public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH for-next 0/4] RDMA/erdma: erdma updates
@ 2024-08-23  7:50 Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 1/4] RDMA/erdma: Make the device probe process more robust Cheng Xu
                   ` (3 more replies)
  0 siblings, 4 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-23  7:50 UTC (permalink / raw)
  To: jgg, leon; +Cc: linux-rdma, KaiShen

Hi,

This series has some updates for erdma driver:
- #1 always issues a reset request in probe routing to ensure that
  the hardware is ready to probe before continuing.
- #2 refactors the initialization and destruction process of EQ to
  make the code cleaner.
- #3 adds disassociate ucontext support.
- #4 returns QP state in erdma_query_qp.

Thanks,
Cheng Xu

Cheng Xu (4):
  RDMA/erdma: Make the device probe process more robust
  RDMA/erdma: Refactor the initialization and destruction of EQ
  RDMA/erdma: Add disassociate ucontext support
  RDMA/erdma: Return QP state in erdma_query_qp

 drivers/infiniband/hw/erdma/erdma.h       |  4 +-
 drivers/infiniband/hw/erdma/erdma_cmdq.c  | 26 ++-----
 drivers/infiniband/hw/erdma/erdma_eq.c    | 83 +++++++++++++----------
 drivers/infiniband/hw/erdma/erdma_main.c  | 49 ++++++++++---
 drivers/infiniband/hw/erdma/erdma_verbs.c | 29 +++++++-
 drivers/infiniband/hw/erdma/erdma_verbs.h |  1 +
 6 files changed, 121 insertions(+), 71 deletions(-)

-- 
2.31.1


^ permalink raw reply	[flat|nested] 8+ messages in thread

* [PATCH for-next 1/4] RDMA/erdma: Make the device probe process more robust
  2024-08-23  7:50 [PATCH for-next 0/4] RDMA/erdma: erdma updates Cheng Xu
@ 2024-08-23  7:50 ` Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ Cheng Xu
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-23  7:50 UTC (permalink / raw)
  To: jgg, leon; +Cc: linux-rdma, KaiShen

Driver may probe again while hardware is destroying the internal
resources allocated for previous probing, which will fail the
device probe. To make it more robust, we always issue a reset at the
beginning of the device probe process.

Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
---
 drivers/infiniband/hw/erdma/erdma.h      |  1 +
 drivers/infiniband/hw/erdma/erdma_main.c | 44 +++++++++++++++++++-----
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/drivers/infiniband/hw/erdma/erdma.h b/drivers/infiniband/hw/erdma/erdma.h
index c8bd698e21b0..b5c258f77ca0 100644
--- a/drivers/infiniband/hw/erdma/erdma.h
+++ b/drivers/infiniband/hw/erdma/erdma.h
@@ -94,6 +94,7 @@ enum {
 
 #define ERDMA_CMDQ_TIMEOUT_MS 15000
 #define ERDMA_REG_ACCESS_WAIT_MS 20
+#define ERDMA_WAIT_DEV_REST_CNT 50
 #define ERDMA_WAIT_DEV_DONE_CNT 500
 
 struct erdma_cmdq {
diff --git a/drivers/infiniband/hw/erdma/erdma_main.c b/drivers/infiniband/hw/erdma/erdma_main.c
index 7080f8a71ec4..9199058a0b29 100644
--- a/drivers/infiniband/hw/erdma/erdma_main.c
+++ b/drivers/infiniband/hw/erdma/erdma_main.c
@@ -209,11 +209,30 @@ static void erdma_device_uninit(struct erdma_dev *dev)
 	dma_pool_destroy(dev->resp_pool);
 }
 
-static void erdma_hw_reset(struct erdma_dev *dev)
+static int erdma_hw_reset(struct erdma_dev *dev, bool wait)
 {
 	u32 ctrl = FIELD_PREP(ERDMA_REG_DEV_CTRL_RESET_MASK, 1);
+	int i;
 
 	erdma_reg_write32(dev, ERDMA_REGS_DEV_CTRL_REG, ctrl);
+
+	if (!wait)
+		return 0;
+
+	for (i = 0; i < ERDMA_WAIT_DEV_REST_CNT; i++) {
+		if (erdma_reg_read32_filed(dev, ERDMA_REGS_DEV_ST_REG,
+					   ERDMA_REG_DEV_ST_RESET_DONE_MASK))
+			break;
+
+		msleep(ERDMA_REG_ACCESS_WAIT_MS);
+	}
+
+	if (i == ERDMA_WAIT_DEV_REST_CNT) {
+		dev_err(&dev->pdev->dev, "wait reset done timeout.\n");
+		return -ETIME;
+	}
+
+	return 0;
 }
 
 static int erdma_wait_hw_init_done(struct erdma_dev *dev)
@@ -239,6 +258,17 @@ static int erdma_wait_hw_init_done(struct erdma_dev *dev)
 	return 0;
 }
 
+static int erdma_preinit_check(struct erdma_dev *dev)
+{
+	u32 version = erdma_reg_read32(dev, ERDMA_REGS_VERSION_REG);
+
+	/* we knows that it is a non-functional function. */
+	if (version == 0)
+		return -ENODEV;
+
+	return erdma_hw_reset(dev, true);
+}
+
 static const struct pci_device_id erdma_pci_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_ALIBABA, 0x107f) },
 	{}
@@ -248,7 +278,6 @@ static int erdma_probe_dev(struct pci_dev *pdev)
 {
 	struct erdma_dev *dev;
 	int bars, err;
-	u32 version;
 
 	err = pci_enable_device(pdev);
 	if (err) {
@@ -287,12 +316,9 @@ static int erdma_probe_dev(struct pci_dev *pdev)
 		goto err_release_bars;
 	}
 
-	version = erdma_reg_read32(dev, ERDMA_REGS_VERSION_REG);
-	if (version == 0) {
-		/* we knows that it is a non-functional function. */
-		err = -ENODEV;
+	err = erdma_preinit_check(dev);
+	if (err)
 		goto err_iounmap_func_bar;
-	}
 
 	err = erdma_device_init(dev, pdev);
 	if (err)
@@ -327,7 +353,7 @@ static int erdma_probe_dev(struct pci_dev *pdev)
 	return 0;
 
 err_reset_hw:
-	erdma_hw_reset(dev);
+	erdma_hw_reset(dev, false);
 
 err_uninit_cmdq:
 	erdma_cmdq_destroy(dev);
@@ -364,7 +390,7 @@ static void erdma_remove_dev(struct pci_dev *pdev)
 	struct erdma_dev *dev = pci_get_drvdata(pdev);
 
 	erdma_ceqs_uninit(dev);
-	erdma_hw_reset(dev);
+	erdma_hw_reset(dev, false);
 	erdma_cmdq_destroy(dev);
 	erdma_aeq_destroy(dev);
 	erdma_comm_irq_uninit(dev);
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
  2024-08-23  7:50 [PATCH for-next 0/4] RDMA/erdma: erdma updates Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 1/4] RDMA/erdma: Make the device probe process more robust Cheng Xu
@ 2024-08-23  7:50 ` Cheng Xu
  2024-08-26 13:51   ` kernel test robot
  2024-08-26 16:05   ` kernel test robot
  2024-08-23  7:50 ` [PATCH for-next 3/4] RDMA/erdma: Add disassociate ucontext support Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 4/4] RDMA/erdma: Return QP state in erdma_query_qp Cheng Xu
  3 siblings, 2 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-23  7:50 UTC (permalink / raw)
  To: jgg, leon; +Cc: linux-rdma, KaiShen

We extracted the common parts of the initialization/destruction
process to make the code cleaner.

Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
---
 drivers/infiniband/hw/erdma/erdma.h      |  3 +-
 drivers/infiniband/hw/erdma/erdma_cmdq.c | 26 ++------
 drivers/infiniband/hw/erdma/erdma_eq.c   | 83 +++++++++++++-----------
 drivers/infiniband/hw/erdma/erdma_main.c |  4 +-
 4 files changed, 55 insertions(+), 61 deletions(-)

diff --git a/drivers/infiniband/hw/erdma/erdma.h b/drivers/infiniband/hw/erdma/erdma.h
index b5c258f77ca0..f3b648dcf4b6 100644
--- a/drivers/infiniband/hw/erdma/erdma.h
+++ b/drivers/infiniband/hw/erdma/erdma.h
@@ -275,7 +275,8 @@ void notify_eq(struct erdma_eq *eq);
 void *get_next_valid_eqe(struct erdma_eq *eq);
 
 int erdma_aeq_init(struct erdma_dev *dev);
-void erdma_aeq_destroy(struct erdma_dev *dev);
+int erdma_eq_common_init(struct erdma_dev *dev, struct erdma_eq *eq, u32 depth);
+void erdma_eq_destroy(struct erdma_dev *dev, struct erdma_eq *eq);
 
 void erdma_aeq_event_handler(struct erdma_dev *dev);
 void erdma_ceq_completion_handler(struct erdma_eq_cb *ceq_cb);
diff --git a/drivers/infiniband/hw/erdma/erdma_cmdq.c b/drivers/infiniband/hw/erdma/erdma_cmdq.c
index 43ff40b5a09d..a3d8922d1ad1 100644
--- a/drivers/infiniband/hw/erdma/erdma_cmdq.c
+++ b/drivers/infiniband/hw/erdma/erdma_cmdq.c
@@ -158,20 +158,13 @@ static int erdma_cmdq_eq_init(struct erdma_dev *dev)
 {
 	struct erdma_cmdq *cmdq = &dev->cmdq;
 	struct erdma_eq *eq = &cmdq->eq;
+	int ret;
 
