and corp-speak mumble

[Ben Myers]

Hi Dave,

On Fri, Oct 19, 2012 at 04:08:50PM +1100, Dave Chinner wrote:
> On Thu, Oct 18, 2012 at 05:28:38PM -0500, Ben Myers wrote:
> > On Wed, Oct 17, 2012 at 03:20:59PM -0500, Eric Sandeen wrote:
> > > Dave had concerns that a regression, which, although quickly fixed, was
> > > cited as the reason for missing a merge window.
> > >
> > > This concerns me too, because it's not just SGI's timetables that matter
> > > here; others are also depending on this work getting upstream within certain
> > > deadlines as well.
> > >
> > > Reading back through the list, I'm alarmed that SGI wants some unspecified
> > > "soak time," but not upstream, for new work.  There's no better place than an
> > > -rc1 to get soak & exposure for tested patches.  Bugs get found and fixed.
> > 
> > I think you are referring to my comment in this message:
> > http://oss.sgi.com/archives/xfs/2012-10/msg00122.html
> > 
> > My soak time comment was with regard to a specific patch set that had a history
> > of problems.  My comment was not meant to indicate that SGI plans to soak every
> > patch before we pull it in.  We do run the regression test suite on new patches
> > before pulling them in though.
> 
> I think the misunderstanding stems from the approach taken to
> resolve this particular case combined it with abstract references to
> stabilty and avoiding all regressions in the dev tree.  We need to
> keep communication clear and direct...
> 
> > > I don't think the XFS developer community needs a lecture on patch submission
> > > processes and quality expectations.
> > 
> > Sorry it came off that way.  Maybe not everyone else sees it, but we have had a
> > few situations recently where a patch was posted which clearly hadn't been
> > adequately tested by the developer.  E.g. something crashes right away, hits an
> > assert, or whatever.
> 
> This is something that the review process is supposed to cover. If
> you don't think a patch set is adequately tested, then that should
> be part of your review feedback.
> 
> i.e. timely communication is very helpful here, so we all should be
> trying to review code within a couple of days of it being posted and
> not leaving it for a week or two to rot before looking at it....
> 
> > It's not a huge deal, but that's why we wanted to remind
> > everyone to test before posting.  I hope that most everyone is not running into
> > those kinds of problems because they are usually caught before commit time.
> 
> Developers do test their code - they test them to the best of their
> ability and available hardware. That doesn't mean the code is
> perfect or bug free - we know from years of experience that a change
> that passes xfstests on several developers machines might cause a
> repeated ASSERT failure on one specific machine with a specific
> configuration somewhere else.
> 
> Such a problem is not the devloper's fault, though, and keeping the
> code out of the dev tree because of such a problem is a harsh
> penalty. And it's apenalty that just places more burden on
> developers and reviewers.
> 
> i.e. Instead of having to just review/track a single patch for the
> regression, the entire patch set has to be kept under management and
> reposted and updated over and over again while the regression is
> triaged and fixes are tested.  Reviewers have to look at the whole
> patch series to see what they need to re-review, it needs to be
> retested, etc. It's messy and time consuming and reduces developer
> productivity because of the additional overhead it imposes on them.
> Dealing with such problems post-integration is much more efficient
> for everyone.
> 
> > > If maintainers get too conservative all it's going to do is slow development,
> > > and slow the discovery & fixes for bugs which will inevitably occur in the
> > > course of active development.
> > 
> > Yep, I understand.  I don't want to slow development either.  I do want to try
> > to get the relevent content in before -rc7 and play by the rules with the
> > upstream tree.  And, I understand that bugs slip through.
> 
> Then, please, say exactly that.  Tell us what your merge cycle plan
> is so we can organise our work flows around that.  It'll make
> everyone's life easier if we know what targets we need to work
> towards.  I know I sound like a broken record, but it's that
> communication thing again...  :/
> 
> > > The master branch is a shared *development* resource.  If SGI is treating it
> > > as something which must never be broken under any circumstances, then I
> > > humbly submit that you're doing it wrong.  It may date me, but I'll point out
> > > that it's meant to be a bazaar, not a cathedral, and that release early,
> > > release often is not a new idea.
> > 
> > I think we're treating the master branch correctly.  If a patch is 1)
> > adequately reviewed and 2) doesn't regress something, we're good to go.  I'll
> > grant you the statement "We simply cannot pull in work with known regressions"
> > was a bit strong.  There are probably some good reasons to pull in known broken
> > code, e.g. the new breakage is preferable to the old, but usually it should be
> > fixed first.
> 
> Sure. But the crux of the concern is that "2) doesn't regress
> something" is new merge criteria because "something" has been
> redefined to have a much wider scope.
> 
> What it comes down to is that minor or isolated regressions in the
> dev tree are an acceptible trade-off for getting code into the dev
> tree quickly and efficiently. Getting the code into the tree fast
> has much wider benefits to the dev community than ensuring that
> merged code does not have any regressions at all.
> 
> > > If you need something solid and slow-moving, that's what forks and branches
> > > and -stable trees are for.
> > 
> > I take your point but I think we also have to be careful not to break stuff
> > that will affect a lot of people.  When we break the master branch it affects
> > maybe tens of people.
> 
> Development trees are for developers. Development trees, by
> definition, are unstable. Developers understand this, and so expect
> to find regressions and breakages in the developement tree.
> 
> > I really don't know how many people use the -rc releases
> > upstream.
> 
> Tens of thousands. Maybe even hundreds of thousands of developer and
> test machines are running -rc kernels. Machines that expect -rc
> kernels to have problems and are running -rc kernels to find those
> problems and get them fixed.  Again, regressions in -rc1 are not the
> end of the world.
> 
> > So here's Ben at v3.6, considering pulling in a broken patchset with
> > a challenging history, along with a one-day-old fix, without much testing, for
> > something that is already fixed in 3.6, in the middle of the merge window,
> > shortly after being told not to do that, when a mistake will affect a lot of
> > people.
> 
> I'm not going get into a he-said, she-said argument over this,
> because....
> 
> > Does that make things any clearer?  Make it worse?
> 
> ... the problem was very clear: a lack of communication.
> 
> (<smash> Oh, look, another broken record. :)
> 
> In the above case, an early "I don't think this is ready for merge"
> statement would have resulted in discussing the issue immediately
> and deciding the way to progress. No surprises, and everyone ends up
> happy with the outcome because they know exactly what is going on.
> 
> The maintainer role is not just being a code monkey - that's the
> easy part. You also have to be a release manager, a front line
> support engineer, a QA manager and - last but not least - a very
> good developer.  All of these tasks require close communication with
> affected parties, so the maintainer must also be able to communicate
> quickly and effectively.
> 
> Put simply, the #1 job of the maintainer is communication and
> co-ordination. If the maintainer is not communicating and/or
> co-ordinating well then everything else breaks down.

