[PATCH] opensm/osm_sm_mad_ctrl.c Changes to some error messages

[PATCH] opensm/osm_sm_mad_ctrl.c Changes to some error messages

[Line Holen]

Added info to some error messages and lowered log level on
subsequent call to osm_dump_dr_smp() to reduce amount of logging
with default log level setting.

Signed-off-by: Line Holen <Line.Holen-xsfywfwIY+M@public.gmane.org>

---

diff --git a/opensm/opensm/osm_sm_mad_ctrl.c b/opensm/opensm/osm_sm_mad_ctrl.c
index 74beb1c..defa014 100644
--- a/opensm/opensm/osm_sm_mad_ctrl.c
+++ b/opensm/opensm/osm_sm_mad_ctrl.c
@@ -3,6 +3,7 @@
  * Copyright (c) 2002-2005 Mellanox Technologies LTD. All rights reserved.
  * Copyright (c) 1996-2003 Intel Corporation. All rights reserved.
  * Copyright (c) 2009 HNR Consulting. All rights reserved.
+ * Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -636,9 +637,38 @@ static void sm_mad_ctrl_rcv_callback(IN osm_madw_t * p_madw,
 		status = p_smp->status;
 
 	if (status != 0) {
+		char buf[BUF_SIZE];
+
+		if (p_smp->mgmt_class == IB_MCLASS_SUBN_DIR) {
+			char ipath[BUF_SIZE], rpath[BUF_SIZE];
+			char tmp[10];
+			uint32_t i;
+
+			for (i = 0; i <= p_smp->hop_count; i++) {
+				if (i == 0) {
+					sprintf(ipath, "%d", p_smp->initial_path[i]);
+					sprintf(rpath, "%d", p_smp->return_path[i]);
+				} else {
+					sprintf(tmp, ",%d", p_smp->initial_path[i]);
+					strcat(ipath, tmp);
+					sprintf(tmp, ",%d", p_smp->return_path[i]);
+					strcat(rpath, tmp);
+				}
+			}
+			sprintf(buf, "\n\t\t\tInitial path: ");
+			strcat(buf, ipath);
+			strcat(buf, " Return path: ");
+			strcat(buf, rpath);
+		}
 		OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3111: "
-			"Error status = 0x%X\n", cl_ntoh16(status));
-		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_ERROR);
+			"Received MAD with error status = 0x%X\n"
+			"\t\t\t%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "%s\n",
+			cl_ntoh16(status), ib_get_sm_method_str(p_smp->method),
+			ib_get_sm_attr_str(p_smp->attr_id),
+			cl_ntoh32(p_smp->attr_mod),
+			cl_ntoh64(p_smp->trans_id),
+		        p_smp->mgmt_class == IB_MCLASS_SUBN_DIR ? buf : "");
+		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
 	}
 
 	switch (p_smp->method) {
@@ -708,9 +738,15 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
 
 	CL_ASSERT(p_madw);
 
+	p_smp = osm_madw_get_smp_ptr(p_madw);
 	OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3113: "
-		"MAD completed in error (%s)\n",
-		ib_get_err_str(p_madw->status));
+		"MAD completed in error (%s): "
+		"%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "\n",
+		ib_get_err_str(p_madw->status),
+		ib_get_sm_method_str(p_smp->method),
+		ib_get_sm_attr_str(p_smp->attr_id),
+		cl_ntoh32(p_smp->attr_mod),
+		cl_ntoh64(p_smp->trans_id));
 
 	/*
 	   If this was a SubnSet MAD, then this error might indicate a problem
@@ -718,7 +754,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
 	   such a problem. The subnet will not be up, and the next sweep should
 	   be a heavy sweep as well.
 	 */
-	p_smp = osm_madw_get_smp_ptr(p_madw);
 	if (p_smp->method == IB_MAD_METHOD_SET &&
 	    (p_smp->attr_id == IB_MAD_ATTR_PORT_INFO ||
 	     p_smp->attr_id == IB_MAD_ATTR_MCAST_FWD_TBL ||
@@ -734,6 +769,8 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
 		p_ctrl->p_subn->subnet_initialization_error = TRUE;
 	}
 
