Searched hist:"40 a5f1b19bacb2de7a051be952dee85e38c9e5f5" (Results 1 – 4 of 4) sorted by relevance
| /linux/include/linux/sunrpc/ |
| H A D | sched.h | diff 40a5f1b19bacb2de7a051be952dee85e38c9e5f5 Fri May 27 16:39:50 CEST 2016 Trond Myklebust <trond.myklebust@primarydata.com> SUNRPC: RPC transport queue must be low latency
rpciod can easily get congested due to the long list of queued rpc_tasks.
Having the receive queue wait in turn for those tasks to complete can
therefore be a bottleneck.
Address the problem by separating the workqueues into:
- rpciod: manages rpc_tasks
- xprtiod: manages transport related work.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
|
| /linux/net/sunrpc/ |
| H A D | sched.c | diff 40a5f1b19bacb2de7a051be952dee85e38c9e5f5 Fri May 27 16:39:50 CEST 2016 Trond Myklebust <trond.myklebust@primarydata.com> SUNRPC: RPC transport queue must be low latency
rpciod can easily get congested due to the long list of queued rpc_tasks.
Having the receive queue wait in turn for those tasks to complete can
therefore be a bottleneck.
Address the problem by separating the workqueues into:
- rpciod: manages rpc_tasks
- xprtiod: manages transport related work.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
|
| H A D | xprt.c | diff 40a5f1b19bacb2de7a051be952dee85e38c9e5f5 Fri May 27 16:39:50 CEST 2016 Trond Myklebust <trond.myklebust@primarydata.com> SUNRPC: RPC transport queue must be low latency
rpciod can easily get congested due to the long list of queued rpc_tasks.
Having the receive queue wait in turn for those tasks to complete can
therefore be a bottleneck.
Address the problem by separating the workqueues into:
- rpciod: manages rpc_tasks
- xprtiod: manages transport related work.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
|
| H A D | xprtsock.c | diff 40a5f1b19bacb2de7a051be952dee85e38c9e5f5 Fri May 27 16:39:50 CEST 2016 Trond Myklebust <trond.myklebust@primarydata.com> SUNRPC: RPC transport queue must be low latency
rpciod can easily get congested due to the long list of queued rpc_tasks.
Having the receive queue wait in turn for those tasks to complete can
therefore be a bottleneck.
Address the problem by separating the workqueues into:
- rpciod: manages rpc_tasks
- xprtiod: manages transport related work.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
|