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

From: John Garry
Date: Tue Nov 20 2018 - 09:58:37 EST

From: Xiang Chen <chenxiang66@xxxxxxxxxxxxx>

For v3 hw, we support DIF/DIX operation for SAS, but not SATA.

In addition, DIF CRC16 is supported.

This patchset adds the SW support for the described features. The main
components are as follows:
- Allocate memory for PI
- Fill PI fields
- Fill related to DIF/DIX in DQ and protection iu memories

Signed-off-by: Xiang Chen <chenxiang66@xxxxxxxxxxxxx>
Signed-off-by: John Garry <john.garry@xxxxxxxxxx>
---
drivers/scsi/hisi_sas/hisi_sas.h | 18 ++++
drivers/scsi/hisi_sas/hisi_sas_main.c | 99 ++++++++++++++---
drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 189 ++++++++++++++++++++++++++++++++-
3 files changed, 289 insertions(+), 17 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index 535c613..a73aad6 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -55,6 +55,11 @@
#define hisi_sas_sge_addr_mem(slot) hisi_sas_sge_addr(slot->buf)
#define hisi_sas_sge_addr_dma(slot) hisi_sas_sge_addr(slot->buf_dma)

+#define hisi_sas_sge_dif_addr(buf) \
+ (buf + offsetof(struct hisi_sas_slot_dif_buf_table, sge_dif_page))
+#define hisi_sas_sge_dif_addr_mem(slot) hisi_sas_sge_dif_addr(slot->buf)
+#define hisi_sas_sge_dif_addr_dma(slot) hisi_sas_sge_dif_addr(slot->buf_dma)
+
#define HISI_SAS_MAX_SSP_RESP_SZ (sizeof(struct ssp_frame_hdr) + 1024)
#define HISI_SAS_MAX_SMP_RESP_SZ 1028
#define HISI_SAS_MAX_STP_RESP_SZ 28
@@ -197,6 +202,7 @@ struct hisi_sas_slot {
struct sas_task *task;
struct hisi_sas_port *port;
u64 n_elem;
+ u64 n_elem_dif;
int dlvry_queue;
int dlvry_queue_slot;
int cmplt_queue;
@@ -268,6 +274,8 @@ struct hisi_hba {
struct pci_dev *pci_dev;
struct device *dev;

+ bool enable_dif_dix;
+
void __iomem *regs;
void __iomem *sgpio_regs;
struct regmap *ctrl;
@@ -422,6 +430,11 @@ struct hisi_sas_sge_page {
struct hisi_sas_sge sge[HISI_SAS_SGE_PAGE_CNT];
} __aligned(16);

+#define HISI_SAS_SGE_DIF_PAGE_CNT SG_CHUNK_SIZE
+struct hisi_sas_sge_dif_page {
+ struct hisi_sas_sge sge[HISI_SAS_SGE_DIF_PAGE_CNT];
+} __aligned(16);
+
struct hisi_sas_command_table_ssp {
struct ssp_frame_hdr hdr;
union {
@@ -452,6 +465,11 @@ struct hisi_sas_slot_buf_table {
struct hisi_sas_sge_page sge_page;
};

+struct hisi_sas_slot_dif_buf_table {
+ struct hisi_sas_slot_buf_table slot_buf;
+ struct hisi_sas_sge_dif_page sge_dif_page;
+};
+
extern struct scsi_transport_template *hisi_sas_stt;
extern void hisi_sas_stop_phys(struct hisi_hba *hisi_hba);
extern int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost);
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index cbda48e..d0b693b 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -252,14 +252,21 @@ void hisi_sas_slot_task_free(struct hisi_hba *hisi_hba, struct sas_task *task,

task->lldd_task = NULL;

- if (!sas_protocol_ata(task->task_proto))
+ if (!sas_protocol_ata(task->task_proto)) {
+ struct sas_ssp_task *ssp_task = &task->ssp_task;
+ struct scsi_cmnd *scsi_cmnd = ssp_task->cmd;
+
if (slot->n_elem)
dma_unmap_sg(dev, task->scatter,
task->num_scatter,
task->data_dir);
+ if (slot->n_elem_dif)
+ dma_unmap_sg(dev, scsi_prot_sglist(scsi_cmnd),
+ scsi_prot_sg_count(scsi_cmnd),
+ task->data_dir);
+ }
}

