Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] Fwd: [OMPI users] OpenMPI with openib partitions
From: Pavel Shamis (Pasha) (pasha_at_[hidden])
Date: 2008-10-07 10:15:55


Matt,
Can you please try attached patch ? I guess it will resolve this issue.

Thanks,
Pasha

Matt Burgess wrote:
> Lenny,
>
> Thanks for the info. It doesn't seem to be be working still. My
> command line is:
>
> /opt/openmpi-ib/1.2.6/bin/mpirun -np 2 -H d2-ib,d3-ib -mca btl
> openib,self -mca btl_openib_of_pkey_val 33033
> /cluster/pallas/x86_64-ib/IMB-MPI1
>
> I don't have a "/sys/class/infiniband/mthca0/ports/1/pkeys/" but I do
> have "/sys/class/infiniband/mlx4_0/ports/1/pkeys/". It's contents are:
>
> 0 106 114 122 16 24 32 40 49 57 65 73 81 9 98
> 1 107 115 123 17 25 33 41 5 58 66 74 82 90 99
> 10 108 116 124 18 26 34 42 50 59 67 75 83 91
> 100 109 117 125 19 27 35 43 51 6 68 76 84 92
> 101 11 118 126 2 28 36 44 52 60 69 77 85 93
> 102 110 119 127 20 29 37 45 53 61 7 78 86 94
> 103 111 12 13 21 3 38 46 54 62 70 79 87 95
> 104 112 120 14 22 30 39 47 55 63 71 8 88 96
> 105 113 121 15 23 31 4 48 56 64 72 80 89 97
>
> We aren't using the opensm, but voltaire's SM on a 2012 switch.
>
> Thanks again,
> Matt
>
>
> On Tue, Oct 7, 2008 at 9:37 AM, Lenny Verkhovsky
> <lenny.verkhovsky_at_[hidden] <mailto:lenny.verkhovsky_at_[hidden]>> wrote:
>
> Hi Matt,
>
> It seems that the right way to do it is the fallowing:
>
> -mca btl openib,self -mca btl_openib_ib_pkey_val 33033
>
> when the value is a decimal number of the pkey, in your case
> 0x8109 = 33033, and no need for btl_openib_ib_pkey_ix value.
>
> ex.
>
> mpirun -np 2 -H witch2,witch3 -mca btl openib,self -mca
> btl_openib_ib_pkey_val 32769 ./mpi_p1_4_1_2 -t lt
> LT (2) (size min max avg) 1 3.511429 3.511429 3.511429
>
> if it's not working check cat
> /sys/class/infiniband/mthca0/ports/1/pkeys/* for pkeys ans SM,
> maybe it's a setup.
>
> Pasha is currently checking this issue.
>
> Best regards,
>
> Lenny.
>
>
>
>
>
> On 10/7/08, *Jeff Squyres* <jsquyres_at_[hidden]
> <mailto:jsquyres_at_[hidden]>> wrote:
>
> FWIW, if this configuration is for all of your users, you
> might want to specify these MCA params in the default MCA
> param file, or the environment, ...etc. Just so that you
> don't have to specify it on every mpirun command line.
>
> See
> http://www.open-mpi.org/faq/?category=tuning#setting-mca-params.
>
>
>
> On Oct 7, 2008, at 5:43 AM, Lenny Verkhovsky wrote:
>
> Sorry, misunderstood the question,
>
> thanks for Pasha the right command line will be
>
> -mca btl openib,self -mca btl_openib_of_pkey_val 0x8109
> -mca btl_openib_of_pkey_ix 1
>
> ex.
>
> #mpirun -np 2 -H witch2,witch3 -mca btl openib,self -mca
> btl_openib_of_pkey_val 0x8001 -mca btl_openib_of_pkey_ix 1
> ./mpi_p1_4_TRUNK -t lt
> LT (2) (size min max avg) 1 3.443480 3.443480 3.443480
>
>
> Best regards
>
> Lenny.
>
>
> On 10/6/08, Jeff Squyres <jsquyres_at_[hidden]
> <mailto:jsquyres_at_[hidden]>> wrote: On Oct 5, 2008, at
> 1:22 PM, Lenny Verkhovsky wrote:
>
> you should probably use -mca tcp,self -mca
> btl_openib_if_include ib0.8109
>
>
> Really? I thought we only took OpenFabrics device names
> in the openib_if_include MCA param...? It looks like
> ib0.8109 is an IPoIB device name.
>
>
>
> Lenny.
>
>
>
> On 10/3/08, Matt Burgess <burgess.matt_at_[hidden]
> <mailto:burgess.matt_at_[hidden]>> wrote:
> Hi,
>
>
> I'm trying to get openmpi working over openib partitions.
> On this cluster, the partition number is 0x109. The ib
> interfaces are pingable over the appropriate ib0.8109
> interface:
>
> d2:/opt/openmpi-ib # ifconfig ib0.8109
> ib0.8109 Link encap:UNSPEC HWaddr
> 80-00-00-4A-FE-80-00-00-00-00-00-00-00-00-00-00
> inet addr:10.21.48.2 <http://10.21.48.2>
> Bcast:10.21.255.255 <http://10.21.255.255>
> Mask:255.255.0.0 <http://255.255.0.0>
> inet6 addr: fe80::202:c902:26:ca01/64 Scope:Link
> UP BROADCAST RUNNING MULTICAST MTU:65520 Metric:1
> RX packets:16811 errors:0 dropped:0 overruns:0 frame:0
> TX packets:15848 errors:0 dropped:1 overruns:0
> carrier:0
> collisions:0 txqueuelen:256
> RX bytes:102229428 (97.4 Mb) TX bytes:102324172
> (97.5 Mb)
>
>
> I have tried the following:
>
> /opt/openmpi-ib/1.2.6/bin/mpirun -np 2 -machinefile
> machinefile -mca btl openib,self -mca btl_openib_max_btls
> 1 -mca btl_openib_ib_pkey_val 0x8109 -mca
> btl_openib_ib_pkey_ix 1 /cluster/pallas/x86_64-ib/IMB-MPI1
>
> but I just get a RETRY EXCEEDED ERROR. Is there a MCA
> parameter I am missing?
>
> I was successful using tcp only:
>
> /opt/openmpi-ib/1.2.6/bin/mpirun -np 2 -machinefile
> machinefile -mca btl tcp,self -mca btl_openib_max_btls 1
> -mca btl_openib_ib_pkey_val 0x8109
> /cluster/pallas/x86_64-ib/IMB-MPI1
>
>
>
> Thanks,
> Matt Burgess
>
> _______________________________________________
> users mailing list
> users_at_[hidden] <mailto:users_at_[hidden]>
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
> _______________________________________________
> users mailing list
> users_at_[hidden] <mailto:users_at_[hidden]>
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
>
> --
> Jeff Squyres
> Cisco Systems
>
>
> _______________________________________________
> users mailing list
> users_at_[hidden] <mailto:users_at_[hidden]>
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
>
>
> --
> Jeff Squyres
> Cisco Systems
>
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel

-- 
--
Pavel Shamis (Pasha)
Mellanox Technologies LTD.

Index: ompi/mca/btl/openib/btl_openib_component.c
===================================================================
--- ompi/mca/btl/openib/btl_openib_component.c (revision 19490)
+++ ompi/mca/btl/openib/btl_openib_component.c (working copy)
@@ -558,7 +558,7 @@ static int init_one_hca(opal_list_t *btl
          goto dealloc_pd;
     }
    
- ret = OMPI_SUCCESS;
+ ret = OMPI_SUCCESS;
     /* Note ports are 1 based hence j = 1 */
     for(i = 1; i <= hca->ib_dev_attr.phys_port_cnt; i++){
         struct ibv_port_attr ib_port_attr;
@@ -580,7 +580,7 @@ static int init_one_hca(opal_list_t *btl
                 uint16_t pkey,j;
                 for (j=0; j < hca->ib_dev_attr.max_pkeys; j++) {
                     ibv_query_pkey(hca->ib_dev_context, i, j, &pkey);
- pkey=ntohs(pkey);
+ pkey=ntohs(pkey) & 0x7fff;
                     if(pkey == mca_btl_openib_component.ib_pkey_val){
                         ret = init_one_port(btl_list, hca, i, j, &ib_port_attr);
                         break;
Index: ompi/mca/btl/openib/btl_openib_ini.c
===================================================================
--- ompi/mca/btl/openib/btl_openib_ini.c (revision 19490)
+++ ompi/mca/btl/openib/btl_openib_ini.c (working copy)
@@ -90,8 +90,6 @@ static int parse_line(parsed_section_val
 static void reset_section(bool had_previous_value, parsed_section_values_t *s);
 static void reset_values(ompi_btl_openib_ini_values_t *v);
 static int save_section(parsed_section_values_t *s);
-static int intify(char *string);
-static int intify_list(char *str, uint32_t **values, int *len);
 static inline void show_help(const char *topic);
 
 
@@ -364,14 +362,14 @@ static int parse_line(parsed_section_val
        all whitespace at the beginning and ending of the value. */
 
     if (0 == strcasecmp(key_buffer, "vendor_id")) {
- if (OMPI_SUCCESS != (ret = intify_list(value, &sv->vendor_ids,
+ if (OMPI_SUCCESS != (ret = ompi_btl_openib_ini_intify_list(value, &sv->vendor_ids,
                                                &sv->vendor_ids_len))) {
             return ret;
         }
     }
 
     else if (0 == strcasecmp(key_buffer, "vendor_part_id")) {
- if (OMPI_SUCCESS != (ret = intify_list(value, &sv->vendor_part_ids,
+ if (OMPI_SUCCESS != (ret = ompi_btl_openib_ini_intify_list(value, &sv->vendor_part_ids,
                                                &sv->vendor_part_ids_len))) {
             return ret;
         }
@@ -379,13 +377,13 @@ static int parse_line(parsed_section_val
 
     else if (0 == strcasecmp(key_buffer, "mtu")) {
         /* Single value */
- sv->values.mtu = (uint32_t) intify(value);
+ sv->values.mtu = (uint32_t) ompi_btl_openib_ini_intify(value);
         sv->values.mtu_set = true;
     }
 
     else if (0 == strcasecmp(key_buffer, "use_eager_rdma")) {
         /* Single value */
- sv->values.use_eager_rdma = (uint32_t) intify(value);
+ sv->values.use_eager_rdma = (uint32_t) ompi_btl_openib_ini_intify(value);
         sv->values.use_eager_rdma_set = true;
     }
 
