aboutsummaryrefslogtreecommitdiff
path: root/net/sunrpc/xprtsock.c
AgeCommit message (Collapse)Author
2006-12-07Merge branch 'master' of /home/trondmy/kernel/linux-2.6/ into merge_linusTrond Myklebust
2006-12-07Merge branch 'master' of /home/trondmy/kernel/linux-2.6/ into merge_linusTrond Myklebust
Conflicts: include/linux/sunrpc/xprt.h net/sunrpc/xprtsock.c Fix up conflicts with the workqueue changes.
2006-12-07[PATCH] lockdep: annotate nfs/nfsd in-kernel socketsPeter Zijlstra
Stick NFS sockets in their own class to avoid some lockdep warnings. NFS sockets are never exposed to user-space, and will hence not trigger certain code paths that would otherwise pose deadlock scenarios. [akpm@osdl.org: cleanups] Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Steven Dickson <SteveD@redhat.com> Acked-by: Ingo Molnar <mingo@elte.hu> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Acked-by: Neil Brown <neilb@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@osdl.org> [ Fixed patch corruption by quilt, pointed out by Peter Zijlstra ] Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-06SUNRPC: relocate the creation of socket-specific tunablesChuck Lever
Clean-up: The RPC client currently creates some sysctls that are specific to the socket transport. Move those entirely into xprtsock.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: create stubs for xprtsock init and cleanupChuck Lever
Over time we will want to add some specific init and cleanup logic for the xprtsock implementation. Add stub routines for initialization and exit processing. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Rename skb_reader_t and friendsChuck Lever
Clean-up: hch suggested that the RPC client shouldn't pollute the name space used by the generic skb manipulation routines in net/core/skbuff.c. Rename a couple of types in xdr.h to adhere to this convention. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: skb_read_bits is the same as xs_tcp_copy_dataChuck Lever
Clean-up: eliminate xs_tcp_copy_data -- it's exactly the same logic as the common routine skb_read_bits. The UDP and TCP socket read code now share the same routine for copying data into an xdr_buf. Now that skb_read_bits() is exported, rename it to avoid confusing it with a generic skb_* function. As these functions are XDR-specific, they should not have names that suggest they are of generic use. Also rename skb_read_and_csum_bits() to be consistent. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Make address format buffers more genericChuck Lever
For now we will assume that all transports will use the address format buffers in the rpc_xprt struct to store their addresses. Change rpc_peer2str() to be a generic routine to handle this, and get rid of the print_address() op in the rpc_xprt_ops vector. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: move saved socket callback functions to a private data structureChuck Lever
Move the three fields for saving socket callback functions out of the rpc_xprt structure and into a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Move the UDP socket bufsize parameters to a private data structureChuck Lever
Move the socket-specific buffer size parameters for UDP sockets to a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Move rpc_xprt socket connect fields into private data structureChuck Lever
Move the socket-specific connection management fields out of the generic rpc_xprt structure into a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Move TCP state flags into xprtsock.cChuck Lever
Move "XPRT_LAST_FRAG" and friends from xprt.h into xprtsock.c, and rename them to use the naming scheme in use in xprtsock.c. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Move TCP receive state variables into private data structureChuck Lever
Move the TCP receive state variables from the generic rpc_xprt structure to a private structure maintained inside net/sunrpc/xprtsock.c. Also rename a function/variable pair to refer to RPC fragment headers instead of record markers, to be consistent with types defined in sunrpc/*.h. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Remove sock and inet fields from rpc_xprtChuck Lever
The "sock" and "inet" fields are socket-specific. Move them to a private data structure maintained entirely within net/sunrpc/xprtsock.c Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Allocate a private data area for socket-specific rpc_xprt fieldsChuck Lever
When setting up a new transport instance, allocate enough memory for an rpc_xprt and a private area. As part of the same memory allocation, it will be easy to find one, given a pointer to the other. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Make the transport-specific setup routine allocate rpc_xprtChuck Lever
Change the location where the rpc_xprt structure is allocated so each transport implementation can allocate a private area from the same chunk of memory. Note also that xprt->ops->destroy, rather than xprt_destroy, is now responsible for freeing rpc_xprt when the transport is destroyed. Test plan: Connectathon. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-12-06SUNRPC: Clean up xs_send_pages()Trond Myklebust
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-11-22WorkStruct: Pass the work_struct pointer instead of context dataDavid Howells
