Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1583970 > unrolled thread

[PATCH 00/29] IB/mlx: Fine-tuning for several function implementations

Started bySF Markus Elfring <elfring@users.sourceforge.net>
First post2017-02-18 21:50 +0100
Last post2017-02-18 22:50 +0100
Articles 20 on this page of 49 — 2 participants

Back to article view | Back to linux.kernel

Contents

  [PATCH 00/29] IB/mlx: Fine-tuning for several function  implementations SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 21:50 +0100
    [PATCH 04/29] IB/mlx4: Improve another size determination in  alloc_pv_object() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 04/29] IB/mlx4: Improve another size determination in  alloc_pv_object() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 09/29] IB/mlx4: Move an assignment out of a check in  forward_trap() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 09/29] IB/mlx4: Move an assignment out of a check in  forward_trap() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:20 +0100
    [PATCH 03/29] IB/mlx4: Improve another size determination in  mlx4_ib_alloc_demux_ctx() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 03/29] IB/mlx4: Improve another size determination in  mlx4_ib_alloc_demux_ctx() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 06/29] IB/mlx4: Delete three unnecessary return statements SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 06/29] IB/mlx4: Delete three unnecessary return statements Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 05/29] IB/mlx4: Fix a typo in a comment line SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 05/29] IB/mlx4: Fix a typo in a comment line Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 07/29] IB/mlx4: Split a condition check in  handle_slaves_guid_change() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 07/29] IB/mlx4: Split a condition check in  handle_slaves_guid_change() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:10 +0100
    [PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof operator  by parentheses SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:00 +0100
      Re: [PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof  operator by parentheses Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:30 +0100
    [PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in  do_slave_init() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in  do_slave_init() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 23:40 +0100
    [PATCH 11/29] IB/mlx4: Use kmalloc_array() in three functions SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 11/29] IB/mlx4: Use kmalloc_array() in three functions Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:30 +0100
    [PATCH 13/29] IB/mlx4: Split a condition check in five functions SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 13/29] IB/mlx4: Split a condition check in five functions Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:30 +0100
    [PATCH 12/29] IB/mlx4: Enclose 17 expressions for the sizeof operator  by parentheses SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 12/29] IB/mlx4: Enclose 17 expressions for the sizeof  operator by parentheses Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:30 +0100
    [PATCH 18/29] IB/mlx4: Delete an unnecessary variable initialisation  in mlx4_ib_add() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
    [PATCH 19/29] IB/mlx4: Delete an unnecessary variable assignment in  mlx4_ib_add() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 19/29] IB/mlx4: Delete an unnecessary variable assignment  in mlx4_ib_add() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 20:50 +0100
    [PATCH 14/29] IB/mlx4: Delete an unnecessary variable in  __mlx4_ib_query_gid() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
    [PATCH 17/29] IB/mlx4: Improve another size determination in  mlx4_ib_add() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:10 +0100
      Re: [PATCH 17/29] IB/mlx4: Improve another size determination in  mlx4_ib_add() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:40 +0100
      Re: [PATCH 17/29] IB/mlx4: Improve another size determination in  mlx4_ib_add() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:40 +0100
    [PATCH 22/29] IB/mlx4: Use kmalloc_array() in alloc_proxy_bufs() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
      Re: [PATCH 22/29] IB/mlx4: Use kmalloc_array() in alloc_proxy_bufs() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 21:00 +0100
    [PATCH 20/29] IB/mlx4: Delete an error message for a failed memory  allocation in mlx4_ib_add() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
      Re: [PATCH 20/29] IB/mlx4: Delete an error message for a failed  memory allocation in mlx4_ib_add() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 20:50 +0100
    [PATCH 21/29] IB/mlx4: Delete unnecessary braces in mlx4_ib_add() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
      Re: [PATCH 21/29] IB/mlx4: Delete unnecessary braces in mlx4_ib_add() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 21:10 +0100
    [PATCH 29/29] IB/mlx5: Use kmalloc_array() in create_srq_kernel() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
    [PATCH 24/29] IB/mlx4: Delete unwanted spaces behind usages of the  sizeof operator SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
    [PATCH 28/29] IB/mlx5: Less function calls in create_kernel_qp()  after error detection SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
    [PATCH 27/29] IB/mlx5: Use kmalloc_array() in create_kernel_qp() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
    [PATCH 23/29] IB/mlx4: Improve size determinations in  create_qp_common() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
      Re: [PATCH 23/29] IB/mlx4: Improve size determinations in  create_qp_common() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 21:00 +0100
    [PATCH 16/29] IB/mlx4: Improve another size determination in  do_slave_init() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:20 +0100
    [PATCH 08/29] IB/mlx4: Delete an unnecessary check before the  function call "kfree" in free_pv_object() SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:30 +0100
      Re: [PATCH 08/29] IB/mlx4: Delete an unnecessary check before the  function call "kfree" in free_pv_object() Majd Dibbiny <majd@mellanox.com> - 2017-02-19 18:20 +0100
    [PATCH 26/29] IB/mlx4: Enclose 14 expressions for the sizeof operator  by parentheses SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:30 +0100
    [PATCH 25/29] IB/mlx4: Add spaces for better code readability SF Markus Elfring <elfring@users.sourceforge.net> - 2017-02-18 22:50 +0100

Page 1 of 3  [1] 2 3  Next page →

#1583970 — [PATCH 00/29] IB/mlx: Fine-tuning for several function implementations

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 21:50 +0100
Subject[PATCH 00/29] IB/mlx: Fine-tuning for several function implementations
Message-ID<tck77-6Ef-3@gated-at.bofh.it>
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 18 Feb 2017 21:34:32 +0100

Several update suggestions were taken into account
from static source code analysis.

Markus Elfring (29):
  Use kcalloc() in mlx4_ib_alloc_pv_bufs()
  Improve another size determination in mlx4_ib_alloc_pv_bufs()
  Improve another size determination in mlx4_ib_alloc_demux_ctx()
  Improve another size determination in alloc_pv_object()
  Fix a typo in a comment line
  Delete three unnecessary return statements
  Split a condition check in handle_slaves_guid_change()
  Delete an unnecessary check before the function call "kfree" in free_pv_object()
  Move an assignment out of a check in forward_trap()
  Enclose 15 expressions for the sizeof operator by parentheses
  Use kmalloc_array() in three functions
  Enclose 17 expressions for the sizeof operator by parentheses
  Split a condition check in five functions
  Delete an unnecessary variable in __mlx4_ib_query_gid()
  Delete an unnecessary return statement in do_slave_init()
  Improve another size determination in do_slave_init()
  Improve another size determination in mlx4_ib_add()
  Delete an unnecessary variable initialisation in mlx4_ib_add()
  Delete an unnecessary variable assignment in mlx4_ib_add()
  Delete an error message for a failed memory allocation in mlx4_ib_add()
  Delete unnecessary braces in mlx4_ib_add()
  Use kmalloc_array() in alloc_proxy_bufs()
  Improve size determinations in create_qp_common()
  Delete unwanted spaces behind usages of the sizeof operator
  Add spaces for better code readability
  Enclose 14 expressions for the sizeof operator by parentheses
  Use kmalloc_array() in create_kernel_qp()
  Less function calls in create_kernel_qp() after error detection
  Use kmalloc_array() in create_srq_kernel()

 drivers/infiniband/hw/mlx4/mad.c  |  72 ++++++++--------
 drivers/infiniband/hw/mlx4/main.c | 164 ++++++++++++++++++++----------------
 drivers/infiniband/hw/mlx4/qp.c   | 173 +++++++++++++++++++-------------------
 drivers/infiniband/hw/mlx5/qp.c   |  66 ++++++++++-----
 drivers/infiniband/hw/mlx5/srq.c  |   5 +-
 5 files changed, 261 insertions(+), 219 deletions(-)

-- 
2.11.1

[toc] | [next] | [standalone]

#1583972 — [PATCH 04/29] IB/mlx4: Improve another size determination in alloc_pv_object()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 04/29] IB/mlx4: Improve another size determination in alloc_pv_object()
Message-ID<tckgN-6HX-1@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 20:20:43 +0100

Replace the specification of a data structure by a pointer dereference
as the parameter for the operator "sizeof" to make the corresponding size
determination a bit safer according to the Linux coding style convention.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 284634d36b1c..ea4892b0f39e 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -1948,7 +1948,7 @@ static int alloc_pv_object(struct mlx4_ib_dev *dev, int slave, int port,
 	struct mlx4_ib_demux_pv_ctx *ctx;
 
 	*ret_ctx = NULL;
-	ctx = kzalloc(sizeof (struct mlx4_ib_demux_pv_ctx), GFP_KERNEL);
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;
 
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584187 — Re: [PATCH 04/29] IB/mlx4: Improve another size determination in alloc_pv_object()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 04/29] IB/mlx4: Improve another size determination in alloc_pv_object()
Message-ID<tcD9M-1CL-31@gated-at.bofh.it>
In reply to#1583972
> On Feb 18, 2017, at 10:52 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 20:20:43 +0100
> 
> Replace the specification of a data structure by a pointer dereference
> as the parameter for the operator "sizeof" to make the corresponding size
> determination a bit safer according to the Linux coding style convention.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index 284634d36b1c..ea4892b0f39e 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -1948,7 +1948,7 @@ static int alloc_pv_object(struct mlx4_ib_dev *dev, int slave, int port,
>    struct mlx4_ib_demux_pv_ctx *ctx;
> 
>    *ret_ctx = NULL;
> -    ctx = kzalloc(sizeof (struct mlx4_ib_demux_pv_ctx), GFP_KERNEL);
> +    ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
>    if (!ctx)
>        return -ENOMEM;
> 
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583973 — [PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs()
Message-ID<tckgN-6HX-3@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 20:00:34 +0100

* A multiplication for the size determination of a memory allocation
  indicated that an array data structure should be processed.
  Thus use the corresponding function "kcalloc".

  This issue was detected by using the Coccinelle software.

