[PATCH 0/2] IB/mlx4: Enable debug print of SMPs and enhance output for MADs

[PATCH 0/2] IB/mlx4: Enable debug print of SMPs and enhance output for MADs

[Håkon Bugge]

SMPs were not printed at all. Added printing of port number and TID to
all MADs





Håkon Bugge (2):
  IB/mlx4: Enable debug print of SMPs
  IB/mlx4: Add port and TID to MAD debug print

 drivers/infiniband/hw/mlx4/mad.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

--
2.14.3

[PATCH 1/2] IB/mlx4: Enable debug print of SMPs

[Håkon Bugge]

IB Subnet Management Packets (SMPs) were excluded from debug prints.

Fixed by enabling print even on QP0 MADs.

Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
---
 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 e5466d786bb1..1718715cc094 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -807,7 +807,7 @@ static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	int err;
 	struct ib_port_attr pattr;
 
-	if (in_wc && in_wc->qp->qp_num) {
+	if (in_wc && in_wc->qp) {
 		pr_debug("received MAD: slid:%d sqpn:%d "
 			"dlid_bits:%d dqpn:%d wc_flags:0x%x, cls %x, mtd %x, atr %x\n",
 			in_wc->slid, in_wc->src_qp,
-- 
2.14.3

Re: [PATCH 1/2] IB/mlx4: Enable debug print of SMPs

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 405 bytes --]

On Mon, Sep 17, 2018 at 04:53:08PM +0200, Håkon Bugge wrote:
> IB Subnet Management Packets (SMPs) were excluded from debug prints.
>
> Fixed by enabling print even on QP0 MADs.
>
> Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
> ---
>  drivers/infiniband/hw/mlx4/mad.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>

Thanks,
Acked-by: Leon Romanovsky <leonro@mellanox.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 801 bytes --]

[PATCH 2/2] IB/mlx4: Add port and TID to MAD debug print

[Håkon Bugge]

Add said information and make the debug print format consistent.

Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
---
 drivers/infiniband/hw/mlx4/mad.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 1718715cc094..2b3e8a959dd9 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -808,14 +808,16 @@ static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	struct ib_port_attr pattr;
 
 	if (in_wc && in_wc->qp) {
-		pr_debug("received MAD: slid:%d sqpn:%d "
-			"dlid_bits:%d dqpn:%d wc_flags:0x%x, cls %x, mtd %x, atr %x\n",
-			in_wc->slid, in_wc->src_qp,
-			in_wc->dlid_path_bits,
-			in_wc->qp->qp_num,
-			in_wc->wc_flags,
-			in_mad->mad_hdr.mgmt_class, in_mad->mad_hdr.method,
-			be16_to_cpu(in_mad->mad_hdr.attr_id));
+		pr_debug("received MAD: port:%d slid:%d sqpn:%d "
+			 "dlid_bits:%d dqpn:%d wc_flags:0x%x tid:%08x cls:%x mtd:%x atr:%x\n",
+			 port_num,
+			 in_wc->slid, in_wc->src_qp,
+			 in_wc->dlid_path_bits,
+			 in_wc->qp->qp_num,
+			 in_wc->wc_flags,
+			 be64_to_cpu(in_mad->mad_hdr.tid),
+			 in_mad->mad_hdr.mgmt_class, in_mad->mad_hdr.method,
+			 be16_to_cpu(in_mad->mad_hdr.attr_id));
 		if (in_wc->wc_flags & IB_WC_GRH) {
 			pr_debug("sgid_hi:0x%016llx sgid_lo:0x%016llx\n",
 				 be64_to_cpu(in_grh->sgid.global.subnet_prefix),
-- 
2.14.3

Re: [PATCH 2/2] IB/mlx4: Add port and TID to MAD debug print

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 7738 bytes --]

Hi Håkon,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on linus/master]
[also build test WARNING on v4.19-rc4 next-20180913]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/H-kon-Bugge/IB-mlx4-Enable-debug-print-of-SMPs/20180917-232956
config: i386-randconfig-x002-201837 (attached as .config)
compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All warnings (new ones prefixed by >>):

   In file included from include/linux/kernel.h:14:0,
                    from include/linux/list.h:9,
                    from include/rdma/ib_mad.h:40,
                    from drivers/infiniband//hw/mlx4/mad.c:33:
   drivers/infiniband//hw/mlx4/mad.c: In function 'ib_process_mad':
>> include/linux/kern_levels.h:5:18: warning: format '%x' expects argument of type 'unsigned int', but argument 9 has type 'long long unsigned int' [-Wformat=]
    #define KERN_SOH "\001"  /* ASCII Start Of Header */
                     ^
   include/linux/printk.h:136:10: note: in definition of macro 'no_printk'
      printk(fmt, ##__VA_ARGS__);  \
             ^~~
   include/linux/kern_levels.h:15:20: note: in expansion of macro 'KERN_SOH'
    #define KERN_DEBUG KERN_SOH "7" /* debug-level messages */
                       ^~~~~~~~
   include/linux/printk.h:346:12: note: in expansion of macro 'KERN_DEBUG'
     no_printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__)
               ^~~~~~~~~~
>> drivers/infiniband//hw/mlx4/mad.c:811:3: note: in expansion of macro 'pr_debug'
      pr_debug("received MAD: port:%d slid:%d sqpn:%d "
      ^~~~~~~~
   drivers/infiniband//hw/mlx4/mad.c:812:48: note: format string is defined here
        "dlid_bits:%d dqpn:%d wc_flags:0x%x tid:%08x cls:%x mtd:%x atr:%x\n",
                                                ~~~^
                                                %08llx
--
   In file included from include/linux/kernel.h:14:0,
                    from include/linux/list.h:9,
                    from include/rdma/ib_mad.h:40,
                    from drivers/infiniband/hw/mlx4/mad.c:33:
   drivers/infiniband/hw/mlx4/mad.c: In function 'ib_process_mad':
>> include/linux/kern_levels.h:5:18: warning: format '%x' expects argument of type 'unsigned int', but argument 9 has type 'long long unsigned int' [-Wformat=]
    #define KERN_SOH "\001"  /* ASCII Start Of Header */
                     ^
   include/linux/printk.h:136:10: note: in definition of macro 'no_printk'
      printk(fmt, ##__VA_ARGS__);  \
             ^~~
   include/linux/kern_levels.h:15:20: note: in expansion of macro 'KERN_SOH'
    #define KERN_DEBUG KERN_SOH "7" /* debug-level messages */
                       ^~~~~~~~
   include/linux/printk.h:346:12: note: in expansion of macro 'KERN_DEBUG'
     no_printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__)
               ^~~~~~~~~~
   drivers/infiniband/hw/mlx4/mad.c:811:3: note: in expansion of macro 'pr_debug'
      pr_debug("received MAD: port:%d slid:%d sqpn:%d "
      ^~~~~~~~
   drivers/infiniband/hw/mlx4/mad.c:812:48: note: format string is defined here
        "dlid_bits:%d dqpn:%d wc_flags:0x%x tid:%08x cls:%x mtd:%x atr:%x\n",
                                                ~~~^
                                                %08llx

vim +/pr_debug +811 drivers/infiniband//hw/mlx4/mad.c

   801	
   802	static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
   803				const struct ib_wc *in_wc, const struct ib_grh *in_grh,
   804				const struct ib_mad *in_mad, struct ib_mad *out_mad)
   805	{
   806		u16 slid, prev_lid = 0;
   807		int err;
   808		struct ib_port_attr pattr;
   809	
   810		if (in_wc && in_wc->qp) {
 > 811			pr_debug("received MAD: port:%d slid:%d sqpn:%d "
   812				 "dlid_bits:%d dqpn:%d wc_flags:0x%x tid:%08x cls:%x mtd:%x atr:%x\n",
   813				 port_num,
   814				 in_wc->slid, in_wc->src_qp,
   815				 in_wc->dlid_path_bits,
   816				 in_wc->qp->qp_num,
   817				 in_wc->wc_flags,
   818				 be64_to_cpu(in_mad->mad_hdr.tid),
   819				 in_mad->mad_hdr.mgmt_class, in_mad->mad_hdr.method,
   820				 be16_to_cpu(in_mad->mad_hdr.attr_id));
   821			if (in_wc->wc_flags & IB_WC_GRH) {
   822				pr_debug("sgid_hi:0x%016llx sgid_lo:0x%016llx\n",
   823					 be64_to_cpu(in_grh->sgid.global.subnet_prefix),
   824					 be64_to_cpu(in_grh->sgid.global.interface_id));
   825				pr_debug("dgid_hi:0x%016llx dgid_lo:0x%016llx\n",
   826					 be64_to_cpu(in_grh->dgid.global.subnet_prefix),
   827					 be64_to_cpu(in_grh->dgid.global.interface_id));
   828			}
   829		}
   830	
   831		slid = in_wc ? ib_lid_cpu16(in_wc->slid) : be16_to_cpu(IB_LID_PERMISSIVE);
   832	
   833		if (in_mad->mad_hdr.method == IB_MGMT_METHOD_TRAP && slid == 0) {
   834			forward_trap(to_mdev(ibdev), port_num, in_mad);
   835			return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_CONSUMED;
   836		}
   837	
   838		if (in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED ||
   839		    in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) {
   840			if (in_mad->mad_hdr.method   != IB_MGMT_METHOD_GET &&
   841			    in_mad->mad_hdr.method   != IB_MGMT_METHOD_SET &&
   842			    in_mad->mad_hdr.method   != IB_MGMT_METHOD_TRAP_REPRESS)
   843				return IB_MAD_RESULT_SUCCESS;
   844	
   845			/*
   846			 * Don't process SMInfo queries -- the SMA can't handle them.
   847			 */
   848			if (in_mad->mad_hdr.attr_id == IB_SMP_ATTR_SM_INFO)
   849				return IB_MAD_RESULT_SUCCESS;
   850		} else if (in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT ||
   851			   in_mad->mad_hdr.mgmt_class == MLX4_IB_VENDOR_CLASS1   ||
   852			   in_mad->mad_hdr.mgmt_class == MLX4_IB_VENDOR_CLASS2   ||
   853			   in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_CONG_MGMT) {
   854			if (in_mad->mad_hdr.method  != IB_MGMT_METHOD_GET &&
   855			    in_mad->mad_hdr.method  != IB_MGMT_METHOD_SET)
   856				return IB_MAD_RESULT_SUCCESS;
   857		} else
   858			return IB_MAD_RESULT_SUCCESS;
   859	
   860		if ((in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED ||
   861		     in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) &&
   862		    in_mad->mad_hdr.method == IB_MGMT_METHOD_SET &&
   863		    in_mad->mad_hdr.attr_id == IB_SMP_ATTR_PORT_INFO &&
   864		    !ib_query_port(ibdev, port_num, &pattr))
   865			prev_lid = ib_lid_cpu16(pattr.lid);
   866	
   867		err = mlx4_MAD_IFC(to_mdev(ibdev),
   868				   (mad_flags & IB_MAD_IGNORE_MKEY ? MLX4_MAD_IFC_IGNORE_MKEY : 0) |
   869				   (mad_flags & IB_MAD_IGNORE_BKEY ? MLX4_MAD_IFC_IGNORE_BKEY : 0) |
   870				   MLX4_MAD_IFC_NET_VIEW,
   871				   port_num, in_wc, in_grh, in_mad, out_mad);
   872		if (err)
   873			return IB_MAD_RESULT_FAILURE;
   874	
   875		if (!out_mad->mad_hdr.status) {
   876			smp_snoop(ibdev, port_num, in_mad, prev_lid);
   877			/* slaves get node desc from FW */
   878			if (!mlx4_is_slave(to_mdev(ibdev)->dev))
   879				node_desc_override(ibdev, out_mad);
   880		}
   881	
   882		/* set return bit in status of directed route responses */
   883		if (in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)
   884			out_mad->mad_hdr.status |= cpu_to_be16(1 << 15);
   885	
   886		if (in_mad->mad_hdr.method == IB_MGMT_METHOD_TRAP_REPRESS)
   887			/* no response for trap repress */
   888			return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_CONSUMED;
   889	
   890		return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
   891	}
   892	

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 33886 bytes --]

Re: [PATCH 2/2] IB/mlx4: Add port and TID to MAD debug print

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 373 bytes --]

On Mon, Sep 17, 2018 at 04:53:09PM +0200, Håkon Bugge wrote:
> Add said information and make the debug print format consistent.
>
> Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
> ---
>  drivers/infiniband/hw/mlx4/mad.c | 18 ++++++++++--------
>  1 file changed, 10 insertions(+), 8 deletions(-)
>