Pass the work_struct pointer to the work function rather than context data. The work function can use container_of() to work out the data. For the cases where the container of the work_struct may go away the moment the pending bit is cleared, it is made possible to defer the release of the structure by deferring the clearing of the pending bit. To make this work, an extra flag is introduced into the management side of the work_struct. This governs auto-release of the structure upon execution. Ordinarily, the work queue executor would release the work_struct for further scheduling or deallocation by clearing the pending bit prior to jumping to the work function. This means that, unless the driver makes some guarantee itself that the work_struct won't go away, the work function may not access anything else in the work_struct or its container lest they be deallocated.. This is a problem if the auxiliary data is taken away (as done by the last patch). However, if the pending bit is *not* cleared before jumping to the work function, then the work function *may* access the work_struct and its container with no problems. But then the work function must itself release the work_struct by calling work_release(). In most cases, automatic release is fine, so this is the default. Special initiators exist for the non-auto-release case (ending in _NAR). Signed-Off-By: David Howells <dhowells@redhat.com>
2006-11-22WorkStruct: Separate delayable and non-delayable events.David Howells
Separate delayable work items from non-delayable work items be splitting them into a separate structure (delayed_work), which incorporates a work_struct and the timer_list removed from work_struct. The work_struct struct is huge, and this limits it's usefulness. On a 64-bit architecture it's nearly 100 bytes in size. This reduces that by half for the non-delayable type of event. Signed-Off-By: David Howells <dhowells@redhat.com>
2006-10-20[PATCH] SUNRPC: fix a typoChuck Lever
Yes, this actually passed tests the way it was. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-09-28[SUNRPC]: trivial endianness annotationsAlexey Dobriyan
pure s/u32/__be32/ [AV: large part based on Alexey's patches] Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-09-23Merge git://git.linux-nfs.org/pub/linux/nfs-2.6Linus Torvalds
* git://git.linux-nfs.org/pub/linux/nfs-2.6: (74 commits) NFS: unmark NFS direct I/O as experimental NFS: add comments clarifying the use of nfs_post_op_update() NFSv4: rpc_mkpipe creating socket inodes w/out sk buffers NFS: Use SEEK_END instead of hardcoded value NFSv4: When mounting with a port=0 argument, substitute port=2049 NFSv4: Poll more aggressively when handling NFS4ERR_DELAY NFSv4: Handle the condition NFS4ERR_FILE_OPEN NFSv4: Retry lease recovery if it failed during a synchronous operation. NFS: Don't invalidate the symlink we just stuffed into the cache NFS: Make read() return an ESTALE if the file has been deleted NFSv4: It's perfectly legal for clp to be NULL here.... NFS: nfs_lookup - don't hash dentry when optimising away the lookup SUNRPC: Fix Oops in pmap_getport_done SUNRPC: Add refcounting to the struct rpc_xprt SUNRPC: Clean up soft task error handling SUNRPC: Handle ENETUNREACH, EHOSTUNREACH and EHOSTDOWN socket errors SUNRPC: rpc_delay() should not clobber the rpc_task->tk_status Fix a referral error Oops NFS: NFS_ROOT should use the new rpc_create API NFS: Fix up compiler warnings on 64-bit platforms in client.c ... Manually resolved conflict in net/sunrpc/xprtsock.c
2006-09-22SUNRPC: Eliminate xprt_create_proto and rpc_create_clientChuck Lever
The two function call API for creating a new RPC client is now obsolete. Remove it. Also, remove an unnecessary check to see whether the caller is capable of using privileged network services. The kernel RPC client always uses a privileged ephemeral port by default; callers are responsible for checking the authority of users to make use of any RPC service, or for specifying that a nonprivileged port is acceptable. Test plan: Repeated runs of Connectathon locking suite. Check network trace to ensure correctness of NLM requests and replies. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-09-22SUNRPC: Use "sockaddr_storage" for storing RPC client's remote peer addressChuck Lever
IPv6 addresses are big (128 bytes). Now that no RPC client consumers treat the addr field in rpc_xprt structs as an opaque, and access it only via the API calls, we can safely widen the field in the rpc_xprt struct to accomodate larger addresses. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-09-22SUNRPC: add xprt switch API for printing formatted remote peer addressesChuck Lever
Add a new method to the transport switch API to provide a way to convert the opaque contents of xprt->addr to a human-readable string. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-09-22SUNRPC: Introduce transport switch callout for pluggable rpcbindChuck Lever
Introduce a clean transport switch API for plugging in different types of rpcbind mechanisms. For instance, rpcbind can cleanly replace the existing portmapper client, or a transport can choose to implement RPC binding any way it likes. Test plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. NFSv2/3 and NFSv4 mounting should be carefully checked. Probably need to rig a server where certain services aren't running, or that returns an error for some typical operation. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-09-22SUNRPC: Create a helper to tell whether a transport is boundChuck Lever
