[PATCH v2 for-rc 0/3] RDMA/bnxt_re: Bug fixes

[PATCH v2 for-rc 0/3] RDMA/bnxt_re: Bug fixes

[Selvin Xavier]

[-- Attachment #1: Type: text/plain, Size: 606 bytes --]

Includes some of the generic bug fixes in bnxt_re driver.
Please review and apply.

Thanks,
Selvin Xavier

v1 -> v2:
 Dropped 7 patches from this series which
 will be posted against for-next branch.

Kalesh AP (3):
  RDMA/bnxt_re: Fix a possible memory leak
  RDMA/bnxt_re: Fix return value of bnxt_re_process_raw_qp_pkt_rx
  RDMA/bnxt_re: Do not enable congestion control on VFs

 drivers/infiniband/hw/bnxt_re/ib_verbs.c |  2 +-
 drivers/infiniband/hw/bnxt_re/main.c     |  4 ++++
 drivers/infiniband/hw/bnxt_re/qplib_fp.c | 11 ++++++-----
 3 files changed, 11 insertions(+), 6 deletions(-)

-- 
2.5.5


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4224 bytes --]

[PATCH v2 for-rc 1/3] RDMA/bnxt_re: Fix a possible memory leak

[Selvin Xavier]

[-- Attachment #1: Type: text/plain, Size: 1660 bytes --]

From: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>

Inside bnxt_qplib_create_cq(), when the check for NULL DPI
fails, driver returns directly without freeing the memory
allocated inside bnxt_qplib_alloc_init_hwq() routine.

Fixed this by moving the check for NULL DPI before invoking
bnxt_qplib_alloc_init_hwq().

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Reviewed-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
---
 drivers/infiniband/hw/bnxt_re/qplib_fp.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
index f139d4c..8974f62 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
@@ -2056,6 +2056,12 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq)
 	u32 pg_sz_lvl;
 	int rc;
 
+	if (!cq->dpi) {
+		dev_err(&rcfw->pdev->dev,
+			"FP: CREATE_CQ failed due to NULL DPI\n");
+		return -EINVAL;
+	}
+
 	hwq_attr.res = res;
 	hwq_attr.depth = cq->max_wqe;
 	hwq_attr.stride = sizeof(struct cq_base);
@@ -2069,11 +2075,6 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq)
 				 CMDQ_BASE_OPCODE_CREATE_CQ,
 				 sizeof(req));
 
-	if (!cq->dpi) {
-		dev_err(&rcfw->pdev->dev,
-			"FP: CREATE_CQ failed due to NULL DPI\n");
-		return -EINVAL;
-	}
 	req.dpi = cpu_to_le32(cq->dpi->dpi);
 	req.cq_handle = cpu_to_le64(cq->cq_handle);
 	req.cq_size = cpu_to_le32(cq->hwq.max_elements);
-- 
2.5.5


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4224 bytes --]

[PATCH v2 for-rc 2/3] RDMA/bnxt_re: Fix return value of bnxt_re_process_raw_qp_pkt_rx

[Selvin Xavier]

[-- Attachment #1: Type: text/plain, Size: 1080 bytes --]

From: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>

bnxt_re_process_raw_qp_pkt_rx() always return 0 and ignores
the return value of bnxt_re_post_send_shadow_qp().

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Reviewed-by: Hongguang Gao <hongguang.gao@broadcom.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
---
 drivers/infiniband/hw/bnxt_re/ib_verbs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index e86afec..0b581c2 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -3343,7 +3343,7 @@ static int bnxt_re_process_raw_qp_pkt_rx(struct bnxt_re_qp *gsi_qp,
 	/* post data received  in the send queue */
 	rc = bnxt_re_post_send_shadow_qp(rdev, gsi_sqp, swr);
 
-	return 0;
+	return rc;
 }
 
 static void bnxt_re_process_res_rawqp1_wc(struct ib_wc *wc,
-- 
2.5.5


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4224 bytes --]

[PATCH v2 for-rc 3/3] RDMA/bnxt_re: Do not enable congestion control on VFs

[Selvin Xavier]

[-- Attachment #1: Type: text/plain, Size: 1086 bytes --]

From: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>

Congestion control needs to be enabled only on the PFs. FW fails
the command if issued on VFs. Avoid sending the command on VFs.

Fixes: f13bcef04ba0 ("RDMA/bnxt_re: Enable congestion control by default")
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Selvin Thyparampil Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
---
 drivers/infiniband/hw/bnxt_re/main.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
index b9e2f89..e34eccd 100644
--- a/drivers/infiniband/hw/bnxt_re/main.c
+++ b/drivers/infiniband/hw/bnxt_re/main.c
@@ -1336,6 +1336,10 @@ static void bnxt_re_setup_cc(struct bnxt_re_dev *rdev, bool enable)
 {
 	struct bnxt_qplib_cc_param cc_param = {};
 
+	/* Do not enable congestion control on VFs */
+	if (rdev->is_virtfn)
+		return;
+
 	/* Currently enabling only for GenP5 adapters */
 	if (!bnxt_qplib_is_chip_gen_p5(rdev->chip_ctx))
 		return;
-- 
2.5.5


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4224 bytes --]

Re: [PATCH v2 for-rc 0/3] RDMA/bnxt_re: Bug fixes

[Jason Gunthorpe]

On Thu, May 18, 2023 at 01:10:58AM -0700, Selvin Xavier wrote:
> Includes some of the generic bug fixes in bnxt_re driver.
> Please review and apply.
> 
> Thanks,
> Selvin Xavier
> 
> v1 -> v2:
>  Dropped 7 patches from this series which
>  will be posted against for-next branch.
> 
> Kalesh AP (3):
>   RDMA/bnxt_re: Fix a possible memory leak
>   RDMA/bnxt_re: Fix return value of bnxt_re_process_raw_qp_pkt_rx
>   RDMA/bnxt_re: Do not enable congestion control on VFs

Applied to for-rc, thanks

Jason