-
spin_lock_irqsave(&dq->lock, flags);
list_del_init(&slot->entry);
spin_unlock_irqrestore(&dq->lock, flags);
@@ -380,6 +387,59 @@ static int hisi_sas_dma_map(struct hisi_hba *hisi_hba,
return rc;
}

+static void hisi_sas_dif_dma_unmap(struct hisi_hba *hisi_hba,
+ struct sas_task *task, int n_elem_dif)
+{
+ struct device *dev = hisi_hba->dev;
+
+ if (n_elem_dif) {
+ struct sas_ssp_task *ssp_task = &task->ssp_task;
+ struct scsi_cmnd *scsi_cmnd = ssp_task->cmd;
+
+ dma_unmap_sg(dev, scsi_prot_sglist(scsi_cmnd),
+ scsi_prot_sg_count(scsi_cmnd),
+ task->data_dir);
+ }
+}
+
+static int hisi_sas_dif_dma_map(struct hisi_hba *hisi_hba,
+ int *n_elem_dif, struct sas_task *task)
+{
+ struct device *dev = hisi_hba->dev;
+ struct sas_ssp_task *ssp_task;
+ struct scsi_cmnd *scsi_cmnd;
+ int rc;
+
+ if (task->num_scatter) {
+ ssp_task = &task->ssp_task;
+ scsi_cmnd = ssp_task->cmd;
+
+ if (scsi_prot_sg_count(scsi_cmnd)) {
+ *n_elem_dif = dma_map_sg(dev,
+ scsi_prot_sglist(scsi_cmnd),
+ scsi_prot_sg_count(scsi_cmnd),
+ task->data_dir);
+
+ if (!*n_elem_dif)
+ return -ENOMEM;
+
+ if (*n_elem_dif > HISI_SAS_SGE_DIF_PAGE_CNT) {
+ dev_err(dev, "task prep: n_elem_dif(%d) too large\n",
+ *n_elem_dif);
+ rc = -EINVAL;
+ goto err_out_dif_dma_unmap;
+ }
+ }
+ }
+
+ return 0;
+
+err_out_dif_dma_unmap:
+ dma_unmap_sg(dev, scsi_prot_sglist(scsi_cmnd),
+ scsi_prot_sg_count(scsi_cmnd), task->data_dir);
+ return rc;
+}
+
static int hisi_sas_task_prep(struct sas_task *task,
struct hisi_sas_dq **dq_pointer,
bool is_tmf, struct hisi_sas_tmf_task *tmf,
@@ -394,7 +454,7 @@ static int hisi_sas_task_prep(struct sas_task *task,
struct asd_sas_port *sas_port = device->port;
struct device *dev = hisi_hba->dev;
int dlvry_queue_slot, dlvry_queue, rc, slot_idx;
- int n_elem = 0, n_elem_req = 0, n_elem_resp = 0;
+ int n_elem = 0, n_elem_dif = 0, n_elem_req = 0, n_elem_resp = 0;
struct hisi_sas_dq *dq;
unsigned long flags;
int wr_q_index;
@@ -427,6 +487,12 @@ static int hisi_sas_task_prep(struct sas_task *task,
if (rc < 0)
goto prep_out;

+ if (!sas_protocol_ata(task->task_proto)) {
+ rc = hisi_sas_dif_dma_map(hisi_hba, &n_elem_dif, task);
+ if (rc < 0)
+ goto err_out_dma_unmap;
+ }
+
if (hisi_hba->hw->slot_index_alloc)
rc = hisi_hba->hw->slot_index_alloc(hisi_hba, device);
else {
@@ -445,7 +511,7 @@ static int hisi_sas_task_prep(struct sas_task *task,
rc = hisi_sas_slot_index_alloc(hisi_hba, scsi_cmnd);
}
if (rc < 0)
- goto err_out_dma_unmap;
+ goto err_out_dif_dma_unmap;

slot_idx = rc;
slot = &hisi_hba->slot_info[slot_idx];
@@ -466,6 +532,7 @@ static int hisi_sas_task_prep(struct sas_task *task,
dlvry_queue_slot = wr_q_index;

slot->n_elem = n_elem;
+ slot->n_elem_dif = n_elem_dif;
slot->dlvry_queue = dlvry_queue;
slot->dlvry_queue_slot = dlvry_queue_slot;
cmd_hdr_base = hisi_hba->cmd_hdr[dlvry_queue];
@@ -509,6 +576,9 @@ static int hisi_sas_task_prep(struct sas_task *task,

err_out_tag:
hisi_sas_slot_index_free(hisi_hba, slot_idx);
+err_out_dif_dma_unmap:
+ if (!sas_protocol_ata(task->task_proto))
+ hisi_sas_dif_dma_unmap(hisi_hba, task, n_elem_dif);
err_out_dma_unmap:
hisi_sas_dma_unmap(hisi_hba, task, n_elem,
n_elem_req, n_elem_resp);
@@ -2142,21 +2212,26 @@ int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
if (!hisi_hba->slot_info)
goto err_out;

- /* roundup to avoid overly large block size */
+ /* roundup to avoid an overly large block size */
max_command_entries_ru = roundup(max_command_entries, 64);
- sz_slot_buf_ru = roundup(sizeof(struct hisi_sas_slot_buf_table), 64);
+ if (hisi_hba->enable_dif_dix)
+ sz_slot_buf_ru = sizeof(struct hisi_sas_slot_dif_buf_table);
+ else
+ sz_slot_buf_ru = sizeof(struct hisi_sas_slot_buf_table);
+ sz_slot_buf_ru = roundup(sz_slot_buf_ru, 64);
s = lcm(max_command_entries_ru, sz_slot_buf_ru);
blk_cnt = (max_command_entries_ru * sz_slot_buf_ru) / s;
slots_per_blk = s / sz_slot_buf_ru;
+
for (i = 0; i < blk_cnt; i++) {
- struct hisi_sas_slot_buf_table *buf;
- dma_addr_t buf_dma;
int slot_index = i * slots_per_blk;
+ dma_addr_t buf_dma;
+ void *buf;

- buf = dmam_alloc_coherent(dev, s, &buf_dma, GFP_KERNEL);
+ buf = dmam_alloc_coherent(dev, s, &buf_dma,
+ GFP_KERNEL | __GFP_ZERO);
if (!buf)
goto err_out;
- memset(buf, 0, s);

for (j = 0; j < slots_per_blk; j++, slot_index++) {
struct hisi_sas_slot *slot;
@@ -2166,8 +2241,8 @@ int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
slot->buf_dma = buf_dma;
slot->idx = slot_index;

- buf++;
- buf_dma += sizeof(*buf);
+ buf += sz_slot_buf_ru;
+ buf_dma += sz_slot_buf_ru;
}
}

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
index 8a08078..314918c 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
@@ -127,6 +127,8 @@
#define PHY_CTRL (PORT_BASE + 0x14)
#define PHY_CTRL_RESET_OFF 0
#define PHY_CTRL_RESET_MSK (0x1 << PHY_CTRL_RESET_OFF)
+#define CMD_HDR_PIR_OFF 8
+#define CMD_HDR_PIR_MSK (0x1 << CMD_HDR_PIR_OFF)
#define SL_CFG (PORT_BASE + 0x84)
#define AIP_LIMIT (PORT_BASE + 0x90)
#define SL_CONTROL (PORT_BASE + 0x94)
@@ -333,6 +335,16 @@
#define ITCT_HDR_RTOLT_OFF 48
#define ITCT_HDR_RTOLT_MSK (0xffffULL << ITCT_HDR_RTOLT_OFF)

+struct hisi_sas_protect_iu_v3_hw {
+ u32 dw0;
+ u32 lbrtcv;
+ u32 lbrtgv;
+ u32 dw3;
+ u32 dw4;
+ u32 dw5;
+ u32 rsv;
+};
+
struct hisi_sas_complete_v3_hdr {
__le32 dw0;
__le32 dw1;
@@ -372,9 +384,27 @@ struct hisi_sas_err_record_v3 {
((fis.command == ATA_CMD_DEV_RESET) && \
((fis.control & ATA_SRST) != 0)))

+#define T10_INSRT_EN_OFF 0
+#define T10_INSRT_EN_MSK (1 << T10_INSRT_EN_OFF)
+#define T10_RMV_EN_OFF 1
+#define T10_RMV_EN_MSK (1 << T10_RMV_EN_OFF)
+#define T10_RPLC_EN_OFF 2
+#define T10_RPLC_EN_MSK (1 << T10_RPLC_EN_OFF)
+#define T10_CHK_EN_OFF 3
+#define T10_CHK_EN_MSK (1 << T10_CHK_EN_OFF)
+#define INCR_LBRT_OFF 5
+#define INCR_LBRT_MSK (1 << INCR_LBRT_OFF)
+#define USR_DATA_BLOCK_SZ_OFF 20
+#define USR_DATA_BLOCK_SZ_MSK (0x3 << USR_DATA_BLOCK_SZ_OFF)
+#define T10_CHK_MSK_OFF 16
+
static bool hisi_sas_intr_conv;
MODULE_PARM_DESC(intr_conv, "interrupt converge enable (0-1)");

+static bool enable_dif_dix;
+module_param(enable_dif_dix, bool, 0444);
+MODULE_PARM_DESC(enable_dif_dix, "DIF/DIX enable (0-1)");
+
static u32 hisi_sas_read32(struct hisi_hba *hisi_hba, u32 off)
{
void __iomem *regs = hisi_hba->regs + off;
@@ -937,7 +967,107 @@ static void prep_prd_sge_v3_hw(struct hisi_hba *hisi_hba,

hdr->prd_table_addr = cpu_to_le64(hisi_sas_sge_addr_dma(slot));

- hdr->sg_len = cpu_to_le32(n_elem << CMD_HDR_DATA_SGL_LEN_OFF);
+ hdr->sg_len |= cpu_to_le32(n_elem << CMD_HDR_DATA_SGL_LEN_OFF);
+}
+
+static void prep_prd_sge_dif_v3_hw(struct hisi_hba *hisi_hba,
+ struct hisi_sas_slot *slot,
+ struct hisi_sas_cmd_hdr *hdr,
+ struct scatterlist *scatter,
+ int n_elem)
+{
+ struct hisi_sas_sge_dif_page *sge_dif_page;
+ struct scatterlist *sg;
+ int i;
+
+ sge_dif_page = hisi_sas_sge_dif_addr_mem(slot);
+
+ for_each_sg(scatter, sg, n_elem, i) {
+ struct hisi_sas_sge *entry = &sge_dif_page->sge[i];
+
+ entry->addr = cpu_to_le64(sg_dma_address(sg));
+ entry->page_ctrl_0 = entry->page_ctrl_1 = 0;
+ entry->data_len = cpu_to_le32(sg_dma_len(sg));
+ entry->data_off = 0;
+ }
+
+ hdr->dif_prd_table_addr = cpu_to_le64(
+ hisi_sas_sge_dif_addr_dma(slot));
+
+ hdr->sg_len |= cpu_to_le32(n_elem << CMD_HDR_DIF_SGL_LEN_OFF);
+}
+
+static void fill_prot_v3_hw(struct scsi_cmnd *scsi_cmnd,
+ struct hisi_sas_protect_iu_v3_hw *prot)
+{
+ u8 prot_type = scsi_get_prot_type(scsi_cmnd);
+ u8 prot_op = scsi_get_prot_op(scsi_cmnd);
+ unsigned int interval = scsi_prot_interval(scsi_cmnd);
+ u32 lbrt_chk_val;
+
+ if (interval == 4096)
+ lbrt_chk_val = (u32)(scsi_get_lba(scsi_cmnd) >> 3);
+ else
+ lbrt_chk_val = (u32)scsi_get_lba(scsi_cmnd);
+
+ switch (prot_op) {
+ case SCSI_PROT_READ_INSERT:
+ prot->dw0 |= T10_INSRT_EN_MSK;
+ prot->lbrtgv = lbrt_chk_val;
+ break;
+ case SCSI_PROT_READ_STRIP:
+ prot->dw0 |= (T10_RMV_EN_MSK | T10_CHK_EN_MSK);
+ prot->lbrtcv = lbrt_chk_val;
+ if (prot_type == SCSI_PROT_DIF_TYPE1)
+ prot->dw4 |= (0xc << 16);
+ else if (prot_type == SCSI_PROT_DIF_TYPE3)
+ prot->dw4 |= (0xfc << 16);
+ break;
+ case SCSI_PROT_READ_PASS:
+ prot->dw0 |= T10_CHK_EN_MSK;
+ prot->lbrtcv = lbrt_chk_val;
+ if (prot_type == SCSI_PROT_DIF_TYPE1)
+ prot->dw4 |= (0xc << 16);
+ else if (prot_type == SCSI_PROT_DIF_TYPE3)
+ prot->dw4 |= (0xfc << 16);
+ break;
+ case SCSI_PROT_WRITE_INSERT:
+ prot->dw0 |= T10_INSRT_EN_MSK;
+ prot->lbrtgv = lbrt_chk_val;
+ break;
+ case SCSI_PROT_WRITE_STRIP:
+ prot->dw0 |= (T10_RMV_EN_MSK | T10_CHK_EN_MSK);
+ prot->lbrtcv = lbrt_chk_val;
+ break;
+ case SCSI_PROT_WRITE_PASS:
+ prot->dw0 |= T10_CHK_EN_MSK;
+ prot->lbrtcv = lbrt_chk_val;
+ if (prot_type == SCSI_PROT_DIF_TYPE1)
+ prot->dw4 |= (0xc << 16);
+ else if (prot_type == SCSI_PROT_DIF_TYPE3)
+ prot->dw4 |= (0xfc << 16);
+ break;
+ default:
+ WARN(1, "prot_op(0x%x) is not valid\n", prot_op);
+ break;
+ }
+
+ switch (interval) {
+ case 512:
+ break;
+ case 4096:
+ prot->dw0 |= (0x1 << USR_DATA_BLOCK_SZ_OFF);
+ break;
+ case 520:
+ prot->dw0 |= (0x2 << USR_DATA_BLOCK_SZ_OFF);
+ break;
+ default:
+ WARN(1, "protection interval (0x%x) invalid\n",
+ interval);
+ break;
+ }
+
+ prot->dw0 |= INCR_LBRT_MSK;
}

static void prep_ssp_v3_hw(struct hisi_hba *hisi_hba,
@@ -953,7 +1083,7 @@ static void prep_ssp_v3_hw(struct hisi_hba *hisi_hba,
struct hisi_sas_tmf_task *tmf = slot->tmf;
int has_data = 0, priority = !!tmf;
u8 *buf_cmd;
- u32 dw1 = 0, dw2 = 0;
+ u32 dw1 = 0, dw2 = 0, len = 0;

hdr->dw0 = cpu_to_le32((1 << CMD_HDR_RESP_REPORT_OFF) |
(2 << CMD_HDR_TLR_CTRL_OFF) |
@@ -992,11 +1122,16 @@ static void prep_ssp_v3_hw(struct hisi_hba *hisi_hba,
hdr->dw2 = cpu_to_le32(dw2);
hdr->transfer_tags = cpu_to_le32(slot->idx);

- if (has_data)
+ if (has_data) {
prep_prd_sge_v3_hw(hisi_hba, slot, hdr, task->scatter,
- slot->n_elem);
+ slot->n_elem);
+
+ if (scsi_prot_sg_count(scsi_cmnd))
+ prep_prd_sge_dif_v3_hw(hisi_hba, slot, hdr,
+ scsi_prot_sglist(scsi_cmnd),
+ slot->n_elem_dif);
+ }

- hdr->data_transfer_len = cpu_to_le32(task->total_xfer_len);
hdr->cmd_table_addr = cpu_to_le64(hisi_sas_cmd_hdr_addr_dma(slot));
hdr->sts_buffer_addr = cpu_to_le64(hisi_sas_status_buf_addr_dma(slot));

@@ -1021,6 +1156,35 @@ static void prep_ssp_v3_hw(struct hisi_hba *hisi_hba,
break;
}
}
+
+ if (has_data && !scsi_prot_op_normal(scsi_cmnd)) {
+ struct hisi_sas_protect_iu_v3_hw prot;
+ unsigned char prot_op = scsi_get_prot_op(scsi_cmnd);
+ u8 *buf_cmd_prot;
+
+ hdr->dw7 |= 1 << CMD_HDR_ADDR_MODE_SEL_OFF;
+ hdr->dw1 |= CMD_HDR_PIR_MSK;
+ buf_cmd_prot = hisi_sas_cmd_hdr_addr_mem(slot) +
+ sizeof(struct ssp_frame_hdr) +
+ sizeof(struct ssp_command_iu);
+
+ memset(&prot, 0, sizeof(struct hisi_sas_protect_iu_v3_hw));
+ fill_prot_v3_hw(scsi_cmnd, &prot);
+ memcpy(buf_cmd_prot, &prot,
+ sizeof(struct hisi_sas_protect_iu_v3_hw));
+
+ if ((prot_op == SCSI_PROT_READ_INSERT) ||
+ (prot_op == SCSI_PROT_WRITE_INSERT) ||
+ (prot_op == SCSI_PROT_WRITE_PASS) ||
+ (prot_op == SCSI_PROT_READ_PASS)) {
+ unsigned int interval = scsi_prot_interval(scsi_cmnd);
+ unsigned int ilog2_interval = ilog2(interval);
+
+ len = (task->total_xfer_len >> ilog2_interval) * 8;
+ }
+ }
+
+ hdr->data_transfer_len = cpu_to_le32(task->total_xfer_len + len);
}

static void prep_smp_v3_hw(struct hisi_hba *hisi_hba,
@@ -2225,6 +2389,7 @@ struct device_attribute *host_attrs_v3_hw[] = {
.bios_param = sas_bios_param,
.this_id = -1,
.sg_tablesize = HISI_SAS_SGE_PAGE_CNT,
+ .sg_prot_tablesize = HISI_SAS_SGE_PAGE_CNT,
.max_sectors = SCSI_DEFAULT_MAX_SECTORS,
.use_clustering = ENABLE_CLUSTERING,
.eh_device_reset_handler = sas_eh_device_reset_handler,
@@ -2284,6 +2449,7 @@ struct device_attribute *host_attrs_v3_hw[] = {
hisi_hba->dev = dev;
hisi_hba->shost = shost;
SHOST_TO_SAS_HA(shost) = &hisi_hba->sha;
+ hisi_hba->enable_dif_dix = enable_dif_dix;

timer_setup(&hisi_hba->timer, NULL, 0);

@@ -2395,6 +2561,19 @@ struct device_attribute *host_attrs_v3_hw[] = {
if (rc)
goto err_out_register_ha;

+ if (hisi_hba->enable_dif_dix) {
+ dev_info(dev, "Registering for DIF/DIX type 1/2/3 protection.\n");
+ scsi_host_set_prot(hisi_hba->shost,
+ SHOST_DIF_TYPE1_PROTECTION |
+ SHOST_DIF_TYPE2_PROTECTION |
+ SHOST_DIF_TYPE3_PROTECTION |
+ SHOST_DIX_TYPE1_PROTECTION |
+ SHOST_DIX_TYPE2_PROTECTION |
+ SHOST_DIX_TYPE3_PROTECTION);
+ scsi_host_set_guard(hisi_hba->shost,
+ SHOST_DIX_GUARD_CRC);
+ }
+
scsi_scan_host(shost);

return 0;
--
1.9.1