-	eq->depth = cmdq->max_outstandings;
-	eq->qbuf = dma_alloc_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT,
-				      &eq->qbuf_dma_addr, GFP_KERNEL);
-	if (!eq->qbuf)
-		return -ENOMEM;
-
-	spin_lock_init(&eq->lock);
-	atomic64_set(&eq->event_num, 0);
+	ret = erdma_eq_common_init(dev, eq, cmdq->max_outstandings);
+	if (ret)
+		return ret;
 
 	eq->db = dev->func_bar + ERDMA_REGS_CEQ_DB_BASE_REG;
-	eq->dbrec = dma_pool_zalloc(dev->db_pool, GFP_KERNEL, &eq->dbrec_dma);
-	if (!eq->dbrec)
-		goto err_out;
 
 	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_EQ_ADDR_H_REG,
 			  upper_32_bits(eq->qbuf_dma_addr));
@@ -181,12 +174,6 @@ static int erdma_cmdq_eq_init(struct erdma_dev *dev)
 	erdma_reg_write64(dev, ERDMA_CMDQ_EQ_DB_HOST_ADDR_REG, eq->dbrec_dma);
 
 	return 0;
-
-err_out:
-	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
-			  eq->qbuf_dma_addr);
-
-	return -ENOMEM;
 }
 
 int erdma_cmdq_init(struct erdma_dev *dev)
@@ -247,10 +234,7 @@ void erdma_cmdq_destroy(struct erdma_dev *dev)
 
 	clear_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state);
 
-	dma_free_coherent(&dev->pdev->dev, cmdq->eq.depth << EQE_SHIFT,
-			  cmdq->eq.qbuf, cmdq->eq.qbuf_dma_addr);
-
-	dma_pool_free(dev->db_pool, cmdq->eq.dbrec, cmdq->eq.dbrec_dma);
+	erdma_eq_destroy(dev, &cmdq->eq);
 
 	dma_free_coherent(&dev->pdev->dev, cmdq->sq.depth << SQEBB_SHIFT,
 			  cmdq->sq.qbuf, cmdq->sq.qbuf_dma_addr);
diff --git a/drivers/infiniband/hw/erdma/erdma_eq.c b/drivers/infiniband/hw/erdma/erdma_eq.c
index 84ccdd8144c9..227c742c43df 100644
--- a/drivers/infiniband/hw/erdma/erdma_eq.c
+++ b/drivers/infiniband/hw/erdma/erdma_eq.c
@@ -80,25 +80,53 @@ void erdma_aeq_event_handler(struct erdma_dev *dev)
 	notify_eq(&dev->aeq);
 }
 
-int erdma_aeq_init(struct erdma_dev *dev)
+int erdma_eq_common_init(struct erdma_dev *dev, struct erdma_eq *eq, u32 depth)
 {
-	struct erdma_eq *eq = &dev->aeq;
-
-	eq->depth = ERDMA_DEFAULT_EQ_DEPTH;
+	u32 buf_size = depth << EQE_SHIFT;
 
-	eq->qbuf = dma_alloc_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT,
-				      &eq->qbuf_dma_addr, GFP_KERNEL);
+	eq->qbuf = dma_alloc_coherent(&dev->pdev->dev, buf_size,
+				      &eq->qbuf_dma_addr,
+				      GFP_KERNEL | __GFP_ZERO);
 	if (!eq->qbuf)
 		return -ENOMEM;
 
+	eq->dbrec = dma_pool_alloc(dev->db_pool, GFP_KERNEL | __GFP_ZERO,
+				   &eq->dbrec_dma);
+	if (!eq->dbrec)
+		goto err_free_qbuf;
+
 	spin_lock_init(&eq->lock);
 	atomic64_set(&eq->event_num, 0);
 	atomic64_set(&eq->notify_num, 0);
+	eq->ci = 0;
+	eq->depth = depth;
+
+	return 0;
+
+err_free_qbuf:
+	dma_free_coherent(&dev->pdev->dev, buf_size, eq->qbuf,
+			  eq->qbuf_dma_addr);
+
+	return -ENOMEM;
+}
+
+void erdma_eq_destroy(struct erdma_dev *dev, struct erdma_eq *eq)
+{
+	dma_pool_free(dev->db_pool, eq->dbrec, eq->dbrec_dma);
+	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
+			  eq->qbuf_dma_addr);
+}
+
+int erdma_aeq_init(struct erdma_dev *dev)
+{
+	struct erdma_eq *eq = &dev->aeq;
+	int ret;
+
+	ret = erdma_eq_common_init(dev, &dev->aeq, ERDMA_DEFAULT_EQ_DEPTH);
+	if (ret)
+		return ret;
 
 	eq->db = dev->func_bar + ERDMA_REGS_AEQ_DB_REG;
-	eq->dbrec = dma_pool_zalloc(dev->db_pool, GFP_KERNEL, &eq->dbrec_dma);
-	if (!eq->dbrec)
-		goto err_out;
 