+	osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
+
 	/*
 	   Since we did not get any response we suspect the DR path
 	   used for the target port.
@@ -764,10 +801,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
 	   An error occurred.  No response was received to a request MAD.
 	   Retire the original request MAD.
 	 */
-
-	osm_dump_dr_smp(p_ctrl->p_log, osm_madw_get_smp_ptr(p_madw),
-			OSM_LOG_ERROR);
-
 	sm_mad_ctrl_update_wire_stats(p_ctrl);
 
 	if (osm_madw_get_err_msg(p_madw) != CL_DISP_MSGID_NONE) {

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] opensm/osm_sm_mad_ctrl.c Changes to some error messages

[Sasha Khapyorsky]

Hi Line,

On 14:39 Fri 27 Nov     , Line Holen wrote:
> Added info to some error messages and lowered log level on
> subsequent call to osm_dump_dr_smp() to reduce amount of logging
> with default log level setting.
> 
> Signed-off-by: Line Holen <Line.Holen-xsfywfwIY+M@public.gmane.org>
> 
> ---
> 
> diff --git a/opensm/opensm/osm_sm_mad_ctrl.c b/opensm/opensm/osm_sm_mad_ctrl.c
> index 74beb1c..defa014 100644
> --- a/opensm/opensm/osm_sm_mad_ctrl.c
> +++ b/opensm/opensm/osm_sm_mad_ctrl.c
> @@ -3,6 +3,7 @@
>   * Copyright (c) 2002-2005 Mellanox Technologies LTD. All rights reserved.
>   * Copyright (c) 1996-2003 Intel Corporation. All rights reserved.
>   * Copyright (c) 2009 HNR Consulting. All rights reserved.
> + * Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
>   *
>   * This software is available to you under a choice of one of two
>   * licenses.  You may choose to be licensed under the terms of the GNU
> @@ -636,9 +637,38 @@ static void sm_mad_ctrl_rcv_callback(IN osm_madw_t * p_madw,
>  		status = p_smp->status;
>  
>  	if (status != 0) {
> +		char buf[BUF_SIZE];
> +
> +		if (p_smp->mgmt_class == IB_MCLASS_SUBN_DIR) {
> +			char ipath[BUF_SIZE], rpath[BUF_SIZE];
> +			char tmp[10];
> +			uint32_t i;
> +
> +			for (i = 0; i <= p_smp->hop_count; i++) {
> +				if (i == 0) {
> +					sprintf(ipath, "%d", p_smp->initial_path[i]);
> +					sprintf(rpath, "%d", p_smp->return_path[i]);
> +				} else {
> +					sprintf(tmp, ",%d", p_smp->initial_path[i]);
> +					strcat(ipath, tmp);
> +					sprintf(tmp, ",%d", p_smp->return_path[i]);
> +					strcat(rpath, tmp);
> +				}
> +			}
> +			sprintf(buf, "\n\t\t\tInitial path: ");
> +			strcat(buf, ipath);
> +			strcat(buf, " Return path: ");
> +			strcat(buf, rpath);
> +		}
>  		OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3111: "
> -			"Error status = 0x%X\n", cl_ntoh16(status));
> -		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_ERROR);
> +			"Received MAD with error status = 0x%X\n"
> +			"\t\t\t%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "%s\n",
> +			cl_ntoh16(status), ib_get_sm_method_str(p_smp->method),
> +			ib_get_sm_attr_str(p_smp->attr_id),
> +			cl_ntoh32(p_smp->attr_mod),
> +			cl_ntoh64(p_smp->trans_id),
> +		        p_smp->mgmt_class == IB_MCLASS_SUBN_DIR ? buf : "");
> +		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
>  	}

What about moving this logging code into separate function?

