From: David Edelsohn <hidden> Date: 2000-08-04 15:18:28
quoted
quoted
quoted
quoted
Gabriel Paubert writes:
Gabriel> I plan to send a lot of small obvious bugfixes in the next 3-4 weeks, the
Gabriel> diffs between my tree and the official one now run in the megabyte range,
Gabriel> and this is too much (would the BitKeeper tree accept the VME patches and
Gabriel> PrePboot BTW ?)
Gabriel> What I would for one like to see is that people that write a patch that
Gabriel> may affect other machines than their own first publish it on this list and
Gabriel> wait a couple of days for comment and test results before pushing it to
Gabriel> the official source tree, somewhat like what happens on gcc-patches.
linuxppc-dev is not necessarily a patch submission mailinglist.
Until PowerPC Linux development has such a thing, I think that it is
better to coordinate with and directly send patches to someone with
write-access who can apply the patches instead of throwing them up in the
air and hoping that someone catches them.
David
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Kevin B. Hendricks <hidden> Date: 2000-08-04 15:28:42
Hi,
But if they are not cc'd to the linuxppc-dev list then we do need some sort of
CHANGELOG process going on. None of the rsync trees seem to do any versioning.
Possibly we should have our own kernel mailing lists for patches and things.
Kevin
David Edelsohn wrote:
quoted
quoted
quoted
quoted
quoted
Gabriel Paubert writes:
Gabriel> I plan to send a lot of small obvious bugfixes in the next 3-4 weeks, the
Gabriel> diffs between my tree and the official one now run in the megabyte range,
Gabriel> and this is too much (would the BitKeeper tree accept the VME patches and
Gabriel> PrePboot BTW ?)
Gabriel> What I would for one like to see is that people that write a patch that
Gabriel> may affect other machines than their own first publish it on this list and
Gabriel> wait a couple of days for comment and test results before pushing it to
Gabriel> the official source tree, somewhat like what happens on gcc-patches.
linuxppc-dev is not necessarily a patch submission mailinglist.
Until PowerPC Linux development has such a thing, I think that it is
better to coordinate with and directly send patches to someone with
write-access who can apply the patches instead of throwing them up in the
air and hoping that someone catches them.
David
Gabriel> I plan to send a lot of small obvious bugfixes in the next 3-4 weeks, the
Gabriel> diffs between my tree and the official one now run in the megabyte range,
Gabriel> and this is too much (would the BitKeeper tree accept the VME patches and
Gabriel> PrePboot BTW ?)
Gabriel> What I would for one like to see is that people that write a patch that
Gabriel> may affect other machines than their own first publish it on this list and
Gabriel> wait a couple of days for comment and test results before pushing it to
Gabriel> the official source tree, somewhat like what happens on gcc-patches.
I must confess this is one of those areas where I wish we were on CVS
instead of BK.
If you want feel free to send your patches my way and I'll be sure
that they get in. I can at least test things on my prep, chrp rs6000s
and even my power mac.
linuxppc-dev is not necessarily a patch submission mailinglist.
Until PowerPC Linux development has such a thing, I think that it is
better to coordinate with and directly send patches to someone with
write-access who can apply the patches instead of throwing them up in the
air and hoping that someone catches them.
This can be a good thing or a bad thing tho. I'd rather like to see
patches posted to the list. Granted there's some line there where they
simply get too big and a link is better. At least with posting to the
list everyone has the chance to comment and better yet test them out!
Regards,
Tom
David
--
Hakuna Matata,
Tom
-----------------------------------------------------------
Linux Guy "My heart is human, my blood is boiling,
tgall@uswest.net my brain IBM" -- Mr Roboto, Styxx
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Benjamin Herrenschmidt <hidden> Date: 2000-08-04 17:49:53
I must confess this is one of those areas where I wish we were on CVS
instead of BK.
If you want feel free to send your patches my way and I'll be sure
that they get in. I can at least test things on my prep, chrp rs6000s
and even my power mac.
quoted
linuxppc-dev is not necessarily a patch submission mailinglist.
Until PowerPC Linux development has such a thing, I think that it is
better to coordinate with and directly send patches to someone with
write-access who can apply the patches instead of throwing them up in the
air and hoping that someone catches them.
This can be a good thing or a bad thing tho. I'd rather like to see
patches posted to the list. Granted there's some line there where they
simply get too big and a link is better. At least with posting to the
list everyone has the chance to comment and better yet test them out!
I'm an horribly bad web designer, but one cool thing to do if someone is
good at it, would be a web-pased patch repository. You send patches with
a web form (eventually an upload mecanism) along with your email, a
comment, etc..., they get assigned a patch number automatically, anybody
can consult the database, and people with write access to BK can mark
them (accepted, refused, postponed, ... along with a comment) once they
have been handled.
Any volunteer to design such a site ?
Ben.
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
I'm an horribly bad web designer, but one cool thing to do if someone is
good at it, would be a web-pased patch repository. You send patches with
a web form (eventually an upload mecanism) along with your email, a
comment, etc..., they get assigned a patch number automatically, anybody
can consult the database, and people with write access to BK can mark
them (accepted, refused, postponed, ... along with a comment) once they
have been handled.
Any volunteer to design such a site ?
What happened to the BK mailing list? Is it still closed for non-write
access members of community? Or does it still exist?
If the BK list were open to people who want to keep track of things, it
can also serve as a patch bucket, since everybody who has the write access
is subscribed to the list automatically. Also, the archive of the list
can be used as CHANGELOG for rsync only people.
I'm not really against the web site idea, but it's just that there won't
be too many people submitting patches in the first place to warrent the
extra administrative burden on developers with write access to BK, I would
think.
Takashi Oe
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Hendricks, Kevin <hidden> Date: 2000-08-04 18:52:14
Hi,
I'm not really against the web site idea, but it's just that there won't
be too many people submitting patches in the first place to warrent the
extra administrative burden on developers with write access to BK, I would
think.
I don't agree. A separate website is important. I don't work on the kernel
directly but I have submitted patches here and there and have run into problems
when testing/debugging the JDK that are kernel issues.
I don't want write access to any tree, but I do need to see what patches are
going in, what changed, the status of my own patches, etc.
Also, I think we could easily modify any of the simple bugdatabase programs like
bugzilla to do exactly what we want (instead of calling them bugs, call them
patches, instead of fixed, call them accepted, etc) You could also create
categories so that driver patches are separate from other patches easily with
this type of software.
We are getting more and more people contributing to the kernel process
I think a simple modified bugtracking system linked to http://penguinppc.org
could be easily made to serve this purpose without having/needing a whole new
site.
My two cents:
Kevin
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
I'm not really against the web site idea, but it's just that there won't
be too many people submitting patches in the first place to warrent the
extra administrative burden on developers with write access to BK, I would
think.
I don't agree. A separate website is important. I don't work on the kernel
directly but I have submitted patches here and there and have run into problems
when testing/debugging the JDK that are kernel issues.
Ok, on a second thought, I feel a patch repository is a good idea for
users. I was originally not too excited about the idea because people who
actually make the final decisions like Cort or Paul haven't said anything,
and I'm not really positive that they can spend time on dealing with the
patches accumulated in the repository in timely manner. Unless we have
their blessing, the repository won't work for many of the patches, and
we'll have to default to current method of non-method. [Well, at least
Ben is willing, so I'm sure the repository will work for pmac specific
patches, actually.] However, the repository will be good for users who
compile kernels themselves or distribution makers. They can search for
fixes for specific problems they are facing, and we can point users to
look at the repository for patches.
I don't want write access to any tree, but I do need to see what patches are
going in, what changed, the status of my own patches, etc.
Also, I think we could easily modify any of the simple bugdatabase programs like
bugzilla to do exactly what we want (instead of calling them bugs, call them
patches, instead of fixed, call them accepted, etc) You could also create
categories so that driver patches are separate from other patches easily with
this type of software.
I agree. It should be fairly easy to set up.
We are getting more and more people contributing to the kernel process
I feel it's not many more than ten at any given time, if we only count
kernel hackers without the write access who submit patches occasionally or
regularly.
Takashi Oe
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Michael Schmitz <hidden> Date: 2000-08-04 20:10:47
quoted
I'm not really against the web site idea, but it's just that there won't
be too many people submitting patches in the first place to warrent the
extra administrative burden on developers with write access to BK, I would
think.
I don't agree. A separate website is important. I don't work on the kernel
directly but I have submitted patches here and there and have run into problems
when testing/debugging the JDK that are kernel issues.
I don't want write access to any tree, but I do need to see what patches are
going in, what changed, the status of my own patches, etc.
Same for me. So far I've been sending patches to Paul or BenH but some
list or web site as patch bucket would allow others to test that stuff.
Michael
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
From: Paul Mackerras <hidden> Date: 2000-08-05 05:42:01
Benjamin Herrenschmidt writes:
I'm an horribly bad web designer, but one cool thing to do if someone is
good at it, would be a web-pased patch repository. You send patches with
a web form (eventually an upload mecanism) along with your email, a
comment, etc..., they get assigned a patch number automatically, anybody
can consult the database, and people with write access to BK can mark
them (accepted, refused, postponed, ... along with a comment) once they
have been handled.
Sounds exactly like the jitterbug site that Andrew Tridgell set up for
the Linux kernel. Linus tried it for a while but unfortunately
eventually decided he didn't like it enough. But for the rest of us
it was great. Tridge even put in a "test patch" button for Linus that
would rsync the current kernel tree, automagically unpack/decompress
the patch, apply it and report any errors. I could ask Tridge to set
up something like this for us.
Paul.
--
Paul Mackerras, Senior Open Source Researcher, Linuxcare, Inc.
+61 2 6262 8990 tel, +61 2 6262 8991 fax
paulus@linuxcare.com.au, http://www.linuxcare.com.au/
Linuxcare. Support for the revolution.
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/
I'm an horribly bad web designer, but one cool thing to do if someone is
good at it, would be a web-pased patch repository. You send patches with
a web form (eventually an upload mecanism) along with your email, a
comment, etc..., they get assigned a patch number automatically, anybody
can consult the database, and people with write access to BK can mark
them (accepted, refused, postponed, ... along with a comment) once they
have been handled.
Any volunteer to design such a site ?
It already exists for Samba, and once it was in a testing period for the
Linux kernel (as maintained by Linus): Jitterbug.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/