RE: [kbuild-all] [PATCH 4/4] scsi: hisi_sas: Add support for DIF/DIX feature for v3 hw

From: Li, Philip
Date: Tue Dec 04 2018 - 19:50:24 EST

Next message: Darwin Dingel: "[PATCH v2 1/2] serial: 8250: Add dts binding to rate limit serial port during overruns"
Previous message: Matthew Wilcox: "Re: [PATCH] /proc/kpagecount: return 0 for special pages that are never mapped"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Subject: Re: [kbuild-all] [PATCH 4/4] scsi: hisi_sas: Add support for DIF/DIX
> feature for v3 hw
>
> On 24/11/2018 17:50, kbuild test robot wrote:
> > Hi Xiang,
> >
> > Thank you for the patch! Perhaps something to improve:
> >
> > [auto build test WARNING on mkp-scsi/for-next]
> > [also build test WARNING on next-20181123]
> > [cannot apply to v4.20-rc3]
> > [if your patch is applied to the wrong git tree, please drop us a note to help
> improve the system]
> >
> > url: https://github.com/0day-ci/linux/commits/John-Garry/hisi_sas-DIF-DIX-
> support/20181123-202805
> > base: https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git for-next
> > config: x86_64-allmodconfig (attached as .config)
> > compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
> > reproduce:
> > # save the attached .config to linux build tree
> > make ARCH=x86_64
>
> For sparse errors - which are detected, below - we need C=2 option, right?
thanks John for the input, we will look into this to enhance the reproducing steps.

>
> It would be good to include that in the instructions.
>
> Thanks,
> John
>
> >
> > All warnings (new ones prefixed by >>):
> >
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: incorrect type in
> argument 1 (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: expected unsigned long
> long [unsigned] [usertype] val
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: got restricted __le64
> [usertype] sas_addr
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:26: warning: cast from restricted
> __le64
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:24: warning: incorrect type in
> assignment (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:24: expected restricted __le64
> [usertype] sas_addr
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:694:24: got unsigned long long
> >>> drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1165:26: warning: invalid assignment:
> |=
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1165:26: left side has type restricted
> __le32
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1165:26: right side has type int
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1166:26: warning: invalid assignment:
> |=
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1166:26: left side has type restricted
> __le32
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1166:26: right side has type int
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1759:37: warning: incorrect type in
> initializer (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1759:37: expected unsigned int
> [unsigned] [usertype] dma_rx_err_type
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1759:37: got restricted __le32
> [usertype] dma_rx_err_type
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1760:40: warning: incorrect type in
> initializer (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1760:40: expected unsigned int
> [unsigned] [usertype] trans_tx_fail_type
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1760:40: got restricted __le32
> [usertype] trans_tx_fail_type
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1767:40: warning: restricted __le32
> degrades to integer
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1781:40: warning: restricted __le32
> degrades to integer
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1840:30: warning: restricted __le32
> degrades to integer
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1864:26: warning: restricted __le32
> degrades to integer
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:1976:37: warning: restricted __le32
> degrades to integer
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2372:25: warning: symbol
> 'host_attrs_v3_hw' was not declared. Should it be static?