* Replace the specification of a data structure by a pointer dereference
  to make the corresponding size determination a bit safer according to
  the Linux coding style convention.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index db564ccc0f92..61bd81baeb29 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -1600,8 +1600,8 @@ static int mlx4_ib_alloc_pv_bufs(struct mlx4_ib_demux_pv_ctx *ctx,
 		return -EINVAL;
 
 	tun_qp = &ctx->qp[qp_type];
-
-	tun_qp->ring = kzalloc(sizeof (struct mlx4_ib_buf) * MLX4_NUM_TUNNEL_BUFS,
+	tun_qp->ring = kcalloc(MLX4_NUM_TUNNEL_BUFS,
+			       sizeof(*tun_qp->ring),
 			       GFP_KERNEL);
 	if (!tun_qp->ring)
 		return -ENOMEM;
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584185 — Re: [PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 01/29] IB/mlx4: Use kcalloc() in mlx4_ib_alloc_pv_bufs()
Message-ID<tcD9M-1CL-27@gated-at.bofh.it>
In reply to#1583973
> On Feb 18, 2017, at 10:55 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 20:00:34 +0100
> 
> * A multiplication for the size determination of a memory allocation
>  indicated that an array data structure should be processed.
>  Thus use the corresponding function "kcalloc".
> 
>  This issue was detected by using the Coccinelle software.
> 
> * Replace the specification of a data structure by a pointer dereference
>  to make the corresponding size determination a bit safer according to
>  the Linux coding style convention.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index db564ccc0f92..61bd81baeb29 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -1600,8 +1600,8 @@ static int mlx4_ib_alloc_pv_bufs(struct mlx4_ib_demux_pv_ctx *ctx,
>        return -EINVAL;
> 
>    tun_qp = &ctx->qp[qp_type];
> -
> -    tun_qp->ring = kzalloc(sizeof (struct mlx4_ib_buf) * MLX4_NUM_TUNNEL_BUFS,
> +    tun_qp->ring = kcalloc(MLX4_NUM_TUNNEL_BUFS,
> +                   sizeof(*tun_qp->ring),
>                   GFP_KERNEL);
>    if (!tun_qp->ring)
>        return -ENOMEM;
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583975 — [PATCH 09/29] IB/mlx4: Move an assignment out of a check in forward_trap()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 09/29] IB/mlx4: Move an assignment out of a check in forward_trap()
Message-ID<tckgN-6HX-9@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 22:22:53 +0100

The script "checkpatch.pl" pointed information out like the following.

ERROR: do not use assignment in if condition

Thus fix the affected source code place.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 9518b7244269..860fec8b9601 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -401,7 +401,8 @@ static void forward_trap(struct mlx4_ib_dev *dev, u8 port_num, const struct ib_m
 		 */
 		spin_lock_irqsave(&dev->sm_lock, flags);
 		memcpy(send_buf->mad, mad, sizeof *mad);
-		if ((send_buf->ah = dev->sm_ah[port_num - 1]))
+		send_buf->ah = dev->sm_ah[port_num - 1];
+		if (send_buf->ah)
 			ret = ib_post_send_mad(send_buf, NULL);
 		else
 			ret = -EINVAL;
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584192 — Re: [PATCH 09/29] IB/mlx4: Move an assignment out of a check in forward_trap()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:20 +0100
SubjectRe: [PATCH 09/29] IB/mlx4: Move an assignment out of a check in forward_trap()
Message-ID<tcDjr-1Gc-9@gated-at.bofh.it>
In reply to#1583975
> On Feb 18, 2017, at 10:58 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 22:22:53 +0100
> 
> The script "checkpatch.pl" pointed information out like the following.
> 
> ERROR: do not use assignment in if condition
> 
> Thus fix the affected source code place.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index 9518b7244269..860fec8b9601 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -401,7 +401,8 @@ static void forward_trap(struct mlx4_ib_dev *dev, u8 port_num, const struct ib_m
>         */
>        spin_lock_irqsave(&dev->sm_lock, flags);
>        memcpy(send_buf->mad, mad, sizeof *mad);
> -        if ((send_buf->ah = dev->sm_ah[port_num - 1]))
> +        send_buf->ah = dev->sm_ah[port_num - 1];
> +        if (send_buf->ah)
>            ret = ib_post_send_mad(send_buf, NULL);
>        else
>            ret = -EINVAL;
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583976 — [PATCH 03/29] IB/mlx4: Improve another size determination in mlx4_ib_alloc_demux_ctx()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 03/29] IB/mlx4: Improve another size determination in mlx4_ib_alloc_demux_ctx()
Message-ID<tckgO-6HX-11@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 20:15:02 +0100

Replace the specification of a data type by a pointer dereference
as the parameter for the operator "sizeof" to make the corresponding size
determination a bit safer according to the Linux coding style convention.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 858da8746d49..284634d36b1c 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -2150,7 +2150,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
 	int i;
 
 	ctx->tun = kcalloc(dev->dev->caps.sqp_demux,
-			   sizeof (struct mlx4_ib_demux_pv_ctx *), GFP_KERNEL);
+			   sizeof(*ctx->tun), GFP_KERNEL);
 	if (!ctx->tun)
 		return -ENOMEM;
 
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584182 — Re: [PATCH 03/29] IB/mlx4: Improve another size determination in mlx4_ib_alloc_demux_ctx()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 03/29] IB/mlx4: Improve another size determination in mlx4_ib_alloc_demux_ctx()
Message-ID<tcD9L-1CL-11@gated-at.bofh.it>
In reply to#1583976
> On Feb 18, 2017, at 10:50 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 20:15:02 +0100
> 
> Replace the specification of a data type by a pointer dereference
> as the parameter for the operator "sizeof" to make the corresponding size
> determination a bit safer according to the Linux coding style convention.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index 858da8746d49..284634d36b1c 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -2150,7 +2150,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
>    int i;
> 
>    ctx->tun = kcalloc(dev->dev->caps.sqp_demux,
> -               sizeof (struct mlx4_ib_demux_pv_ctx *), GFP_KERNEL);
> +               sizeof(*ctx->tun), GFP_KERNEL);
>    if (!ctx->tun)
>        return -ENOMEM;
> 
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583977 — [PATCH 06/29] IB/mlx4: Delete three unnecessary return statements

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 06/29] IB/mlx4: Delete three unnecessary return statements
Message-ID<tckgO-6HX-13@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 21:34:27 +0100

The script "checkpatch.pl" pointed information out like the following.

WARNING: void function return statements are not generally useful

Thus remove such a statement in the affected functions.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index d8c27e3ed69a..cf33efce69d2 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -1173,7 +1173,6 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
 out:
 	kfree(in_mad);
 	kfree(out_mad);
-	return;
 }
 
 void handle_port_mgmt_change_event(struct work_struct *work)
@@ -2140,7 +2139,6 @@ void mlx4_ib_tunnels_update_work(struct work_struct *work)
 	mlx4_ib_tunnels_update(dmxw->dev, dmxw->slave, (int) dmxw->port,
 			       dmxw->do_init);
 	kfree(dmxw);
-	return;
 }
 
 static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
@@ -2268,7 +2266,6 @@ static void mlx4_ib_master_tunnels(struct mlx4_ib_dev *dev, int do_init)
 	/* initialize or tear down tunnel QPs for the master */
 	for (i = 0; i < dev->dev->caps.num_ports; i++)
 		mlx4_ib_tunnels_update(dev, mlx4_master_func_num(dev->dev), i + 1, do_init);
-	return;
 }
 
 int mlx4_ib_init_sriov(struct mlx4_ib_dev *dev)
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584183 — Re: [PATCH 06/29] IB/mlx4: Delete three unnecessary return statements

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 06/29] IB/mlx4: Delete three unnecessary return statements
Message-ID<tcD9M-1CL-23@gated-at.bofh.it>
In reply to#1583977
> On Feb 18, 2017, at 10:54 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 21:34:27 +0100
> 
> The script "checkpatch.pl" pointed information out like the following.
> 
> WARNING: void function return statements are not generally useful
> 
> Thus remove such a statement in the affected functions.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 3 ---
> 1 file changed, 3 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index d8c27e3ed69a..cf33efce69d2 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -1173,7 +1173,6 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
> out:
>    kfree(in_mad);
>    kfree(out_mad);
> -    return;
> }
> 
> void handle_port_mgmt_change_event(struct work_struct *work)
> @@ -2140,7 +2139,6 @@ void mlx4_ib_tunnels_update_work(struct work_struct *work)
>    mlx4_ib_tunnels_update(dmxw->dev, dmxw->slave, (int) dmxw->port,
>                   dmxw->do_init);
>    kfree(dmxw);
> -    return;
> }
> 
> static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
> @@ -2268,7 +2266,6 @@ static void mlx4_ib_master_tunnels(struct mlx4_ib_dev *dev, int do_init)
>    /* initialize or tear down tunnel QPs for the master */
>    for (i = 0; i < dev->dev->caps.num_ports; i++)
>        mlx4_ib_tunnels_update(dev, mlx4_master_func_num(dev->dev), i + 1, do_init);
> -    return;
> }
> 
> int mlx4_ib_init_sriov(struct mlx4_ib_dev *dev)
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583978 — [PATCH 05/29] IB/mlx4: Fix a typo in a comment line

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 05/29] IB/mlx4: Fix a typo in a comment line
Message-ID<tckgO-6HX-15@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 20:48:59 +0100

Add a missing character in this description and adjust
the comment formatting.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index ea4892b0f39e..d8c27e3ed69a 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -1559,8 +1559,10 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc
 		}
 	}
 
-	/* We are using standard ib_core services to send the mad, so generate a
-	 * stadard address handle by decoding the tunnelled mlx4_ah fields */
+	/*
+	 * We are using standard ib_core services to send the mad, so generate
+	 * a standard address handle by decoding the tunnelled mlx4_ah fields.
+	 */
 	memcpy(&ah.av, &tunnel->hdr.av, sizeof (struct mlx4_av));
 	ah.ibah.device = ctx->ib_dev;
 
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584180 — Re: [PATCH 05/29] IB/mlx4: Fix a typo in a comment line

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 05/29] IB/mlx4: Fix a typo in a comment line
Message-ID<tcD9L-1CL-7@gated-at.bofh.it>
In reply to#1583978
> On Feb 18, 2017, at 10:53 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 20:48:59 +0100
> 
> Add a missing character in this description and adjust
> the comment formatting.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index ea4892b0f39e..d8c27e3ed69a 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -1559,8 +1559,10 @@ static void mlx4_ib_multiplex_mad(struct mlx4_ib_demux_pv_ctx *ctx, struct ib_wc
>        }
>    }
> 
> -    /* We are using standard ib_core services to send the mad, so generate a
> -     * stadard address handle by decoding the tunnelled mlx4_ah fields */
> +    /*
> +     * We are using standard ib_core services to send the mad, so generate
> +     * a standard address handle by decoding the tunnelled mlx4_ah fields.
> +     */
>    memcpy(&ah.av, &tunnel->hdr.av, sizeof (struct mlx4_av));
>    ah.ibah.device = ctx->ib_dev;
> 
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583980 — [PATCH 07/29] IB/mlx4: Split a condition check in handle_slaves_guid_change()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 07/29] IB/mlx4: Split a condition check in handle_slaves_guid_change()
Message-ID<tckgO-6HX-19@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Fri, 17 Feb 2017 21:41:25 +0100

The kfree() function was called in up to two cases by the
handle_slaves_guid_change() function during error handling even if
the passed variable contained a null pointer.

