On Mon, Oct 28, 2013 at 11:02:59AM +0800, Dave Young wrote: > On 10/28/13 at 09:12am, Dave Young wrote: > > > > + size = nr_maps * sizeof(struct efi_mem_descriptor) + sizeof(struct efi_setup_data); > > > > + sd = malloc(sizeof(struct setup_data) + size); > > > > + if (!sd) { > > > > + free(esd); > > > > + return; > > > > + } > > > > > > This function appears to leak sd. > > > > Will fix > > sd will be passed to add_buffer, it should be freed after kexec_load, > is it really necessary, or I misunderstand something? Thanks and sorry for missing that. I now think the current handling of sd in this patch seems fine.