summaryrefslogtreecommitdiff
path: root/fs/dlm/lowcomms.c
AgeCommit message (Collapse)Author
2011-07-06dlm: dump address of unknown nodeMasatake YAMATO
When the dlm fails to make a network connection to another node, include the address of the node in the error message. Signed-off-by: Masatake YAMATO <yamato@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2011-03-31Fix common misspellingsLucas De Marchi
Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
2011-03-10dlm: use alloc_workqueue functionDavid Teigland
Replaces deprecated create_singlethread_workqueue(). Signed-off-by: David Teigland <teigland@redhat.com>
2011-02-11dlm: use single thread workqueuesDavid Teigland
The recent commit to use cmwq for send and recv threads dcce240ead802d42b1e45ad2fcb2ed4a399cb255 introduced problems, apparently due to multiple workqueue threads. Single threads make the problems go away, so return to that until we fully understand the concurrency issues with multiple threads. Signed-off-by: David Teigland <teigland@redhat.com>
2010-12-13dlm: sanitize work_start() in lowcomms.cNamhyung Kim
The create_workqueue() returns NULL if failed rather than ERR_PTR(). Fix error checking and remove unnecessary variable 'error'. Signed-off-by: Namhyung Kim <namhyung@gmail.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: David Teigland <teigland@redhat.com>
2010-11-12dlm: reduce cond_resched during sendBob Peterson
Calling cond_resched() after every send can unnecessarily degrade performance. Go back to an old method of scheduling after 25 messages. Signed-off-by: Bob Peterson <rpeterso@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2010-11-12dlm: use TCP_NODELAYDavid Teigland
Nagling doesn't help and can sometimes hurt dlm comms. Signed-off-by: David Teigland <teigland@redhat.com>
2010-11-12dlm: Use cmwq for send and receive workqueuesSteven Whitehouse
So far as I can tell, there is no reason to use a single-threaded send workqueue for dlm, since it may need to send to several sockets concurrently. Both workqueues are set to WQ_MEM_RECLAIM to avoid any possible deadlocks, WQ_HIGHPRI since locking traffic is highly latency sensitive (and to avoid a priority inversion wrt GFS2's glock_workqueue) and WQ_FREEZABLE just in case someone needs to do that (even though with current cluster infrastructure, it doesn't make sense as the node will most likely land up ejected from the cluster) in the future. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: David Teigland <teigland@redhat.com>
2010-11-11dlm: Handle application limited situations properly.David Miller
In the normal regime where an application uses non-blocking I/O writes on a socket, they will handle -EAGAIN and use poll() to wait for send space. They don't actually sleep on the socket I/O write. But kernel level RPC layers that do socket I/O operations directly and key off of -EAGAIN on the write() to "try again later" don't use poll(), they instead have their own sleeping mechanism and rely upon ->sk_write_space() to trigger the wakeup. So they do effectively sleep on the write(), but this mechanism alone does not let the socket layers know what's going on. Therefore they must emulate what would have happened, otherwise TCP cannot possibly see that the connection is application window size limited. Handle this, therefore, like SUNRPC by setting SOCK_NOSPACE and bumping the ->sk_write_count as needed when we hit the send buffer limits. This should make TCP send buffer size auto-tuning and the ->sk_write_space() callback invocations actually happen. Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: David Teigland <teigland@redhat.com>
2010-08-05fs/dlm: Drop unnecessary null testJulia Lawall
hlist_for_each_entry binds its first argument to a non-null value, and thus any null test on the value of that argument is superfluous. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ iterator I; expression x,E,E1,E2; statement S,S1,S2; @@ I(x,...) { <... - (x != NULL) && E ...> } // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: David Teigland <teigland@redhat.com>
2010-03-30include cleanup: Update gfp.h and slab.h includes to prepare for breaking ↵Tejun Heo
implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2009-11-30dlm: always use GFP_NOFSDavid Teigland
Replace all GFP_KERNEL and ls_allocation with GFP_NOFS. ls_allocation would be GFP_KERNEL for userland lockspaces and GFP_NOFS for file system lockspaces. It was discovered that any lockspaces on the system can affect all others by triggering memory reclaim in the file system which could in turn call back into the dlm to acquire locks, deadlocking dlm threads that were shared by all lockspaces, like dlm_recv. Signed-off-by: David Teigland <teigland@redhat.com>
2009-09-30dlm: fix socket fd translationDavid Teigland
The code to set up sctp sockets was not using the sockfd_lookup() and sockfd_put() routines to translate an fd to a socket. The direct fget and fput calls were resulting in error messages from alloc_fd(). Also clean up two log messages and remove a third, related to setting up sctp associations. Signed-off-by: David Teigland <teigland@redhat.com>
2009-09-30dlm: fix lowcomms_connect_node for sctpDavid Teigland
The recently added dlm_lowcomms_connect_node() from 391fbdc5d527149578490db2f1619951d91f3561 does not work when using SCTP instead of TCP. The sctp connection code has nothing to do without data to send. Check for no data in the sctp connection code and do nothing instead of triggering a BUG. Also have connect_node() do nothing when the protocol is sctp. Signed-off-by: David Teigland <teigland@redhat.com>
2009-08-24dlm: use kernel_sendpagePaolo Bonzini
Using kernel_sendpage() is cleaner and safer than following sock->ops ourselves. Signed-off-by: Paolo Bonzini <bonzini@gnu.org> Signed-off-by: David Teigland <teigland@redhat.com>
2009-08-24dlm: fix connection close handlingLars Marowsky-Bree
Closing a connection to a node can create problems if there are outstanding messages for that node. The problems include dlm_send spinning attempting to reconnect, or BUG from tcp_connect_to_sock() attempting to use a partially closed connection. To cleanly close a connection, we now first attempt to send any pending messages, cancel any remaining workqueue work, and flag the connection as closed to avoid reconnect attempts. Signed-off-by: Lars Marowsky-Bree <lmb@suse.de> Signed-off-by: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-08-18dlm: fix double-release of socket in error exit pathCasey Dahlin
The last correction to the tcp_connect_to_sock error exit path, commit a89d63a159b1ba5833be2bef00adf8ad8caac8be, can free an already freed socket, due to collision with a previous (incomplete) attempt to fix the same issue, commit 311f6fc77c51926dbdfbeab0a5d88d70f01fa3f4. Signed-off-by: Casey Dahlin <cdahlin@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-07-14dlm: free socket in error exit pathCasey Dahlin
In the tcp_connect_to_sock() error exit path, the socket allocated at the top of the function was not being freed. Signed-off-by: Casey Dahlin <cdahlin@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-05-15dlm: use more NOFS allocationDavid Teigland
Change some GFP_KERNEL allocations to use either GFP_NOFS or ls_allocation (when available) which the fs sets to GFP_NOFS. The point is to prevent allocations from going back into the cluster fs in places where that might lead to deadlock. Signed-off-by: David Teigland <teigland@redhat.com>
2009-05-15dlm: connect to nodes earlierChristine Caulfield
Make network connections to other nodes earlier, in the context of dlm_recoverd. This avoids connecting to nodes from dlm_send where we try to avoid allocations which could possibly deadlock if memory reclaim goes into the cluster fs which may try to do a dlm operation. Signed-off-by: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-03-11dlm: replace idr with hash table for connectionsChristine Caulfield
Integer nodeids can be too large for the idr code; use a hash table instead. Signed-off-by: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-01-28dlm: comment typo fixesJoe Perches
Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David Teigland <teigland@redhat.com>
2009-01-28dlm: use ipv6_addr_copyJoe Perches
Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David Teigland <teigland@redhat.com>
2008-12-23dlm: remove kmap/kunmapSteven Whitehouse
The pages used in lowcomms are not highmem, so kmap is not necessary. Cc: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2008-12-23dlm: fix up memory allocation flagsSteven Whitehouse
Use ls_allocation for memory allocations, which a cluster fs sets to GFP_NOFS. Use GFP_NOFS for allocations when no lockspace struct is available. Taking dlm locks needs to avoid calling back into the cluster fs because write-out can require taking dlm locks. Cc: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2008-07-14dlm: release socket on errorMasatake YAMATO
It seems that `sock' allocated by sock_create_kern in tcp_connect_to_sock() of dlm/fs/lowcomms.c is not released if dlm_nodeid_to_addr an error. Acked-by: Christine Caulfield <ccaulfie@redhat.com> Signed-off-by: Masatake YAMATO <yamato@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2008-05-19dlm: tcp_connect_to_sock should check for -EINVAL, not EINVALMarcin Slusarz
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> Cc: Christine Caulfield <ccaulfie@redhat.com> Cc: David Teigland <teigland@redhat.com> Cc: cluster-devel@redhat.com Signed-off-by: David Teigland <teigland@redhat.com>
2008-05-19dlm: convert connections_lock in a mutexMatthias Kaehlcke
The semaphore connections_lock is used as a mutex. Convert it to the mutex API. Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> Cc: Christine Caulfield <ccaulfie@redhat.com> Cc: David Teigland <teigland@redhat.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David Teigland <teigland@redhat.com>
2008-01-29dlm: close otherconsPatrick Caulfeld
This patch addresses a problem introduced with the last round of lowcomms patches where the 'othercon' connections do not get freed when the DLM shuts down. This results in the error message "slab error in kmem_cache_destroy(): cache `dlm_conn': Can't free all objects" and the DLM cannot be restarted without a system reboot. See bz#428119 Signed-off-by: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Fabio M. Di Nitto <fabbione@ubuntu.com> Signed-off-by: David Teigland <teigland@redhat.com>
2008-01-29dlm: bind connections from known local address when using TCPLon Hohberger
A common problem occurs when multiple IP addresses within the same subnet are assigned to the same NIC. If we make a connection attempt to another address on the same subnet as one of those addresses, the connection attempt will not necessarily be routed from the address we want. In the case of the DLM, the other nodes will quickly drop the connection attempt, causing problems. This patch makes the DLM bind to the local address it acquired from the cluster manager when using TCP prior to making a connection, obviating the need for administrators to "fix" their systems or use clever routing tricks. Signed-off-by: Lon Hohberger <lhh@redhat.com> Signed-off-by: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
2007-11-07[DLM] lowcomms: Do not muck with sysctl_rmem_max.David S. Miller
Use SO_RCVBUFFORCE instead. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-10[DLM] Make dlm_sendd cond_resched morePatrick Caulfield
Under high recovery loads dlm_sendd can monopolise the CPU and cause soft lockups. This one extra and one moved cond_resched() make it yield a little more during such times keeping work moving. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-10-10[DLM] Fix lowcomms socket closingPatrick Caulfield
This patch fixes the slight mess made in lowcomms closing by previous patches and fixes all sorts of DLM hangs. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-08-14[DLM] More othercon fixesPatrick Caulfield
The last patch to clean out 'othercon' structures only fixed half the problem. The attached addresses the other situations too, and fixes bz#238490 Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-08-14[DLM] zero unused parts of sockaddr_storagePatrick Caulfield
When we build a sockaddr_storage for an IP address, clear the unused parts as they could be used for node comparisons. I have seen this occasionally make sctp connections fail. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-08-14[DLM] Clear othercon pointers when a connection is closedPatrick Caulfield
This patch clears the othercon pointer and frees the memory when a connnection is closed. This could cause a small memory leak when nodes leave the cluster. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-07-20mm: Remove slab destructors from kmem_cache_create().Paul Mundt
Slab destructors were no longer supported after Christoph's c59def9f222d44bb7e2f0a559f2906191a0862d7 change. They've been BUGs for both slab and slub, and slob never supported them either. This rips out support for the dtor pointer from kmem_cache_create() completely and fixes up every single callsite in the kernel (there were about 224, not including the slab allocator definitions themselves, or the documentation references). Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2007-07-09[GFS2] git-gfs2-nmw-build-fixakpm@linux-foundation.org
Cc: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-07-09[DLM] Telnet to port 21064 can stop all lockspacesPatrick Caulfield
This patch fixes Red Hat bz#245892 Opening a tcp connection from a cluster member to another cluster member targeting the dlm port it is enough to stop every dlm operation in the cluster. This means that GFS and rgmanager will hang. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-07-09[DLM] fix socket shutdownPatrick Caulfield
This patch clears the user_data of active sockets as part of cleanup. This prevents any late-arriving data from trying to add jobs to the work queue while we are tidying up. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-Off-By: David Teigland <teigland@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-01[DLM] lowcomms styleDavid Teigland
Replace some printk with log_print, and fix some simple cases of lines over 80. Also, return -ENOTCONN if lowcomms_start fails due to no local IP address being available. Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-01[DLM] Lowcomms nodeid range & initialisation fixesPatrick Caulfield
Fix a few range & initialization bugs in lowcomms. - max_nodeid is really the highest nodeid encountered, so all loops must include it in their iterations. - clean dlm_local_count & connection_idr so we can do a clean restart. - Remove a spurious BUG_ON Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-01[DLM] Fix dlm_lowcoms_stop hangJosef Bacik
When you attempt to release a lockspace in DLM, it will hang trying to down a semaphore that has already been downed. The attached patch fixes the problem. Signed-off-by: Josef Bacik <jwhiter@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com> Cc: Patrick Caulfield <pcaulfie@redhat.com>
2007-05-01[DLM] Consolidate transport protocolsPatrick Caulfield
This patch consolidates the TCP & SCTP protocols for the DLM into a single file and makes it switchable at run-time (well, at least before the DLM actually starts up!) For RHEL5 this patch requires Neil Horman's patch that expands the in-kernel socket API but that has already been twice ACKed so it should be OK. The patch adds a new lowcomms.c file that replaces the existing lowcomms-sctp.c & lowcomms-tcp.c files. Signed-off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-11-30[DLM] Add support for tcp communicationsPatrick Caulfield
The following patch adds a TCP based communications layer to the DLM which is compile time selectable. The existing SCTP layer gives the advantage of allowing multihoming, whereas the TCP layer has been heavily tested in previous versions of the DLM and is known to be robust and therefore can be used as a baseline for performance testing. Signed-off-by: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-10-20[DLM] fix iovec length in recvmsgPatrick Caulfield
I didn't spot that the msg_iovlen was set to 2 if there were two elements in the iovec but left at zero if not :( I think this might be why bob was still seeing trouble. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-10-12[DLM] fix iovec length in recvmsgPatrick Caulfield
The DLM always passes the iovec length as 1, this is wrong when the circular buffer wraps round. Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-10-09[PATCH] dlm gfp_t annotationsAl Viro
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-08-11[DLM] move kmap to after spin_unlockDavid Teigland
Doing the kmap() while holding the spinlock was causing recursive spinlock problems. It seems the kmap was scheduling, although there was no warning as I'd expect. Patrick, do we need locking around the kmap? Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-06-19[DLM] init rwsem earlierDavid Teigland
The nodeinfo_lock rwsem needs to be initialized when the module is loaded instead of when the dlm is first used. Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>