* Only one copy of bitbake should be run against a build directory
@ 2012-10-04 12:07 Phil Blundell
2012-10-04 12:27 ` Martin Jansa
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Phil Blundell @ 2012-10-04 12:07 UTC (permalink / raw)
To: openembedded-core
Since updating my copy of bitbake to one which does this extra locking,
I've come to realise that the constraint of having only one copy of
bitbake running is a bit of a nuisance when making use of devshells. I
used to quite often have one or two long-running devshells for packages
that I was actively working on, and then in parallel with that would use
bitbake to recompile other things. With the new locking mechanism, as
soon as I have a single devshell open I am now prohibited from using
bitbake for anything else in that same build directory.
Would it be reasonable to exempt devshells from that locking or is there
some compelling reason why they need to be serialised?
thanks
p.
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: Only one copy of bitbake should be run against a build directory
2012-10-04 12:07 Only one copy of bitbake should be run against a build directory Phil Blundell
@ 2012-10-04 12:27 ` Martin Jansa
2012-10-04 16:52 ` Khem Raj
2012-10-04 12:36 ` Morten Minde Neergaard
2012-10-04 12:53 ` Richard Purdie
2 siblings, 1 reply; 6+ messages in thread
From: Martin Jansa @ 2012-10-04 12:27 UTC (permalink / raw)
To: Phil Blundell; +Cc: openembedded-core
[-- Attachment #1: Type: text/plain, Size: 1152 bytes --]
On Thu, Oct 04, 2012 at 01:07:46PM +0100, Phil Blundell wrote:
> Since updating my copy of bitbake to one which does this extra locking,
> I've come to realise that the constraint of having only one copy of
> bitbake running is a bit of a nuisance when making use of devshells. I
> used to quite often have one or two long-running devshells for packages
> that I was actively working on, and then in parallel with that would use
> bitbake to recompile other things. With the new locking mechanism, as
> soon as I have a single devshell open I am now prohibited from using
> bitbake for anything else in that same build directory.
>
> Would it be reasonable to exempt devshells from that locking or is there
> some compelling reason why they need to be serialised?
The same does apply to bitbake-diffsigs now after IIRC this patch
http://git.openembedded.org/bitbake/commit/?id=cc70181659c07e04c205e17832846acf1ff31d28
before that I could use bitbake-diffsigs from any directory (not only
TOPDIR) and also when build in the same directory was still running.
Cheers,
--
Martin 'JaMa' Jansa jabber: Martin.Jansa@gmail.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Only one copy of bitbake should be run against a build directory
2012-10-04 12:27 ` Martin Jansa
@ 2012-10-04 16:52 ` Khem Raj
2012-10-04 22:15 ` Paul Eggleton
0 siblings, 1 reply; 6+ messages in thread
From: Khem Raj @ 2012-10-04 16:52 UTC (permalink / raw)
To: Martin Jansa; +Cc: Phil Blundell, openembedded-core
On Thu, Oct 4, 2012 at 5:27 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> On Thu, Oct 04, 2012 at 01:07:46PM +0100, Phil Blundell wrote:
>> Since updating my copy of bitbake to one which does this extra locking,
>> I've come to realise that the constraint of having only one copy of
>> bitbake running is a bit of a nuisance when making use of devshells. I
>> used to quite often have one or two long-running devshells for packages
>> that I was actively working on, and then in parallel with that would use
>> bitbake to recompile other things. With the new locking mechanism, as
>> soon as I have a single devshell open I am now prohibited from using
>> bitbake for anything else in that same build directory.
>>
>> Would it be reasonable to exempt devshells from that locking or is there
>> some compelling reason why they need to be serialised?
>
> The same does apply to bitbake-diffsigs now after IIRC this patch
> http://git.openembedded.org/bitbake/commit/?id=cc70181659c07e04c205e17832846acf1ff31d28
> before that I could use bitbake-diffsigs from any directory (not only
> TOPDIR) and also when build in the same directory was still running.