 	erdma_reg_write32(dev, ERDMA_REGS_AEQ_ADDR_H_REG,
 			  upper_32_bits(eq->qbuf_dma_addr));
@@ -108,12 +136,6 @@ int erdma_aeq_init(struct erdma_dev *dev)
 	erdma_reg_write64(dev, ERDMA_AEQ_DB_HOST_ADDR_REG, eq->dbrec_dma);
 
 	return 0;
-
-err_out:
-	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
-			  eq->qbuf_dma_addr);
-
-	return -ENOMEM;
 }
 
 void erdma_aeq_destroy(struct erdma_dev *dev)
@@ -234,32 +256,21 @@ static int erdma_ceq_init_one(struct erdma_dev *dev, u16 ceqn)
 	struct erdma_eq *eq = &dev->ceqs[ceqn].eq;
 	int ret;
 
-	eq->depth = ERDMA_DEFAULT_EQ_DEPTH;
-	eq->qbuf = dma_alloc_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT,
-				      &eq->qbuf_dma_addr, GFP_KERNEL);
-	if (!eq->qbuf)
-		return -ENOMEM;
-
-	spin_lock_init(&eq->lock);
-	atomic64_set(&eq->event_num, 0);
-	atomic64_set(&eq->notify_num, 0);
+	ret = erdma_eq_common_init(dev, eq, ERDMA_DEFAULT_EQ_DEPTH);
+	if (ret)
+		return ret;
 
 	eq->db = dev->func_bar + ERDMA_REGS_CEQ_DB_BASE_REG +
 		 (ceqn + 1) * ERDMA_DB_SIZE;
-
-	eq->dbrec = dma_pool_zalloc(dev->db_pool, GFP_KERNEL, &eq->dbrec_dma);
-	if (!eq->dbrec) {
-		dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT,
-				  eq->qbuf, eq->qbuf_dma_addr);
-		return -ENOMEM;
-	}
-
-	eq->ci = 0;
 	dev->ceqs[ceqn].dev = dev;
+	dev->ceqs[ceqn].ready = true;
 
 	/* CEQ indexed from 1, 0 rsvd for CMDQ-EQ. */
 	ret = create_eq_cmd(dev, ceqn + 1, eq);
-	dev->ceqs[ceqn].ready = ret ? false : true;
+	if (ret) {
+		erdma_eq_destroy(dev, eq);
+		dev->ceqs[ceqn].ready = false;
+	}
 
 	return ret;
 }
@@ -283,9 +294,7 @@ static void erdma_ceq_uninit_one(struct erdma_dev *dev, u16 ceqn)
 	if (err)
 		return;
 
-	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
-			  eq->qbuf_dma_addr);
-	dma_pool_free(dev->db_pool, eq->dbrec, eq->dbrec_dma);
+	erdma_eq_destroy(dev, eq);
 }
 
 int erdma_ceqs_init(struct erdma_dev *dev)
diff --git a/drivers/infiniband/hw/erdma/erdma_main.c b/drivers/infiniband/hw/erdma/erdma_main.c
index 9199058a0b29..d1cb488e7ad4 100644
--- a/drivers/infiniband/hw/erdma/erdma_main.c
+++ b/drivers/infiniband/hw/erdma/erdma_main.c
@@ -359,7 +359,7 @@ static int erdma_probe_dev(struct pci_dev *pdev)
 	erdma_cmdq_destroy(dev);
 
 err_uninit_aeq:
-	erdma_aeq_destroy(dev);
+	erdma_eq_destroy(dev, &dev->aeq);
 
 err_uninit_comm_irq:
 	erdma_comm_irq_uninit(dev);
@@ -392,7 +392,7 @@ static void erdma_remove_dev(struct pci_dev *pdev)
 	erdma_ceqs_uninit(dev);
 	erdma_hw_reset(dev, false);
 	erdma_cmdq_destroy(dev);
-	erdma_aeq_destroy(dev);
+	erdma_eq_destroy(dev, &dev->aeq);
 	erdma_comm_irq_uninit(dev);
 	pci_free_irq_vectors(dev->pdev);
 	erdma_device_uninit(dev);
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [PATCH for-next 3/4] RDMA/erdma: Add disassociate ucontext support
  2024-08-23  7:50 [PATCH for-next 0/4] RDMA/erdma: erdma updates Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 1/4] RDMA/erdma: Make the device probe process more robust Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ Cheng Xu
@ 2024-08-23  7:50 ` Cheng Xu
  2024-08-23  7:50 ` [PATCH for-next 4/4] RDMA/erdma: Return QP state in erdma_query_qp Cheng Xu
  3 siblings, 0 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-23  7:50 UTC (permalink / raw)
  To: jgg, leon; +Cc: linux-rdma, KaiShen

All IO pages mapped to user space are handled by rdma_user_mmap_io,
so add empty stub for disassociate ucontext.

Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
---
 drivers/infiniband/hw/erdma/erdma_main.c  | 1 +
 drivers/infiniband/hw/erdma/erdma_verbs.c | 4 ++++
 drivers/infiniband/hw/erdma/erdma_verbs.h | 1 +
 3 files changed, 6 insertions(+)

diff --git a/drivers/infiniband/hw/erdma/erdma_main.c b/drivers/infiniband/hw/erdma/erdma_main.c
index d1cb488e7ad4..1ccf1b65c02c 100644
--- a/drivers/infiniband/hw/erdma/erdma_main.c
+++ b/drivers/infiniband/hw/erdma/erdma_main.c
@@ -516,6 +516,7 @@ static const struct ib_device_ops erdma_device_ops = {
 	.dereg_mr = erdma_dereg_mr,
 	.destroy_cq = erdma_destroy_cq,
 	.destroy_qp = erdma_destroy_qp,
+	.disassociate_ucontext = erdma_disassociate_ucontext,
 	.get_dma_mr = erdma_get_dma_mr,
 	.get_hw_stats = erdma_get_hw_stats,
 	.get_port_immutable = erdma_get_port_immutable,
diff --git a/drivers/infiniband/hw/erdma/erdma_verbs.c b/drivers/infiniband/hw/erdma/erdma_verbs.c
index 40c9b6e46b82..48b08a15e6a8 100644
--- a/drivers/infiniband/hw/erdma/erdma_verbs.c
+++ b/drivers/infiniband/hw/erdma/erdma_verbs.c
@@ -1700,6 +1700,10 @@ int erdma_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 	return ret;
 }
 
+void erdma_disassociate_ucontext(struct ib_ucontext *ibcontext)
+{
+}
+
 void erdma_set_mtu(struct erdma_dev *dev, u32 mtu)
 {
 	struct erdma_cmdq_config_mtu_req req;
diff --git a/drivers/infiniband/hw/erdma/erdma_verbs.h b/drivers/infiniband/hw/erdma/erdma_verbs.h
index 4f02ba06b210..b7478376eb80 100644
--- a/drivers/infiniband/hw/erdma/erdma_verbs.h
+++ b/drivers/infiniband/hw/erdma/erdma_verbs.h
@@ -344,6 +344,7 @@ int erdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int mask,
 		    struct ib_udata *data);
 int erdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata);
 int erdma_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata);
+void erdma_disassociate_ucontext(struct ib_ucontext *ibcontext);
 int erdma_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags);
 struct ib_mr *erdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
 				u64 virt, int access, struct ib_udata *udata);
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [PATCH for-next 4/4] RDMA/erdma: Return QP state in erdma_query_qp
  2024-08-23  7:50 [PATCH for-next 0/4] RDMA/erdma: erdma updates Cheng Xu
                   ` (2 preceding siblings ...)
  2024-08-23  7:50 ` [PATCH for-next 3/4] RDMA/erdma: Add disassociate ucontext support Cheng Xu
@ 2024-08-23  7:50 ` Cheng Xu
  3 siblings, 0 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-23  7:50 UTC (permalink / raw)
  To: jgg, leon; +Cc: linux-rdma, KaiShen

Fix qp_state and cur_qp_state to return correct values in
struct ib_qp_attr.

Fixes: 155055771704 ("RDMA/erdma: Add verbs implementation")
Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
---
 drivers/infiniband/hw/erdma/erdma_verbs.c | 25 ++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/erdma/erdma_verbs.c b/drivers/infiniband/hw/erdma/erdma_verbs.c
index 48b08a15e6a8..de11f0f1adb1 100644
--- a/drivers/infiniband/hw/erdma/erdma_verbs.c
+++ b/drivers/infiniband/hw/erdma/erdma_verbs.c
@@ -1544,11 +1544,31 @@ int erdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
 	return ret;
 }
 
+static inline enum ib_qp_state query_qp_state(struct erdma_qp *qp)
+{
+	switch (qp->attrs.state) {
+	case ERDMA_QP_STATE_IDLE:
+		return IB_QPS_INIT;
+	case ERDMA_QP_STATE_RTR:
+		return IB_QPS_RTR;
+	case ERDMA_QP_STATE_RTS:
+		return IB_QPS_RTS;
+	case ERDMA_QP_STATE_CLOSING:
+		return IB_QPS_ERR;
+	case ERDMA_QP_STATE_TERMINATE:
+		return IB_QPS_ERR;
+	case ERDMA_QP_STATE_ERROR:
+		return IB_QPS_ERR;
+	default:
+		return IB_QPS_ERR;
+	}
+}
+
 int erdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
 		   int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr)
 {
-	struct erdma_qp *qp;
 	struct erdma_dev *dev;
+	struct erdma_qp *qp;
 
 	if (ibqp && qp_attr && qp_init_attr) {
 		qp = to_eqp(ibqp);
@@ -1575,6 +1595,9 @@ int erdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
 
 	qp_init_attr->cap = qp_attr->cap;
 
