On Mon, Feb 06, 2012 at 04:11:45AM -0500, Alex Jia wrote: > Hi all, > I found the following issues: [...] > 2. test cases fail: > > TEST: virnetsockettest > ........!!!.!!! 15 FAIL > > TEST: qemuxml2argvtest > ....................!!.................. 40 > ......................................!. 80 > ........................................ 120 > ........................................ 160 > ...!.!!!!!!.!..!!.!.................... 199 FAIL Weird that definitely worked fine for me. Try to do an rpmbuild --rebuild libvirt-0.9.10-rc1.tar.gz to get a list of possible missing build packages. However I don't see how that could possibly affect qemuxml2argvtest ...

Daniel

On 02/06/2012 03:11 AM, Alex Jia wrote: >>> TEST: qemuxml2argvtest >>> ....................!!.................. 40 >>> ......................................!. 80 >>> ........................................ 120 >>> ........................................ 160 >>> ...!.!!!!!!.!..!!.!.................... 199 FAIL >> Weird that definitely worked fine for me. >> Try to do an rpmbuild --rebuild libvirt-0.9.10-rc1.tar.gz to get >> a list of possible missing build packages. However I don't see how >> that could possibly affect qemuxml2argvtest ... I haven't reproduced this issue. > It's okay if I installed related dev packages. Do you know which dev package made the difference? > Other issues: > 1. > GEN probes.o > /tmp/tmpOSBnp3.c:1: warning: return type defaults to 'int' > /tmp/tmpOSBnp3.c:1: warning: '__dtrace' defined but not used > CC libvirt_qemu_la-libvirt-qemu.lo > > Notes, maybe, we should silence the warning. I'd like to; but doing that requires either patching systemtap-sdt-devel, or else post-processing the systemtap generated files prior to passing them to the compiler. In other words, the warning is not coming from libvirt source code. > 2. > CC libvirtmod_qemu_la-libvirt-qemu-override.lo > libvirt-qemu-override.c:53: warning: 'py_str' defined but not used > CC libvirtmod_qemu_la-libvirt-qemu.lo > > Notes, it should be a useful function, maybe, we will use it later ... We should fix this one. > 3. > CCLD libvirt_test.la > > *** Warning: Linking the shared library libvirt.la against the non-libtool > *** objects probes.o is not portable! > > 4. > *** Warning: Linking the shared library libvirt_test.la against the > non-libtool > *** objects probes.o is not portable! > CCLD libvirt-qemu.la > > > Notes, I often meet the item 3 and 4 warnings when compiling, although I > saw gcc book said > they were common error and should use .o instead of .la, we can ignore > these 2 warnings > in here, right? Not the gcc book. But this has previously come up on this list, and the

answer is still the same - libtool doesn't have a way to let us shut it up when we _know_ that we are doing something that works on Linux, and where we are not doing the non-portable action of using probes.o on non-Linux because systemtap is Linux-specific. You can ignore the

warning, and any patch to silence it would have to come from upstream libtool, or else finding a way to create a .lo file that wraps the generated probes.o file but which libtool can still link with.

On 02/07/2012 02:41 AM, Eric Blake wrote: > On 02/06/2012 03:11 AM, Alex Jia wrote: > >>>> TEST: qemuxml2argvtest >>>> ....................!!.................. 40 >>>> ......................................!. 80 >>>> ........................................ 120 >>>> ........................................ 160 >>>> ...!.!!!!!!.!..!!.!.................... 199 FAIL >>> Weird that definitely worked fine for me. >>> Try to do an rpmbuild --rebuild libvirt-0.9.10-rc1.tar.gz to get >>> a list of possible missing build packages. However I don't see how >>> that could possibly affect qemuxml2argvtest ... > I haven't reproduced this issue. > >> It's okay if I installed related dev packages. > Do you know which dev package made the difference?

> >> Other issues: >> 1. >> GEN probes.o >> /tmp/tmpOSBnp3.c:1: warning: return type defaults to 'int' >> /tmp/tmpOSBnp3.c:1: warning: '__dtrace' defined but not used >> CC libvirt_qemu_la-libvirt-qemu.lo >> >> Notes, maybe, we should silence the warning. > I'd like to; but doing that requires either patching > systemtap-sdt-devel, or else post-processing the systemtap generated > files prior to passing them to the compiler. In other words, the > warning is not coming from libvirt source code. > >> 2. >> CC libvirtmod_qemu_la-libvirt-qemu-override.lo >> libvirt-qemu-override.c:53: warning: 'py_str' defined but not used >> CC libvirtmod_qemu_la-libvirt-qemu.lo >> >> Notes, it should be a useful function, maybe, we will use it later ... > We should fix this one. > >> 3. >> CCLD libvirt_test.la >> >> *** Warning: Linking the shared library libvirt.la against the >> non-libtool >> *** objects probes.o is not portable! >> >> 4. >> *** Warning: Linking the shared library libvirt_test.la against the >> non-libtool >> *** objects probes.o is not portable! >> CCLD libvirt-qemu.la >> >> >> Notes, I often meet the item 3 and 4 warnings when compiling, >> although I >> saw gcc book said >> they were common error and should use .o instead of .la, we can ignore >> these 2 warnings >> in here, right? > Not the gcc book. But this has previously come up on this list, and the The common error is introduced in ch8 of "The Definitive Guide to GCC" v1 book. but I think it should be a warning not error. > answer is still the same - libtool doesn't have a way to let us shut it > up when we _know_ that we are doing something that works on Linux, and > where we are not doing the non-portable action of using probes.o on > non-Linux because systemtap is Linux-specific. You can ignore the Yeah, indeed. > warning, and any patch to silence it would have to come from upstream > libtool, or else finding a way to create a .lo file that wraps the > generated probes.o file but which libtool can still link with. However, libtool manual said we may silence this warning if specify link library the path with '-lm' option, I'm not sure whether we need to follow this, as you said, after all, libvirt also supports non-linux platform. http://www.gnu.org/software/libtool/manual/libtool.html Thanks & Regards, Alex -- libvir-list mailing list libvir-list(a)redhat.com https://www.redhat.com/mailman/listinfo/libvir-list

We are now entering the freeze for libvirt-0.9.10 Hopefully all the API changes needed for that version are already commited to git head. I have made a release candidate 1 tarball (and associated rpms) at ftp://libvirt.org/libvirt/libvirt-0.9.10-rc1.tar.gz and the git tree is tagged. I think I will make an release candidate 2 tarball on Wednesday, and I'm hoping for a final release next Monday the 13th. Please give it a try !

On Mon, Feb 06, 2012 at 03:21:21PM +0800, Daniel Veillard wrote: > We are now entering the freeze for libvirt-0.9.10 > Hopefully all the API changes needed for that version are > already commited to git head. Okay I made an rc2 release with the python bindings fix (thanks !) ftp://libvirt.org/libvirt/libvirt-0.9.10-rc2.tar.gz the rpms are available too and the git tree is tagged. Hopefully that one is more usable, and I'm still hoping for a release on Monday, thanks ! Daniel

On 02/08/2012 02:59 PM, Daniel Veillard wrote: > On Mon, Feb 06, 2012 at 03:21:21PM +0800, Daniel Veillard wrote: >> We are now entering the freeze for libvirt-0.9.10 >> Hopefully all the API changes needed for that version are >> already commited to git head. > Okay I made an rc2 release with the python bindings fix (thanks !) The python bindings work well for me on rhel, and unit test cases also are fine except for valgrind found a memory leak: ==21503== 112 (32 direct, 80 indirect) bytes in 1 blocks are definitely lost in loss record 37 of 40 ==21503== at 0x4A04A28: calloc (vg_replace_malloc.c:467) ==21503== by 0x4A8991: virAlloc (memory.c:101) ==21503== by 0x505A6C: x86DataCopy (cpu_x86.c:247) ==21503== by 0x507B34: x86Compute (cpu_x86.c:1225) ==21503== by 0x43103C: qemuBuildCommandLine (qemu_command.c:3561) ==21503== by 0x41C9F7: testCompareXMLToArgvHelper (qemuxml2argvtest.c:183) ==21503== by 0x41E10D: virtTestRun (testutils.c:141) ==21503== by 0x41B942: mymain (qemuxml2argvtest.c:705) ==21503== by 0x41D7E7: virtTestMain (testutils.c:696)

ftp://libvirt.org/libvirt/libvirt-0.9.10-rc2.tar.gz the rpms are available too and the git tree is tagged. Hopefully that one is more usable, and I'm still hoping for a release on Monday,

We are now entering the freeze for libvirt-0.9.10 Hopefully all the API changes needed for that version are already commited to git head. I have made a release candidate 1 tarball (and associated rpms) at ftp://libvirt.org/libvirt/libvirt-0.9.10-rc1.tar.gz and the git tree is tagged. I think I will make an release candidate 2 tarball on Wednesday, and I'm hoping for a final release next Monday the 13th. Please give it a try !

On 02/06/2012 12:21 AM, Daniel Veillard wrote: > We are now entering the freeze for libvirt-0.9.10 > Hopefully all the API changes needed for that version are > already commited to git head. > > I have made a release candidate 1 tarball (and associated rpms) at > ftp://libvirt.org/libvirt/libvirt-0.9.10-rc1.tar.gz > and the git tree is tagged. > > I think I will make an release candidate 2 tarball on Wednesday, and > I'm hoping for a final release next Monday the 13th. > > Please give it a try ! I'm wondering if we're too late for one more API to round out the API we added in this release. Qemu is proposing a new monitor command system_wakeup which lets the host inject a power-button wakeup to a guest that has gone into S3 sleep. Since we added the virDomainPMSuspendForDuration command to put the guest into S3 from the host, I think we need a counterpart API that wakes the guest back up, rather than relying on mouse clicks, serial input, or timeouts doing the job for us. It's a shame that virDomainResume() has no flags argument, and that it is already tied to virDomainSuspend() (aka guest pause), so I'm thinking something like: /* Inject a wakeup into the guest that previously used * virDomainPMSuspendForDuration, rather than waiting for the * previously requested duration (if any) to elapse. */ int virDomainPMWakeup(virDomainPtr dom, unsigned int flags);

-- Eric Blake eblake(a)redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

-- libvir-list mailing list libvir-list(a)redhat.com https://www.redhat.com/mailman/listinfo/libvir-list

On the other hand, it probably also points out that we are missing a counterpart: virDomainGetPowerState(domain, flags) after all, with NodeSuspendForDuration, we always know what state we are in (the host is active if you can query, and all other states the host is non-responsive); but with guests, the hypervisor can indeed be queried to see whether the guest is currently active or in S3.

On Fri, Feb 10, 2012 at 13:33:47 -0700, Eric Blake wrote: > On 02/10/2012 12:21 PM, Dave Allan wrote: > >> I'm wondering if we're too late for one more API to round out the API we > >> added in this release. Qemu is proposing a new monitor command > >> system_wakeup which lets the host inject a power-button wakeup to a > >> guest that has gone into S3 sleep. Since we added the > >> virDomainPMSuspendForDuration command to put the guest into S3 from the > >> host, I think we need a counterpart API that wakes the guest back up, > >> rather than relying on mouse clicks, serial input, or timeouts doing the > >> job for us. It's a shame that virDomainResume() has no flags argument, > >> and that it is already tied to virDomainSuspend() (aka guest pause), so > >> I'm thinking something like: > >> > >> /* Inject a wakeup into the guest that previously used > >> * virDomainPMSuspendForDuration, rather than waiting for the > >> * previously requested duration (if any) to elapse. */ > >> int virDomainPMWakeup(virDomainPtr dom, unsigned int flags); > > > > Just to throw this out there, could we not simply implement the rest > > of the power states in this API? Kind of late to be asking this > > question, I suppose, but we haven't released this API yet, could we > > change its name to virDomainPMSetPowerStateForDuration? > > Interesting idea! It means we have one call, instead of 2, and 4 PM > states instead of 3 (suspend, hibernate, hybrid, and now wakeup). Of > course, wakeup for duration doesn't make much sense, so we could > explicitly require a 0 duration for the wakeup case. I'm not a big fan of this idea. There are two different things here: suspending (in whatever way) and waking up. I think it makes sense to have separate APIs for them. After all we don't have a virDomainSetState API which we could use to transition domain between lifecycle states (running, shutoff, paused, ...). We have separate APIs for each transition, each doing one "simple" thing and possibly taking different data needed for that transition.

Jirka