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2858:22: warning: incorrect type in
> assignment (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2858:22: expected unsigned int
> [unsigned] [usertype] device_state
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2858:22: got restricted pci_power_t
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2863:35: warning: incorrect type in
> argument 2 (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2863:35: expected restricted
> pci_power_t [usertype] state
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2863:35: got unsigned int [unsigned]
> [usertype] device_state
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2878:32: warning: incorrect type in
> initializer (different base types)
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2878:32: expected unsigned int
> [unsigned] [usertype] device_state
> > drivers/scsi/hisi_sas/hisi_sas_v3_hw.c:2878:32: got restricted pci_power_t
> [usertype] current_state
> >
> > vim +1165 drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
> >
> > 1072
> > 1073 static void prep_ssp_v3_hw(struct hisi_hba *hisi_hba,
> > 1074 struct hisi_sas_slot *slot)
> > 1075 {
> > 1076 struct sas_task *task = slot->task;
> > 1077 struct hisi_sas_cmd_hdr *hdr = slot->cmd_hdr;
> > 1078 struct domain_device *device = task->dev;
> > 1079 struct hisi_sas_device *sas_dev = device->lldd_dev;
> > 1080 struct hisi_sas_port *port = slot->port;
> > 1081 struct sas_ssp_task *ssp_task = &task->ssp_task;
> > 1082 struct scsi_cmnd *scsi_cmnd = ssp_task->cmd;
> > 1083 struct hisi_sas_tmf_task *tmf = slot->tmf;
> > 1084 int has_data = 0, priority = !!tmf;
> > 1085 u8 *buf_cmd;
> > 1086 u32 dw1 = 0, dw2 = 0, len = 0;
> > 1087
> > 1088 hdr->dw0 = cpu_to_le32((1 <<
> CMD_HDR_RESP_REPORT_OFF) |
> > 1089 (2 << CMD_HDR_TLR_CTRL_OFF) |
> > 1090 (port->id << CMD_HDR_PORT_OFF)
> |
> > 1091 (priority <<
> CMD_HDR_PRIORITY_OFF) |
> > 1092 (1 << CMD_HDR_CMD_OFF)); /*
> ssp */
> > 1093
> > 1094 dw1 = 1 << CMD_HDR_VDTL_OFF;
> > 1095 if (tmf) {
> > 1096 dw1 |= 2 << CMD_HDR_FRAME_TYPE_OFF;
> > 1097 dw1 |= DIR_NO_DATA << CMD_HDR_DIR_OFF;
> > 1098 } else {
> > 1099 dw1 |= 1 << CMD_HDR_FRAME_TYPE_OFF;
> > 1100 switch (scsi_cmnd->sc_data_direction) {
> > 1101 case DMA_TO_DEVICE:
> > 1102 has_data = 1;
> > 1103 dw1 |= DIR_TO_DEVICE <<
> CMD_HDR_DIR_OFF;
> > 1104 break;
> > 1105 case DMA_FROM_DEVICE:
> > 1106 has_data = 1;
> > 1107 dw1 |= DIR_TO_INI <<
> CMD_HDR_DIR_OFF;
> > 1108 break;
> > 1109 default:
> > 1110 dw1 &= ~CMD_HDR_DIR_MSK;
> > 1111 }
> > 1112 }
> > 1113
> > 1114 /* map itct entry */
> > 1115 dw1 |= sas_dev->device_id << CMD_HDR_DEV_ID_OFF;
> > 1116 hdr->dw1 = cpu_to_le32(dw1);
> > 1117
> > 1118 dw2 = (((sizeof(struct ssp_command_iu) + sizeof(struct
> ssp_frame_hdr)
> > 1119 + 3) / 4) << CMD_HDR_CFL_OFF) |
> > 1120 ((HISI_SAS_MAX_SSP_RESP_SZ / 4) <<
> CMD_HDR_MRFL_OFF) |
> > 1121 (2 << CMD_HDR_SG_MOD_OFF);
> > 1122 hdr->dw2 = cpu_to_le32(dw2);
> > 1123 hdr->transfer_tags = cpu_to_le32(slot->idx);
> > 1124
> > 1125 if (has_data) {
> > 1126 prep_prd_sge_v3_hw(hisi_hba, slot, hdr, task-
> >scatter,
> > 1127 slot->n_elem);
> > 1128
> > 1129 if (scsi_prot_sg_count(scsi_cmnd))
> > 1130 prep_prd_sge_dif_v3_hw(hisi_hba, slot,
> hdr,
> > 1131
> scsi_prot_sglist(scsi_cmnd),
> > 1132 slot->n_elem_dif);
> > 1133 }
> > 1134
> > 1135 hdr->cmd_table_addr =
> cpu_to_le64(hisi_sas_cmd_hdr_addr_dma(slot));
> > 1136 hdr->sts_buffer_addr =
> cpu_to_le64(hisi_sas_status_buf_addr_dma(slot));
> > 1137
> > 1138 buf_cmd = hisi_sas_cmd_hdr_addr_mem(slot) +
> > 1139 sizeof(struct ssp_frame_hdr);
> > 1140
> > 1141 memcpy(buf_cmd, &task->ssp_task.LUN, 8);
> > 1142 if (!tmf) {
> > 1143 buf_cmd[9] = ssp_task->task_attr | (ssp_task-
> >task_prio << 3);
> > 1144 memcpy(buf_cmd + 12, scsi_cmnd->cmnd,
> scsi_cmnd->cmd_len);
> > 1145 } else {
> > 1146 buf_cmd[10] = tmf->tmf;
> > 1147 switch (tmf->tmf) {
> > 1148 case TMF_ABORT_TASK:
> > 1149 case TMF_QUERY_TASK:
> > 1150 buf_cmd[12] =
> > 1151 (tmf-
> >tag_of_task_to_be_managed >> 8) & 0xff;
> > 1152 buf_cmd[13] =
> > 1153 tmf->tag_of_task_to_be_managed
> & 0xff;
> > 1154 break;
> > 1155 default:
> > 1156 break;
> > 1157 }
> > 1158 }
> > 1159
> > 1160 if (has_data && !scsi_prot_op_normal(scsi_cmnd)) {
> > 1161 struct hisi_sas_protect_iu_v3_hw prot;
> > 1162 unsigned char prot_op =
> scsi_get_prot_op(scsi_cmnd);
> > 1163 u8 *buf_cmd_prot;
> > 1164
> >> 1165 hdr->dw7 |= 1 <<
> CMD_HDR_ADDR_MODE_SEL_OFF;
> > 1166 hdr->dw1 |= CMD_HDR_PIR_MSK;
> > 1167 buf_cmd_prot =
> hisi_sas_cmd_hdr_addr_mem(slot) +
> > 1168 sizeof(struct ssp_frame_hdr) +
> > 1169 sizeof(struct ssp_command_iu);
> > 1170
> > 1171 memset(&prot, 0, sizeof(struct
> hisi_sas_protect_iu_v3_hw));
> > 1172 fill_prot_v3_hw(scsi_cmnd, &prot);
> > 1173 memcpy(buf_cmd_prot, &prot,
> > 1174 sizeof(struct hisi_sas_protect_iu_v3_hw));
> > 1175
> > 1176 if ((prot_op == SCSI_PROT_READ_INSERT) ||
> > 1177 (prot_op == SCSI_PROT_WRITE_INSERT) ||
> > 1178 (prot_op == SCSI_PROT_WRITE_PASS) ||
> > 1179 (prot_op == SCSI_PROT_READ_PASS)) {
> > 1180 unsigned int interval =
> scsi_prot_interval(scsi_cmnd);
> > 1181 unsigned int ilog2_interval =
> ilog2(interval);
> > 1182
> > 1183 len = (task->total_xfer_len >>
> ilog2_interval) * 8;
> > 1184 }
> > 1185 }
> > 1186
> > 1187 hdr->data_transfer_len = cpu_to_le32(task->total_xfer_len
> + len);
> > 1188 }
> > 1189
> >
> > ---
> > 0-DAY kernel test infrastructure Open Source Technology Center
> > https://lists.01.org/pipermail/kbuild-all Intel Corporation
> >
>
>
> _______________________________________________
> kbuild-all mailing list
> kbuild-all@xxxxxxxxxxxx
> https://lists.01.org/mailman/listinfo/kbuild-all

Next message: Darwin Dingel: "[PATCH v2 1/2] serial: 8250: Add dts binding to rate limit serial port during overruns"
Previous message: Matthew Wilcox: "Re: [PATCH] /proc/kpagecount: return 0 for special pages that are never mapped"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]