[PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[John Garry]

This patchset introduces some misc bug fixes, improvements,
and new features to the HiSilicon SAS driver.

The new features include:
- ECC and AXI bus fatal error handling for v2 hw
- PHY (min and max) linkrate setting

John Garry (1):
  hisi_sas: use atomic64_t for hisi_sas_device.running_req

Xiang Chen (10):
  hisi_sas: add v2 hw support for ECC and AXI bus fatal error
  hisi_sas: alloc queue id of slot according to device id
  hisi_sas: only process broadcast change in phy_bcast_v2_hw()
  hisi_sas: fix port form bug in hisi_sas_port_notify_formed()
  hisi_sas: replace WARN_ON() with dev_warn() for internal abort
  hisi_sas: modify return value of hisi_sas_query_task()
  hisi_sas: delete repeated configuration in free_device_v2_hw()
  hisi_sas: modify some values in get_ata_protocol()
  hisi_sas: check SATA FIS when directly attaching SATA device
  hisi_sas: add PHY set linkrate support for v1 and v2 hw

 drivers/scsi/hisi_sas/hisi_sas.h       |   9 +-
 drivers/scsi/hisi_sas/hisi_sas_main.c  |  37 ++-
 drivers/scsi/hisi_sas/hisi_sas_v1_hw.c |  79 +++--
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 539 +++++++++++++++++++++++++++++++--
 4 files changed, 599 insertions(+), 65 deletions(-)

-- 
1.9.1

[PATCH 01/11] hisi_sas: add v2 hw support for ECC and AXI bus fatal error

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

For ECC 1bit error, logic can recover it, so we only print
a warning.
For ECC multi-bit and AXI bus fatal error, we panic.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 425 ++++++++++++++++++++++++++++++++-
 1 file changed, 420 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 0763b47..9e70e6d 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -55,10 +55,44 @@
 #define HGC_DFX_CFG2			0xc0
 #define HGC_IOMB_PROC1_STATUS	0x104
 #define CFG_1US_TIMER_TRSH		0xcc
+#define HGC_LM_DFX_STATUS2		0x128
+#define HGC_LM_DFX_STATUS2_IOSTLIST_OFF		0
+#define HGC_LM_DFX_STATUS2_IOSTLIST_MSK	(0xfff << \
+					 HGC_LM_DFX_STATUS2_IOSTLIST_OFF)
+#define HGC_LM_DFX_STATUS2_ITCTLIST_OFF		12
+#define HGC_LM_DFX_STATUS2_ITCTLIST_MSK	(0x7ff << \
+					 HGC_LM_DFX_STATUS2_ITCTLIST_OFF)
+#define HGC_CQE_ECC_ADDR		0x13c
+#define HGC_CQE_ECC_1B_ADDR_OFF	0
+#define HGC_CQE_ECC_1B_ADDR_MSK	(0x3f < HGC_CQE_ECC_1B_ADDR_OFF)
+#define HGC_CQE_ECC_MB_ADDR_OFF	8
+#define HGC_CQE_ECC_MB_ADDR_MSK (0x3f < HGC_CQE_ECC_MB_ADDR_OFF)
+#define HGC_IOST_ECC_ADDR		0x140
+#define HGC_IOST_ECC_1B_ADDR_OFF	0
+#define HGC_IOST_ECC_1B_ADDR_MSK	(0x3ff < HGC_IOST_ECC_1B_ADDR_OFF)
+#define HGC_IOST_ECC_MB_ADDR_OFF	16
+#define HGC_IOST_ECC_MB_ADDR_MSK	(0x3ff < HGC_IOST_ECC_MB_ADDR_OFF)
+#define HGC_DQE_ECC_ADDR		0x144
+#define HGC_DQE_ECC_1B_ADDR_OFF	0
+#define HGC_DQE_ECC_1B_ADDR_MSK	(0xfff < HGC_DQE_ECC_1B_ADDR_OFF)
+#define HGC_DQE_ECC_MB_ADDR_OFF	16
+#define HGC_DQE_ECC_MB_ADDR_MSK (0xfff < HGC_DQE_ECC_MB_ADDR_OFF)
 #define HGC_INVLD_DQE_INFO		0x148
 #define HGC_INVLD_DQE_INFO_FB_CH0_OFF	9
 #define HGC_INVLD_DQE_INFO_FB_CH0_MSK	(0x1 << HGC_INVLD_DQE_INFO_FB_CH0_OFF)
 #define HGC_INVLD_DQE_INFO_FB_CH3_OFF	18
+#define HGC_ITCT_ECC_ADDR		0x150
+#define HGC_ITCT_ECC_1B_ADDR_OFF		0
+#define HGC_ITCT_ECC_1B_ADDR_MSK		(0x3ff << \
+						 HGC_ITCT_ECC_1B_ADDR_OFF)
+#define HGC_ITCT_ECC_MB_ADDR_OFF		16
+#define HGC_ITCT_ECC_MB_ADDR_MSK		(0x3ff << \
+						 HGC_ITCT_ECC_MB_ADDR_OFF)
+#define HGC_AXI_FIFO_ERR_INFO	0x154
+#define AXI_ERR_INFO_OFF		0
+#define AXI_ERR_INFO_MSK		(0xff << AXI_ERR_INFO_OFF)
+#define FIFO_ERR_INFO_OFF		8
+#define FIFO_ERR_INFO_MSK		(0xff << FIFO_ERR_INFO_OFF)
 #define INT_COAL_EN			0x19c
 #define OQ_INT_COAL_TIME		0x1a0
 #define OQ_INT_COAL_CNT			0x1a4
@@ -73,13 +107,41 @@
 #define ENT_INT_SRC1_D2H_FIS_CH1_MSK	(0x1 << ENT_INT_SRC1_D2H_FIS_CH1_OFF)
 #define ENT_INT_SRC2			0x1bc
 #define ENT_INT_SRC3			0x1c0
+#define ENT_INT_SRC3_WP_DEPTH_OFF		8
+#define ENT_INT_SRC3_IPTT_SLOT_NOMATCH_OFF	9
+#define ENT_INT_SRC3_RP_DEPTH_OFF		10
+#define ENT_INT_SRC3_AXI_OFF			11
+#define ENT_INT_SRC3_FIFO_OFF			12
+#define ENT_INT_SRC3_LM_OFF				14
 #define ENT_INT_SRC3_ITC_INT_OFF	15
 #define ENT_INT_SRC3_ITC_INT_MSK	(0x1 << ENT_INT_SRC3_ITC_INT_OFF)
+#define ENT_INT_SRC3_ABT_OFF		16
 #define ENT_INT_SRC_MSK1		0x1c4
 #define ENT_INT_SRC_MSK2		0x1c8
 #define ENT_INT_SRC_MSK3		0x1cc
 #define ENT_INT_SRC_MSK3_ENT95_MSK_OFF	31
 #define ENT_INT_SRC_MSK3_ENT95_MSK_MSK	(0x1 << ENT_INT_SRC_MSK3_ENT95_MSK_OFF)
+#define SAS_ECC_INTR			0x1e8
+#define SAS_ECC_INTR_DQE_ECC_1B_OFF		0
+#define SAS_ECC_INTR_DQE_ECC_MB_OFF		1
+#define SAS_ECC_INTR_IOST_ECC_1B_OFF	2
+#define SAS_ECC_INTR_IOST_ECC_MB_OFF	3
+#define SAS_ECC_INTR_ITCT_ECC_MB_OFF	4
+#define SAS_ECC_INTR_ITCT_ECC_1B_OFF	5
+#define SAS_ECC_INTR_IOSTLIST_ECC_MB_OFF	6
+#define SAS_ECC_INTR_IOSTLIST_ECC_1B_OFF	7
+#define SAS_ECC_INTR_ITCTLIST_ECC_1B_OFF	8
+#define SAS_ECC_INTR_ITCTLIST_ECC_MB_OFF	9
+#define SAS_ECC_INTR_CQE_ECC_1B_OFF		10
+#define SAS_ECC_INTR_CQE_ECC_MB_OFF		11
+#define SAS_ECC_INTR_NCQ_MEM0_ECC_MB_OFF	12
+#define SAS_ECC_INTR_NCQ_MEM0_ECC_1B_OFF	13
+#define SAS_ECC_INTR_NCQ_MEM1_ECC_MB_OFF	14
+#define SAS_ECC_INTR_NCQ_MEM1_ECC_1B_OFF	15
+#define SAS_ECC_INTR_NCQ_MEM2_ECC_MB_OFF	16
+#define SAS_ECC_INTR_NCQ_MEM2_ECC_1B_OFF	17
+#define SAS_ECC_INTR_NCQ_MEM3_ECC_MB_OFF	18
+#define SAS_ECC_INTR_NCQ_MEM3_ECC_1B_OFF	19
 #define SAS_ECC_INTR_MSK		0x1ec
 #define HGC_ERR_STAT_EN			0x238
 #define DLVRY_Q_0_BASE_ADDR_LO		0x260
@@ -94,7 +156,20 @@
 #define COMPL_Q_0_DEPTH			0x4e8
 #define COMPL_Q_0_WR_PTR		0x4ec
 #define COMPL_Q_0_RD_PTR		0x4f0
-
+#define HGC_RXM_DFX_STATUS14	0xae8
+#define HGC_RXM_DFX_STATUS14_MEM0_OFF		0
+#define HGC_RXM_DFX_STATUS14_MEM0_MSK		(0x1ff << \
+						 HGC_RXM_DFX_STATUS14_MEM0_OFF)
+#define HGC_RXM_DFX_STATUS14_MEM1_OFF		9
+#define HGC_RXM_DFX_STATUS14_MEM1_MSK		(0x1ff << \
+						 HGC_RXM_DFX_STATUS14_MEM1_OFF)
+#define HGC_RXM_DFX_STATUS14_MEM2_OFF		18
+#define HGC_RXM_DFX_STATUS14_MEM2_MSK		(0x1ff << \
+						 HGC_RXM_DFX_STATUS14_MEM2_OFF)
+#define HGC_RXM_DFX_STATUS15	0xaec
+#define HGC_RXM_DFX_STATUS15_MEM3_OFF		0
+#define HGC_RXM_DFX_STATUS15_MEM3_MSK		(0x1ff << \
+						 HGC_RXM_DFX_STATUS15_MEM3_OFF)
 /* phy registers need init */
 #define PORT_BASE			(0x2000)
 
@@ -267,6 +342,8 @@
 #define ITCT_HDR_RTOLT_OFF		48
 #define ITCT_HDR_RTOLT_MSK		(0xffffULL << ITCT_HDR_RTOLT_OFF)
 
+#define HISI_SAS_FATAL_INT_NR	2
+
 struct hisi_sas_complete_v2_hdr {
 	__le32 dw0;
 	__le32 dw1;
@@ -808,7 +885,7 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK1, 0x7efefefe);
 	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK2, 0x7efefefe);
 	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK3, 0x7ffffffe);
-	hisi_sas_write32(hisi_hba, SAS_ECC_INTR_MSK, 0xfffff3c0);
+	hisi_sas_write32(hisi_hba, SAS_ECC_INTR_MSK, 0xfff00c30);
 	for (i = 0; i < hisi_hba->queue_count; i++)
 		hisi_sas_write32(hisi_hba, OQ0_INT_SRC_MSK+0x4*i, 0);
 
@@ -824,7 +901,7 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 		hisi_sas_phy_write32(hisi_hba, i, DONE_RECEIVED_TIME, 0x10);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT0, 0xffffffff);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT1, 0xffffffff);
-		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2, 0xffffffff);
+		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2, 0xfff87fff);
 		hisi_sas_phy_write32(hisi_hba, i, RXOP_CHECK_CFG_H, 0x1000);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT1_MSK, 0xffffffff);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2_MSK, 0x8ffffbff);
@@ -2007,8 +2084,9 @@ static irqreturn_t int_chnl_int_v2_hw(int irq_no, void *p)
 			if (irq_value1) {
 				if (irq_value1 & (CHL_INT1_DMAC_RX_ECC_ERR_MSK |
 						  CHL_INT1_DMAC_TX_ECC_ERR_MSK))
-					panic("%s: DMAC RX/TX ecc bad error! (0x%x)",
-						dev_name(dev), irq_value1);
+					panic("%s: DMAC RX/TX ecc bad error!\
+					       (0x%x)",
+					      dev_name(dev), irq_value1);
 
 				hisi_sas_phy_write32(hisi_hba, phy_no,
 						     CHL_INT1, irq_value1);
@@ -2039,6 +2117,318 @@ static irqreturn_t int_chnl_int_v2_hw(int irq_no, void *p)
 	return IRQ_HANDLED;
 }
 
+static void
+one_bit_ecc_error_process_v2_hw(struct hisi_hba *hisi_hba, u32 irq_value)
+{
+	struct device *dev = &hisi_hba->pdev->dev;
+	u32 reg_val;
+
+	if (irq_value & BIT(SAS_ECC_INTR_DQE_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_DQE_ECC_ADDR);
+		dev_warn(dev, "hgc_dqe_acc1b_intr found: \
+			       Ram address is 0x%08X\n",
+			 (reg_val & HGC_DQE_ECC_1B_ADDR_MSK) >>
+			 HGC_DQE_ECC_1B_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_IOST_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_IOST_ECC_ADDR);
+		dev_warn(dev, "hgc_iost_acc1b_intr found: \
+			       Ram address is 0x%08X\n",
+			 (reg_val & HGC_IOST_ECC_1B_ADDR_MSK) >>
+			 HGC_IOST_ECC_1B_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_ITCT_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_ITCT_ECC_ADDR);
+		dev_warn(dev, "hgc_itct_acc1b_intr found: \
+			       Ram address is 0x%08X\n",
+			 (reg_val & HGC_ITCT_ECC_1B_ADDR_MSK) >>
+			 HGC_ITCT_ECC_1B_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_IOSTLIST_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
+		dev_warn(dev, "hgc_iostl_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_LM_DFX_STATUS2_IOSTLIST_MSK) >>
+			 HGC_LM_DFX_STATUS2_IOSTLIST_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_ITCTLIST_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
+		dev_warn(dev, "hgc_itctl_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_LM_DFX_STATUS2_ITCTLIST_MSK) >>
+			 HGC_LM_DFX_STATUS2_ITCTLIST_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_CQE_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_CQE_ECC_ADDR);
+		dev_warn(dev, "hgc_cqe_acc1b_intr found: \
+			       Ram address is 0x%08X\n",
+			 (reg_val & HGC_CQE_ECC_1B_ADDR_MSK) >>
+			 HGC_CQE_ECC_1B_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM0_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		dev_warn(dev, "rxm_mem0_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_RXM_DFX_STATUS14_MEM0_MSK) >>
+			 HGC_RXM_DFX_STATUS14_MEM0_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM1_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		dev_warn(dev, "rxm_mem1_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_RXM_DFX_STATUS14_MEM1_MSK) >>
+			 HGC_RXM_DFX_STATUS14_MEM1_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM2_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		dev_warn(dev, "rxm_mem2_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_RXM_DFX_STATUS14_MEM2_MSK) >>
+			 HGC_RXM_DFX_STATUS14_MEM2_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM3_ECC_1B_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS15);
+		dev_warn(dev, "rxm_mem3_acc1b_intr found: \
+			       memory address is 0x%08X\n",
+			 (reg_val & HGC_RXM_DFX_STATUS15_MEM3_MSK) >>
+			 HGC_RXM_DFX_STATUS15_MEM3_OFF);
+	}
+
+}
+
+static void multi_bit_ecc_error_process_v2_hw(struct hisi_hba *hisi_hba,
+		u32 irq_value)
+{
+	u32 reg_val;
+	struct device *dev = &hisi_hba->pdev->dev;
+
+	if (irq_value & BIT(SAS_ECC_INTR_DQE_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_DQE_ECC_ADDR);
+		panic("%s: hgc_dqe_accbad_intr (0x%x) found: \
+		       Ram address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_DQE_ECC_MB_ADDR_MSK) >>
+		      HGC_DQE_ECC_MB_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_IOST_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_IOST_ECC_ADDR);
+		panic("%s: hgc_iost_accbad_intr (0x%x) found: \
+		       Ram address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_IOST_ECC_MB_ADDR_MSK) >>
+		      HGC_IOST_ECC_MB_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_ITCT_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_ITCT_ECC_ADDR);
+		panic("%s: hgc_itct_accbad_intr (0x%x) found: \
+		       Ram address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_ITCT_ECC_MB_ADDR_MSK) >>
+		      HGC_ITCT_ECC_MB_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_IOSTLIST_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
+		panic("%s: hgc_iostl_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_LM_DFX_STATUS2_IOSTLIST_MSK) >>
+		      HGC_LM_DFX_STATUS2_IOSTLIST_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_ITCTLIST_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
+		panic("%s: hgc_itctl_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_LM_DFX_STATUS2_ITCTLIST_MSK) >>
+		      HGC_LM_DFX_STATUS2_ITCTLIST_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_CQE_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_CQE_ECC_ADDR);
+		panic("%s: hgc_cqe_accbad_intr (0x%x) found: \
+		       Ram address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_CQE_ECC_MB_ADDR_MSK) >>
+		      HGC_CQE_ECC_MB_ADDR_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM0_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		panic("%s: rxm_mem0_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_RXM_DFX_STATUS14_MEM0_MSK) >>
+		      HGC_RXM_DFX_STATUS14_MEM0_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM1_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		panic("%s: rxm_mem1_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_RXM_DFX_STATUS14_MEM1_MSK) >>
+		      HGC_RXM_DFX_STATUS14_MEM1_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM2_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
+		panic("%s: rxm_mem2_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_RXM_DFX_STATUS14_MEM2_MSK) >>
+		      HGC_RXM_DFX_STATUS14_MEM2_OFF);
+	}
+
+	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM3_ECC_MB_OFF)) {
+		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS15);
+		panic("%s: rxm_mem3_accbad_intr (0x%x) found: \
+		       memory address is 0x%08X\n",
+		      dev_name(dev), irq_value,
+		      (reg_val & HGC_RXM_DFX_STATUS15_MEM3_MSK) >>
+		      HGC_RXM_DFX_STATUS15_MEM3_OFF);
+	}
+
+}
+
+static irqreturn_t fatal_ecc_int_v2_hw(int irq_no, void *p)
+{
+	struct hisi_hba *hisi_hba = p;
+	u32 irq_value, irq_msk;
+
+	irq_msk = hisi_sas_read32(hisi_hba, SAS_ECC_INTR_MSK);
+	hisi_sas_write32(hisi_hba, SAS_ECC_INTR_MSK, irq_msk | 0xffffffff);
+
+	irq_value = hisi_sas_read32(hisi_hba, SAS_ECC_INTR);
+	if (irq_value) {
+		one_bit_ecc_error_process_v2_hw(hisi_hba, irq_value);
+		multi_bit_ecc_error_process_v2_hw(hisi_hba, irq_value);
+	}
+
+	hisi_sas_write32(hisi_hba, SAS_ECC_INTR, irq_value);
+	hisi_sas_write32(hisi_hba, SAS_ECC_INTR_MSK, irq_msk);
+
+	return IRQ_HANDLED;
+}
+
+#define AXI_ERR_NR	8
+static const char axi_err_info[AXI_ERR_NR][32] = {
+	"IOST_AXI_W_ERR",
+	"IOST_AXI_R_ERR",
+	"ITCT_AXI_W_ERR",
+	"ITCT_AXI_R_ERR",
+	"SATA_AXI_W_ERR",
+	"SATA_AXI_R_ERR",
+	"DQE_AXI_R_ERR",
+	"CQE_AXI_W_ERR"
+};
+
+#define FIFO_ERR_NR	5
+static const char fifo_err_info[FIFO_ERR_NR][32] = {
+	"CQE_WINFO_FIFO",
+	"CQE_MSG_FIFIO",
+	"GETDQE_FIFO",
+	"CMDP_FIFO",
+	"AWTCTRL_FIFO"
+};
+
+static irqreturn_t fatal_axi_int_v2_hw(int irq_no, void *p)
+{
+	struct hisi_hba *hisi_hba = p;
+	u32 irq_value, irq_msk, err_value;
+	struct device *dev = &hisi_hba->pdev->dev;
+
+	irq_msk = hisi_sas_read32(hisi_hba, ENT_INT_SRC_MSK3);
+	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK3, irq_msk | 0xfffffffe);
+
+	irq_value = hisi_sas_read32(hisi_hba, ENT_INT_SRC3);
+	if (irq_value) {
+		if (irq_value & BIT(ENT_INT_SRC3_WP_DEPTH_OFF)) {
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					1 << ENT_INT_SRC3_WP_DEPTH_OFF);
+			panic("%s: write pointer and depth error (0x%x) \
+			       found!\n",
+			      dev_name(dev), irq_value);
+		}
+
+		if (irq_value & BIT(ENT_INT_SRC3_IPTT_SLOT_NOMATCH_OFF)) {
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					 1 <<
+					 ENT_INT_SRC3_IPTT_SLOT_NOMATCH_OFF);
+			panic("%s: iptt no match slot error (0x%x) found!\n",
+			      dev_name(dev), irq_value);
+		}
+
+		if (irq_value & BIT(ENT_INT_SRC3_RP_DEPTH_OFF))
+			panic("%s: read pointer and depth error (0x%x) \
+			       found!\n",
+			      dev_name(dev), irq_value);
+
+		if (irq_value & BIT(ENT_INT_SRC3_AXI_OFF)) {
+			int i;
+
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					1 << ENT_INT_SRC3_AXI_OFF);
+			err_value = hisi_sas_read32(hisi_hba,
+						    HGC_AXI_FIFO_ERR_INFO);
+
+			for (i = 0; i < AXI_ERR_NR; i++) {
+				if (err_value & BIT(i))
+					panic("%s: %s (0x%x) found!\n",
+					       dev_name(dev),
+					      axi_err_info[i], irq_value);
+			}
+		}
+
+		if (irq_value & BIT(ENT_INT_SRC3_FIFO_OFF)) {
+			int i;
+
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					1 << ENT_INT_SRC3_FIFO_OFF);
+			err_value = hisi_sas_read32(hisi_hba,
+						    HGC_AXI_FIFO_ERR_INFO);
+
+			for (i = 0; i < FIFO_ERR_NR; i++) {
+				if (err_value & BIT(AXI_ERR_NR + i))
+					panic("%s: %s (0x%x) found!\n",
+					      dev_name(dev),
+					      fifo_err_info[i], irq_value);
+			}
+
+		}
+
+		if (irq_value & BIT(ENT_INT_SRC3_LM_OFF)) {
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					1 << ENT_INT_SRC3_LM_OFF);
+			panic("%s: LM add/fetch list error (0x%x) found!\n",
+			      dev_name(dev), irq_value);
+		}
+
+		if (irq_value & BIT(ENT_INT_SRC3_ABT_OFF)) {
+			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
+					1 << ENT_INT_SRC3_ABT_OFF);
+			panic("%s: SAS_HGC_ABT fetch LM list error (0x%x) found!\n",
+			      dev_name(dev), irq_value);
+		}
+	}
+
+	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK3, irq_msk);
+
+	return IRQ_HANDLED;
+}
+
 static irqreturn_t cq_interrupt_v2_hw(int irq_no, void *p)
 {
 	struct hisi_sas_cq *cq = p;
@@ -2192,6 +2582,11 @@ static irqreturn_t sata_int_v2_hw(int irq_no, void *p)
 	int_chnl_int_v2_hw,
 };
 