Thanks for the talk, coach.  ;)

Well everyone seems to agree that communication is the #1 problem.  So for my
part, that is what I'll be working on.  I'll ask the rest of the SGI XFS geeks
to do so as well.  Other contributors need to do the same, and to remember that
it cuts both ways:  I have a limited amount of time that I can spend running my
yap on this list while still getting some (marginally) useful technical work
done.  I am determined to spend it as best I can.

Gotta go.  Maybe I can pull in some code tonight! 8P

Thanks Dave,
-Ben

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: master branch fast-forwarded to v3.7-rc1, and corp-speak mumble

[Christoph Hellwig]

On Wed, Oct 17, 2012 at 03:20:59PM -0500, Eric Sandeen wrote:
> Dave had concerns that a regression, which, although quickly fixed, was
> cited as the reason for missing a merge window.
> 
> This concerns me too, because it's not just SGI's timetables that matter
> here; others are also depending on this work getting upstream within certain
> deadlines as well.
> 
> Reading back through the list, I'm alarmed that SGI wants some unspecified
> "soak time," but not upstream, for new work.  There's no better place than
> an -rc1 to get soak & exposure for tested patches.  Bugs get found and fixed.
> I don't think the XFS developer community needs a lecture on patch submission
> processes and quality expectations.

The best place is the for-next branch.  We should aim for getting
patches in early in the window rather than last minute, which is way to
common in XFS land.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs