On Tue, 2014-09-23 at 14:42 +0200, Michal Privoznik wrote:
> On 23.09.2014 11:34, Chen Fan wrote:
> > use virDomainNumatuneNodeSetIsAvailable() to verify momory.nodeset
> > whether is out of range. and move up the verification.
> >
> > Signed-off-by: Chen Fan <chen.fan.fnst@xxxxxxxxxxxxxx>
> > ---
> > src/conf/numatune_conf.c | 3 +++
> > src/util/virnuma.c | 15 ---------------
> > 2 files changed, 3 insertions(+), 15 deletions(-)
>
> I'd expect a test case for this.
Ok, I will add it.
thanks,
Chen
>
> >
> > diff --git a/src/conf/numatune_conf.c b/src/conf/numatune_conf.c
> > index a9b20aa..8b43167 100644
> > --- a/src/conf/numatune_conf.c
> > +++ b/src/conf/numatune_conf.c
> > @@ -278,6 +278,9 @@ virDomainNumatuneParseXML(virDomainNumatunePtr *numatunePtr,
> > nodeset) < 0)
> > goto cleanup;
> >
> > + if (!virDomainNumatuneNodeSetIsAvailable(*numatunePtr, -1))
> > + goto cleanup;
> > +
> > if (virDomainNumatuneNodeParseXML(numatunePtr, ncells, ctxt) < 0)
> > goto cleanup;
> >
> > diff --git a/src/util/virnuma.c b/src/util/virnuma.c
> > index 1a34398..4766f16 100644
> > --- a/src/util/virnuma.c
> > +++ b/src/util/virnuma.c
> > @@ -95,31 +95,16 @@ virNumaSetupMemoryPolicy(virDomainNumatunePtr numatune,
> > int ret = -1;
> > int bit = 0;
> > size_t i;
> > - int maxnode = 0;
> > virBitmapPtr tmp_nodemask = NULL;
> >
> > tmp_nodemask = virDomainNumatuneGetNodeset(numatune, nodemask, -1);
> > if (!tmp_nodemask)
> > return 0;
> >
> > - if (numa_available() < 0) {
> > - virReportError(VIR_ERR_INTERNAL_ERROR,
> > - "%s", _("Host kernel is not aware of NUMA."));
> > - return -1;
> > - }
> > -
> > - maxnode = numa_max_node();
> > - maxnode = maxnode < NUMA_NUM_NODES ? maxnode : NUMA_NUM_NODES;
> > -
> > /* Convert nodemask to NUMA bitmask. */
> > nodemask_zero(&mask);
> > bit = -1;
> > while ((bit = virBitmapNextSetBit(tmp_nodemask, bit)) >= 0) {
> > - if (bit > maxnode) {
> > - virReportError(VIR_ERR_INTERNAL_ERROR,
> > - _("NUMA node %d is out of range"), bit);
> > - return -1;
> > - }
> > nodemask_set(&mask, bit);
> > }
> >
> >
>
> Yet again, this suffers the same problem that 1/2 does: domain may be lost.
>
> Michal
--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list