Hide the contents and format of xprt->addr by eliminating direct uses of the xprt->addr.sin_port field. This change is required to support alternate RPC host address formats (eg IPv6). Test-plan: Destructive testing (unplugging the network temporarily). Repeated runs of Connectathon locking suite with UDP and TCP. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-09-22[SUNRPC]: Remove the unnecessary check for highmem in xs_sendpages().Sridhar Samudrala
Just call kernel_sendpage() directly. Signed-off-by: Sridhar Samudrala <sri@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-09-22[SUNRPC]: Update to use in-kernel sockets API.Sridhar Samudrala
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com> Acked-by: James Morris <jmorris@namei.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-08-03RPC: Ensure that we disconnect TCP socket when client requests error outTrond Myklebust
If we're part way through transmitting a TCP request, and the client errors, then we need to disconnect and reconnect the TCP socket in order to avoid confusing the server. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> (cherry picked from 031a50c8b9ea82616abd4a4e18021a25848941ce commit)
2006-07-21[NET]: Conversions from kmalloc+memset to k(z|c)alloc.Panagiotis Issaris
Signed-off-by: Panagiotis Issaris <takis@issaris.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-09SUNRPC: select privileged port numbers at randomChuck Lever
Make the RPC client select privileged ephemeral source ports at random. This improves DRC behavior on the server by using the same port when reconnecting for the same mount point, but using a different port for fresh mounts. The Linux TCP implementation already does this for nonprivileged ports. Note that TCP sockets in TIME_WAIT will prevent quick reuse of a random ephemeral port number by leaving the port INUSE until the connection transitions out of TIME_WAIT. Test plan: Connectathon against every known server implementation using multiple mount points. Locking especially. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-20SUNRPC: introduce per-task RPC iostatsChuck Lever
Account for various things that occur while an RPC task is executed. Separate timers for RPC round trip and RPC execution time show how long RPC requests wait in queue before being sent. Eventually these will be accumulated at xprt_release time in one place where they can be viewed from userland. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-03-20SUNRPC: add a handful of per-xprt countersChuck Lever
Monitor generic transport events. Add a transport switch callout to format transport counters for export to user-land. Test plan: Compile kernel with CONFIG_NFS enabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-01-06SUNRPC: Ensure client closes the socket when server initiates a closeTrond Myklebust
If the server decides to close the RPC socket, we currently don't actually respond until either another RPC call is scheduled, or until xprt_autoclose() gets called by the socket expiry timer (which may be up to 5 minutes later). This patch ensures that xprt_autoclose() is called much sooner if the server closes the socket. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-01-06SUNRPC: transport switch API for setting port numberChuck Lever
At some point, transport endpoint addresses will no longer be IPv4. To hide the structure of the rpc_xprt's address field from ULPs and port mappers, add an API for setting the port number during an RPC bind operation. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. NFSv2/3 and NFSv4 mounting should be carefully checked. Probably need to rig a server where certain services aren't running, or that returns an error for some typical operation. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2006-01-06SUNRPC: switchable buffer allocationChuck Lever
Add RPC client transport switch support for replacing buffer management on a per-transport basis. In the current IPv4 socket transport implementation, RPC buffers are allocated as needed for each RPC message that is sent. Some transport implementations may choose to use pre-allocated buffers for encoding, sending, receiving, and unmarshalling RPC messages, however. For transports capable of direct data placement, the buffers can be carved out of a pre-registered area of memory rather than from a slab cache. Test-plan: Millions of fsx operations. Performance characterization with "sio" and "iozone". Use oprofile and other tools to look for significant regression in CPU utilization. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-12-19RPC: Do not block on skb allocationTrond Myklebust
If we get something like the following, [ 125.300636] [<c04086e1>] schedule_timeout+0x54/0xa5 [ 125.305931] [<c040866e>] io_schedule_timeout+0x29/0x33 [ 125.311495] [<c02880c4>] blk_congestion_wait+0x70/0x85 [ 125.317058] [<c014136b>] throttle_vm_writeout+0x69/0x7d [ 125.322720] [<c014714d>] shrink_zone+0xe0/0xfa [ 125.327560] [<c01471d4>] shrink_caches+0x6d/0x6f [ 125.332581] [<c01472a6>] try_to_free_pages+0xd0/0x1b5 [ 125.338056] [<c013fa4b>] __alloc_pages+0x135/0x2e8 [ 125.343258] [<c03b74ad>] tcp_sendmsg+0xaa0/0xb78 [ 125.348281] [<c03d4666>] inet_sendmsg+0x48/0x53 [ 125.353212] [<c0388716>] sock_sendmsg+0xb8/0xd3 [ 125.358147] [<c0388773>] kernel_sendmsg+0x42/0x4f [ 125.363259] [<c038bc00>] sock_no_sendpage+0x5e/0x77 [ 125.368556] [<c03ee7af>] xs_tcp_send_request+0x2af/0x375 then the socket is blocked until memory is reclaimed, and no progress can ever be made. Try to access the emergency pools by using GFP_ATOMIC. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-11-04SUNRPC: allow sunrpc.o to link when CONFIG_SYSCTL is disabledChuck Lever
The sunrpc module should build properly even when CONFIG_SYSCTL is disabled. Reported by Jan-Benedict Glaw. Test plan: Compile kernel with CONFIG_NFS as a module and built-in, and CONFIG_SYSCTL enabled and disabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: rationalize set_buffer_sizeChuck Lever
In fact, ->set_buffer_size should be completely functionless for non-UDP. Test-plan: Check socket buffer size on UDP sockets over time. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: parametrize various transport connect timeoutsChuck Lever
Each transport implementation can now set unique bind, connect, reestablishment, and idle timeout values. These are variables, allowing the values to be modified dynamically. This permits exponential backoff of any of these values, for instance. As an example, we implement exponential backoff for the connection reestablishment timeout. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: make sure to get the same local port number when reconnectingChuck Lever
Implement a best practice: if the remote end drops our connection, try to reconnect using the same port number. This is important because the NFS server's Duplicate Reply Cache often hashes on the source port number. If the client reuses the port number when it reconnects, the server's DRC will be more effective. Based on suggestions by Mike Eisler, Olaf Kirch, and Alexey Kuznetsky. Test-plan: Destructive testing. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: allow RPC client's port range to be adjustableChuck Lever
Select an RPC client source port between 650 and 1023 instead of between 1 and 800. The old range conflicts with a number of network services. Provide sysctls to allow admins to select a different port range. Note that this doesn't affect user-level RPC library behavior, which still uses 1 to 800. Based on a suggestion by Olaf Kirch <okir@suse.de>. Test-plan: Repeated mount and unmount. Destructive testing. Idle timeouts. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: clean up after nocong was removedChuck Lever
Clean-up: Move some macros that are specific to the Van Jacobson implementation into xprt.c. Get rid of the cong_wait field in rpc_xprt, which is no longer used. Get rid of xprt_clear_backlog. Test-plan: Compile with CONFIG_NFS enabled. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: remove xprt->nocongChuck Lever
Get rid of the "xprt->nocong" variable. Test-plan: Use WAN simulation to cause sporadic bursty packet loss with UDP mounts. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: add a release_rqst callout to the RPC transport switchChuck Lever
The final place where congestion control state is adjusted is in xprt_release, where each request is finally released. Add a callout there to allow transports to perform additional processing when a request is about to be released. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: add generic interface for adjusting the congestion windowChuck Lever
A new interface that allows transports to adjust their congestion window using the Van Jacobson implementation in xprt.c is provided. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: separate xprt_timer implementationsChuck Lever
Allow transports to hook the retransmit timer interrupt. Some transports calculate their congestion window here so that a retransmit timeout has immediate effect on the congestion window. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: expose API for serializing access to RPC transportsChuck Lever
The next method we abstract is the one that releases a transport, allowing another task to have access to the transport. Again, one generic version of this is provided for transports that don't need the RPC client to perform congestion control, and one version is for transports that can use the original Van Jacobson implementation in xprt.c. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2005-09-23[PATCH] RPC: expose API for serializing access to RPC transportsChuck Lever
The next several patches introduce an API that allows transports to choose whether the RPC client provides congestion control or whether the transport itself provides it. The first method we abstract is the one that serializes access to the RPC transport to prevent the bytes from different requests from mingling together. This method provides proper request serialization and the opportunity to prevent new requests from being started because the transport is congested. The normal situation is for the transport to handle congestion control itself. Although NFS over UDP was first, it has been recognized after years of experience that having the transport provide congestion control is much better than doing it in the RPC client. Thus TCP, and probably every future transport implementation, will use the default method, xprt_lock_write, provided in xprt.c, which does not provide any kind of congestion control. UDP can continue using the xprt.c-provided Van Jacobson congestion avoidance implementation. Test-plan: Use WAN simulation to cause sporadic bursty packet loss. Look for significant regression in performance or client stability. Signed-off-by: Chuck Lever <cel@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>