| Message ID | 874n906avo.fsf@linaro.org |
|---|---|
| State | New |
| Headers | show |
On 10/01/2013 04:37 PM, Alex Bennée wrote: > > lmr@redhat.com writes: > >> Hello: >> >> In the daily qemu.git autotest run, I started to see the following >> problem while checking out the qemu.git tree and checking out the >> submodules: >> >>> $ git submodule update >>> <snip> >>> fatal: reference is not a tree: ece025f5980bae88fa677bc9c0d24d2e580e205d >>> <snip> >>> Unable to checkout 'ece025f5980bae88fa677bc9c0d24d2e580e205d' in submodule path 'roms/seabios' > > It looks like a recent commit to update the version of SeaBIOS used > broke the build as the qemu mirror hasn't been updated yet. You can work > around it for now by patching .gitmodules: Ok, thanks for the explanation. Local patching is not really a scalable option when performing automated testing though. I'll keep the daily job running, and when the mirror is updated, the failures will stop. Cheers, Lucas
lmr@redhat.com writes: > On 10/01/2013 04:37 PM, Alex Bennée wrote: >> >> lmr@redhat.com writes: >> >>> Hello: >>> >>> In the daily qemu.git autotest run, I started to see the following >>> problem while checking out the qemu.git tree and checking out the >>> submodules: Oh I just noticed that. Is this Stefan's buildbot stuff or an internal auto-test set-up? I only ask because I noticed this while testing my Travis patches. I'm trying to get a feel for what CI type tasks are running in QEMU world, both public and internal to companies. >> >> It looks like a recent commit to update the version of SeaBIOS used >> broke the build as the qemu mirror hasn't been updated yet. You can work >> around it for now by patching .gitmodules: > > Ok, thanks for the explanation. Local patching is not really a scalable > option when performing automated testing though. I'll keep the daily job > running, and when the mirror is updated, the failures will stop. I'll be submitting a patch tomorrow when I post the next version of the Travis patches up to the mailing list.
On 10/01/2013 04:52 PM, Alex Bennée wrote: > > lmr@redhat.com writes: > >> On 10/01/2013 04:37 PM, Alex Bennée wrote: >>> >>> lmr@redhat.com writes: >>> >>>> Hello: >>>> >>>> In the daily qemu.git autotest run, I started to see the following >>>> problem while checking out the qemu.git tree and checking out the >>>> submodules: > > Oh I just noticed that. Is this Stefan's buildbot stuff or an internal > auto-test set-up? I only ask because I noticed this while testing my > Travis patches. I'm trying to get a feel for what CI type tasks are > running in QEMU world, both public and internal to companies. That's an internal autotest test grid. Once a day we check the latest qemu master and run sort of a smoke style test job, although it's not so, 'smoky', since it takes about a couple of hours to run. Tests covered: 1) Installing Fedora 19 on a host 2) git clone the latest master 3) Install a RHEL 6.4 guest from kickstart 4) Basic boot, reboot, single host migration using tcp, unix socket, fd and exec migration 5) Install a Windows 7 guest from an unattended XML, using the latest virtio drivers that our windows drivers team releases. 6) Perform the same tests outlined in 4) I hope it gives you a better idea of what's going on.
diff --git a/.gitmodules b/.gitmodules index d7e3f3c..d5d5417 100644 --- a/.gitmodules +++ b/.gitmodules @@ -3,7 +3,7 @@ url = git://git.qemu.org/vgabios.git/ [submodule "roms/seabios"] path = roms/seabios - url = git://git.qemu.org/seabios.git/ + url = git://git.seabios.org/seabios.git [submodule "roms/SLOF"] path = roms/SLOF url = git://git.qemu.org/SLOF.git