Re: [libvirt PATCH v2 00/11] virHashNew refactorings

On 7/6/21 2:37 PM, Tim Wiederhake wrote: > "virHashNew" cannot return NULL, yet we check for NULL in various places. > > This series is the first of several that remove these checks. Where > applicable, the functions are refactored to use automatic memory management > by means of g_autoptr etc. as well. > > v1: https://listman.redhat.com/archives/libvir-list/2021-July/msg00074.html > > Changes since v1: > * Fixed a memory leak that was introduced in patch #3 and fixed in patch #4 > * Split up patches 3 and 4 into three patches > > Regards, > Tim > > Tim Wiederhake (11): > qemuMonitorGetAllBlockStatsInfo: Clean up line break > qemuMonitorGetAllBlockStatsInfo: Remove superfluous variable > initialization > qemuMonitorGetAllBlockStatsInfo: Assign hash table only on success > qemuMonitorGetAllBlockStatsInfo: Use automatic memory management > qemuMonitorGetAllBlockStatsInfo: `virHashNew` cannot return NULL > qemuMonitorGetBlockInfo: Remove superfluous variable "ret" > qemuMonitorGetBlockInfo: Use automatic memory management > qemuMonitorGetBlockInfo: `virHashNew` cannot return NULL > qemuMonitorGetChardevInfo: Remove superfluous variable "ret" > qemuMonitorGetChardevInfo: Use automatic memory management > qemuMonitorGetChardevInfo: `virHashNew` cannot return NULL > > src/qemu/qemu_monitor.c | 53 ++++++++++++----------------------------- > 1 file changed, 15 insertions(+), 38 deletions(-) > Looks good to me. Please let me know if you agree with my suggestion in 10/11. I can fix that before pushing. Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com&gt;