> > what's SC and what's SC-to-VL? aren't you using SM to do such mappings? if this > and the below sysfs work-around/s for the maybe current lack of SM to > configure the port? why do the upstream kernel has to carry ~hundreds of sysfs > entries forever for such non-documented things? > This sysfs entries are used by PSM2 to form packets from user space. > > > > >> +HFI1_SL2SC_ATTR(0); > > same comment on SL2SC > > >> +HFI1_VL2MTU_ATTR(0); > > same comment on VL2MTU, why in sysfs?! > Same answer. > >> +HFI1_DIAGC_ATTR(rc_resends); > >> +HFI1_DIAGC_ATTR_PCPU(rc_acks, C_SW_CPU_RC_ACKS, > CNTR_INVALID_VL); > >> +HFI1_DIAGC_ATTR_PCPU(rc_qacks, C_SW_CPU_RC_QACKS, > CNTR_INVALID_VL); > >> +HFI1_DIAGC_ATTR_PCPU(rc_delayed_comp, > C_SW_CPU_RC_DELAYED_COMP, > >> + CNTR_INVALID_VL); > >> +HFI1_DIAGC_ATTR(seq_naks); > >> +HFI1_DIAGC_ATTR(rdma_seq); > >> +HFI1_DIAGC_ATTR(rnr_naks); > >> +HFI1_DIAGC_ATTR(other_naks); > >> +HFI1_DIAGC_ATTR(rc_timeouts); > >> +HFI1_DIAGC_ATTR(loop_pkts); > >> +HFI1_DIAGC_ATTR(pkt_drops); > >> +HFI1_DIAGC_ATTR(dmawait); > >> +HFI1_DIAGC_ATTR(unaligned); > >> +HFI1_DIAGC_ATTR(rc_dupreq); > >> +HFI1_DIAGC_ATTR(rc_seqnak); > The counters can be metered via other mechanisms. They will be removed in v3. ��.n��������+%������w��{.n�����{���fk��ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f