Re: [libvirt] [PATCH python] include usable memory in virDomainMemoryStats

On 08/02/2017 01:17 PM, Martin Kletzander wrote: > On Wed, Aug 02, 2017 at 12:52:32PM +0200, Michal Privoznik wrote: >> On 08/01/2017 03:23 PM, Tomáš Golembiovský wrote: >>> Signed-off-by: Tomáš Golembiovský <tgolembi(a)redhat.com&gt; >>> --- >>> libvirt-override.c | 3 +++ >>> 1 file changed, 3 insertions(+) >>> >>> diff --git a/libvirt-override.c b/libvirt-override.c >>> index 0abfc37..832e05c 100644 >>> --- a/libvirt-override.c >>> +++ b/libvirt-override.c >>> @@ -398,6 +398,9 @@ libvirt_virDomainMemoryStats(PyObject *self >>> ATTRIBUTE_UNUSED, >>> case VIR_DOMAIN_MEMORY_STAT_RSS: >>> key = libvirt_constcharPtrWrap("rss"); >>> break; >>> + case VIR_DOMAIN_MEMORY_STAT_USABLE: >>> + key = libvirt_constcharPtrWrap("usable"); >>> + break; >>> default: >>> continue; >>> } >>> >> >> Almost. Firstly, there's VIR_DOMAIN_MEMORY_STAT_LAST_UPDATE which is not >> handled either. Secondly, these two macros were introduced in libvirt >> 2.1.0. So we need a check that enables them iff building with 2.1.0 or >> newer. I'm fixing both issues and pushing. I wonder if there's something >> we can do to not forget about macros like these again. >> > > You mean like this? Or am I missing a reason why this won't work? > > diff --git a/libvirt-override.c b/libvirt-override.c > index 0abfc379c528..6e678d2efb2d 100644 > --- a/libvirt-override.c > +++ b/libvirt-override.c > @@ -373,7 +373,7 @@ libvirt_virDomainMemoryStats(PyObject *self > ATTRIBUTE_UNUSED, > return NULL; > > for (i = 0; i < nr_stats; i++) { > - switch (stats[i].tag) { > + switch ((virDomainMemoryStatTags) stats[i].tag) { > case VIR_DOMAIN_MEMORY_STAT_SWAP_IN: > key = libvirt_constcharPtrWrap("swap_in"); > break; > @@ -398,7 +398,8 @@ libvirt_virDomainMemoryStats(PyObject *self > ATTRIBUTE_UNUSED, > case VIR_DOMAIN_MEMORY_STAT_RSS: > key = libvirt_constcharPtrWrap("rss"); > break; > - default: > + case VIR_DOMAIN_MEMORY_STAT_NR: > + case VIR_DOMAIN_MEMORY_STAT_LAST: > continue; > } > val = libvirt_ulonglongWrap(stats[i].val); > -- Oh right. Apparently some time off is in place :-) Either of you care posting the patch?