Sasha

>  
>  	switch (p_smp->method) {
> @@ -708,9 +738,15 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>  
>  	CL_ASSERT(p_madw);
>  
> +	p_smp = osm_madw_get_smp_ptr(p_madw);
>  	OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3113: "
> -		"MAD completed in error (%s)\n",
> -		ib_get_err_str(p_madw->status));
> +		"MAD completed in error (%s): "
> +		"%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "\n",
> +		ib_get_err_str(p_madw->status),
> +		ib_get_sm_method_str(p_smp->method),
> +		ib_get_sm_attr_str(p_smp->attr_id),
> +		cl_ntoh32(p_smp->attr_mod),
> +		cl_ntoh64(p_smp->trans_id));
>  
>  	/*
>  	   If this was a SubnSet MAD, then this error might indicate a problem
> @@ -718,7 +754,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>  	   such a problem. The subnet will not be up, and the next sweep should
>  	   be a heavy sweep as well.
>  	 */
> -	p_smp = osm_madw_get_smp_ptr(p_madw);
>  	if (p_smp->method == IB_MAD_METHOD_SET &&
>  	    (p_smp->attr_id == IB_MAD_ATTR_PORT_INFO ||
>  	     p_smp->attr_id == IB_MAD_ATTR_MCAST_FWD_TBL ||
> @@ -734,6 +769,8 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>  		p_ctrl->p_subn->subnet_initialization_error = TRUE;
>  	}
>  
> +	osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
> +
>  	/*
>  	   Since we did not get any response we suspect the DR path
>  	   used for the target port.
> @@ -764,10 +801,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>  	   An error occurred.  No response was received to a request MAD.
>  	   Retire the original request MAD.
>  	 */
> -
> -	osm_dump_dr_smp(p_ctrl->p_log, osm_madw_get_smp_ptr(p_madw),
> -			OSM_LOG_ERROR);
> -
>  	sm_mad_ctrl_update_wire_stats(p_ctrl);
>  
>  	if (osm_madw_get_err_msg(p_madw) != CL_DISP_MSGID_NONE) {
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] opensm/osm_sm_mad_ctrl.c Changes to some error messages

[Line Holen]

On 12/ 1/09 05:13 PM, Sasha Khapyorsky wrote:
> Hi Line,
> 
> On 14:39 Fri 27 Nov     , Line Holen wrote:
>> Added info to some error messages and lowered log level on
>> subsequent call to osm_dump_dr_smp() to reduce amount of logging
>> with default log level setting.
>>
>> Signed-off-by: Line Holen <Line.Holen-xsfywfwIY+M@public.gmane.org>
>>
>> ---
>>
>> diff --git a/opensm/opensm/osm_sm_mad_ctrl.c b/opensm/opensm/osm_sm_mad_ctrl.c
>> index 74beb1c..defa014 100644
>> --- a/opensm/opensm/osm_sm_mad_ctrl.c
>> +++ b/opensm/opensm/osm_sm_mad_ctrl.c
>> @@ -3,6 +3,7 @@
>>   * Copyright (c) 2002-2005 Mellanox Technologies LTD. All rights reserved.
>>   * Copyright (c) 1996-2003 Intel Corporation. All rights reserved.
>>   * Copyright (c) 2009 HNR Consulting. All rights reserved.
>> + * Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
>>   *
>>   * This software is available to you under a choice of one of two
>>   * licenses.  You may choose to be licensed under the terms of the GNU
>> @@ -636,9 +637,38 @@ static void sm_mad_ctrl_rcv_callback(IN osm_madw_t * p_madw,
>>  		status = p_smp->status;
>>  
>>  	if (status != 0) {
>> +		char buf[BUF_SIZE];
>> +
>> +		if (p_smp->mgmt_class == IB_MCLASS_SUBN_DIR) {
>> +			char ipath[BUF_SIZE], rpath[BUF_SIZE];
>> +			char tmp[10];
>> +			uint32_t i;
>> +
>> +			for (i = 0; i <= p_smp->hop_count; i++) {
>> +				if (i == 0) {
>> +					sprintf(ipath, "%d", p_smp->initial_path[i]);
>> +					sprintf(rpath, "%d", p_smp->return_path[i]);
>> +				} else {
>> +					sprintf(tmp, ",%d", p_smp->initial_path[i]);
>> +					strcat(ipath, tmp);
>> +					sprintf(tmp, ",%d", p_smp->return_path[i]);
>> +					strcat(rpath, tmp);
>> +				}
>> +			}
>> +			sprintf(buf, "\n\t\t\tInitial path: ");
>> +			strcat(buf, ipath);
>> +			strcat(buf, " Return path: ");
>> +			strcat(buf, rpath);
>> +		}
>>  		OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3111: "
>> -			"Error status = 0x%X\n", cl_ntoh16(status));
>> -		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_ERROR);
>> +			"Received MAD with error status = 0x%X\n"
>> +			"\t\t\t%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "%s\n",
>> +			cl_ntoh16(status), ib_get_sm_method_str(p_smp->method),
>> +			ib_get_sm_attr_str(p_smp->attr_id),
>> +			cl_ntoh32(p_smp->attr_mod),
>> +			cl_ntoh64(p_smp->trans_id),
>> +		        p_smp->mgmt_class == IB_MCLASS_SUBN_DIR ? buf : "");
>> +		osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
>>  	}
> 
> What about moving this logging code into separate function?

