[libvirt] [PATCH 05/23] cpu: Use virDomainCapsCPUModelsPtr in cpu driver APIs
Jiri Denemark
jdenemar at redhat.com
Fri Oct 13 14:47:30 UTC 2017
On Thu, Oct 12, 2017 at 07:27:22 -0400, John Ferlan wrote:
>
>
> On 10/04/2017 10:58 AM, Jiri Denemark wrote:
> > All APIs which expect a list of CPU models supported by hypervisors were
> > switched from char **models and int models to just accept a pointer to
> > virDomainCapsCPUModels object stored in domain capabilities. This avoids
> > the need to transform virDomainCapsCPUModelsPtr into a NULL-terminated
> > list of model names and also allows the various cpu driver APIs to
> > access additional details (such as its usability) about each CPU model.
...
> > @@ -514,21 +503,20 @@ virCPUProbeHost(virArch arch)
> > virCPUDefPtr
> > cpuBaseline(virCPUDefPtr *cpus,
> > unsigned int ncpus,
> > - const char **models,
> > - unsigned int nmodels,
> > + virDomainCapsCPUModelsPtr models,
> > bool migratable)
> > {
> > struct cpuArchDriver *driver;
> > size_t i;
> >
> > - VIR_DEBUG("ncpus=%u, nmodels=%u", ncpus, nmodels);
> > + VIR_DEBUG("ncpus=%u", ncpus);
>
> You could add models=%p... not that it's necessary
Done.
> > diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
> > index 7ddc6cafd4..225cee4ef9 100644
> > --- a/src/qemu/qemu_capabilities.c
> > +++ b/src/qemu/qemu_capabilities.c
> [...]
> > -int
> > +virDomainCapsCPUModelsPtr
> > virQEMUCapsGetCPUDefinitions(virQEMUCapsPtr qemuCaps,
> > - virDomainVirtType type,
> > - char ***names,
> > - size_t *count)
> > + virDomainVirtType type)
> > {
> > - size_t i;
> > - char **models = NULL;
> > - virDomainCapsCPUModelsPtr cpus;
> > -
> > - *count = 0;
> > - if (names)
> > - *names = NULL;
> > -
> > if (type == VIR_DOMAIN_VIRT_KVM)
> > - cpus = qemuCaps->kvmCPUModels;
> > + return qemuCaps->kvmCPUModels;
> > else
> > - cpus = qemuCaps->tcgCPUModels;
> > + return qemuCaps->tcgCPUModels;
> >
...
> > + return NULL;
>
> Nice overall simplification, but how do we reach here?
Eh, the "return NULL;" shouldn't be there.
Jirka
More information about the libvir-list
mailing list