@@ -547,7 +545,7 @@ static int save_section(parsed_section_v
 /*
  * Do string-to-integer conversion, for both hex and decimal numbers
  */
-static int intify(char *str)
+int ompi_btl_openib_ini_intify(char *str)
 {
     while (isspace(*str)) {
         ++str;
@@ -568,7 +566,7 @@ static int intify(char *str)
 /*
  * Take a comma-delimited list and infity them all
  */
-static int intify_list(char *value, uint32_t **values, int *len)
+int ompi_btl_openib_ini_intify_list(char *value, uint32_t **values, int *len)
 {
     char *comma;
     char *str = value;
@@ -584,7 +582,7 @@ static int intify_list(char *value, uint
         if (NULL == *values) {
             return OMPI_ERR_OUT_OF_RESOURCE;
         }
- *values[0] = (uint32_t) intify(str);
+ *values[0] = (uint32_t) ompi_btl_openib_ini_intify(str);
         *len = 1;
     } else {
         /* If we found a comma, loop over all the values. Be a
@@ -594,7 +592,7 @@ static int intify_list(char *value, uint
         do {
             *comma = '\0';
             *values = realloc(*values, sizeof(uint32_t) * (*len + 2));
- (*values)[*len] = (int32_t) intify(str);
+ (*values)[*len] = (int32_t) ompi_btl_openib_ini_intify(str);
             ++(*len);
             str = comma + 1;
             comma = strchr(str, ',');
@@ -602,7 +600,7 @@ static int intify_list(char *value, uint
         /* Get the last value (i.e., the value after the last
            comma, because it won't have been snarfed in the
            loop) */
- (*values)[*len] = (uint32_t) intify(str);
+ (*values)[*len] = (uint32_t) ompi_btl_openib_ini_intify(str);
         ++(*len);
     }
 
Index: ompi/mca/btl/openib/btl_openib_ini.h
===================================================================
--- ompi/mca/btl/openib/btl_openib_ini.h (revision 19490)
+++ ompi/mca/btl/openib/btl_openib_ini.h (working copy)
@@ -49,6 +49,9 @@ extern "C" {
      */
     int ompi_btl_openib_ini_finalize(void);
 
+ int ompi_btl_openib_ini_intify(char *string);
+ int ompi_btl_openib_ini_intify_list(char *str, uint32_t **values, int *len);
+
 #if defined(c_plusplus) || defined(__cplusplus)
 }
 #endif
Index: ompi/mca/btl/openib/btl_openib_mca.c
===================================================================
--- ompi/mca/btl/openib/btl_openib_mca.c (revision 19490)
+++ ompi/mca/btl/openib/btl_openib_mca.c (working copy)
@@ -27,6 +27,7 @@
 #include "opal/mca/base/mca_base_param.h"
 #include "btl_openib.h"
 #include "btl_openib_mca.h"
+#include "btl_openib_ini.h"
 
 /*
  * Local flags
@@ -97,7 +98,7 @@ static inline int reg_int(const char* pa
  */
 int btl_openib_register_mca_params(void)
 {
- char *msg, *str;
+ char *msg, *str, *pkey;
     int ival, ival2, ret, tmp;
 
     ret = OMPI_SUCCESS;
@@ -192,13 +193,15 @@ int btl_openib_register_mca_params(void)
                   0, &ival, REGINT_GE_ZERO));
     mca_btl_openib_component.ib_pkey_ix = (uint32_t) ival;
 
- CHECK(reg_int("ib_pkey_val", "InfiniBand pkey value"
+ CHECK(reg_string("ib_pkey_val", "InfiniBand pkey value"
                   "(must be > 0 and < 0xffff)",
- 0, &ival, REGINT_GE_ZERO));
- if (ival > 0xffff) {
+ "0", &pkey, 0));
+ mca_btl_openib_component.ib_pkey_val = ompi_btl_openib_ini_intify(pkey) & 0x7fff;
+ if (mca_btl_openib_component.ib_pkey_val > 0xffff ||
+ mca_btl_openib_component.ib_pkey_val < 0) {
         ret = OMPI_ERR_BAD_PARAM;
     }
- mca_btl_openib_component.ib_pkey_val = (uint32_t) ival;
+ free(pkey);
 
     CHECK(reg_int("ib_psn", "InfiniBand packet sequence starting number "
                   "(must be >= 0)",