The common practice in IB fabrics is to set the description of an HCA to be the hostname of the machine plus a description (i.e., "myhost hca-1", "myhost hca-2", etc..)
This has a limitation, however. The first is that if the machine's hostname is set via DHCP, the HCA description may be set before the hostname is, leading to an incorrect description. This can also occur if the machine's hostname changes for some other reason after boot. This can cause difficulties and confusion when trying to maintain a large fabric - if all your nodes are described as "localhost HCA-1" it can be very difficult to figure out which node is suffering from symbol errors. This patch addresses the problem by providing a function to build the node description. If the provided source string for the description contains a '%h' it will be replaced at read time with the hostname of the node. If the provided source string contains a '%d' it will be replaced at read time with the name of the HCA. For example, if the node description of the second HCA on node-a13 is set to'%h: %d', at read time this will be expanded to read "node-a2: qib1". This ensures that even after a fabric has been completely initialized, if a node's hostname changes, that change will be reflected in the next sweep of the SM, but also maintains compatibility with existing code since the behavior is unchanged if the description string does not contain a '%' character. Changes from Version 1: Replaced the original substitution of the hostname for '@' with an implementation of the '%h' and '%d' fields, and changed the default node description for qib and ipath HCA types to default to "%h: %d <description>". Signed-off-by: Michael Heinz <[email protected]> --- drivers/infiniband/core/mad.c | 33 +++++++++++++++++++++++++++++++++ include/rdma/ib_mad.h | 9 +++++++++ 2 files changed, 42 insertions(+), 0 deletions(-) diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 822cfdc..d5468e1 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -41,6 +41,7 @@ #include "mad_rmpp.h" #include "smi.h" #include "agent.h" +#include "linux/utsname.h" MODULE_LICENSE("Dual BSD/GPL"); MODULE_DESCRIPTION("kernel IB MAD API"); @@ -932,6 +933,38 @@ int ib_get_mad_data_offset(u8 mgmt_class) } EXPORT_SYMBOL(ib_get_mad_data_offset); +void ib_build_node_desc(struct ib_smp *smp, struct ib_device *dev) +{ + char *dest = smp->data; + char *end = dest + IB_SMP_DATA_SIZE; + char *src = dev->node_desc; + char *field; + + while (*src && (dest < end)) { + if (*src != '%') { + *dest++ = *src++; + } else { + src++; + switch (*src) { + case 'h': + field = init_utsname()->nodename; + src++; + for (; *field && (*field != '.') && + (dest < end);) + *dest++ = *field++; + break; + case 'd': + field = dev->name; + src++; + for (; *field && (dest < end);) + *dest++ = *field++; + break; + } + } + } +} +EXPORT_SYMBOL(ib_build_node_desc); + int ib_is_mad_class_rmpp(u8 mgmt_class) { if ((mgmt_class == IB_MGMT_CLASS_SUBN_ADM) || diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h index d3b9401..417a371 100644 --- a/include/rdma/ib_mad.h +++ b/include/rdma/ib_mad.h @@ -40,6 +40,7 @@ #include <linux/list.h> #include <rdma/ib_verbs.h> +#include <rdma/ib_smi.h> /* Management base version */ #define IB_MGMT_BASE_VERSION 1 @@ -637,6 +638,14 @@ int ib_is_mad_class_rmpp(u8 mgmt_class); int ib_get_mad_data_offset(u8 mgmt_class); /** + * ib_build_node_desc - copies the node description and replaces + * any @ markers with the present system node name. + * @dest: destination + * @src: source + */ +void ib_build_node_desc(struct ib_smp *smp, struct ib_device *dev); + +/** * ib_get_rmpp_segment - returns the data buffer for a given RMPP segment. * @send_buf: Previously allocated send data buffer. * @seg_num: number of segment to return -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