+static irq_handler_t fatal_interrupts[HISI_SAS_FATAL_INT_NR] = {
+	fatal_ecc_int_v2_hw,
+	fatal_axi_int_v2_hw
+};
+
 /**
  * There is a limitation in the hip06 chipset that we need
  * to map in all mbigen interrupts, even if they are not used.
@@ -2247,6 +2642,26 @@ static int interrupt_init_v2_hw(struct hisi_hba *hisi_hba)
 		}
 	}
 
+	for (i = 0; i < HISI_SAS_FATAL_INT_NR; i++) {
+		int idx = i;
+
+		irq = irq_map[idx + 81];
+		if (!irq) {
+			dev_err(dev, "irq init: fail map fatal interrupt %d\n",
+				idx);
+			return -ENOENT;
+		}
+
+		rc = devm_request_irq(dev, irq, fatal_interrupts[i], 0,
+				      DRV_NAME " fatal", hisi_hba);
+		if (rc) {
+			dev_err(dev,
+				"irq init: could not request fatal interrupt %d, rc=%d\n",
+				irq, rc);
+			return -ENOENT;
+		}
+	}
+
 	for (i = 0; i < hisi_hba->queue_count; i++) {
 		int idx = i + 96; /* First cq interrupt is irq96 */
 
-- 
1.9.1

[PATCH 02/11] hisi_sas: alloc queue id of slot according to device id

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Currently slots are allocated from queues in a round-robin fashion.
This causes a problem for internal commands in device mode. For this
mode, we should ensure that the internal abort command is the last
command seen in the host for that device. We can only ensure this when
we place the internal abort command after the preceding commands for
device that in the same queue, as there is no order in which the host
will select a queue to execute the next command.

This queue restriction makes supporting scsi mq more tricky in
the future, but should not be a blocker.

Note: Even though v1 hw does not support internal abort, the
      allocation method is chosen to be the same for consistency.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas.h       |  4 ++--
 drivers/scsi/hisi_sas/hisi_sas_main.c  |  8 ++++----
 drivers/scsi/hisi_sas/hisi_sas_v1_hw.c | 30 ++++++++++++------------------
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 29 ++++++++++++-----------------
 4 files changed, 30 insertions(+), 41 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index 64046c5..bf59fab 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -150,7 +150,8 @@ struct hisi_sas_hw {
 				struct domain_device *device);
 	struct hisi_sas_device *(*alloc_dev)(struct domain_device *device);
 	void (*sl_notify)(struct hisi_hba *hisi_hba, int phy_no);
-	int (*get_free_slot)(struct hisi_hba *hisi_hba, int *q, int *s);
+	int (*get_free_slot)(struct hisi_hba *hisi_hba, u32 dev_id,
+			int *q, int *s);
 	void (*start_delivery)(struct hisi_hba *hisi_hba);
 	int (*prep_ssp)(struct hisi_hba *hisi_hba,
 			struct hisi_sas_slot *slot, int is_tmf,
@@ -207,7 +208,6 @@ struct hisi_hba {
 	struct hisi_sas_port port[HISI_SAS_MAX_PHYS];
 
 	int	queue_count;
-	int	queue;
 	struct hisi_sas_slot	*slot_prep;
 
 	struct dma_pool *sge_page_pool;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 9afc697..9f5ccc5 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -232,8 +232,8 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 		rc = hisi_sas_slot_index_alloc(hisi_hba, &slot_idx);
 	if (rc)
 		goto err_out;
-	rc = hisi_hba->hw->get_free_slot(hisi_hba, &dlvry_queue,
-					 &dlvry_queue_slot);
+	rc = hisi_hba->hw->get_free_slot(hisi_hba, sas_dev->device_id,
+					&dlvry_queue, &dlvry_queue_slot);
 	if (rc)
 		goto err_out_tag;
 
@@ -987,8 +987,8 @@ static int hisi_sas_query_task(struct sas_task *task)
 	rc = hisi_sas_slot_index_alloc(hisi_hba, &slot_idx);
 	if (rc)
 		goto err_out;
-	rc = hisi_hba->hw->get_free_slot(hisi_hba, &dlvry_queue,
-					 &dlvry_queue_slot);
+	rc = hisi_hba->hw->get_free_slot(hisi_hba, sas_dev->device_id,
+					&dlvry_queue, &dlvry_queue_slot);
 	if (rc)
 		goto err_out_tag;
 
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index c0ac49d..bbc5760 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -862,29 +862,23 @@ static int get_wideport_bitmap_v1_hw(struct hisi_hba *hisi_hba, int port_id)
  * The callpath to this function and upto writing the write
  * queue pointer should be safe from interruption.
  */
-static int get_free_slot_v1_hw(struct hisi_hba *hisi_hba, int *q, int *s)
+static int get_free_slot_v1_hw(struct hisi_hba *hisi_hba, u32 dev_id,
+				int *q, int *s)
 {
 	struct device *dev = &hisi_hba->pdev->dev;
 	struct hisi_sas_dq *dq;
 	u32 r, w;
-	int queue = hisi_hba->queue;
-
-	while (1) {
-		dq = &hisi_hba->dq[queue];
-		w = dq->wr_point;
-		r = hisi_sas_read32_relaxed(hisi_hba,
-				    DLVRY_Q_0_RD_PTR + (queue * 0x14));
-		if (r == (w+1) % HISI_SAS_QUEUE_SLOTS) {
-			queue = (queue + 1) % hisi_hba->queue_count;
-			if (queue == hisi_hba->queue) {
-				dev_warn(dev, "could not find free slot\n");
-				return -EAGAIN;
-			}
-			continue;
-		}
-		break;
+	int queue = dev_id % hisi_hba->queue_count;
+
+	dq = &hisi_hba->dq[queue];
+	w = dq->wr_point;
+	r = hisi_sas_read32_relaxed(hisi_hba,
+				DLVRY_Q_0_RD_PTR + (queue * 0x14));
+	if (r == (w+1) % HISI_SAS_QUEUE_SLOTS) {
+		dev_warn(dev, "could not find free slot\n");
+		return -EAGAIN;
 	}
-	hisi_hba->queue = (queue + 1) % hisi_hba->queue_count;
+
 	*q = queue;
 	*s = w;
 	return 0;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 9e70e6d..cb19e73 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -1089,29 +1089,24 @@ static int get_wideport_bitmap_v2_hw(struct hisi_hba *hisi_hba, int port_id)
  * The callpath to this function and upto writing the write
  * queue pointer should be safe from interruption.
  */
-static int get_free_slot_v2_hw(struct hisi_hba *hisi_hba, int *q, int *s)
+static int get_free_slot_v2_hw(struct hisi_hba *hisi_hba, u32 dev_id,
+				int *q, int *s)
 {
 	struct device *dev = &hisi_hba->pdev->dev;
 	struct hisi_sas_dq *dq;
 	u32 r, w;
-	int queue = hisi_hba->queue;
+	int queue = dev_id % hisi_hba->queue_count;
 
-	while (1) {
-		dq = &hisi_hba->dq[queue];
-		w = dq->wr_point;
-		r = hisi_sas_read32_relaxed(hisi_hba,
-					    DLVRY_Q_0_RD_PTR + (queue * 0x14));
-		if (r == (w+1) % HISI_SAS_QUEUE_SLOTS) {
-			queue = (queue + 1) % hisi_hba->queue_count;
-			if (queue == hisi_hba->queue) {
-				dev_warn(dev, "could not find free slot\n");
-				return -EAGAIN;
-			}
-			continue;
-		}
-		break;
+	dq = &hisi_hba->dq[queue];
+	w = dq->wr_point;
+	r = hisi_sas_read32_relaxed(hisi_hba,
+				DLVRY_Q_0_RD_PTR + (queue * 0x14));
+	if (r == (w+1) % HISI_SAS_QUEUE_SLOTS) {
+		dev_warn(dev, "full queue=%d r=%d w=%d\n\n",
+				queue, r, w);
+		return -EAGAIN;
 	}
-	hisi_hba->queue = (queue + 1) % hisi_hba->queue_count;
+
 	*q = queue;
 	*s = w;
 	return 0;
-- 
1.9.1

[PATCH 03/11] hisi_sas: only process broadcast change in phy_bcast_v2_hw()

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

There are many BROADCAST primitives generated by the host.
We are only interested in BROADCAST (CHANGE) primitives currently,
so only process this.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index cb19e73..7567193 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -194,6 +194,9 @@
 #define SL_CONTROL_NOTIFY_EN_MSK	(0x1 << SL_CONTROL_NOTIFY_EN_OFF)
 #define SL_CONTROL_CTA_OFF		17
 #define SL_CONTROL_CTA_MSK		(0x1 << SL_CONTROL_CTA_OFF)
+#define RX_PRIMS_STATUS         (PORT_BASE + 0x98)
+#define RX_BCAST_CHG_OFF        1
+#define RX_BCAST_CHG_MSK        (0x1 << RX_BCAST_CHG_OFF)
 #define TX_ID_DWORD0			(PORT_BASE + 0x9c)
 #define TX_ID_DWORD1			(PORT_BASE + 0xa0)
 #define TX_ID_DWORD2			(PORT_BASE + 0xa4)
@@ -2044,9 +2047,12 @@ static void phy_bcast_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
 	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
 	struct asd_sas_phy *sas_phy = &phy->sas_phy;
 	struct sas_ha_struct *sas_ha = &hisi_hba->sha;
+	u32 bcast_status;
 
 	hisi_sas_phy_write32(hisi_hba, phy_no, SL_RX_BCAST_CHK_MSK, 1);
-	sas_ha->notify_port_event(sas_phy, PORTE_BROADCAST_RCVD);
+	bcast_status = hisi_sas_phy_read32(hisi_hba, phy_no, RX_PRIMS_STATUS);
+	if (bcast_status & RX_BCAST_CHG_MSK)
+		sas_ha->notify_port_event(sas_phy, PORTE_BROADCAST_RCVD);
 	hisi_sas_phy_write32(hisi_hba, phy_no, CHL_INT0,
 			     CHL_INT0_SL_RX_BCST_ACK_MSK);
 	hisi_sas_phy_write32(hisi_hba, phy_no, SL_RX_BCAST_CHK_MSK, 0);
-- 
1.9.1

[PATCH 04/11] hisi_sas: fix port form bug in hisi_sas_port_notify_formed()

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

When we form a wideport, we should use hardware PHY port_id instead
of sas_phy->id.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 9f5ccc5..486aa92 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -537,7 +537,7 @@ static void hisi_sas_port_notify_formed(struct asd_sas_phy *sas_phy)
 	struct hisi_hba *hisi_hba = sas_ha->lldd_ha;
 	struct hisi_sas_phy *phy = sas_phy->lldd_phy;
 	struct asd_sas_port *sas_port = sas_phy->port;
-	struct hisi_sas_port *port = &hisi_hba->port[sas_phy->id];
+	struct hisi_sas_port *port = &hisi_hba->port[phy->port_id];
 	unsigned long flags;
 
 	if (!sas_port)
-- 
1.9.1

[PATCH 05/11] hisi_sas: replace WARN_ON() with dev_warn() for internal abort

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Replace WARN_ON() with dev_warn() print when internal abort fails.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 486aa92..9133238 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -764,7 +764,8 @@ static int hisi_sas_exec_internal_tmf_task(struct domain_device *device,
 		task = NULL;
 	}
 ex_err:
-	WARN_ON(retry == TASK_RETRY);
+	if (retry == TASK_RETRY)
+		dev_warn(dev, "abort tmf: executing internal task failed!\n");
 	sas_free_task(task);
 	return res;
 }
-- 
1.9.1

[PATCH 06/11] hisi_sas: modify return value of hisi_sas_query_task()

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

sas_scsi_find_task() only deals with return value
TMF_RESP_FUNC_FAILED/TMF_RESP_FUNC_SUCC/TMF_RESP_FUNC_COMPLETE of
query task. So for LLDD errors just return TMF_RESP_FUNC_FAILED.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 9133238..504cbcf 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -961,6 +961,9 @@ static int hisi_sas_query_task(struct sas_task *task)
 		case TMF_RESP_FUNC_FAILED:
 		case TMF_RESP_FUNC_COMPLETE:
 			break;
+		default:
+			rc = TMF_RESP_FUNC_FAILED;
+			break;
 		}
 	}
 	return rc;
-- 
1.9.1

[PATCH 07/11] hisi_sas: delete repeated configuration in free_device_v2_hw()

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Delete repeated configuration items for hisi_sas_device() when
we free a device. These items are now only set in
hisi_sas_dev_gone().

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 7567193..3081eec 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -739,8 +739,6 @@ static void free_device_v2_hw(struct hisi_hba *hisi_hba,
 			qw0 &= ~(1 << ITCT_HDR_VALID_OFF);
 			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
 					 ENT_INT_SRC3_ITC_INT_MSK);
-			hisi_hba->devices[dev_id].dev_type = SAS_PHY_UNUSED;
-			hisi_hba->devices[dev_id].dev_status = HISI_SAS_DEV_NORMAL;
 
 			/* clear the itct */
 			hisi_sas_write32(hisi_hba, ITCT_CLR, 0);
-- 
1.9.1

[PATCH 08/11] hisi_sas: modify some values in get_ata_protocol()

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Modify and add some SATA commands according to SATA protocol.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 3081eec..a5faa4d 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -1750,6 +1750,7 @@ static u8 get_ata_protocol(u8 cmd, int direction)
 	case ATA_CMD_NCQ_NON_DATA:
 	return SATA_PROTOCOL_FPDMA;
 
+	case ATA_CMD_DOWNLOAD_MICRO:
 	case ATA_CMD_ID_ATA:
 	case ATA_CMD_PMP_READ:
 	case ATA_CMD_READ_LOG_EXT:
@@ -1761,18 +1762,27 @@ static u8 get_ata_protocol(u8 cmd, int direction)
 	case ATA_CMD_PIO_WRITE_EXT:
 	return SATA_PROTOCOL_PIO;
 
+	case ATA_CMD_DSM:
+	case ATA_CMD_DOWNLOAD_MICRO_DMA:
+	case ATA_CMD_PMP_READ_DMA:
+	case ATA_CMD_PMP_WRITE_DMA:
 	case ATA_CMD_READ:
 	case ATA_CMD_READ_EXT:
 	case ATA_CMD_READ_LOG_DMA_EXT:
+	case ATA_CMD_READ_STREAM_DMA_EXT:
+	case ATA_CMD_TRUSTED_RCV_DMA:
+	case ATA_CMD_TRUSTED_SND_DMA:
 	case ATA_CMD_WRITE:
 	case ATA_CMD_WRITE_EXT:
+	case ATA_CMD_WRITE_FUA_EXT:
 	case ATA_CMD_WRITE_QUEUED:
 	case ATA_CMD_WRITE_LOG_DMA_EXT:
+	case ATA_CMD_WRITE_STREAM_DMA_EXT:
 	return SATA_PROTOCOL_DMA;
 
-	case ATA_CMD_DOWNLOAD_MICRO:
-	case ATA_CMD_DEV_RESET:
 	case ATA_CMD_CHK_POWER:
+	case ATA_CMD_DEV_RESET:
+	case ATA_CMD_EDD:
 	case ATA_CMD_FLUSH:
 	case ATA_CMD_FLUSH_EXT:
 	case ATA_CMD_VERIFY:
-- 
1.9.1

[PATCH 09/11] hisi_sas: check SATA FIS when directly attaching SATA device

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Check ERR bit of status to decide whether there is something wrong with
initial register-D2H FIS. If error exists, PHY reset the channel to
restart OOB.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index a5faa4d..cda3baf 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -2537,6 +2537,16 @@ static irqreturn_t sata_int_v2_hw(int irq_no, void *p)
 		goto end;
 	}
 
+	/* check ERR bit of Status Register */
+	if (fis->status & ATA_ERR) {
+		dev_warn(dev, "sata int: phy%d FIS status: 0x%x\n", phy_no,
+				fis->status);
+		disable_phy_v2_hw(hisi_hba, phy_no);
+		enable_phy_v2_hw(hisi_hba, phy_no);
+		res = IRQ_NONE;
+		goto end;
+	}
+
 	if (unlikely(phy_no == 8)) {
 		u32 port_state = hisi_sas_read32(hisi_hba, PORT_STATE);
 
-- 
1.9.1

[PATCH 10/11] hisi_sas: use atomic64_t for hisi_sas_device.running_req

[John Garry]

Sometimes the value of hisi_sas_device.running_req
would go negative unless we have the check for
running_req >= 0 before trying to decrement.

This is because using running_req is not thread-safe.

As such, the value for running_req may be actually incorrect,
so use atomic64_t instead.

Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Xiang Chen <chenxiang66@hisilicon.com>
---
 drivers/scsi/hisi_sas/hisi_sas.h       | 2 +-
 drivers/scsi/hisi_sas/hisi_sas_main.c  | 9 +++++----
 drivers/scsi/hisi_sas/hisi_sas_v1_hw.c | 4 ++--
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 4 ++--
 4 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index bf59fab..8b5ecc6 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -111,7 +111,7 @@ struct hisi_sas_device {
 	struct domain_device	*sas_device;
 	u64 attached_phy;
 	u64 device_id;
-	u64 running_req;
+	atomic64_t running_req;
 	u8 dev_status;
 };
 
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 504cbcf..18e2194 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -162,8 +162,8 @@ static void hisi_sas_slot_abort(struct work_struct *work)
 	hisi_sas_slot_task_free(hisi_hba, task, abort_slot);
 	if (task->task_done)
 		task->task_done(task);
-	if (sas_dev && sas_dev->running_req)
-		sas_dev->running_req--;
+	if (sas_dev)
+		atomic64_dec(&sas_dev->running_req);
 }
 
 static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
@@ -303,7 +303,7 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 
 	hisi_hba->slot_prep = slot;
 
-	sas_dev->running_req++;
+	atomic64_inc(&sas_dev->running_req);
 	++(*pass);
 
 	return 0;
@@ -1027,7 +1027,8 @@ static int hisi_sas_query_task(struct sas_task *task)
 
 	hisi_hba->slot_prep = slot;
 
-	sas_dev->running_req++;
+	atomic64_inc(&sas_dev->running_req);
+
 	/* send abort command to our chip */
 	hisi_hba->hw->start_delivery(hisi_hba);
 
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index bbc5760..05177fc 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -1366,8 +1366,8 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba,
 	}
 
 out:
-	if (sas_dev && sas_dev->running_req)
-		sas_dev->running_req--;
+	if (sas_dev)
+		atomic64_dec(&sas_dev->running_req);
 
 	hisi_sas_slot_task_free(hisi_hba, task, slot);
 	sts = ts->stat;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index cda3baf..e88113a 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -1728,8 +1728,8 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 	}
 
 out:
-	if (sas_dev && sas_dev->running_req)
-		sas_dev->running_req--;
+	if (sas_dev)
+		atomic64_dec(&sas_dev->running_req);
 
 	hisi_sas_slot_task_free(hisi_hba, task, slot);
 	sts = ts->stat;
-- 
1.9.1

[PATCH 11/11] hisi_sas: add PHY set linkrate support for v1 and v2 hw

[John Garry]

From: Xiang Chen <chenxiang66@hisilicon.com>

Add the function to set PHY min and max linkrate through
sysfs interface.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
---
 drivers/scsi/hisi_sas/hisi_sas.h       |  3 +++
 drivers/scsi/hisi_sas/hisi_sas_main.c  | 12 +++++++--
 drivers/scsi/hisi_sas/hisi_sas_v1_hw.c | 45 ++++++++++++++++++++++++++++++++++
 drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 45 ++++++++++++++++++++++++++++++++++
 4 files changed, 103 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index 8b5ecc6..c0cd505 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -168,6 +168,9 @@ struct hisi_sas_hw {
 	void (*phy_enable)(struct hisi_hba *hisi_hba, int phy_no);
 	void (*phy_disable)(struct hisi_hba *hisi_hba, int phy_no);
 	void (*phy_hard_reset)(struct hisi_hba *hisi_hba, int phy_no);
+	void (*phy_set_linkrate)(struct hisi_hba *hisi_hba, int phy_no,
+			struct sas_phy_linkrates *linkrates);
+	enum sas_linkrate (*phy_get_max_linkrate)(void);
 	void (*free_device)(struct hisi_hba *hisi_hba,
 			    struct hisi_sas_device *dev);
 	int (*get_wideport_bitmap)(struct hisi_hba *hisi_hba, int port_id);
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 18e2194..6d6f150 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -369,9 +369,14 @@ static void hisi_sas_bytes_dmaed(struct hisi_hba *hisi_hba, int phy_no)
 		struct sas_phy *sphy = sas_phy->phy;
 
 		sphy->negotiated_linkrate = sas_phy->linkrate;
-		sphy->minimum_linkrate = phy->minimum_linkrate;
 		sphy->minimum_linkrate_hw = SAS_LINK_RATE_1_5_GBPS;
-		sphy->maximum_linkrate = phy->maximum_linkrate;
+		sphy->maximum_linkrate_hw =
+			hisi_hba->hw->phy_get_max_linkrate();
+		if (sphy->minimum_linkrate == SAS_LINK_RATE_UNKNOWN)
+			sphy->minimum_linkrate = phy->minimum_linkrate;
+
+		if (sphy->maximum_linkrate == SAS_LINK_RATE_UNKNOWN)
+			sphy->maximum_linkrate = phy->maximum_linkrate;
 	}
 
 	if (phy->phy_type & PORT_TYPE_SAS) {
@@ -645,6 +650,9 @@ static int hisi_sas_control_phy(struct asd_sas_phy *sas_phy, enum phy_func func,
 		break;
 
 	case PHY_FUNC_SET_LINK_RATE:
+		hisi_hba->hw->phy_set_linkrate(hisi_hba, phy_no, funcdata);
+		break;
+
 	case PHY_FUNC_RELEASE_SPINUP_HOLD:
 	default:
 		return -EOPNOTSUPP;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index 05177fc..8a1be0b 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -843,6 +843,49 @@ static void sl_notify_v1_hw(struct hisi_hba *hisi_hba, int phy_no)
 	hisi_sas_phy_write32(hisi_hba, phy_no, SL_CONTROL, sl_control);
 }
 
+static enum sas_linkrate phy_get_max_linkrate_v1_hw(void)
+{
+	return SAS_LINK_RATE_6_0_GBPS;
+}
+
+static void phy_set_linkrate_v1_hw(struct hisi_hba *hisi_hba, int phy_no,
+		struct sas_phy_linkrates *r)
+{
+	u32 prog_phy_link_rate =
+		hisi_sas_phy_read32(hisi_hba, phy_no, PROG_PHY_LINK_RATE);
+	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
+	struct asd_sas_phy *sas_phy = &phy->sas_phy;
+	int i;
+	enum sas_linkrate min, max;
+	u32 rate_mask = 0;
+
+	if (r->maximum_linkrate == SAS_LINK_RATE_UNKNOWN) {
+		max = sas_phy->phy->maximum_linkrate;
+		min = r->minimum_linkrate;
+	} else if (r->minimum_linkrate == SAS_LINK_RATE_UNKNOWN) {
+		max = r->maximum_linkrate;
+		min = sas_phy->phy->minimum_linkrate;
+	} else
+		return;
+
+	sas_phy->phy->maximum_linkrate = max;
+	sas_phy->phy->minimum_linkrate = min;
+
+	min -= SAS_LINK_RATE_1_5_GBPS;
+	max -= SAS_LINK_RATE_1_5_GBPS;
+
+	for (i = 0; i <= max; i++)
+		rate_mask |= 1 << (i * 2);
+
+	prog_phy_link_rate &= ~0xff;
+	prog_phy_link_rate |= rate_mask;
+
+	hisi_sas_phy_write32(hisi_hba, phy_no, PROG_PHY_LINK_RATE,
+			prog_phy_link_rate);
+
+	phy_hard_reset_v1_hw(hisi_hba, phy_no);
+}
+
 static int get_wideport_bitmap_v1_hw(struct hisi_hba *hisi_hba, int port_id)
 {
 	int i, bitmap = 0;
@@ -1818,6 +1861,8 @@ static int hisi_sas_v1_init(struct hisi_hba *hisi_hba)
 	.phy_enable = enable_phy_v1_hw,
 	.phy_disable = disable_phy_v1_hw,
 	.phy_hard_reset = phy_hard_reset_v1_hw,
+	.phy_set_linkrate = phy_set_linkrate_v1_hw,
+	.phy_get_max_linkrate = phy_get_max_linkrate_v1_hw,
 	.get_wideport_bitmap = get_wideport_bitmap_v1_hw,
 	.max_command_entries = HISI_SAS_COMMAND_ENTRIES_V1_HW,
 	.complete_hdr_size = sizeof(struct hisi_sas_complete_v1_hdr),
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index e88113a..15487f2 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -1060,6 +1060,49 @@ static void sl_notify_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
 	hisi_sas_phy_write32(hisi_hba, phy_no, SL_CONTROL, sl_control);
 }
 
+static enum sas_linkrate phy_get_max_linkrate_v2_hw(void)
+{
+	return SAS_LINK_RATE_12_0_GBPS;
+}
+
+static void phy_set_linkrate_v2_hw(struct hisi_hba *hisi_hba, int phy_no,
+		struct sas_phy_linkrates *r)
+{
+	u32 prog_phy_link_rate =
+		hisi_sas_phy_read32(hisi_hba, phy_no, PROG_PHY_LINK_RATE);
+	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
+	struct asd_sas_phy *sas_phy = &phy->sas_phy;
+	int i;
+	enum sas_linkrate min, max;
+	u32 rate_mask = 0;
+
+	if (r->maximum_linkrate == SAS_LINK_RATE_UNKNOWN) {
+		max = sas_phy->phy->maximum_linkrate;
+		min = r->minimum_linkrate;
+	} else if (r->minimum_linkrate == SAS_LINK_RATE_UNKNOWN) {
+		max = r->maximum_linkrate;
+		min = sas_phy->phy->minimum_linkrate;
+	} else
+		return;
+
+	sas_phy->phy->maximum_linkrate = max;
+	sas_phy->phy->minimum_linkrate = min;
+
+	min -= SAS_LINK_RATE_1_5_GBPS;
+	max -= SAS_LINK_RATE_1_5_GBPS;
+
+	for (i = 0; i <= max; i++)
+		rate_mask |= 1 << (i * 2);
+
+	prog_phy_link_rate &= ~0xff;
+	prog_phy_link_rate |= rate_mask;
+
+	hisi_sas_phy_write32(hisi_hba, phy_no, PROG_PHY_LINK_RATE,
+			prog_phy_link_rate);
+
+	phy_hard_reset_v2_hw(hisi_hba, phy_no);
+}
+
 static int get_wideport_bitmap_v2_hw(struct hisi_hba *hisi_hba, int port_id)
 {
 	int i, bitmap = 0;
@@ -2739,6 +2782,8 @@ static int hisi_sas_v2_init(struct hisi_hba *hisi_hba)
 	.phy_enable = enable_phy_v2_hw,
 	.phy_disable = disable_phy_v2_hw,
 	.phy_hard_reset = phy_hard_reset_v2_hw,
+	.phy_set_linkrate = phy_set_linkrate_v2_hw,
+	.phy_get_max_linkrate = phy_get_max_linkrate_v2_hw,
 	.max_command_entries = HISI_SAS_COMMAND_ENTRIES_V2_HW,
 	.complete_hdr_size = sizeof(struct hisi_sas_complete_v2_hdr),
 };
-- 
1.9.1

Re: [PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[Martin K. Petersen]

>>>>> "John" == John Garry <john.garry@huawei.com> writes:

John> This patchset introduces some misc bug fixes, improvements, and
John> new features to the HiSilicon SAS driver.

This series needs review.