+	qp_attr->qp_state = query_qp_state(qp);
+	qp_attr->cur_qp_state = query_qp_state(qp);
+
 	return 0;
 }
 
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* Re: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
  2024-08-23  7:50 ` [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ Cheng Xu
@ 2024-08-26 13:51   ` kernel test robot
  2024-08-27  1:25     ` Cheng Xu
  2024-08-26 16:05   ` kernel test robot
  1 sibling, 1 reply; 8+ messages in thread
From: kernel test robot @ 2024-08-26 13:51 UTC (permalink / raw)
  To: Cheng Xu, jgg, leon; +Cc: oe-kbuild-all, linux-rdma, KaiShen

Hi Cheng,

kernel test robot noticed the following build warnings:

[auto build test WARNING on rdma/for-next]
[also build test WARNING on linus/master v6.11-rc5 next-20240826]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Cheng-Xu/RDMA-erdma-Make-the-device-probe-process-more-robust/20240826-123256
base:   https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git for-next
patch link:    https://lore.kernel.org/r/20240823075058.89488-3-chengyou%40linux.alibaba.com
patch subject: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
config: s390-allyesconfig (https://download.01.org/0day-ci/archive/20240826/202408262144.SpsbTKs7-lkp@intel.com/config)
compiler: s390-linux-gcc (GCC) 14.1.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408262144.SpsbTKs7-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408262144.SpsbTKs7-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> drivers/infiniband/hw/erdma/erdma_eq.c:141:6: warning: no previous prototype for 'erdma_aeq_destroy' [-Wmissing-prototypes]
     141 | void erdma_aeq_destroy(struct erdma_dev *dev)
         |      ^~~~~~~~~~~~~~~~~


vim +/erdma_aeq_destroy +141 drivers/infiniband/hw/erdma/erdma_eq.c

f2a0a630b95345 Cheng Xu 2022-07-27  140  
f2a0a630b95345 Cheng Xu 2022-07-27 @141  void erdma_aeq_destroy(struct erdma_dev *dev)
f2a0a630b95345 Cheng Xu 2022-07-27  142  {
f2a0a630b95345 Cheng Xu 2022-07-27  143  	struct erdma_eq *eq = &dev->aeq;
f2a0a630b95345 Cheng Xu 2022-07-27  144  
f0697bf078368d Boshi Yu 2024-03-11  145  	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
f2a0a630b95345 Cheng Xu 2022-07-27  146  			  eq->qbuf_dma_addr);
f0697bf078368d Boshi Yu 2024-03-11  147  
fdb09ed15f272a Boshi Yu 2024-03-11  148  	dma_pool_free(dev->db_pool, eq->dbrec, eq->dbrec_dma);
f2a0a630b95345 Cheng Xu 2022-07-27  149  }
f2a0a630b95345 Cheng Xu 2022-07-27  150  

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
  2024-08-23  7:50 ` [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ Cheng Xu
  2024-08-26 13:51   ` kernel test robot
@ 2024-08-26 16:05   ` kernel test robot
  1 sibling, 0 replies; 8+ messages in thread
From: kernel test robot @ 2024-08-26 16:05 UTC (permalink / raw)
  To: Cheng Xu, jgg, leon; +Cc: llvm, oe-kbuild-all, linux-rdma, KaiShen

Hi Cheng,

kernel test robot noticed the following build warnings:

[auto build test WARNING on rdma/for-next]
[also build test WARNING on linus/master v6.11-rc5 next-20240826]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Cheng-Xu/RDMA-erdma-Make-the-device-probe-process-more-robust/20240826-123256
base:   https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git for-next
patch link:    https://lore.kernel.org/r/20240823075058.89488-3-chengyou%40linux.alibaba.com
patch subject: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
config: s390-allmodconfig (https://download.01.org/0day-ci/archive/20240826/202408262310.5jeNjXWm-lkp@intel.com/config)
compiler: clang version 20.0.0git (https://github.com/llvm/llvm-project 08e5a1de8227512d4774a534b91cb2353cef6284)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408262310.5jeNjXWm-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202408262310.5jeNjXWm-lkp@intel.com/

All warnings (new ones prefixed by >>):