* Split a condition check for memory allocation failures.

* Adjust jump targets according to the Linux coding style convention.

* Delete an initialisation for these variables at the beginning
  which became unnecessary with this refactoring.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index cf33efce69d2..75b6522b3a8f 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -1128,17 +1128,20 @@ static void propagate_pkey_ev(struct mlx4_ib_dev *dev, int port_num,
 static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
 				      u32 guid_tbl_blk_num, u32 change_bitmap)
 {
-	struct ib_smp *in_mad  = NULL;
-	struct ib_smp *out_mad  = NULL;
+	struct ib_smp *in_mad;
+	struct ib_smp *out_mad;
 	u16 i;
 
 	if (!mlx4_is_mfunc(dev->dev) || !mlx4_is_master(dev->dev))
 		return;
 
 	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	if (!in_mad)
+		return;
+
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
-	if (!in_mad || !out_mad)
-		goto out;
+	if (!out_mad)
+		goto free_in_mad;
 
 	guid_tbl_blk_num  *= 4;
 
@@ -1171,8 +1174,9 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
 	}
 
 out:
-	kfree(in_mad);
 	kfree(out_mad);
+free_in_mad:
+	kfree(in_mad);
 }
 
 void handle_port_mgmt_change_event(struct work_struct *work)
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584184 — Re: [PATCH 07/29] IB/mlx4: Split a condition check in handle_slaves_guid_change()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:10 +0100
SubjectRe: [PATCH 07/29] IB/mlx4: Split a condition check in handle_slaves_guid_change()
Message-ID<tcD9M-1CL-19@gated-at.bofh.it>
In reply to#1583980
> On Feb 18, 2017, at 10:55 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Fri, 17 Feb 2017 21:41:25 +0100
> 
> The kfree() function was called in up to two cases by the
> handle_slaves_guid_change() function during error handling even if
> the passed variable contained a null pointer.
> 
> * Split a condition check for memory allocation failures.
> 
> * Adjust jump targets according to the Linux coding style convention.
> 
> * Delete an initialisation for these variables at the beginning
>  which became unnecessary with this refactoring.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index cf33efce69d2..75b6522b3a8f 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -1128,17 +1128,20 @@ static void propagate_pkey_ev(struct mlx4_ib_dev *dev, int port_num,
> static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
>                      u32 guid_tbl_blk_num, u32 change_bitmap)
> {
> -    struct ib_smp *in_mad  = NULL;
> -    struct ib_smp *out_mad  = NULL;
> +    struct ib_smp *in_mad;
> +    struct ib_smp *out_mad;
>    u16 i;
> 
>    if (!mlx4_is_mfunc(dev->dev) || !mlx4_is_master(dev->dev))
>        return;
> 
>    in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
> +    if (!in_mad)
> +        return;
> +
>    out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
> -    if (!in_mad || !out_mad)
> -        goto out;
> +    if (!out_mad)
> +        goto free_in_mad;
> 
>    guid_tbl_blk_num  *= 4;
> 
> @@ -1171,8 +1174,9 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
>    }
> 
> out:
> -    kfree(in_mad);
>    kfree(out_mad);
> +free_in_mad:
> +    kfree(in_mad);
> }
> 
> void handle_port_mgmt_change_event(struct work_struct *work)
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583981 — [PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof operator by parentheses

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:00 +0100
Subject[PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof operator by parentheses
Message-ID<tckgO-6HX-25@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 18 Feb 2017 09:54:15 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The script "checkpatch.pl" pointed information out like the following.

WARNING: sizeof … should be sizeof(…)

Thus fix the affected source code places.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/mad.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 860fec8b9601..b26817f0669f 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -195,7 +195,7 @@ static void update_sm_ah(struct mlx4_ib_dev *dev, u8 port_num, u16 lid, u8 sl)
 	if (!dev->send_agent[port_num - 1][0])
 		return;
 
-	memset(&ah_attr, 0, sizeof ah_attr);
+	memset(&ah_attr, 0, sizeof(ah_attr));
 	ah_attr.dlid     = lid;
 	ah_attr.sl       = sl;
 	ah_attr.port_num = port_num;
@@ -400,7 +400,7 @@ static void forward_trap(struct mlx4_ib_dev *dev, u8 port_num, const struct ib_m
 		 * it's OK for our devices).
 		 */
 		spin_lock_irqsave(&dev->sm_lock, flags);
-		memcpy(send_buf->mad, mad, sizeof *mad);
+		memcpy(send_buf->mad, mad, sizeof(*mad));
 		send_buf->ah = dev->sm_ah[port_num - 1];
 		if (send_buf->ah)
 			ret = ib_post_send_mad(send_buf, NULL);
@@ -555,7 +555,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
 