Thanks,
Acked-by: Leon Romanovsky <leonro@mellanox.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 801 bytes --]

Re: [PATCH 2/2] IB/mlx4: Add port and TID to MAD debug print

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 2529 bytes --]

Hi Håkon,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on linus/master]
[also build test WARNING on v4.19-rc4 next-20180913]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/H-kon-Bugge/IB-mlx4-Enable-debug-print-of-SMPs/20180917-232956
config: i386-allmodconfig (attached as .config)
compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All warnings (new ones prefixed by >>):

   In file included from drivers/infiniband/hw/mlx4/mad.c:46:0:
   drivers/infiniband/hw/mlx4/mad.c: In function 'ib_process_mad':
>> drivers/infiniband/hw/mlx4/mlx4_ib.h:57:21: warning: format '%x' expects argument of type 'unsigned int', but argument 10 has type 'long long unsigned int' [-Wformat=]
    #define pr_fmt(fmt) "<" MLX4_IB_DRV_NAME "> %s: " fmt, __func__
                        ^
   include/linux/dynamic_debug.h:127:35: note: in expansion of macro 'pr_fmt'
      __dynamic_pr_debug(&descriptor, pr_fmt(fmt), \
                                      ^~~~~~
   include/linux/printk.h:340:2: note: in expansion of macro 'dynamic_pr_debug'
     dynamic_pr_debug(fmt, ##__VA_ARGS__)
     ^~~~~~~~~~~~~~~~
   drivers/infiniband/hw/mlx4/mad.c:811:3: note: in expansion of macro 'pr_debug'
      pr_debug("received MAD: port:%d slid:%d sqpn:%d "
      ^~~~~~~~
   drivers/infiniband/hw/mlx4/mad.c:812:48: note: format string is defined here
        "dlid_bits:%d dqpn:%d wc_flags:0x%x tid:%08x cls:%x mtd:%x atr:%x\n",
                                                ~~~^
                                                %08llx

vim +57 drivers/infiniband/hw/mlx4/mlx4_ib.h

b1d8eb5a Jack Morgenstein 2012-06-19  53  
b1d8eb5a Jack Morgenstein 2012-06-19  54  #ifdef pr_fmt
b1d8eb5a Jack Morgenstein 2012-06-19  55  #undef pr_fmt
b1d8eb5a Jack Morgenstein 2012-06-19  56  #endif
b1d8eb5a Jack Morgenstein 2012-06-19 @57  #define pr_fmt(fmt)	"<" MLX4_IB_DRV_NAME "> %s: " fmt, __func__
b1d8eb5a Jack Morgenstein 2012-06-19  58  

:::::: The code at line 57 was first introduced by commit
:::::: b1d8eb5a213640f1be98a90e73a241d15b70045c IB/mlx4: Add debug prints

:::::: TO: Jack Morgenstein <jackm@dev.mellanox.co.il>
:::::: CC: Roland Dreier <roland@purestorage.com>

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 65199 bytes --]

[PATCH v2 0/2] IB/mlx4: Enable debug print of SMPs and enhance output for MADs

[Håkon Bugge]

SMPs were not printed at all. Added printing of port number and TID to
all MADs





Håkon Bugge (2):
  IB/mlx4: Enable debug print of SMPs
  IB/mlx4: Add port and TID to MAD debug print

 drivers/infiniband/hw/mlx4/mad.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

--

v1 -> v2
    * Fixed incorrect format for printing the TID for the second patch

2.14.3

[PATCH 1/2] IB/mlx4: Enable debug print of SMPs

[Håkon Bugge]

IB Subnet Management Packets (SMPs) were excluded from debug prints.

Fixed by enabling print even on QP0 MADs.

Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
---
 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 e5466d786bb1..1718715cc094 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -807,7 +807,7 @@ static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	int err;
 	struct ib_port_attr pattr;
 
-	if (in_wc && in_wc->qp->qp_num) {
+	if (in_wc && in_wc->qp) {
 		pr_debug("received MAD: slid:%d sqpn:%d "
 			"dlid_bits:%d dqpn:%d wc_flags:0x%x, cls %x, mtd %x, atr %x\n",
 			in_wc->slid, in_wc->src_qp,
-- 
2.14.3