That's fine by me. I assume you mean all the logging code within the "if (status != 0)"
section. I'll send a new patch.

> 
> Sasha
> 
>>  
>>  	switch (p_smp->method) {
>> @@ -708,9 +738,15 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>>  
>>  	CL_ASSERT(p_madw);
>>  
>> +	p_smp = osm_madw_get_smp_ptr(p_madw);
>>  	OSM_LOG(p_ctrl->p_log, OSM_LOG_ERROR, "ERR 3113: "
>> -		"MAD completed in error (%s)\n",
>> -		ib_get_err_str(p_madw->status));
>> +		"MAD completed in error (%s): "
>> +		"%s(%s), attr_mod 0x%x, TID 0x%" PRIx64 "\n",
>> +		ib_get_err_str(p_madw->status),
>> +		ib_get_sm_method_str(p_smp->method),
>> +		ib_get_sm_attr_str(p_smp->attr_id),
>> +		cl_ntoh32(p_smp->attr_mod),
>> +		cl_ntoh64(p_smp->trans_id));
>>  
>>  	/*
>>  	   If this was a SubnSet MAD, then this error might indicate a problem
>> @@ -718,7 +754,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>>  	   such a problem. The subnet will not be up, and the next sweep should
>>  	   be a heavy sweep as well.
>>  	 */
>> -	p_smp = osm_madw_get_smp_ptr(p_madw);
>>  	if (p_smp->method == IB_MAD_METHOD_SET &&
>>  	    (p_smp->attr_id == IB_MAD_ATTR_PORT_INFO ||
>>  	     p_smp->attr_id == IB_MAD_ATTR_MCAST_FWD_TBL ||
>> @@ -734,6 +769,8 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>>  		p_ctrl->p_subn->subnet_initialization_error = TRUE;
>>  	}
>>  
>> +	osm_dump_dr_smp(p_ctrl->p_log, p_smp, OSM_LOG_VERBOSE);
>> +
>>  	/*
>>  	   Since we did not get any response we suspect the DR path
>>  	   used for the target port.
>> @@ -764,10 +801,6 @@ static void sm_mad_ctrl_send_err_cb(IN void *context, IN osm_madw_t * p_madw)
>>  	   An error occurred.  No response was received to a request MAD.
>>  	   Retire the original request MAD.
>>  	 */
>> -
>> -	osm_dump_dr_smp(p_ctrl->p_log, osm_madw_get_smp_ptr(p_madw),
>> -			OSM_LOG_ERROR);
>> -
>>  	sm_mad_ctrl_update_wire_stats(p_ctrl);
>>  
>>  	if (osm_madw_get_err_msg(p_madw) != CL_DISP_MSGID_NONE) {
>>

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html