* [PATCH] opensm/osm_get_port_by_lid(): use faster cl_ptr_vector_get()
@ 2009-09-29 11:02 Sasha Khapyorsky
2009-09-29 11:02 ` [PATCH] opensm/osm_get_port_by_lid(): speedup a port lookup Sasha Khapyorsky
0 siblings, 1 reply; 3+ messages in thread
From: Sasha Khapyorsky @ 2009-09-29 11:02 UTC (permalink / raw)
To: OpenIB, linux-rdma; +Cc: Hal Rosenstock
Use faster cl_ptr_vector_get() call instead of cl_ptr_vector_at(). In
this way eliminate 'stat' variable needs.
Signed-off-by: Sasha Khapyorsky <sashak-smomgflXvOZWk0Htik3J/w@public.gmane.org>
---
opensm/opensm/osm_subnet.c | 8 ++++----
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/opensm/opensm/osm_subnet.c b/opensm/opensm/osm_subnet.c
index b475031..67bc7e1 100644
--- a/opensm/opensm/osm_subnet.c
+++ b/opensm/opensm/osm_subnet.c
@@ -646,22 +646,22 @@ osm_port_t *osm_get_port_by_guid(IN osm_subn_t const *p_subn, IN ib_net64_t guid
osm_port_t *osm_get_port_by_lid(IN osm_subn_t const * subn, IN ib_net16_t lid)
{
osm_port_t *port = NULL;
- ib_api_status_t stat;
uint16_t base_lid;
uint8_t lmc;
lid = cl_ntoh16(lid);
+ if (lid >= cl_ptr_vector_get_size(&subn->port_lid_tbl))
+ return NULL;
/* Loop on lmc from 0 up through max LMC possible */
for (lmc = 0; lmc <= IB_PORT_LMC_MAX; lmc++) {
/* Calculate a base LID assuming this is the real LMC */
base_lid = lid & ~((1 << lmc) - 1);
- stat = cl_ptr_vector_at(&subn->port_lid_tbl, base_lid,
- (void *)&port);
+ port = cl_ptr_vector_get(&subn->port_lid_tbl, base_lid);
/* Determine if base LID "tested" is the real base LID */
/* This is true if the LMC "tested" is the port's actual LMC */
- if (stat == CL_SUCCESS && port && lmc == osm_port_get_lmc(port))
+ if (port && lmc == osm_port_get_lmc(port))
return port;
}
--
1.6.5.rc1
--
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
^ permalink raw reply related [flat|nested] 3+ messages in thread* [PATCH] opensm/osm_get_port_by_lid(): speedup a port lookup
2009-09-29 11:02 [PATCH] opensm/osm_get_port_by_lid(): use faster cl_ptr_vector_get() Sasha Khapyorsky
@ 2009-09-29 11:02 ` Sasha Khapyorsky
2009-09-29 11:03 ` [PATCH] opensm/osm_get_port_by_lid(): don't bother with lmc Sasha Khapyorsky
0 siblings, 1 reply; 3+ messages in thread
From: Sasha Khapyorsky @ 2009-09-29 11:02 UTC (permalink / raw)
To: OpenIB, linux-rdma; +Cc: Hal Rosenstock
Speedup a port lookup over LMC array - it is not necessary to match LMC
exactly for found port because base lid should be equal to requested lid
masked value, so '<=' comparison should be enough and we don't need to
loop up to an actual port's lmc value match.
Signed-off-by: Sasha Khapyorsky <sashak-smomgflXvOZWk0Htik3J/w@public.gmane.org>
---
opensm/opensm/osm_subnet.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/opensm/opensm/osm_subnet.c b/opensm/opensm/osm_subnet.c
index 67bc7e1..30f8af5 100644
--- a/opensm/opensm/osm_subnet.c
+++ b/opensm/opensm/osm_subnet.c
@@ -661,7 +661,7 @@ osm_port_t *osm_get_port_by_lid(IN osm_subn_t const * subn, IN ib_net16_t lid)
port = cl_ptr_vector_get(&subn->port_lid_tbl, base_lid);
/* Determine if base LID "tested" is the real base LID */
/* This is true if the LMC "tested" is the port's actual LMC */
- if (port && lmc == osm_port_get_lmc(port))
+ if (port && lmc <= osm_port_get_lmc(port))
return port;
}
--
1.6.5.rc1
--
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
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH] opensm/osm_get_port_by_lid(): don't bother with lmc
2009-09-29 11:02 ` [PATCH] opensm/osm_get_port_by_lid(): speedup a port lookup Sasha Khapyorsky
@ 2009-09-29 11:03 ` Sasha Khapyorsky
0 siblings, 0 replies; 3+ messages in thread
From: Sasha Khapyorsky @ 2009-09-29 11:03 UTC (permalink / raw)
To: OpenIB, linux-rdma; +Cc: Hal Rosenstock
Since subn->port_lid_tbl vector is filled for all port's LIDs in
accordance with its LMC value, so we don't need to bother with LMC
tracking and instead can just return a pointer indexed by requested lid.
Obviously it speeds this helper up significantly.
Signed-off-by: Sasha Khapyorsky <sashak-smomgflXvOZWk0Htik3J/w@public.gmane.org>
---
opensm/opensm/osm_subnet.c | 21 ++-------------------
1 files changed, 2 insertions(+), 19 deletions(-)
diff --git a/opensm/opensm/osm_subnet.c b/opensm/opensm/osm_subnet.c
index 30f8af5..97b62c2 100644
--- a/opensm/opensm/osm_subnet.c
+++ b/opensm/opensm/osm_subnet.c
@@ -645,26 +645,9 @@ osm_port_t *osm_get_port_by_guid(IN osm_subn_t const *p_subn, IN ib_net64_t guid
**********************************************************************/
osm_port_t *osm_get_port_by_lid(IN osm_subn_t const * subn, IN ib_net16_t lid)
{
- osm_port_t *port = NULL;
- uint16_t base_lid;
- uint8_t lmc;
-
lid = cl_ntoh16(lid);
- if (lid >= cl_ptr_vector_get_size(&subn->port_lid_tbl))
- return NULL;
-
- /* Loop on lmc from 0 up through max LMC possible */
- for (lmc = 0; lmc <= IB_PORT_LMC_MAX; lmc++) {
- /* Calculate a base LID assuming this is the real LMC */
- base_lid = lid & ~((1 << lmc) - 1);
-
- port = cl_ptr_vector_get(&subn->port_lid_tbl, base_lid);
- /* Determine if base LID "tested" is the real base LID */
- /* This is true if the LMC "tested" is the port's actual LMC */
- if (port && lmc <= osm_port_get_lmc(port))
- return port;
- }
-
+ if (lid < cl_ptr_vector_get_size(&subn->port_lid_tbl))
+ return cl_ptr_vector_get(&subn->port_lid_tbl, lid);
return NULL;
}
--
1.6.5.rc1
--
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
^ permalink raw reply related [flat|nested] 3+ messages in thread
end of thread, other threads:[~2009-09-29 11:03 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-09-29 11:02 [PATCH] opensm/osm_get_port_by_lid(): use faster cl_ptr_vector_get() Sasha Khapyorsky
2009-09-29 11:02 ` [PATCH] opensm/osm_get_port_by_lid(): speedup a port lookup Sasha Khapyorsky
2009-09-29 11:03 ` [PATCH] opensm/osm_get_port_by_lid(): don't bother with lmc Sasha Khapyorsky
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox