On Tue, Jun 05, 2018 at 03:36:14PM +0200, Michal Privoznik wrote: > On 06/05/2018 03:08 PM, Erik Skultety wrote: > > On Tue, Jun 05, 2018 at 01:41:02PM +0200, Tomáš Golembiovský wrote: > >> Signed-off-by: Tomáš Golembiovský <tgolembi@xxxxxxxxxx> > >> --- > >> include/libvirt/libvirt-domain.h | 9 ++++++++- > >> src/qemu/qemu_monitor_json.c | 2 ++ > >> 2 files changed, 10 insertions(+), 1 deletion(-) > >> > >> diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h > >> index da773b76cb..b96c018a90 100644 > >> --- a/include/libvirt/libvirt-domain.h > >> +++ b/include/libvirt/libvirt-domain.h > >> @@ -628,11 +628,18 @@ typedef enum { > >> /* Timestamp of the last update of statistics, in seconds. */ > >> VIR_DOMAIN_MEMORY_STAT_LAST_UPDATE = 9, > >> > >> + /* > >> + * The amount of memory, in bytes, that can be quickly reclaimed without > >> + * additional I/O. Typically these pages are used for caching files from > >> + * disk. > >> + */ > >> + VIR_DOMAIN_MEMORY_STAT_DISK_CACHES = 10, > >> + > >> /* > >> * The number of statistics supported by this version of the interface. > >> * To add new statistics, add them to the enum and increase this value. > >> */ > >> - VIR_DOMAIN_MEMORY_STAT_NR = 10, > >> + VIR_DOMAIN_MEMORY_STAT_NR = 11, > > > > this is a public header, this must never change, otherwise you break backwards > > compatibility... > > Not true. This is meant to work roughly like this: > > virDomainMemoryStatStruct stats[VIR_DOMAIN_MEMORY_STAT_NR]; > > nr_stats = virDomainMemoryStats(dom, stats, VIR_DOMAIN_MEMORY_STAT_NR, 0); > > And depending what version you are compiled with you will get different > number of results. Important to say that to maintain backward > compatibility we have a rule that says _NR can only grow and never > shrink. Tomas' patch is actually correct (in this aspect). > View examples: Doh! If only I read the commentary right above the enum it would tell me the very same thing, I have to admit, I didn't even read properly, I saw a value being changed and I immediately responded, sorry, I take it back. Erik -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list