Hi Bart, On Sun, Jun 14, 2020 at 03:39:19PM -0700, Bart Van Assche wrote: > @@ -1063,7 +1063,8 @@ qla2100_fw_dump(scsi_qla_host_t *vha) > } > > if (rval == QLA_SUCCESS) > - qla2xxx_copy_queues(ha, &fw->risc_ram[cnt]); > + qla2xxx_copy_queues(ha, (char *)fw + > + offsetof(typeof(*fw), risc_ram) + cnt); This looks pretty ugly to me. Any chance to write this in a way it's understandable by humans and coverity is not annoyed? Do I understand it correctly, it's valid to read after the end of risc_ram? Thanks, Daniel