 	/* create ah. Just need an empty one with the port num for the post send.
 	 * The driver will set the force loopback bit in post_send */
-	memset(&attr, 0, sizeof attr);
+	memset(&attr, 0, sizeof(attr));
 	attr.port_num = port;
 	if (is_eth) {
 		union ib_gid sgid;
@@ -590,8 +590,8 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
 
 	/* copy over to tunnel buffer */
 	if (grh)
-		memcpy(&tun_mad->grh, grh, sizeof *grh);
-	memcpy(&tun_mad->mad, mad, sizeof *mad);
+		memcpy(&tun_mad->grh, grh, sizeof(*grh));
+	memcpy(&tun_mad->mad, mad, sizeof(*mad));
 
 	/* adjust tunnel data */
 	tun_mad->hdr.pkey_index = cpu_to_be16(tun_pkey_ix);
@@ -961,7 +961,7 @@ static int iboe_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	}
 	mutex_unlock(&dev->counters_table[port_num - 1].mutex);
 	if (stats_avail) {
-		memset(out_mad->data, 0, sizeof out_mad->data);
+		memset(out_mad->data, 0, sizeof(out_mad->data));
 		switch (counter_stats.counter_mode & 0xf) {
 		case 0:
 			edit_counter(&counter_stats,
@@ -1136,11 +1136,11 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
 	if (!mlx4_is_mfunc(dev->dev) || !mlx4_is_master(dev->dev))
 		return;
 
-	in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
+	in_mad  = kmalloc(sizeof(*in_mad), GFP_KERNEL);
 	if (!in_mad)
 		return;
 
-	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
+	out_mad = kmalloc(sizeof(*out_mad), GFP_KERNEL);
 	if (!out_mad)
 		goto free_in_mad;
 
@@ -1149,8 +1149,8 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
 	for (i = 0; i < 4; i++) {
 		if (change_bitmap && (!((change_bitmap >> (8 * i)) & 0xff)))
 			continue;
-		memset(in_mad, 0, sizeof *in_mad);
-		memset(out_mad, 0, sizeof *out_mad);
+		memset(in_mad, 0, sizeof(*in_mad));
+		memset(out_mad, 0, sizeof(*out_mad));
 
 		in_mad->base_version  = 1;
 		in_mad->mgmt_class    = IB_MGMT_CLASS_SUBN_LID_ROUTED;
@@ -1421,7 +1421,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port,
 				   sizeof (struct mlx4_mad_snd_buf),
 				   DMA_TO_DEVICE);
 
-	memcpy(&sqp_mad->payload, mad, sizeof *mad);
+	memcpy(&sqp_mad->payload, mad, sizeof(*mad));
 
 	ib_dma_sync_single_for_device(&dev->ib_dev,
 				      sqp->tx_ring[wire_tx_ix].buf.map,
@@ -1804,7 +1804,7 @@ static int create_pv_sqp(struct mlx4_ib_demux_pv_ctx *ctx,
 
 	tun_qp = &ctx->qp[qp_type];
 
-	memset(&qp_init_attr, 0, sizeof qp_init_attr);
+	memset(&qp_init_attr, 0, sizeof(qp_init_attr));
 	qp_init_attr.init_attr.send_cq = ctx->cq;
 	qp_init_attr.init_attr.recv_cq = ctx->cq;
 	qp_init_attr.init_attr.sq_sig_type = IB_SIGNAL_ALL_WR;
@@ -1837,7 +1837,7 @@ static int create_pv_sqp(struct mlx4_ib_demux_pv_ctx *ctx,
 		return ret;
 	}
 
-	memset(&attr, 0, sizeof attr);
+	memset(&attr, 0, sizeof(attr));
 	attr.qp_state = IB_QPS_INIT;
 	ret = 0;
 	if (create_tun)
@@ -2184,7 +2184,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
 		goto err_mcg;
 	}
 
-	snprintf(name, sizeof name, "mlx4_ibt%d", port);
+	snprintf(name, sizeof(name), "mlx4_ibt%d", port);
 	ctx->wq = alloc_ordered_workqueue(name, WQ_MEM_RECLAIM);
 	if (!ctx->wq) {
 		pr_err("Failed to create tunnelling WQ for port %d\n", port);
@@ -2192,7 +2192,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
 		goto err_wq;
 	}
 
-	snprintf(name, sizeof name, "mlx4_ibud%d", port);
+	snprintf(name, sizeof(name), "mlx4_ibud%d", port);
 	ctx->ud_wq = alloc_ordered_workqueue(name, WQ_MEM_RECLAIM);
 	if (!ctx->ud_wq) {
 		pr_err("Failed to create up/down WQ for port %d\n", port);
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584199 — Re: [PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof operator by parentheses

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 18:30 +0100
SubjectRe: [PATCH 10/29] IB/mlx4: Enclose 15 expressions for the sizeof operator by parentheses
Message-ID<tcDt8-1Jy-35@gated-at.bofh.it>
In reply to#1583981
> On Feb 18, 2017, at 10:59 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Sat, 18 Feb 2017 09:54:15 +0100
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
> 
> The script "checkpatch.pl" pointed information out like the following.
> 
> WARNING: sizeof … should be sizeof(…)
> 
> Thus fix the affected source code places.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/mad.c | 30 +++++++++++++++---------------
> 1 file changed, 15 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
> index 860fec8b9601..b26817f0669f 100644
> --- a/drivers/infiniband/hw/mlx4/mad.c
> +++ b/drivers/infiniband/hw/mlx4/mad.c
> @@ -195,7 +195,7 @@ static void update_sm_ah(struct mlx4_ib_dev *dev, u8 port_num, u16 lid, u8 sl)
>    if (!dev->send_agent[port_num - 1][0])
>        return;
> 
> -    memset(&ah_attr, 0, sizeof ah_attr);
> +    memset(&ah_attr, 0, sizeof(ah_attr));
>    ah_attr.dlid     = lid;
>    ah_attr.sl       = sl;
>    ah_attr.port_num = port_num;
> @@ -400,7 +400,7 @@ static void forward_trap(struct mlx4_ib_dev *dev, u8 port_num, const struct ib_m
>         * it's OK for our devices).
>         */
>        spin_lock_irqsave(&dev->sm_lock, flags);
> -        memcpy(send_buf->mad, mad, sizeof *mad);
> +        memcpy(send_buf->mad, mad, sizeof(*mad));
>        send_buf->ah = dev->sm_ah[port_num - 1];
>        if (send_buf->ah)
>            ret = ib_post_send_mad(send_buf, NULL);
> @@ -555,7 +555,7 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
> 
>    /* create ah. Just need an empty one with the port num for the post send.
>     * The driver will set the force loopback bit in post_send */
> -    memset(&attr, 0, sizeof attr);
> +    memset(&attr, 0, sizeof(attr));
>    attr.port_num = port;
>    if (is_eth) {
>        union ib_gid sgid;
> @@ -590,8 +590,8 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
> 
>    /* copy over to tunnel buffer */
>    if (grh)
> -        memcpy(&tun_mad->grh, grh, sizeof *grh);
> -    memcpy(&tun_mad->mad, mad, sizeof *mad);
> +        memcpy(&tun_mad->grh, grh, sizeof(*grh));
> +    memcpy(&tun_mad->mad, mad, sizeof(*mad));
> 
>    /* adjust tunnel data */
>    tun_mad->hdr.pkey_index = cpu_to_be16(tun_pkey_ix);
> @@ -961,7 +961,7 @@ static int iboe_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
>    }
>    mutex_unlock(&dev->counters_table[port_num - 1].mutex);
>    if (stats_avail) {
> -        memset(out_mad->data, 0, sizeof out_mad->data);
> +        memset(out_mad->data, 0, sizeof(out_mad->data));
>        switch (counter_stats.counter_mode & 0xf) {
>        case 0:
>            edit_counter(&counter_stats,
> @@ -1136,11 +1136,11 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
>    if (!mlx4_is_mfunc(dev->dev) || !mlx4_is_master(dev->dev))
>        return;
> 
> -    in_mad  = kmalloc(sizeof *in_mad, GFP_KERNEL);
> +    in_mad  = kmalloc(sizeof(*in_mad), GFP_KERNEL);
>    if (!in_mad)
>        return;
> 
> -    out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
> +    out_mad = kmalloc(sizeof(*out_mad), GFP_KERNEL);
>    if (!out_mad)
>        goto free_in_mad;
> 
> @@ -1149,8 +1149,8 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num,
>    for (i = 0; i < 4; i++) {
>        if (change_bitmap && (!((change_bitmap >> (8 * i)) & 0xff)))
>            continue;
> -        memset(in_mad, 0, sizeof *in_mad);
> -        memset(out_mad, 0, sizeof *out_mad);
> +        memset(in_mad, 0, sizeof(*in_mad));
> +        memset(out_mad, 0, sizeof(*out_mad));
> 
>        in_mad->base_version  = 1;
>        in_mad->mgmt_class    = IB_MGMT_CLASS_SUBN_LID_ROUTED;
> @@ -1421,7 +1421,7 @@ int mlx4_ib_send_to_wire(struct mlx4_ib_dev *dev, int slave, u8 port,
>                   sizeof (struct mlx4_mad_snd_buf),
>                   DMA_TO_DEVICE);
> 
> -    memcpy(&sqp_mad->payload, mad, sizeof *mad);
> +    memcpy(&sqp_mad->payload, mad, sizeof(*mad));
> 
>    ib_dma_sync_single_for_device(&dev->ib_dev,
>                      sqp->tx_ring[wire_tx_ix].buf.map,
> @@ -1804,7 +1804,7 @@ static int create_pv_sqp(struct mlx4_ib_demux_pv_ctx *ctx,
> 
>    tun_qp = &ctx->qp[qp_type];
> 
> -    memset(&qp_init_attr, 0, sizeof qp_init_attr);
> +    memset(&qp_init_attr, 0, sizeof(qp_init_attr));
>    qp_init_attr.init_attr.send_cq = ctx->cq;
>    qp_init_attr.init_attr.recv_cq = ctx->cq;
>    qp_init_attr.init_attr.sq_sig_type = IB_SIGNAL_ALL_WR;
> @@ -1837,7 +1837,7 @@ static int create_pv_sqp(struct mlx4_ib_demux_pv_ctx *ctx,
>        return ret;
>    }
> 
> -    memset(&attr, 0, sizeof attr);
> +    memset(&attr, 0, sizeof(attr));
>    attr.qp_state = IB_QPS_INIT;
>    ret = 0;
>    if (create_tun)
> @@ -2184,7 +2184,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
>        goto err_mcg;
>    }
> 
> -    snprintf(name, sizeof name, "mlx4_ibt%d", port);
> +    snprintf(name, sizeof(name), "mlx4_ibt%d", port);
>    ctx->wq = alloc_ordered_workqueue(name, WQ_MEM_RECLAIM);
>    if (!ctx->wq) {
>        pr_err("Failed to create tunnelling WQ for port %d\n", port);
> @@ -2192,7 +2192,7 @@ static int mlx4_ib_alloc_demux_ctx(struct mlx4_ib_dev *dev,
>        goto err_wq;
>    }
> 
> -    snprintf(name, sizeof name, "mlx4_ibud%d", port);
> +    snprintf(name, sizeof(name), "mlx4_ibud%d", port);
>    ctx->ud_wq = alloc_ordered_workqueue(name, WQ_MEM_RECLAIM);
>    if (!ctx->ud_wq) {
>        pr_err("Failed to create up/down WQ for port %d\n", port);
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583982 — [PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in do_slave_init()

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:10 +0100
Subject[PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in do_slave_init()
Message-ID<tckqt-70u-1@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 18 Feb 2017 15:16:54 +0100

The script "checkpatch.pl" pointed information out like the following.

WARNING: void function return statements are not generally useful

Thus remove such a statement here.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 89bbb61d62f5..b2b472977a39 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -3131,7 +3131,6 @@ static void do_slave_init(struct mlx4_ib_dev *ibdev, int slave, int do_init)
 	}
 out:
 	kfree(dm);
-	return;
 }
 
 static void mlx4_ib_handle_catas_error(struct mlx4_ib_dev *ibdev)
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1584270 — Re: [PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in do_slave_init()

FromMajd Dibbiny <majd@mellanox.com>
Date2017-02-19 23:40 +0100
SubjectRe: [PATCH 15/29] IB/mlx4: Delete an unnecessary return statement in do_slave_init()
Message-ID<tcIj8-4Ho-13@gated-at.bofh.it>
In reply to#1583982
> On Feb 18, 2017, at 11:04 PM, SF Markus Elfring <elfring@users.sourceforge.net> wrote:
> 
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Sat, 18 Feb 2017 15:16:54 +0100
> 
> The script "checkpatch.pl" pointed information out like the following.
> 
> WARNING: void function return statements are not generally useful
> 
> Thus remove such a statement here.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
> ---
> drivers/infiniband/hw/mlx4/main.c | 1 -
> 1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
> index 89bbb61d62f5..b2b472977a39 100644
> --- a/drivers/infiniband/hw/mlx4/main.c
> +++ b/drivers/infiniband/hw/mlx4/main.c
> @@ -3131,7 +3131,6 @@ static void do_slave_init(struct mlx4_ib_dev *ibdev, int slave, int do_init)
>    }
> out:
>    kfree(dm);
> -    return;
> }
> 
> static void mlx4_ib_handle_catas_error(struct mlx4_ib_dev *ibdev)
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thanks,
Reviewed-by: Majd Dibbiny
<majd@mellanox.com>

[toc] | [prev] | [next] | [standalone]

#1583983 — [PATCH 11/29] IB/mlx4: Use kmalloc_array() in three functions

FromSF Markus Elfring <elfring@users.sourceforge.net>
Date2017-02-18 22:10 +0100
Subject[PATCH 11/29] IB/mlx4: Use kmalloc_array() in three functions
Message-ID<tckqt-70u-3@gated-at.bofh.it>
In reply to#1583970
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sat, 18 Feb 2017 10:32:32 +0100

Multiplications for the size determination of memory allocations
indicated that array data structures should be processed.
Thus use the corresponding function "kmalloc_array".

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 drivers/infiniband/hw/mlx4/main.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index ba6af84cc236..350c9148340e 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -304,7 +304,9 @@ static int mlx4_ib_add_gid(struct ib_device *device,
 		ctx->refcount++;
 	}
 	if (!ret && hw_update) {
-		gids = kmalloc(sizeof(*gids) * MLX4_MAX_PORT_GIDS, GFP_ATOMIC);
+		gids = kmalloc_array(MLX4_MAX_PORT_GIDS,
+				     sizeof(*gids),
+				     GFP_ATOMIC);
 		if (!gids) {
 			ret = -ENOMEM;
 		} else {
@@ -359,7 +361,9 @@ static int mlx4_ib_del_gid(struct ib_device *device,
 	if (!ret && hw_update) {
 		int i;
 
-		gids = kmalloc(sizeof(*gids) * MLX4_MAX_PORT_GIDS, GFP_ATOMIC);
+		gids = kmalloc_array(MLX4_MAX_PORT_GIDS,
+				     sizeof(*gids),
+				     GFP_ATOMIC);
 		if (!gids) {
 			ret = -ENOMEM;
 		} else {
@@ -2828,9 +2832,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 			goto err_counter;
 
 		ibdev->ib_uc_qpns_bitmap =
-			kmalloc(BITS_TO_LONGS(ibdev->steer_qpn_count) *
-				sizeof(long),
-				GFP_KERNEL);
+			kmalloc_array(BITS_TO_LONGS(ibdev->steer_qpn_count),
+				      sizeof(long),
+				      GFP_KERNEL);
 		if (!ibdev->ib_uc_qpns_bitmap)
 			goto err_steer_qp_release;
 
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | linux.kernel

csiph-web