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 02/11] hisi_sas: alloc queue id of slot according to device id

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Currently slots are allocated from queues in a round-robin fashion.
> This causes a problem for internal commands in device mode. For this
> mode, we should ensure that the internal abort command is the last
> command seen in the host for that device. We can only ensure this when
> we place the internal abort command after the preceding commands for
> device that in the same queue, as there is no order in which the host
> will select a queue to execute the next command.

Is there performance penalty, since only one queue is supported for a device.

>
> This queue restriction makes supporting scsi mq more tricky in
> the future, but should not be a blocker.
>
> Note: Even though v1 hw does not support internal abort, the
>       allocation method is chosen to be the same for consistency.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 04/11] hisi_sas: fix port form bug in hisi_sas_port_notify_formed()

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> When we form a wideport, we should use hardware PHY port_id instead
> of sas_phy->id.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 07/11] hisi_sas: delete repeated configuration in free_device_v2_hw()

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Delete repeated configuration items for hisi_sas_device() when
> we free a device. These items are now only set in
> hisi_sas_dev_gone().
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 10/11] hisi_sas: use atomic64_t for hisi_sas_device.running_req

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> Sometimes the value of hisi_sas_device.running_req
> would go negative unless we have the check for
> running_req >= 0 before trying to decrement.
>
> This is because using running_req is not thread-safe.
>
> As such, the value for running_req may be actually incorrect,
> so use atomic64_t instead.
>
> Signed-off-by: John Garry <john.garry@huawei.com>
> Reviewed-by: Xiang Chen <chenxiang66@hisilicon.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 08/11] hisi_sas: modify some values in get_ata_protocol()

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Modify and add some SATA commands according to SATA protocol.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 01/11] hisi_sas: add v2 hw support for ECC and AXI bus fatal error

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> For ECC 1bit error, logic can recover it, so we only print
> a warning.
> For ECC multi-bit and AXI bus fatal error, we panic.

Is it possible to recover via resetting phy and device etc instead of panic?

Thanks

Re: [PATCH 03/11] hisi_sas: only process broadcast change in phy_bcast_v2_hw()

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> There are many BROADCAST primitives generated by the host.
> We are only interested in BROADCAST (CHANGE) primitives currently,
> so only process this.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 11/11] hisi_sas: add PHY set linkrate support for v1 and v2 hw

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Add the function to set PHY min and max linkrate through
> sysfs interface.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 09/11] hisi_sas: check SATA FIS when directly attaching SATA device

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Check ERR bit of status to decide whether there is something wrong with
> initial register-D2H FIS. If error exists, PHY reset the channel to
> restart OOB.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 06/11] hisi_sas: modify return value of hisi_sas_query_task()

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> sas_scsi_find_task() only deals with return value
> TMF_RESP_FUNC_FAILED/TMF_RESP_FUNC_SUCC/TMF_RESP_FUNC_COMPLETE of
> query task. So for LLDD errors just return TMF_RESP_FUNC_FAILED.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Re: [PATCH 02/11] hisi_sas: alloc queue id of slot according to device id

[John Garry]

On 16/11/2016 01:41, Zhangfei Gao wrote:
> On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
>> From: Xiang Chen <chenxiang66@hisilicon.com>
>>
>> Currently slots are allocated from queues in a round-robin fashion.
>> This causes a problem for internal commands in device mode. For this
>> mode, we should ensure that the internal abort command is the last
>> command seen in the host for that device. We can only ensure this when
>> we place the internal abort command after the preceding commands for
>> device that in the same queue, as there is no order in which the host
>> will select a queue to execute the next command.
>
> Is there performance penalty, since only one queue is supported for a device.

Hi Zhangfei,

 From testing I have not seen any noteable performance change. However, 
please note the comment on mq, below.

Cheers,
John

>
>>
>> This queue restriction makes supporting scsi mq more tricky in
>> the future, but should not be a blocker.
>>
>> Note: Even though v1 hw does not support internal abort, the
>>       allocation method is chosen to be the same for consistency.
>>
>> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
>> Signed-off-by: John Garry <john.garry@huawei.com>
>
> Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
> .
>

Re: [PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[John Garry]

On 14/11/2016 23:49, Martin K. Petersen wrote:
>>>>>> "John" == John Garry <john.garry@huawei.com> writes:
>
> John> This patchset introduces some misc bug fixes, improvements, and
> John> new features to the HiSilicon SAS driver.
>
> This series needs review.
>

Hi Martin,

Are you happy with this patchset now that I've got an external review?

Thanks,
John

Re: [PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[Martin K. Petersen]

>>>>> "John" == John Garry <john.garry@huawei.com> writes:

John,

John> Are you happy with this patchset now that I've got an external
John> review?

Zhangfei Geo asked a question about patch 1/11 that has yet to be
answered. Patch 5/11 is still unreviewed.

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 05/11] hisi_sas: replace WARN_ON() with dev_warn() for internal abort

[Zhangfei Gao]

On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
> From: Xiang Chen <chenxiang66@hisilicon.com>
>
> Replace WARN_ON() with dev_warn() print when internal abort fails.
>
> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
> Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

Sorry, miss this one.

Re: [PATCH 01/11] hisi_sas: add v2 hw support for ECC and AXI bus fatal error

[John Garry]

On 16/11/2016 01:47, Zhangfei Gao wrote:
> On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
>> From: Xiang Chen <chenxiang66@hisilicon.com>
>>
>> For ECC 1bit error, logic can recover it, so we only print
>> a warning.
>> For ECC multi-bit and AXI bus fatal error, we panic.
>
> Is it possible to recover via resetting phy and device etc instead of panic?
>
> Thanks
>
>


Hi Zhangfei,

We are actually now working on supporting controller reset for certain 
AXI/ECC errors, so that we will not need to panic.

Thanks,
John

Re: [PATCH 01/11] hisi_sas: add v2 hw support for ECC and AXI bus fatal error

[Zhangfei Gao]

On Wed, Nov 23, 2016 at 4:59 PM, John Garry <john.garry@huawei.com> wrote:
> On 16/11/2016 01:47, Zhangfei Gao wrote:
>>
>> On Mon, Nov 7, 2016 at 8:48 PM, John Garry <john.garry@huawei.com> wrote:
>>>
>>> From: Xiang Chen <chenxiang66@hisilicon.com>

Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>

>>>
>>> For ECC 1bit error, logic can recover it, so we only print
>>> a warning.
>>> For ECC multi-bit and AXI bus fatal error, we panic.
>>
>>
>> Is it possible to recover via resetting phy and device etc instead of
>> panic?
>>
>> Thanks
>>
>>
>
>
> Hi Zhangfei,
>
> We are actually now working on supporting controller reset for certain
> AXI/ECC errors, so that we will not need to panic.

Got it, thanks for the info.

Thanks

Re: [PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[John Garry]

On 22/11/2016 22:01, Martin K. Petersen wrote:
>>>>>> "John" == John Garry <john.garry@huawei.com> writes:
>
> John,
>
> John> Are you happy with this patchset now that I've got an external
> John> review?
>
> Zhangfei Geo asked a question about patch 1/11 that has yet to be
> answered. Patch 5/11 is still unreviewed.
>

Hi Martin,

I think that these 2 outstanding issues have been addressed. Please let 
me know if there is anything else.

Thanks alot,
John

Re: [PATCH 00/11] hisi_sas: some fixes, improvements, and new features

[Martin K. Petersen]

>>>>> "John" == John Garry <john.garry@huawei.com> writes:

John> I think that these 2 outstanding issues have been
John> addressed. Please let me know if there is anything else.

Series applied to 4.10/scsi-queue.

-- 
Martin K. Petersen	Oracle Linux Engineering