>
and also to use runqemu
> Cheers,
>
> --
> Martin 'JaMa' Jansa jabber: Martin.Jansa@gmail.com
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Only one copy of bitbake should be run against a build directory
2012-10-04 16:52 ` Khem Raj
@ 2012-10-04 22:15 ` Paul Eggleton
0 siblings, 0 replies; 6+ messages in thread
From: Paul Eggleton @ 2012-10-04 22:15 UTC (permalink / raw)
To: openembedded-core; +Cc: Martin Jansa
On Thursday 04 October 2012 09:52:45 Khem Raj wrote:
> On Thu, Oct 4, 2012 at 5:27 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > The same does apply to bitbake-diffsigs now after IIRC this patch
> > http://git.openembedded.org/bitbake/commit/?id=cc70181659c07e04c205e178328
> > 46acf1ff31d28 before that I could use bitbake-diffsigs from any directory
> > (not only TOPDIR) and also when build in the same directory was still
> > running.
>
> and also to use runqemu
Leaving the issue of devshell aside for a second, I agree that this is fairly
unnecessary for the utilities; if we can avoid the checks for those we should.
I'll look into this.
Cheers,
Paul
--
Paul Eggleton
Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Only one copy of bitbake should be run against a build directory
2012-10-04 12:07 Only one copy of bitbake should be run against a build directory Phil Blundell
2012-10-04 12:27 ` Martin Jansa
@ 2012-10-04 12:36 ` Morten Minde Neergaard
2012-10-04 12:53 ` Richard Purdie
2 siblings, 0 replies; 6+ messages in thread
From: Morten Minde Neergaard @ 2012-10-04 12:36 UTC (permalink / raw)
To: Phil Blundell; +Cc: openembedded-core
At 13:07, Thu 2012-10-04, Phil Blundell wrote:
> Since updating my copy of bitbake to one which does this extra locking,
> I've come to realise that the constraint of having only one copy of
> bitbake running is a bit of a nuisance when making use of devshells.
In Message-ID: <20120924074452.GH2254@kattastrofe.rd.tandberg.com>,
Subject: [OE-core] Customize devshell terminal command / environment, I
suggested the patch found in https://gist.github.com/3774736
This patch would allow you to customize the terminal used, including
spawning a terminal that detaches itself from the bitbake. So, my setup
is: patch plus bitbake running in a screen instance plus these settings:
OE_TERMINAL = "custom"
OE_TERMINAL_CUSTOMCMD = "screen"
OE_TERMINAL_EXPORTS += "STY"
(Should I submit this (non-obtrusive) patch somewhere? (-: )
Smiles,
--
Morten Minde Neergaard
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Only one copy of bitbake should be run against a build directory
2012-10-04 12:07 Only one copy of bitbake should be run against a build directory Phil Blundell
2012-10-04 12:27 ` Martin Jansa
2012-10-04 12:36 ` Morten Minde Neergaard
@ 2012-10-04 12:53 ` Richard Purdie
2 siblings, 0 replies; 6+ messages in thread
From: Richard Purdie @ 2012-10-04 12:53 UTC (permalink / raw)
To: Phil Blundell; +Cc: openembedded-core
On Thu, 2012-10-04 at 13:07 +0100, Phil Blundell wrote:
> Since updating my copy of bitbake to one which does this extra locking,
> I've come to realise that the constraint of having only one copy of
> bitbake running is a bit of a nuisance when making use of devshells. I
> used to quite often have one or two long-running devshells for packages
> that I was actively working on, and then in parallel with that would use
> bitbake to recompile other things. With the new locking mechanism, as
> soon as I have a single devshell open I am now prohibited from using
> bitbake for anything else in that same build directory.
>
> Would it be reasonable to exempt devshells from that locking or is there
> some compelling reason why they need to be serialised?
The reason it was added was that there were too many people shooting
themselves in the foot with multiple bitbake processes running without
them realising it. I has "saved" me a few times too.
I'm not sure how you'd allow devshell but not anything else,
particularly as there may be tasks that run before the devshell task
gets executed.
I'd be fine with a --ignore-the-lockfile-I-know-what-I-am-doing type
option, then on your own head be it ;-).
Cheers,
Richard
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-10-04 22:28 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-10-04 12:07 Only one copy of bitbake should be run against a build directory Phil Blundell
2012-10-04 12:27 ` Martin Jansa
2012-10-04 16:52 ` Khem Raj
2012-10-04 22:15 ` Paul Eggleton
2012-10-04 12:36 ` Morten Minde Neergaard
2012-10-04 12:53 ` Richard Purdie
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox