buildbot for kvm.git

DISCUSSION Avi Kivity 4 participants 22 messages Nov 10, 2010 – Nov 30, 2010 KVM

Share & tools

Avi Kivity

author

2010-11-10 13:31 UTC #0 lore raw

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.

I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:

x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
s390: defconfig + KVM=m
ia64: defconfig + KVM=m KVM_INTEL=m
powerpc: ppc4x_defconfig + VIRTUALIZATION=y KVM=y KVM_440=y KVM_E500=n
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m
VIRTIO_BLK=m VIRTIO_NET=m
VIRTIO_CONSOLE=m
powerpc: ppc64_defconfig VIRTUALIZATION=y KVM=y KVM_BOOK3S_64=m'
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m VIRTIO_BLK=m VIRTIO_NET=m VIRTIO_CONSOLE=m

The recipe for a build is:
- check out the source
- MAKE defconfig
where MAKE="make ARCH=$ARCH
CROSS_COMPILE=/opt/gcc-4.5.0-nolibc/$ARCH-linux/bin/$ARCH-linux-'
- append the extra options to .config
- MAKE oldconfig
- MAKE

Can you help with this?

Signature · 1 line

error compiling committee.c: too many arguments to function

Avi Kivity

author

2010-11-10 13:33 UTC #1 lore raw

Quoted 3 lines · Avi Kivity

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
I've now a virtual machine with a cross compile environment for s390, ppc, and ia64.

Note: the cross compile environments are from

http://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.0/

in case anyone's interested.

Signature · 1 line

error compiling committee.c: too many arguments to function

Alexander Graf

2010-11-10 15:02 UTC #2 lore raw

In reply to Avi Kivity

Quoted 11 lines · Avi Kivity

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
s390: defconfig + KVM=m
ia64: defconfig + KVM=m KVM_INTEL=m
powerpc: ppc4x_defconfig + VIRTUALIZATION=y KVM=y KVM_440=y KVM_E500=n
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m VIRTIO_BLK=m VIRTIO_NET=m VIRTIO_CONSOLE=m

Could we please also add an e500v2 build here? It compiles different MMU code.

Btw, thanks a lot for doing this!

Alex

Avi Kivity

author

2010-11-10 15:03 UTC #3 lore raw

Quoted 14 lines · Alexander Graf

Quoted 11 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
s390: defconfig + KVM=m
ia64: defconfig + KVM=m KVM_INTEL=m
powerpc: ppc4x_defconfig + VIRTUALIZATION=y KVM=y KVM_440=y KVM_E500=n
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m VIRTIO_BLK=m VIRTIO_NET=m VIRTIO_CONSOLE=m

Quoted 1 line

Could we please also add an e500v2 build here? It compiles different MMU code.

Please post the relevant CONFIG_ settings.

Signature · 1 line

error compiling committee.c: too many arguments to function

Alexander Graf

2010-11-10 15:41 UTC #4 lore raw

Quoted 15 lines · Avi Kivity

Quoted 14 lines · Alexander Graf
Quoted 11 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
s390: defconfig + KVM=m
ia64: defconfig + KVM=m KVM_INTEL=m
powerpc: ppc4x_defconfig + VIRTUALIZATION=y KVM=y KVM_440=y KVM_E500=n
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m VIRTIO_BLK=m VIRTIO_NET=m VIRTIO_CONSOLE=m

Quoted 3 lines

Quoted 1 line
Could we please also add an e500v2 build here? It compiles different MMU code.
Please post the relevant CONFIG_ settings.

Ok, will do as soon as I get into physical contact with my e500 box again. That will hopefully be by tomorrow :).

Alex

Daniel Gollub

2010-11-11 09:32 UTC #5 lore raw

In reply to Avi Kivity

Quoted 3 lines · Avi Kivity

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.

[...]

Quoted 1 line

Can you help with this?

Sure. I'll look into that next week.

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Avi Kivity

author

2010-11-22 09:37 UTC #6 lore raw

Quoted 9 lines · Daniel Gollub

Quoted 3 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
[...]
Quoted 1 line
Can you help with this?
Sure. I'll look into that next week.

Daniel, any news about this?

Signature · 1 line

error compiling committee.c: too many arguments to function

Daniel Gollub

2010-11-23 00:46 UTC #7 lore raw

Quoted 6 lines · Avi Kivity

Quoted 5 lines · Daniel Gollub
Quoted 3 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.

Quoted 7 lines

Quoted 5 lines
[...]
Quoted 1 line
Can you help with this?
Sure. I'll look into that next week.
Daniel, any news about this?

Currently I'm applying your recipe on the buildmaster configuration.

Beside that, buildmaster and a small x86_64 buildslave got setup and is available on:
http://buildbot.b1-systems.de/kvm/

Once I'm done with the buildmaster configuration (and some more testing) kvm.git continuous build testing could be ready within the next days. (I'm travelling right now, but shouldn't block me to get this done)

If you like you can already setup the git post-receive hook in the kvm.git repo to trigger the buildmaster.

Like for qemu-kvm.git you need to copy git_buildbot.py (preferably a copy of the one which is used for qemu-kvm.git) and change the master-port to 9991 ( master = "....:9991" )

In hooks/post-receive you add:
/path/to/git_buildbot.py $1 $2 $3

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Avi Kivity

author

2010-11-23 10:22 UTC #8 lore raw

Quoted 28 lines · Daniel Gollub

Quoted 6 lines · Avi Kivity
Quoted 5 lines · Daniel Gollub
Quoted 3 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
Quoted 7 lines
Quoted 5 lines
[...]
Quoted 1 line
Can you help with this?
Sure. I'll look into that next week.
Daniel, any news about this?
Currently I'm applying your recipe on the buildmaster configuration.
Beside that, buildmaster and a small x86_64 buildslave got setup and is available on:
http://buildbot.b1-systems.de/kvm/
Once I'm done with the buildmaster configuration (and some more testing) kvm.git continuous build testing could be ready within the next days. (I'm travelling right now, but shouldn't block me to get this done)
If you like you can already setup the git post-receive hook in the kvm.git repo to trigger the buildmaster.

Quoted 7 lines

Like for qemu-kvm.git you need to copy git_buildbot.py (preferably a copy of the one which is used for qemu-kvm.git) and change the master-port to 9991 ( master = "....:9991" )
In hooks/post-receive you add:
/path/to/git_buildbot.py $1 $2 $3

Thanks, done.

Will you set up crossbuilders for ppc/ia64/s390, or will I contribute a builder?

Signature · 1 line

error compiling committee.c: too many arguments to function

Daniel Gollub

2010-11-24 10:15 UTC #10 lore raw

In reply to Avi Kivity

Quoted 3 lines · Avi Kivity

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.

First test builds succeed for x86 and ia64:
http://buildbot.b1-systems.de/kvm/grid

with gcc-4.5.0-nolibc toolchain from
http://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.0/

powerpc not ready yet - trivial $ARCH change required in the buildbot master configuration.

s390 compilation seems to run fine and hit the first compilation issue:
http://buildbot.b1-systems.de/kvm/builders/s390/builds/0/steps/compile/logs/stdio

Currently buildbot notification mails don't get send to the kvm list. I would like to do some more testing before spamming with wrong-testing results.

Some more questions inline:

Quoted 6 lines

I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m

Without virtio? intentionally?

Quoted 11 lines

s390: defconfig + KVM=m
ia64: defconfig + KVM=m KVM_INTEL=m
powerpc: ppc4x_defconfig + VIRTUALIZATION=y KVM=y KVM_440=y KVM_E500=n
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m
VIRTIO_BLK=m VIRTIO_NET=m
VIRTIO_CONSOLE=m
powerpc: ppc64_defconfig VIRTUALIZATION=y KVM=y KVM_BOOK3S_64=m'
KVM_EXIT_TIMING=n VIRTIO_PCI=m VIRTIO_BALLOON=m VIRTIO_BLK=m VIRTIO_NET=m VIRTIO_CONSOLE=m

Quoted 2 lines

The recipe for a build is:
- check out the source

Any specific branches beside the master branch?

Quoted 3 lines

- MAKE defconfig
where MAKE="make ARCH=$ARCH
CROSS_COMPILE=/opt/gcc-4.5.0-nolibc/$ARCH-linux/bin/$ARCH-linux-'

The buildbot master configuration requires exactly that path for CROSS_COMPILE on the buildslave. The buildsaves need to be setup the toolchain before hand manually.

Just let me and Christian know if we should add another toolchain or upgrade to a newer one.

Quoted 1 line

- append the extra options to .config

Done this by invoking: ./scripts/config (proposal/attempt from Christian)

(See http://buildbot.b1-systems.de/kvm/builders/x86_64/builds/11/steps/configure_2/logs/stdio )

Quoted 1 line

- MAKE oldconfig

What about make target oldnoconfig instead?
To avoid that the build get stuck with every new config option?

Demo of the issue with "make oldconfig":
http://buildbot.b1-systems.de/kvm/builders/x86_64/builds/10/steps/configure_3/logs/stdio

Quoted 1 line

- MAKE

[...]

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Avi Kivity

author

2010-11-24 10:23 UTC #11 lore raw

Quoted 8 lines · Daniel Gollub

Quoted 3 lines · Avi Kivity
Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.
First test builds succeed for x86 and ia64:
http://buildbot.b1-systems.de/kvm/grid

Quoted 8 lines

with gcc-4.5.0-nolibc toolchain from
http://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.0/
powerpc not ready yet - trivial $ARCH change required in the buildbot master configuration.
s390 compilation seems to run fine and hit the first compilation issue:
http://buildbot.b1-systems.de/kvm/builders/s390/builds/0/steps/compile/logs/stdio

Will work on fixing it - so it's already useful!

I think buildbot has a mechanism where you can send it a patch to be build-tested. This allows developers to build-test on architectures they don't have. Can we activate this?

Quoted 6 lines

Currently buildbot notification mails don't get send to the kvm list. I would like to do some more testing before spamming with wrong-testing results.
Some more questions inline:

Quoted 8 lines

Quoted 6 lines
I've now a virtual machine with a cross compile environment for s390, ppc, and ia64. I'd like buildbot to build for the following archs and configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
Without virtio? intentionally?

No, that's my old configuration I used for build testing. While virtio isn't part of kvm, and is arch-neutral, it would be good to enable it.

Quoted 5 lines

Quoted 3 lines
The recipe for a build is:
- check out the source
Any specific branches beside the master branch?

master and next.

Quoted 9 lines

Quoted 4 lines
- MAKE defconfig
where MAKE="make ARCH=$ARCH
CROSS_COMPILE=/opt/gcc-4.5.0-nolibc/$ARCH-linux/bin/$ARCH-linux-'
The buildbot master configuration requires exactly that path for CROSS_COMPILE on the buildslave. The buildsaves need to be setup the toolchain before hand manually.
Just let me and Christian know if we should add another toolchain or upgrade to a newer one.

I think this is fine. Hopefully we won't upgrade too often, this should last us a year or two.

Quoted 5 lines

Quoted 1 line
- append the extra options to .config
Done this by invoking: ./scripts/config (proposal/attempt from Christian)
(See http://buildbot.b1-systems.de/kvm/builders/x86_64/builds/11/steps/configure_2/logs/stdio )

Ah, better than the hackery I used before.

Quoted 7 lines

Quoted 1 line
- MAKE oldconfig
What about make target oldnoconfig instead?
To avoid that the build get stuck with every new config option?
Demo of the issue with "make oldconfig":
http://buildbot.b1-systems.de/kvm/builders/x86_64/builds/10/steps/configure_3/logs/stdio

I used 'make oldconfig < /dev/null', wasn't aware of oldnoconfig.

Thanks for doing this!

Signature · 1 line

error compiling committee.c: too many arguments to function

Daniel Gollub

2010-11-24 20:49 UTC #12 lore raw

On Wednesday, November 24, 2010 11:23:37 pm Avi Kivity wrote:
[...]

Quoted 3 lines

I think buildbot has a mechanism where you can send it a patch to be build-tested. This allows developers to build-test on architectures they don't have. Can we activate this?

Sure. Just enabled that. Actually it's also available for qemu-kvm.git but never really got used.

Who should get access to this? I wouldn't make it available for anonymous, since someone could checkin an "evil"-patch for a makefile, which includes "rm -rf /this/and/that" and vanish the buildslave.

I'm fine giving (personal) user accounts away for people who just ask for it. Or do you want to make this only available for a small (trusted) group?

Try-Build would get triggerd with the buildbot command (on developers workstation) like this:

$ cd kvm.git
... do some changes ...
$ buildbot try --vc git -c pb -m buildbot.b1-systems.de:8033 -b s390 \
  -u $USER --passwd=$PASSWORD

[...]

Quoted 8 lines

Quoted 5 lines
Quoted 3 lines
configs:
x86_64: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
i386: defconfig + KVM=m KVM_INTEL=m KVM_AMD=m
Without virtio? intentionally?
No, that's my old configuration I used for build testing. While virtio isn't part of kvm, and is arch-neutral, it would be good to enable it.

Ok. I see - was just wondering because it was set in your powerpc configurations but not for others.

Quoted 7 lines

Quoted 5 lines
Quoted 3 lines
The recipe for a build is:
- check out the source
Any specific branches beside the master branch?
master and next.

done.

On the next kvm.git push the buildbot should trigger a build test for
branches: master and next.

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Daniel Gollub

2010-11-24 20:51 UTC #9 lore raw

In reply to Avi Kivity

Quoted 2 lines · Avi Kivity

Will you set up crossbuilders for ppc/ia64/s390, or will I contribute a builder?

Later once the setup is stable it would be nice to have multiple buildslaves, so we get really fast response from the buildbot if a commit breaks something.

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Daniel Gollub

2010-11-24 21:03 UTC #18 lore raw

In reply to Avi Kivity

Quoted 2 lines · Avi Kivity

Quoted 2 lines
powerpc not ready yet - trivial $ARCH change required in the buildbot master configuration.

powerpc builders are now also setup and already spot something.

Build issue with ppc44x:
http://buildbot.b1-systems.de/kvm/builders/ppc44x/builds/2/steps/compile/logs/stdio

Build issue with ppc64:
http://buildbot.b1-systems.de/kvm/builders/ppc64/builds/4/steps/compile/logs/stdio

Quoted 6 lines

Quoted 4 lines
s390 compilation seems to run fine and hit the first compilation issue:
http://buildbot.b1-systems.de/kvm/builders/s390/builds/0/steps/compile/lo
gs/stdio
Will work on fixing it - so it's already useful!

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Alexander Graf

2010-11-25 01:20 UTC #19 lore raw

Quoted 11 lines · Daniel Gollub

Quoted 2 lines · Avi Kivity
Quoted 2 lines
powerpc not ready yet - trivial $ARCH change required in the buildbot master configuration.
powerpc builders are now also setup and already spot something.
Build issue with ppc44x:
http://buildbot.b1-systems.de/kvm/builders/ppc44x/builds/2/steps/compile/logs/stdio
Build issue with ppc64:
http://buildbot.b1-systems.de/kvm/builders/ppc64/builds/4/steps/compile/logs/stdio

Same bug as the one of the S390 ones.

Avi, my (whitespace broken) patch for this is:

include/linux/kvm_host.h +2 -0

index f17beae..da0794f 100644
@@ -673,11 +673,13 @@ static inline int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags)
 
 static inline void kvm_irqfd_release(struct kvm *kvm) {}
 
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
 static inline void kvm_irq_routing_update(struct kvm *kvm,
                                          struct kvm_irq_routing_table *irq_rt)
 {
        rcu_assign_pointer(kvm->irq_routing, irq_rt);
 }
+#endif
 
 static inline int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
 {

Have you fixed this locally in your tree already or do you want to pull from me?

Alex

Avi Kivity

author

2010-11-25 09:17 UTC #20 lore raw

Quoted 38 lines · Alexander Graf

Quoted 11 lines · Daniel Gollub
Quoted 2 lines · Avi Kivity
Quoted 2 lines
powerpc not ready yet - trivial $ARCH change required in the buildbot master configuration.
powerpc builders are now also setup and already spot something.
Build issue with ppc44x:
http://buildbot.b1-systems.de/kvm/builders/ppc44x/builds/2/steps/compile/logs/stdio
Build issue with ppc64:
http://buildbot.b1-systems.de/kvm/builders/ppc64/builds/4/steps/compile/logs/stdio
Same bug as the one of the S390 ones.
Avi, my (whitespace broken) patch for this is:
include/linux/kvm_host.h +2 -0
index f17beae..da0794f 100644
@@ -673,11 +673,13 @@ static inline int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags)
 
  static inline void kvm_irqfd_release(struct kvm *kvm) {}
 
+#ifdef CONFIG_HAVE_KVM_IRQCHIP
  static inline void kvm_irq_routing_update(struct kvm *kvm,
                                           struct kvm_irq_routing_table *irq_rt)
  {
         rcu_assign_pointer(kvm->irq_routing, irq_rt);
  }
+#endif
 
  static inline int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
  {

Quoted 1 line

Have you fixed this locally in your tree already or do you want to pull from me?

I don't have this locally.

Signature · 1 line

error compiling committee.c: too many arguments to function

Avi Kivity

author

2010-11-25 09:23 UTC #13 lore raw

In reply to Daniel Gollub

Quoted 9 lines · Daniel Gollub

On Wednesday, November 24, 2010 11:23:37 pm Avi Kivity wrote:
[...]
Quoted 3 lines
I think buildbot has a mechanism where you can send it a patch to be build-tested. This allows developers to build-test on architectures they don't have. Can we activate this?
Sure. Just enabled that. Actually it's also available for qemu-kvm.git but never really got used.

Quoted 6 lines

Who should get access to this? I wouldn't make it available for anonymous, since someone could checkin an "evil"-patch for a makefile, which includes "rm -rf /this/and/that" and vanish the buildslave.
I'm fine giving (personal) user accounts away for people who just ask for it. Or do you want to make this only available for a small (trusted) group?

I think you can give accounts to kvm contributors (present in kvm.git commit log). I'll start asking people to build-test and point them at you for accounts. Should be rare though, most commits are x86 specific.

I'll start by asking for an account.

Signature · 1 line

error compiling committee.c: too many arguments to function

Daniel Gollub

2010-11-29 11:36 UTC #14 lore raw

Quoted 7 lines · Avi Kivity

Quoted 3 lines
I'm fine giving (personal) user accounts away for people who just ask for it. Or do you want to make this only available for a small (trusted) group?
I think you can give accounts to kvm contributors (present in kvm.git commit log). I'll start asking people to build-test and point them at you for accounts. Should be rare though, most commits are x86 specific.

Ok.

Quoted 2 lines

I'll start by asking for an account.

Unfortunately we're running (in meanwhile) an old version of buildbot as buildbot-master (0.7.8, as shipped by Debian 5). And there seems to be an issue using a recent buildbot version to trigger a "try"-build (e.g. 0.8.1) . Everyone would have to downgrade to buildbot (client) 0.7.8, even if their distro ships a more recent version.

Another (minor) issue with the old buildbot-master is that every time someone performs a try-build of a patch the buildbot-slaves will vanish the source- directory on the next testrun. And perform a complete git clone from scratch, which wastes some minutes of testing time and a lot of traffic. Need to check more recent version of buildbot, lots of stuff got changed/rewritten when it comes to VCS handling.

Newer buildbot version also support now "GitPoller" to detect new changes. So we can avoid the rsync-git.kernel.org-delay issue.

So there are lots of reason to look into new buildbot version, before giving away try-accounts.

FYI, ppc44x is finally building (uboot-mkimage was missing on the slave). Continuous Testing for all requested architectures and the master and next branch should work now. Will add the kvm mailinglist tomorrow to the notification list for build-failures of kvm.git.

Once I reviewed the newer buildbot version I start providing the try-accounts.

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Jan Kiszka

2010-11-29 12:20 UTC #15 lore raw

Quoted 10 lines · Daniel Gollub

Quoted 7 lines · Avi Kivity
Quoted 3 lines
I'm fine giving (personal) user accounts away for people who just ask for it. Or do you want to make this only available for a small (trusted) group?
I think you can give accounts to kvm contributors (present in kvm.git commit log). I'll start asking people to build-test and point them at you for accounts. Should be rare though, most commits are x86 specific.

Quoted 8 lines

Ok.
Quoted 2 lines
I'll start by asking for an account.
Unfortunately we're running (in meanwhile) an old version of buildbot as buildbot-master (0.7.8, as shipped by Debian 5).

Last time I checked (quite a few moons ago, though), that version contained an unfixed security issue. I installed a vanilla version on my server for that reason (and forgot to patch that afterward...).

Jan

Signature · 2 lines

Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux

Daniel Gollub

2010-11-29 19:22 UTC #16 lore raw

Quoted 6 lines · Jan Kiszka

Quoted 2 lines
Unfortunately we're running (in meanwhile) an old version of buildbot as buildbot-master (0.7.8, as shipped by Debian 5).
Last time I checked (quite a few moons ago, though), that version contained an unfixed security issue. I installed a vanilla version on my server for that reason (and forgot to patch that afterward...).

Are you talking about this one?
http://buildbot.net/trac/wiki/SecurityAlert0711

We applied that patch already.
Are you aware of any other unfixed security issues?

Best Regards,
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Jan Kiszka

2010-11-30 09:12 UTC #17 lore raw

Quoted 9 lines · Daniel Gollub

Quoted 6 lines · Jan Kiszka
Quoted 2 lines
Unfortunately we're running (in meanwhile) an old version of buildbot as buildbot-master (0.7.8, as shipped by Debian 5).
Last time I checked (quite a few moons ago, though), that version contained an unfixed security issue. I installed a vanilla version on my server for that reason (and forgot to patch that afterward...).

Quoted 5 lines

Are you talking about this one?
http://buildbot.net/trac/wiki/SecurityAlert0711
We applied that patch already.
Are you aware of any other unfixed security issues?

Not that I recall.

The point was just that, if I already have to patch the package, then I could also use a more recent version.

Jan

Signature · 2 lines

Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux

Daniel Gollub

2010-11-30 10:47 UTC #21 lore raw

buildbot for kvm.git is now live (was: Re: buildbot for kvm.git)

In reply to Avi Kivity

Quoted 3 lines · Avi Kivity

Daniel, the buildbot has been fairly effective in keeping qemu-kvm.git building. I'd like to extend that to kvm.git, especially for non-x86 architectures.

buildbot for kvm.git is now live and is build-testing the master and next-branch for i386, x86_64, ia64, s390, ppc64, ppc44x.

Continuous build-testing is triggered by a git receive hook and a nightly build scheduler. And will notify directly to the kvm list on any build issue.

In near future we will provide access for easy test-building your patches via the buildbot across all available architectures.

The builders spot a recent regression for non-x86 in the next-branch.

common issue across all non-x86:

... error: 'struct kvm' has no member named 'tlbs_dirty'

s390:   http://buildbot.b1-systems.de/kvm/builders/next-s390/builds/3/steps/compile/logs/stdio
ppc64:  http://buildbot.b1-systems.de/kvm/builders/next-ppc64/builds/2/steps/compile/logs/stdio
ppc44x: http://buildbot.b1-systems.de/kvm/builders/next-ppc44x/builds/1/steps/compile/logs/stdio
ia64:   http://buildbot.b1-systems.de/kvm/builders/next-ia64/builds/2/steps/compile/logs/stdio

Good: 9ee00410d82a7c5cab5ae347d97fbf8a95c55506
Bad:  f5921ca4a97a18033a69df128bc4d369d8a2b280

Enjoy the buildbot!
Daniel

Signature · 7 lines

Daniel Gollub
Linux Consultant & Developer
Mail: [email protected]

B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537