[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210507140638.339-1-anand.a.khoje@oracle.com>
Date: Fri, 7 May 2021 19:36:38 +0530
From: Anand Khoje <anand.a.khoje@...cle.com>
To: dledford@...hat.com, jgg@...pe.ca, leon@...nel.org,
avihaih@...dia.com, liangwenpeng@...wei.com,
jackm@....mellanox.co.il, galpress@...zon.com,
kamalheib1@...il.com, mbloch@...dia.com, lee.jones@...aro.org,
maorg@...lanox.com, maxg@...lanox.com, parav@...dia.com,
eli@...lanox.com, ogerlitz@...lanox.com,
linux-rdma@...r.kernel.org, linux-kernel@...r.kernel.org
Cc: Anand Khoje <anand.a.khoje@...cle.com>,
Haakon Bugge <haakon.bugge@...cle.com>
Subject: [PATCH] IB/core: Obtain subnet_prefix from cache in IB devices
ib_query_port() calls device->ops.query_port() to get the port
attributes. The method of querying is device driver specific.
The same function calls device->ops.query_gid() to get the GID and
extract the subnet_prefix (gid_prefix).
The GID and subnet_prefix are stored in a cache. But they do not get
read from the cache if the device is an Infiniband device. The
following change takes advantage of the cached subnet_prefix.
Testing with RDBMS has shown a significant improvement in performance
with this change.
The function ib_cache_is_initialised() is introduced because
ib_query_port() gets called early in the stage when the cache is not
built while reading port immutable property.
In that case, the default GID still gets read from HCA for IB link
layer. The shuffling of netdev_lock in struct ib_port_data is done
such that the size of struct ib_port_data remains the same after
adding flags.
Fixes: fad61ad ("IB/core: Add subnet prefix to port info")
Signed-off-by: Anand Khoje <anand.a.khoje@...cle.com>
Signed-off-by: Haakon Bugge <haakon.bugge@...cle.com>
---
drivers/infiniband/core/cache.c | 7 ++++++-
drivers/infiniband/core/device.c | 11 +++++++++++
include/rdma/ib_cache.h | 7 +++++++
include/rdma/ib_verbs.h | 10 +++++++++-
4 files changed, 33 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index 3b0991f..b580c26 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -1627,6 +1627,8 @@ int ib_cache_setup_one(struct ib_device *device)
err = ib_cache_update(device, p, true);
if (err)
return err;
+ set_bit(IB_PORT_CACHE_INITIALIZED,
+ &device->port_data[p].flags);
}
return 0;
@@ -1642,8 +1644,11 @@ void ib_cache_release_one(struct ib_device *device)
* all the device's resources when the cache could no
* longer be accessed.
*/
- rdma_for_each_port (device, p)
+ rdma_for_each_port (device, p) {
+ clear_bit(IB_PORT_CACHE_INITIALIZED,
+ &device->port_data[p].flags);
kfree(device->port_data[p].cache.pkey);
+ }
gid_table_release_one(device);
}
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index c660cef..6d62023 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -2064,6 +2064,17 @@ static int __ib_query_port(struct ib_device *device,
IB_LINK_LAYER_INFINIBAND)
return 0;
+ if (!ib_cache_is_initialised(device, port_num))
+ goto query_gid_from_device;
+
+ err = ib_get_cached_subnet_prefix(device, port_num,
+ &port_attr->subnet_prefix);
+ if (err)
+ goto query_gid_from_device;
+
+ return 0;
+
+query_gid_from_device:
err = device->ops.query_gid(device, port_num, 0, &gid);
if (err)
return err;
diff --git a/include/rdma/ib_cache.h b/include/rdma/ib_cache.h
index 226ae37..bebeb94 100644
--- a/include/rdma/ib_cache.h
+++ b/include/rdma/ib_cache.h
@@ -114,4 +114,11 @@ ssize_t rdma_query_gid_table(struct ib_device *device,
struct ib_uverbs_gid_entry *entries,
size_t max_entries);
+static inline bool ib_cache_is_initialised(struct ib_device *device,
+ u8 port_num)
+{
+ return test_bit(IB_PORT_CACHE_INITIALIZED,
+ &device->port_data[port_num].flags);
+}
+
#endif /* _IB_CACHE_H */
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 7e2f369..ad2a55e 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -2169,17 +2169,25 @@ struct ib_port_immutable {
u32 max_mad_size;
};
+enum ib_port_data_flags {
+ IB_PORT_CACHE_INITIALIZED = 1 << 0,
+};
+
struct ib_port_data {
struct ib_device *ib_dev;
struct ib_port_immutable immutable;
spinlock_t pkey_list_lock;
+
+ spinlock_t netdev_lock;
+
+ unsigned long flags;
+
struct list_head pkey_list;
struct ib_port_cache cache;
- spinlock_t netdev_lock;
struct net_device __rcu *netdev;
struct hlist_node ndev_hash_link;
struct rdma_port_counter port_counter;
--
1.8.3.1
Powered by blists - more mailing lists