   In file included from include/linux/highmem.h:10:
   In file included from include/linux/mm.h:2228:
   include/linux/vmstat.h:500:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
     500 |         return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~ ^
     501 |                            item];
         |                            ~~~~
   include/linux/vmstat.h:507:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
     507 |         return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~ ^
     508 |                            NR_VM_NUMA_EVENT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~~
   include/linux/vmstat.h:514:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
     514 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
         |                               ~~~~~~~~~~~ ^ ~~~
   include/linux/vmstat.h:519:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
     519 |         return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~ ^
     520 |                            NR_VM_NUMA_EVENT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~~
   include/linux/vmstat.h:528:43: warning: arithmetic between different enumeration types ('enum zone_stat_item' and 'enum numa_stat_item') [-Wenum-enum-conversion]
     528 |         return vmstat_text[NR_VM_ZONE_STAT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~ ^
     529 |                            NR_VM_NUMA_EVENT_ITEMS +
         |                            ~~~~~~~~~~~~~~~~~~~~~~
   In file included from drivers/infiniband/hw/erdma/erdma_eq.c:7:
   In file included from drivers/infiniband/hw/erdma/erdma_verbs.h:10:
   In file included from drivers/infiniband/hw/erdma/erdma.h:11:
   In file included from include/linux/netdevice.h:38:
   In file included from include/net/net_namespace.h:43:
   In file included from include/linux/skbuff.h:28:
   In file included from include/linux/dma-mapping.h:11:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:93:
   include/asm-generic/io.h:548:31: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     548 |         val = __raw_readb(PCI_IOBASE + addr);
         |                           ~~~~~~~~~~ ^
   include/asm-generic/io.h:561:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     561 |         val = __le16_to_cpu((__le16 __force)__raw_readw(PCI_IOBASE + addr));
         |                                                         ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:37:59: note: expanded from macro '__le16_to_cpu'
      37 | #define __le16_to_cpu(x) __swab16((__force __u16)(__le16)(x))
         |                                                           ^
   include/uapi/linux/swab.h:102:54: note: expanded from macro '__swab16'
     102 | #define __swab16(x) (__u16)__builtin_bswap16((__u16)(x))
         |                                                      ^
   In file included from drivers/infiniband/hw/erdma/erdma_eq.c:7:
   In file included from drivers/infiniband/hw/erdma/erdma_verbs.h:10:
   In file included from drivers/infiniband/hw/erdma/erdma.h:11:
   In file included from include/linux/netdevice.h:38:
   In file included from include/net/net_namespace.h:43:
   In file included from include/linux/skbuff.h:28:
   In file included from include/linux/dma-mapping.h:11:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:93:
   include/asm-generic/io.h:574:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     574 |         val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
         |                                                         ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:35:59: note: expanded from macro '__le32_to_cpu'
      35 | #define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
         |                                                           ^
   include/uapi/linux/swab.h:115:54: note: expanded from macro '__swab32'
     115 | #define __swab32(x) (__u32)__builtin_bswap32((__u32)(x))
         |                                                      ^
   In file included from drivers/infiniband/hw/erdma/erdma_eq.c:7:
   In file included from drivers/infiniband/hw/erdma/erdma_verbs.h:10:
   In file included from drivers/infiniband/hw/erdma/erdma.h:11:
   In file included from include/linux/netdevice.h:38:
   In file included from include/net/net_namespace.h:43:
   In file included from include/linux/skbuff.h:28:
   In file included from include/linux/dma-mapping.h:11:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:93:
   include/asm-generic/io.h:585:33: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     585 |         __raw_writeb(value, PCI_IOBASE + addr);
         |                             ~~~~~~~~~~ ^
   include/asm-generic/io.h:595:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     595 |         __raw_writew((u16 __force)cpu_to_le16(value), PCI_IOBASE + addr);
         |                                                       ~~~~~~~~~~ ^
   include/asm-generic/io.h:605:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     605 |         __raw_writel((u32 __force)cpu_to_le32(value), PCI_IOBASE + addr);
         |                                                       ~~~~~~~~~~ ^
   include/asm-generic/io.h:693:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     693 |         readsb(PCI_IOBASE + addr, buffer, count);
         |                ~~~~~~~~~~ ^
   include/asm-generic/io.h:701:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     701 |         readsw(PCI_IOBASE + addr, buffer, count);
         |                ~~~~~~~~~~ ^
   include/asm-generic/io.h:709:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     709 |         readsl(PCI_IOBASE + addr, buffer, count);
         |                ~~~~~~~~~~ ^
   include/asm-generic/io.h:718:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     718 |         writesb(PCI_IOBASE + addr, buffer, count);
         |                 ~~~~~~~~~~ ^
   include/asm-generic/io.h:727:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     727 |         writesw(PCI_IOBASE + addr, buffer, count);
         |                 ~~~~~~~~~~ ^
   include/asm-generic/io.h:736:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
     736 |         writesl(PCI_IOBASE + addr, buffer, count);
         |                 ~~~~~~~~~~ ^
>> drivers/infiniband/hw/erdma/erdma_eq.c:141:6: warning: no previous prototype for function 'erdma_aeq_destroy' [-Wmissing-prototypes]
     141 | void erdma_aeq_destroy(struct erdma_dev *dev)
         |      ^
   drivers/infiniband/hw/erdma/erdma_eq.c:141:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
     141 | void erdma_aeq_destroy(struct erdma_dev *dev)
         | ^
         | static 
   18 warnings generated.


vim +/erdma_aeq_destroy +141 drivers/infiniband/hw/erdma/erdma_eq.c

f2a0a630b953451 Cheng Xu 2022-07-27  140  
f2a0a630b953451 Cheng Xu 2022-07-27 @141  void erdma_aeq_destroy(struct erdma_dev *dev)
f2a0a630b953451 Cheng Xu 2022-07-27  142  {
f2a0a630b953451 Cheng Xu 2022-07-27  143  	struct erdma_eq *eq = &dev->aeq;
f2a0a630b953451 Cheng Xu 2022-07-27  144  
f0697bf078368d7 Boshi Yu 2024-03-11  145  	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
f2a0a630b953451 Cheng Xu 2022-07-27  146  			  eq->qbuf_dma_addr);
f0697bf078368d7 Boshi Yu 2024-03-11  147  
fdb09ed15f272ad Boshi Yu 2024-03-11  148  	dma_pool_free(dev->db_pool, eq->dbrec, eq->dbrec_dma);
f2a0a630b953451 Cheng Xu 2022-07-27  149  }
f2a0a630b953451 Cheng Xu 2022-07-27  150  

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
  2024-08-26 13:51   ` kernel test robot
@ 2024-08-27  1:25     ` Cheng Xu
  0 siblings, 0 replies; 8+ messages in thread
From: Cheng Xu @ 2024-08-27  1:25 UTC (permalink / raw)
  To: kernel test robot, jgg, leon; +Cc: oe-kbuild-all, linux-rdma, KaiShen



On 8/26/24 9:51 PM, kernel test robot wrote:
> Hi Cheng,
> 
> kernel test robot noticed the following build warnings:
> 
> [auto build test WARNING on rdma/for-next]
> [also build test WARNING on linus/master v6.11-rc5 next-20240826]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
> 
> url:    https://github.com/intel-lab-lkp/linux/commits/Cheng-Xu/RDMA-erdma-Make-the-device-probe-process-more-robust/20240826-123256
> base:   https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git for-next
> patch link:    https://lore.kernel.org/r/20240823075058.89488-3-chengyou%40linux.alibaba.com
> patch subject: [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ
> config: s390-allyesconfig (https://download.01.org/0day-ci/archive/20240826/202408262144.SpsbTKs7-lkp@intel.com/config)
> compiler: s390-linux-gcc (GCC) 14.1.0
> reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240826/202408262144.SpsbTKs7-lkp@intel.com/reproduce)
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202408262144.SpsbTKs7-lkp@intel.com/
> 
> All warnings (new ones prefixed by >>):
> 
>>> drivers/infiniband/hw/erdma/erdma_eq.c:141:6: warning: no previous prototype for 'erdma_aeq_destroy' [-Wmissing-prototypes]
>      141 | void erdma_aeq_destroy(struct erdma_dev *dev)
>          |      ^~~~~~~~~~~~~~~~~

Forgot to remove this unused function, will fix.


> 
> vim +/erdma_aeq_destroy +141 drivers/infiniband/hw/erdma/erdma_eq.c
> 
> f2a0a630b95345 Cheng Xu 2022-07-27  140  
> f2a0a630b95345 Cheng Xu 2022-07-27 @141  void erdma_aeq_destroy(struct erdma_dev *dev)
> f2a0a630b95345 Cheng Xu 2022-07-27  142  {
> f2a0a630b95345 Cheng Xu 2022-07-27  143  	struct erdma_eq *eq = &dev->aeq;
> f2a0a630b95345 Cheng Xu 2022-07-27  144  
> f0697bf078368d Boshi Yu 2024-03-11  145  	dma_free_coherent(&dev->pdev->dev, eq->depth << EQE_SHIFT, eq->qbuf,
> f2a0a630b95345 Cheng Xu 2022-07-27  146  			  eq->qbuf_dma_addr);
> f0697bf078368d Boshi Yu 2024-03-11  147  
> fdb09ed15f272a Boshi Yu 2024-03-11  148  	dma_pool_free(dev->db_pool, eq->dbrec, eq->dbrec_dma);
> f2a0a630b95345 Cheng Xu 2022-07-27  149  }
> f2a0a630b95345 Cheng Xu 2022-07-27  150  
> 

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2024-08-27  1:26 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-08-23  7:50 [PATCH for-next 0/4] RDMA/erdma: erdma updates Cheng Xu
2024-08-23  7:50 ` [PATCH for-next 1/4] RDMA/erdma: Make the device probe process more robust Cheng Xu
2024-08-23  7:50 ` [PATCH for-next 2/4] RDMA/erdma: Refactor the initialization and destruction of EQ Cheng Xu
2024-08-26 13:51   ` kernel test robot
2024-08-27  1:25     ` Cheng Xu
2024-08-26 16:05   ` kernel test robot
2024-08-23  7:50 ` [PATCH for-next 3/4] RDMA/erdma: Add disassociate ucontext support Cheng Xu
2024-08-23  7:50 ` [PATCH for-next 4/4] RDMA/erdma: Return QP state in erdma_query_qp Cheng Xu

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox