)]}' { "log": [ { "commit": "405f55712dfe464b3240d7816cc4fe4174831be2", "tree": "96c425ea7fa8b31058b8f83a433c5e5265c8ebc7", "parents": [ "f9fabcb58a6d26d6efde842d1703ac7cfa9427b6" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Sat Jul 11 22:08:37 2009 +0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 12 12:22:34 2009 -0700" }, "message": "headers: smp_lock.h redux\n\n* Remove smp_lock.h from files which don\u0027t need it (including some headers!)\n* Add smp_lock.h to files which do need it\n* Make smp_lock.h include conditional in hardirq.h\n It\u0027s needed only for one kernel_locked() usage which is under CONFIG_PREEMPT\n\n This will make hardirq.h inclusion cheaper for every PREEMPT\u003dn config\n (which includes allmodconfig/allyesconfig, BTW)\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7e0338c0de18c50f09aea1fbef45110cf7d64a3c", "tree": "30a935c1f6eee7125a9fbb802a33292b1f7268fa", "parents": [ "df36b439c5fedefe013d4449cb6a50d15e2f4d70", "47fcb03fefee2501e79176932a4184fc24d6f8ec" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jun 22 12:55:50 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jun 22 12:55:50 2009 -0700" }, "message": "Merge branch \u0027for-2.6.31\u0027 of git://fieldses.org/git/linux-nfsd\n\n* \u0027for-2.6.31\u0027 of git://fieldses.org/git/linux-nfsd: (60 commits)\n SUNRPC: Fix the TCP server\u0027s send buffer accounting\n nfsd41: Backchannel: minorversion support for the back channel\n nfsd41: Backchannel: cleanup nfs4.0 callback encode routines\n nfsd41: Remove ip address collision detection case\n nfsd: optimise the starting of zero threads when none are running.\n nfsd: don\u0027t take nfsd_mutex twice when setting number of threads.\n nfsd41: sanity check client drc maxreqs\n nfsd41: move channel attributes from nfsd4_session to a nfsd4_channel_attr struct\n NFS: kill off complicated macro \u0027PROC\u0027\n sunrpc: potential memory leak in function rdma_read_xdr\n nfsd: minor nfsd_vfs_write cleanup\n nfsd: Pull write-gathering code out of nfsd_vfs_write\n nfsd: track last inode only in use_wgather case\n sunrpc: align cache_clean work\u0027s timer\n nfsd: Use write gathering only with NFSv2\n NFSv4: kill off complicated macro \u0027PROC\u0027\n NFSv4: do exact check about attribute specified\n knfsd: remove unreported filehandle stats counters\n knfsd: fix reply cache memory corruption\n knfsd: reply cache cleanups\n ...\n" }, { "commit": "dd2b63d049480979016b959abc2d141cdddb1389", "tree": "3b1f9a59b5c0a7c6027ec102a034a4236b41ebd6", "parents": [ "281fe15dc1d6ad46992f18b7a6644269ec5f7138" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:28 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:40 2009 -0700" }, "message": "nfs41: Rename rq_received to rq_reply_bytes_recvd\n\nThe \u0027rq_received\u0027 member of \u0027struct rpc_rqst\u0027 is used to track when we\nhave received a reply to our request. With v4.1, the backchannel\ncan now accept callback requests over the existing connection. Rename\nthis field to make it clear that it is only used for tracking reply bytes\nand not all bytes received on the connection.\n\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "9c9f3f5fa62cc4959e4d4d1cf1ec74f2d6ac1197", "tree": "a8fb954b9e707b4afdd822e503fc4857f4d88d5d", "parents": [ "7652e5a09ba319241607b22d9055ce93fd5b8039" ], "author": { "name": "Andy Adamson", "email": "andros@netapp.com", "time": "Wed Apr 01 09:23:10 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:31 2009 -0700" }, "message": "nfs41: sunrpc: add a struct svc_xprt pointer to struct svc_serv for backchannel use\n\nThis svc_xprt is passed on to the callback service thread to be later used\nto processes incoming svc_rqst\u0027s\n\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "7652e5a09ba319241607b22d9055ce93fd5b8039", "tree": "9d1d491af237dffb1593ed8f2351ebda36b32972", "parents": [ "a43cde94feded0f65fce36330614691c650ae8fe" ], "author": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Apr 01 09:23:09 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:30 2009 -0700" }, "message": "nfs41: sunrpc: provide functions to create and destroy a svc_xprt for backchannel use\n\nFor nfs41 callbacks we need an svc_xprt to process requests coming up the\nbackchannel socket as rpc_rqst\u0027s that are transformed into svc_rqst\u0027s that\nneed a rq_xprt to be processed.\n\nThe svc_{udp,tcp}_create methods are too heavy for this job as svc_create_socket\ncreates an actual socket to listen on while for nfs41 we\u0027re \"reusing\" the\nfore channel\u0027s socket.\n\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "4d6bbb6233c9cf23822a2f66f8470c9f40854b77", "tree": "0779a7173d2d33adc83162596e64f5636e9f7d5c", "parents": [ "1cad7ea6fe98dc414bd3df55275c147bd15ebf97" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:07 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:29 2009 -0700" }, "message": "nfs41: Backchannel bc_svc_process()\n\nImplement the NFSv4.1 backchannel service. Invokes the common callback\nprocessing logic svc_process_common() to authenticate the call and\ndispatch the appropriate NFSv4.1 XDR decoder and operation procedure.\nIt then invokes bc_send() to send the reply over the same connection.\nbc_send() is implemented in a separate patch.\n\nAt this time there is no slot validation or reply cache handling.\n\n[nfs41: Preallocate rpc_rqst receive buffer for handling callbacks]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[Move bc_svc_process() declaration to correct patch]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "0d90ba1cd416525c4825c111db862d8b15a02e9b", "tree": "1f1571410b2d1b17eb8e503fc3b60450f93e9dee", "parents": [ "55ae1aabfb108106dd095de2578ceef1c755a8b8" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:04 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:28 2009 -0700" }, "message": "nfs41: Backchannel callback service helper routines\n\nExecutes the backchannel task on the RPC state machine using\nthe existing open connection previously established by the client.\n\nSigned-off-by: Ricardo Labiaga \u003cricardo.labiaga@netapp.com\u003e\n\nnfs41: Add bc_svc.o to sunrpc Makefile.\n\n[nfs41: bc_send() does not need to be exported outside RPC module]\n[nfs41: xprt_free_bc_request() need not be exported outside RPC module]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[Update copyright]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "55ae1aabfb108106dd095de2578ceef1c755a8b8", "tree": "7ccb91f5aac0a3646f7d4e8fbe913955aa300f95", "parents": [ "44b98efdd0a205bdca2cb63493350d06ff6804b1" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:03 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 14:11:24 2009 -0700" }, "message": "nfs41: Add backchannel processing support to RPC state machine\n\nAdds rpc_run_bc_task() which is called by the NFS callback service to\nprocess backchannel requests. It performs similar work to rpc_run_task()\nthough \"schedules\" the backchannel task to be executed starting at the\ncall_trasmit state in the RPC state machine.\n\nIt also introduces some miscellaneous updates to the argument validation,\ncall_transmit, and transport cleanup functions to take into account\nthat there are now forechannel and backchannel tasks.\n\nBackchannel requests do not carry an RPC message structure, since the\npayload has already been XDR encoded using the existing NFSv4 callback\nmechanism.\n\nIntroduce a new transmit state for the client to reply on to backchannel\nrequests. This new state simply reserves the transport and issues the\nreply. In case of a connection related error, disconnects the transport and\ndrops the reply. It requires the forechannel to re-establish the connection\nand the server to retransmit the request, as stated in NFSv4.1 section\n2.9.2 \"Client and Server Transport Behavior\".\n\nNote: There is no need to loop attempting to reserve the transport. If EAGAIN\nis returned by xprt_prepare_transmit(), return with tk_status \u003d\u003d 0,\nsetting tk_action to call_bc_transmit. rpc_execute() will invoke it again\nafter the task is taken off the sleep queue.\n\n[nfs41: rpc_run_bc_task() need not be exported outside RPC module]\n[nfs41: New call_bc_transmit RPC state]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[nfs41: Backchannel: No need to loop in call_bc_transmit()]\nSigned-off-by: Andy Adamson \u003candros@netapp.com\u003e\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[rpc_count_iostats incorrectly exits early]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[Convert rpc_reply_expected() to inline function]\n[Remove unnecessary BUG_ON()]\n[Rename variable]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "4a8d70bfef01f8e6b27785e2625e88e9a80924a5", "tree": "15ed5ffd9210cf10d62abb9c3b78648c3fb50675", "parents": [ "fb7a0b9addbdbbb13b7bc02abf55ee524ea19ce1" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:01 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 13:06:15 2009 -0700" }, "message": "nfs41: New include/linux/sunrpc/bc_xprt.h\n\nContains prototype for backchannel helper routines.\n\nSigned-off-by: Ricardo Labiaga \u003cricardo.labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[nfs41: xprt_setup_backchannel v4.0 only inline]\n Fix compile error when CONFIG_NFS_V4_1 is not set.\nSigned-off-by: Andy Adamson \u003candros@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[Update Copyright notice and fix formatting]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "fb7a0b9addbdbbb13b7bc02abf55ee524ea19ce1", "tree": "58f623e78f74c590e20b48f95897c5857d39728f", "parents": [ "f9acac1a4710ce88871f1ae323fc91c1cb6e9d52" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:23:00 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 13:06:14 2009 -0700" }, "message": "nfs41: New backchannel helper routines\n\nThis patch introduces support to setup the callback xprt on the client side.\nIt allocates/ destroys the preallocated memory structures used to process\nbackchannel requests.\n\nAt setup time, xprt_setup_backchannel() is invoked to allocate one or\nmore rpc_rqst structures and substructures. This ensures that they\nare available when an RPC callback arrives. The rpc_rqst structures\nare maintained in a linked list attached to the rpc_xprt structure.\nWe keep track of the number of allocations so that they can be correctly\nremoved when the channel is destroyed.\n\nWhen an RPC callback arrives, xprt_alloc_bc_request() is invoked to\nobtain a preallocated rpc_rqst structure. An rpc_xprt structure is\nreturned, and its RPC_BC_PREALLOC_IN_USE bit is set in\nrpc_xprt-\u003ebc_flags. The structure is removed from the the list\nsince it is now in use, and it will be later added back when its\nuser is done with it.\n\nAfter the RPC callback replies, the rpc_rqst structure is returned\nby invoking xprt_free_bc_request(). This clears the\nRPC_BC_PREALLOC_IN_USE bit and adds it back to the list, allowing it\nto be reused by a subsequent RPC callback request.\n\nTo be consistent with the reception of RPC messages, the backchannel requests\nshould be placed into the \u0027struct rpc_rqst\u0027 rq_rcv_buf, which is then in turn\ncopied to the \u0027struct rpc_rqst\u0027 rq_private_buf.\n\n[nfs41: Preallocate rpc_rqst receive buffer for handling callbacks]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[Update copyright notice and explain page allocation]\nSigned-off-by: Ricardo Labiaga \u003cRicardo.Labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "56632b5bff5af10eb12d7e9499b5ffcadcb7a7b2", "tree": "41632f7d033a791e1925a84f14488efcb9c0dd83", "parents": [ "7146851376861fe55c7a48ac8fc1354a5fff6cd0" ], "author": { "name": "Ricardo Labiaga", "email": "Ricardo.Labiaga@netapp.com", "time": "Wed Apr 01 09:22:58 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 13:06:13 2009 -0700" }, "message": "nfs41: client callback structures\n\nAdds new list of rpc_xprt structures, and a readers/writers lock to\nprotect the list. The list is used to preallocate resources for\nthe backchannel during backchannel requests. Callbacks are not\nexpected to cause significant latency, so only one callback will\nbe allowed at this time.\n\nIt also adds a pointer to the NFS callback service so that\nrequests can be directed to it for processing.\n\nNew callback members added to svc_serv. The NFSv4.1 callback service will\nsleep on the svc_serv-\u003esvc_cb_waitq until new callback requests arrive.\nThe request will be queued in svc_serv-\u003esvc_cb_list. This patch adds this\nlist, the sleep queue and spinlock to svc_serv.\n\n[nfs41: NFSv4.1 callback support]\nSigned-off-by: Ricardo Labiaga \u003cricardo.labiaga@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n" }, { "commit": "aae2006e9b0c294114915c13022fa348e1a88023", "tree": "2643d2ed1b2d44c5e5d7577e1ae0516eae95f977", "parents": [ "c3fad1b1aaf850bf692642642ace7cd0d64af0a3" ], "author": { "name": "Andy Adamson", "email": "andros@netapp.com", "time": "Wed Apr 01 09:22:40 2009 -0400" }, "committer": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Wed Jun 17 12:25:07 2009 -0700" }, "message": "nfs41: sunrpc: Export the call prepare state for session reset\n\nSigned-off-by: Andy Adamson\u003candros@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "7eef4091a653c243a87e5375c54504cc03bec4d8", "tree": "f65b77f830b2c8f7d014512badfef5df0d591ee9", "parents": [ "0a93a47f042c459f0f46942c3a920e3c81878031", "07a2039b8eb0af4ff464efd3dfd95de5c02648c6" ], "author": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Jun 15 18:08:07 2009 -0700" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Jun 15 18:08:07 2009 -0700" }, "message": "Merge commit \u0027v2.6.30\u0027 into for-2.6.31\n" }, { "commit": "f75e6745aa3084124ae1434fd7629853bdaf6798", "tree": "bfd605c079e8a1dd6ad008acc6031441c220c845", "parents": [ "414772fa496273d1a93cefa6dab790f5fdf9de82" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Apr 21 17:18:20 2009 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 02 16:35:08 2009 -0700" }, "message": "SUNRPC: Fix the problem of EADDRNOTAVAIL syslog floods on reconnect\n\nSee http://bugzilla.kernel.org/show_bug.cgi?id\u003d13034\n\nIf the port gets into a TIME_WAIT state, then we cannot reconnect without\nbinding to a new port.\n\nTested-by: Petr Vandrovec \u003cpetr@vandrovec.name\u003e\nTested-by: Jean Delvare \u003ckhali@linux-fr.org\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8435d34dbbe75678c3cdad3d53b1e7996a79b3bf", "tree": "efce3e51cb67c1d8ce39cd87bc22fbac8ee32cbe", "parents": [ "bfba9ab4c64f0e5c33930711e6c073c285e01fcf" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Thu Apr 23 19:32:40 2009 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Tue Apr 28 13:54:28 2009 -0400" }, "message": "SUNRPC: pass buffer size to svc_sock_names()\n\nAdjust the synopsis of svc_sock_names() to pass in the size of the\noutput buffer. Add a documenting comment.\n\nThis is a cosmetic change for now. A subsequent patch will make sure\nthe buffer length is passed to one_sock_name(), where the length will\nactually be useful.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "bfba9ab4c64f0e5c33930711e6c073c285e01fcf", "tree": "1342b7e7d682fd468f0eb271faf2c15afdc8bd50", "parents": [ "335c54bdc4d3bacdbd619ec95cd0b352435bd37f" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Thu Apr 23 19:32:33 2009 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Tue Apr 28 13:54:28 2009 -0400" }, "message": "SUNRPC: pass buffer size to svc_addsock()\n\nAdjust the synopsis of svc_addsock() to pass in the size of the output\nbuffer. Add a documenting comment.\n\nThis is a cosmetic change for now. A subsequent patch will make sure\nthe buffer length is passed to one_sock_name(), where the length will\nactually be useful.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "335c54bdc4d3bacdbd619ec95cd0b352435bd37f", "tree": "922de1595031ab24a31ba263c1f07bf876797b05", "parents": [ "ea068bad27cefc71ab03230dbf01a8f8d98da5ba" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Thu Apr 23 19:32:25 2009 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Tue Apr 28 13:54:28 2009 -0400" }, "message": "NFSD: Prevent a buffer overflow in svc_xprt_names()\n\nThe svc_xprt_names() function can overflow its buffer if it\u0027s so near\nthe end of the passed in buffer that the \"name too long\" string still\ndoesn\u0027t fit. Of course, it could never tell if it was near the end\nof the passed in buffer, since its only caller passes in zero as the\nbuffer length.\n\nLet\u0027s make this API a little safer.\n\nChange svc_xprt_names() so it *always* checks for a buffer overflow,\nand change its only caller to pass in the correct buffer length.\n\nIf svc_xprt_names() does overflow its buffer, it now fails with an\nENAMETOOLONG errno, instead of trying to write a message at the end\nof the buffer. I don\u0027t like this much, but I can\u0027t figure out a clean\nway that\u0027s always safe to return some of the names, *and* an\nindication that the buffer was not long enough.\n\nThe displayed error when doing a \u0027cat /proc/fs/nfsd/portlist\u0027 is\n\"File name too long\".\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "abc5c44d6284fab8fb21bcfc52c0f16f980637df", "tree": "98825098d3f6488a1a8fb55ec40131313b1d25e9", "parents": [ "dcf1a3573eae69937fb14462369c4d3e6f4a37f1" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Thu Apr 23 19:31:25 2009 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Tue Apr 28 13:54:25 2009 -0400" }, "message": "SUNRPC: Fix error return value of svc_addr_len()\n\nThe svc_addr_len() helper function returns -EAFNOSUPPORT if it doesn\u0027t\nrecognize the address family of the passed-in socket address. However,\nthe return type of this function is size_t, which means -EAFNOSUPPORT\nis turned into a very large positive value in this case.\n\nThe check in svc_udp_recvfrom() to see if the return value is less\nthan zero therefore won\u0027t work at all.\n\nAdditionally, handle_connect_req() passes this value directly to\nmemset(). This could cause memset() to clobber a large chunk of memory\nif svc_addr_len() has returned an error. Currently the address family\nof these addresses, however, is known to be supported long before\nhandle_connect_req() is called, so this isn\u0027t a real risk.\n\nChange the error return value of svc_addr_len() to zero, which fits in\nthe range of size_t, and is safer to pass to memset() directly.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "a63856252d2112e7c452696037a86ceb12f47f80", "tree": "b1ad03fe441349069f80e58de425b3f72af9e5b7", "parents": [ "b24241a09208874d5d770bee30791daae41ad762", "04826f43d4f0a4d56423eb8abb9f2ec9987df5b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 13:25:56 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 06 13:25:56 2009 -0700" }, "message": "Merge branch \u0027for-2.6.30\u0027 of git://linux-nfs.org/~bfields/linux\n\n* \u0027for-2.6.30\u0027 of git://linux-nfs.org/~bfields/linux: (81 commits)\n nfsd41: define nfsd4_set_statp as noop for !CONFIG_NFSD_V4\n nfsd41: define NFSD_DRC_SIZE_SHIFT in set_max_drc\n nfsd41: Documentation/filesystems/nfs41-server.txt\n nfsd41: CREATE_EXCLUSIVE4_1\n nfsd41: SUPPATTR_EXCLCREAT attribute\n nfsd41: support for 3-word long attribute bitmask\n nfsd: dynamically skip encoded fattr bitmap in _nfsd4_verify\n nfsd41: pass writable attrs mask to nfsd4_decode_fattr\n nfsd41: provide support for minor version 1 at rpc level\n nfsd41: control nfsv4.1 svc via /proc/fs/nfsd/versions\n nfsd41: add OPEN4_SHARE_ACCESS_WANT nfs4_stateid bmap\n nfsd41: access_valid\n nfsd41: clientid handling\n nfsd41: check encode size for sessions maxresponse cached\n nfsd41: stateid handling\n nfsd: pass nfsd4_compound_state* to nfs4_preprocess_{state,seq}id_op\n nfsd41: destroy_session operation\n nfsd41: non-page DRC for solo sequence responses\n nfsd41: Add a create session replay cache\n nfsd41: create_session operation\n ...\n" }, { "commit": "c3d06f9ce8544fecfe13e377d1e2c2e47fe18dbc", "tree": "f34acde5f484a8374065295e205921e1860da65c", "parents": [ "074fe897536f095309c5aaffcf912952882ab2cb" ], "author": { "name": "Andy Adamson", "email": "andros@netapp.com", "time": "Fri Apr 03 08:28:18 2009 +0300" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Fri Apr 03 17:41:17 2009 -0700" }, "message": "nfsd41: hard page limit for DRC\n\nUse no more than 1/128th of the number of free pages at nfsd startup for the\nv4.1 DRC.\n\nThis is an arbitrary default which should probably end up under the control\nof an administrator.\n\nSigned-off-by: Andy Adamson \u003candros@netapp.com\u003e\n[moved added fields in struct svc_serv under CONFIG_NFSD_V4_1]\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[fix set_max_drc calculation of sv_drc_max_pages]\n[moved NFSD_DRC_SIZE_SHIFT\u0027s declaration up in header file]\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "2f425878b6a71571341dcd3f9e9d1a6f6355da9c", "tree": "52735958c1614458bd6b459040f977dfc3c3e9a9", "parents": [ "20766016329eb4985c2c8b2a1b2333e0f865fdf9" ], "author": { "name": "Andy Adamson", "email": "andros@netapp.com", "time": "Fri Apr 03 08:27:32 2009 +0300" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Fri Apr 03 17:41:12 2009 -0700" }, "message": "nfsd: don\u0027t use the deferral service, return NFS4ERR_DELAY\n\nOn an NFSv4.1 server cache miss that causes an upcall, NFS4ERR_DELAY will be\nreturned. It is up to the NFSv4.1 client to resend only the operations that\nhave not been processed.\n\nInitialize rq_usedeferral to 1 in svc_process(). It sill be turned off in\nnfsd4_proc_compound() only when NFSv4.1 Sessions are used.\n\nNote: this isn\u0027t an adequate solution on its own. It\u0027s acceptable as a way\nto get some minimal 4.1 up and working, but we\u0027re going to have to find a\nway to avoid returning DELAY in all common cases before 4.1 can really be\nconsidered ready.\n\nSigned-off-by: Andy Adamson \u003candros@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[nfsd41: reverse rq_nodeferral negative logic]\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\n[sunrpc: initialize rq_usedeferral]\nSigned-off-by: Andy Adamson \u003candros@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "49a9072f29a1039f142ec98b44a72d7173651c02", "tree": "d2a87541cb699fb6b1fbdef492f3dc6c305aa541", "parents": [ "9652ada3fb5914a67d8422114e8a76388330fa79" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Mar 18 20:46:29 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Mar 28 15:54:48 2009 -0400" }, "message": "SUNRPC: Remove @family argument from svc_create() and svc_create_pooled()\n\nSince an RPC service listener\u0027s protocol family is specified now via\nsvc_create_xprt(), it no longer needs to be passed to svc_create() or\nsvc_create_pooled(). Remove that argument from the synopsis of those\nfunctions, and remove the sv_family field from the svc_serv struct.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "9652ada3fb5914a67d8422114e8a76388330fa79", "tree": "b37d77bf972dda04b46794c4e50ec6f6d044dc79", "parents": [ "baf01caf09e87579c2d157e5ee29975db8551522" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Mar 18 20:46:21 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Mar 28 15:54:36 2009 -0400" }, "message": "SUNRPC: Change svc_create_xprt() to take a @family argument\n\nThe sv_family field is going away. Pass a protocol family argument to\nsvc_create_xprt() instead of extracting the family from the passed-in\nsvc_serv struct.\n\nAgain, as this is a listener socket and not an address, we make this\nnew argument an \"int\" protocol family, instead of an \"sa_family_t.\"\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "4b62e58cccff9c5e7ffc7023f7ec24c75fbd549b", "tree": "34965810fe2a9aff001dd193a3cb925c2c3e1abd", "parents": [ "156e62094a74cf43f02f56ef96b6cda567501357" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Mar 18 20:46:06 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Mar 28 15:54:12 2009 -0400" }, "message": "SUNRPC: Pass a family argument to svc_register()\n\nThe sv_family field is going away. Instead of using sv_family, have\nthe svc_register() function take a protocol family argument.\n\nSince this argument represents a protocol family, and not an address\nfamily, this argument takes an int, as this is what is passed to\nsock_create_kern(). Also make sure svc_register\u0027s helpers are\nchecking for PF_FOO instead of AF_FOO. The value of [AP]F_FOO are\nequivalent; this is simply a symbolic change to reflect the semantics\nof the value stored in that variable.\n\nsock_create_kern() should return EPFNOSUPPORT if the passed-in\nprotocol family isn\u0027t supported, but it uses EAFNOSUPPORT for this\ncase. We will stick with that tradition here, as svc_register()\nis called by the RPC server in the same path as sock_create_kern().\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "156e62094a74cf43f02f56ef96b6cda567501357", "tree": "100fdf25a7f13f8c2ebebd241d02a9ed46e89292", "parents": [ "adbbe929569e6eec8ff9feca23f1f2b40b42853d" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Mar 18 20:45:58 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Mar 28 15:53:57 2009 -0400" }, "message": "SUNRPC: Clean up svc_find_xprt() calling sequence\n\nClean up: add documentating comment and use appropriate data types for\nsvc_find_xprt()\u0027s arguments.\n\nThis also eliminates a mixed sign comparison: @port was an int, while\nthe return value of svc_xprt_local_port() is an unsigned short.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "efb3288b423d7e3533a68dccecaa05a56a281a4e", "tree": "083d0f507c5e3552cf0ab5ae1efadbe7e2bf2d8f", "parents": [ "776bd5c7a207de546918f805090bfc823d2660c8" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Mar 18 20:45:43 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Mar 28 15:53:22 2009 -0400" }, "message": "SUNRPC: Clean up static inline functions in svc_xprt.h\n\nClean up: Enable the use of const arguments in higher level svc_ APIs\nby adding const to the arguments of the helper functions in svc_xprt.h\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "7d1e8255cf959fba7ee2317550dfde39f0b936ae", "tree": "0a046b83e5ac29501849c34408059480d4c210be", "parents": [ "5e3771ce2d6a69e10fcc870cdf226d121d868491" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 11 14:38:03 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Mar 19 15:17:34 2009 -0400" }, "message": "SUNRPC: Add the equivalent of the linger and linger2 timeouts to RPC sockets\n\nThis fixes a regression against FreeBSD servers as reported by Tomas\nKasparek. Apparently when using RPC over a TCP socket, the FreeBSD servers\ndon\u0027t ever react to the client closing the socket, and so commit\ne06799f958bf7f9f8fae15f0c6f519953fb0257c (SUNRPC: Use shutdown() instead of\nclose() when disconnecting a TCP socket) causes the setup to hang forever\nwhenever the client attempts to close and then reconnect.\n\nWe break the deadlock by adding a \u0027linger2\u0027 style timeout to the socket,\nafter which, the client will abort the connection using a TCP \u0027RST\u0027.\n\nThe default timeout is set to 15 seconds. A subsequent patch will put it\nunder user control by means of a systctl.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "2795e53b4ed5d1f49d2283f416c922f55ec7d461", "tree": "cba9e5c2bcbbf7da75b4c73c7398b20a770262d2", "parents": [ "0953e620de0538cbd081f1b45126f6098112a598" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Thu Mar 12 12:07:14 2009 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Mar 18 18:13:28 2009 -0400" }, "message": "SUNRPC: Clean up static inline functions in svc_xprt.h\n\nClean up: Enable the use of const arguments in higher level svc_ APIs\nby adding const to the arguments of the helper functions in svc_xprt.h\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "03cf6c9f49a8fea953d38648d016e3f46e814991", "tree": "c7e639bd9fb775af1919d2442d0bf2306dfbfca7", "parents": [ "59a252ff8c0f2fa32c896f69d56ae33e641ce7ad" ], "author": { "name": "Greg Banks", "email": "gnb@sgi.com", "time": "Tue Jan 13 21:26:36 2009 +1100" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Mar 18 17:38:42 2009 -0400" }, "message": "knfsd: add file to export stats about nfsd pools\n\nAdd /proc/fs/nfsd/pool_stats to export to userspace various\nstatistics about the operation of rpc server thread pools.\n\nThis patch is based on a forward-ported version of\nknfsd-add-pool-thread-stats which has been shipping in the SGI\n\"Enhanced NFS\" product since 2006 and which was previously\nposted:\n\nhttp://article.gmane.org/gmane.linux.nfs/10375\n\nIt has also been updated thus:\n\n * moved EXPORT_SYMBOL() to near the function it exports\n * made the new struct struct seq_operations const\n * used SEQ_START_TOKEN instead of ((void *)1)\n * merged fix from SGI PV 990526 \"sunrpc: use dprintk instead of\n printk in svc_pool_stats_*()\" by Harshula Jayasuriya.\n * merged fix from SGI PV 964001 \"Crash reading pool_stats before\n nfsds are started\".\n\nSigned-off-by: Greg Banks \u003cgnb@sgi.com\u003e\nSigned-off-by: Harshula Jayasuriya \u003charshula@sgi.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "59a252ff8c0f2fa32c896f69d56ae33e641ce7ad", "tree": "f79089d44737e5f050cc4869b42829650096747f", "parents": [ "8bbfa9f3889b643fc7de82c0c761ef17097f8faf" ], "author": { "name": "Greg Banks", "email": "gnb@sgi.com", "time": "Tue Jan 13 21:26:35 2009 +1100" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Mar 18 17:38:41 2009 -0400" }, "message": "knfsd: avoid overloading the CPU scheduler with enormous load averages\n\nAvoid overloading the CPU scheduler with enormous load averages\nwhen handling high call-rate NFS loads. When the knfsd bottom half\nis made aware of an incoming call by the socket layer, it tries to\nchoose an nfsd thread and wake it up. As long as there are idle\nthreads, one will be woken up.\n\nIf there are lot of nfsd threads (a sensible configuration when\nthe server is disk-bound or is running an HSM), there will be many\nmore nfsd threads than CPUs to run them. Under a high call-rate\nlow service-time workload, the result is that almost every nfsd is\nrunnable, but only a handful are actually able to run. This situation\ncauses two significant problems:\n\n1. The CPU scheduler takes over 10% of each CPU, which is robbing\n the nfsd threads of valuable CPU time.\n\n2. At a high enough load, the nfsd threads starve userspace threads\n of CPU time, to the point where daemons like portmap and rpc.mountd\n do not schedule for tens of seconds at a time. Clients attempting\n to mount an NFS filesystem timeout at the very first step (opening\n a TCP connection to portmap) because portmap cannot wake up from\n select() and call accept() in time.\n\nDisclaimer: these effects were observed on a SLES9 kernel, modern\nkernels\u0027 schedulers may behave more gracefully.\n\nThe solution is simple: keep in each svc_pool a counter of the number\nof threads which have been woken but have not yet run, and do not wake\nany more if that count reaches an arbitrary small threshold.\n\nTesting was on a 4 CPU 4 NIC Altix using 4 IRIX clients, each with 16\nsynthetic client threads simulating an rsync (i.e. recursive directory\nlisting) workload reading from an i386 RH9 install image (161480\nregular files in 10841 directories) on the server. That tree is small\nenough to fill in the server\u0027s RAM so no disk traffic was involved.\nThis setup gives a sustained call rate in excess of 60000 calls/sec\nbefore being CPU-bound on the server. The server was running 128 nfsds.\n\nProfiling showed schedule() taking 6.7% of every CPU, and __wake_up()\ntaking 5.2%. This patch drops those contributions to 3.0% and 2.2%.\nLoad average was over 120 before the patch, and 20.9 after.\n\nThis patch is a forward-ported version of knfsd-avoid-nfsd-overload\nwhich has been shipping in the SGI \"Enhanced NFS\" product since 2006.\nIt has been posted before:\n\nhttp://article.gmane.org/gmane.linux.nfs/10374\n\nSigned-off-by: Greg Banks \u003cgnb@sgi.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "77f18f5e4ebdea35ec3d92343b0ed7546dc87637", "tree": "dd9a97818130a8596b1833c82a625b3ed9cf9186", "parents": [ "4ac35c2f794503d3acda20d98e89cf63f6e94332" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Wed Feb 11 17:16:58 2009 -0800" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Mar 18 17:30:51 2009 -0400" }, "message": "nfs: replace uses of __constant_{endian}\n\nThe base versions handle constant folding now, none of these headers\nare exported to userspace, so the __ prefixed versions are not\nnecessary.\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nReviewed-by: NeilBrown \u003cneilb@suse.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "441e3e242903f9b190d5764bed73edb58f977413", "tree": "3898c80feadee3cdeb3c99cb7d62861127f1b4ef", "parents": [ "b38ab40ad58c1fc43ea590d6342f6a6763ac8fb6" ], "author": { "name": "Tom Talpey", "email": "tmtalpey@gmail.com", "time": "Wed Mar 11 14:37:56 2009 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 11 14:37:56 2009 -0400" }, "message": "SUNRPC: dynamically load RPC transport modules on-demand\n\nProvide an api to attempt to load any necessary kernel RPC\nclient transport module automatically. By convention, the\ndesired module name is \"xprt\"+\"transport name\". For example,\nwhen NFS mounting with \"-o proto\u003drdma\", attempt to load the\n\"xprtrdma\" module.\n\nSigned-off-by: Tom Talpey \u003ctmtalpey@gmail.com\u003e\nCc: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "c9233eb7b0b11ef176d4bf68da2ce85464b6ec39", "tree": "22f9f999e0d48a5f3aae6b32f9ff702eb4db05cc", "parents": [ "548eaca46b3cf4419b6c2be839a106d8641ffb70" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Oct 20 11:51:57 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Tue Jan 06 11:53:47 2009 -0500" }, "message": "sunrpc: add sv_maxconn field to svc_serv (try #3)\n\nsvc_check_conn_limits() attempts to prevent denial of service attacks\nby having the service close old connections once it reaches a\nthreshold. This threshold is based on the number of threads in the\nservice:\n\n\t(serv-\u003esv_nrthreads + 3) * 20\n\nOnce we reach this, we drop the oldest connections and a printk pops\nto warn the admin that they should increase the number of threads.\n\nIncreasing the number of threads isn\u0027t an option however for services\nlike lockd. We don\u0027t want to eliminate this check entirely for such\nservices but we need some way to increase this limit.\n\nThis patch adds a sv_maxconn field to the svc_serv struct. When it\u0027s\nset to 0, we use the current method to calculate the max number of\nconnections. RPC services can then set this on an as-needed basis.\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nAcked-by: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "08cc36cbd1ee7d86422713bb21551eed1326b894", "tree": "52cc683387f903b34a7f6f798dcdbae385b58db8", "parents": [ "3c92ec8ae91ecf59d88c798301833d7cf83f2179", "46f72f57d279688c4524df78edb5738db730eeef" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 30 16:51:43 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 30 16:51:43 2008 -0500" }, "message": "Merge branch \u0027devel\u0027 into next\n" }, { "commit": "608207e8884e083ad8b8d33eda868da70f0d63e8", "tree": "86073f5f2e4a550bb1efe395954abe3fa91a555e", "parents": [ "68e76ad0baf8f5d5060377c2423ee6eed5c63057" ], "author": { "name": "Olga Kornievskaia", "email": "aglo@citi.umich.edu", "time": "Tue Dec 23 16:17:40 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 16:17:40 2008 -0500" }, "message": "rpc: pass target name down to rpc level on callbacks\n\nThe rpc client needs to know the principal that the setclientid was done\nas, so it can tell gssd who to authenticate to.\n\nSigned-off-by: Olga Kornievskaia \u003caglo@citi.umich.edu\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "68e76ad0baf8f5d5060377c2423ee6eed5c63057", "tree": "baab3e30cc3b50bc49a553ee884885189ce85a8e", "parents": [ "34769fc488b463cb753fc632f8f5ba56c918b7cb" ], "author": { "name": "Olga Kornievskaia", "email": "aglo@citi.umich.edu", "time": "Tue Dec 23 16:17:15 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 16:17:15 2008 -0500" }, "message": "nfsd: pass client principal name in rsc downcall\n\nTwo principals are involved in krb5 authentication: the target, who we\nauthenticate *to* (normally the name of the server, like\nnfs/server.citi.umich.edu@CITI.UMICH.EDU), and the source, we we\nauthenticate *as* (normally a user, like bfields@UMICH.EDU)\n\nIn the case of NFSv4 callbacks, the target of the callback should be the\nsource of the client\u0027s setclientid call, and the source should be the\nnfs server\u0027s own principal.\n\nTherefore we allow svcgssd to pass down the name of the principal that\njust authenticated, so that on setclientid we can store that principal\nname with the new client, to be used later on callbacks.\n\nSigned-off-by: Olga Kornievskaia \u003caglo@citi.umich.edu\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "c381060869317b3c84430d4f54965d409cbfe65f", "tree": "eca752e4fe61fd6e497fede253f83a01a36a365a", "parents": [ "db75b3d6b5b0dad29860370618ea94d2726641b4" ], "author": { "name": "\\\"J. Bruce Fields\\", "email": "bfields@citi.umich.edu", "time": "Tue Dec 23 16:08:32 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 16:08:32 2008 -0500" }, "message": "rpc: add an rpc_pipe_open method\n\nWe want to transition to a new gssd upcall which is text-based and more\neasily extensible.\n\nTo simplify upgrades, as well as testing and debugging, it will help if\nwe can upgrade gssd (to a version which understands the new upcall)\nwithout having to choose at boot (or module-load) time whether we want\nthe new or the old upcall.\n\nWe will do this by providing two different pipes: one named, as\ncurrently, after the mechanism (normally \"krb5\"), and supporting the\nold upcall. One named \"gssd\" and supporting the new upcall version.\n\nWe allow gssd to indicate which version it supports by its choice of\nwhich pipe to open.\n\nAs we have no interest in supporting *simultaneous* use of both\nversions, we\u0027ll forbid opening both pipes at the same time.\n\nSo, add a new pipe_open callback to the rpc_pipefs api, which the gss\ncode can use to track which pipes have been open, and to refuse opens of\nincompatible pipes.\n\nWe only need this to be called on the first open of a given pipe.\n\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "c977a2ef40a38c45537ad03823d0a004f06373f0", "tree": "66c640c46f6a434d94e1bbde6b867cbd8b73aad4", "parents": [ "027b6ca02192f381a5a91237ba8a8cf625dc6f6a" ], "author": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Tue Dec 23 16:06:13 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 16:06:13 2008 -0500" }, "message": "sunrpc: get rid of rpc_rqst.rq_bufsize\n\nrq_bufsize is not used.\n\nSigned-off-by: Mike Sager \u003cMike.Sager@netapp.com\u003e\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "88a9fe8cae3bb52e82489447f45e8d7ba1409ca8", "tree": "acef9d0b7d4c8c2e98faf852ab8bc0173842fbe5", "parents": [ "136221fc3219b3805c48db5da065e8e3467175d4" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 15:21:31 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Dec 23 15:21:31 2008 -0500" }, "message": "SUNRPC: Remove the last remnant of the BKL...\n\nSomehow, this escaped the previous purge. There should be no need to keep\nany extra locks in the XDR callbacks.\n\nThe NFS client XDR code only writes into private objects, whereas all reads\nof shared objects are confined to fields that do not change, such as\nfilehandles...\n\nDitto for lockd, the NFSv2/v3 client mount code, and rpcbind.\n\nThe nfsd XDR code may require the BKL, but since it does a synchronous RPC\ncall from a thread that already holds the lock, that issue is moot.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "3685f25de1b0447fff381c420de1e25bd57c9efb", "tree": "90a5d87c0ced1b27f654606d50a9ff13ded6f862", "parents": [ "be859405487324ed548f1ba11dc949b8230ab991" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Fri Oct 31 00:56:49 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Oct 31 00:56:49 2008 -0700" }, "message": "misc: replace NIPQUAD()\n\nUsing NIPQUAD() with NIPQUAD_FMT, %d.%d.%d.%d or %u.%u.%u.%u\ncan be replaced with %pI4\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5b095d98928fdb9e3b75be20a54b7a6cbf6ca9ad", "tree": "b6caa0cdbaac016447a790881ad4a6c5dfce6900", "parents": [ "4b7a4274ca63dadd9c4f17fc953f3a5d19855c4c" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Wed Oct 29 12:52:50 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 29 12:52:50 2008 -0700" }, "message": "net: replace %p6 with %pI6\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0c6ce78abf6e228d44c3840edb8a4ae0c1299825", "tree": "55f228de41dd191609310c03633579eec05b1f33", "parents": [ "38ff4fa49bef77e86b21d95d9ce341a098f02839" ], "author": { "name": "Harvey Harrison", "email": "harvey.harrison@gmail.com", "time": "Tue Oct 28 16:09:23 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 28 23:02:31 2008 -0700" }, "message": "net: replace uses of NIP6_FMT with %p6\n\nSigned-off-by: Harvey Harrison \u003charvey.harrison@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6925bac120097b823fc990c143b9789c21cc60b5", "tree": "0f92e1b22122fba623aeff4c271c16df673244eb", "parents": [ "04ab591808565f968d4406f6435090ad671ebdab", "011935a0a710c20bb7ae63523b78856848db1926" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Oct 15 15:54:56 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Oct 15 15:54:56 2008 -0400" }, "message": "Merge branch \u0027next\u0027\n" }, { "commit": "5675add36e76b9487e7f9e689f854cb8d6afd9b4", "tree": "414b5e26b099c6aa42ee989925c0dfb8b717366e", "parents": [ "1a954051b0cf79bd67e5f9db40333e3a9b1d05d2" ], "author": { "name": "Tom Talpey", "email": "talpey@netapp.com", "time": "Thu Oct 09 15:01:41 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Oct 10 15:13:31 2008 -0400" }, "message": "RPC/RDMA: harden connection logic against missing/late rdma_cm upcalls.\n\nAdd defensive timeouts to wait_for_completion() calls in RDMA\naddress resolution, and make them interruptible. Fix the timeout\nunits to milliseconds (formerly jiffies) and move to private header.\n\nSigned-off-by: Tom Talpey \u003ctalpey@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "fe9053b30bb48b99f7b45541249f5cfe96bdf7f7", "tree": "41904e855b60678ae8788e68e8a5944d686b31e3", "parents": [ "8d4ba0347ccfea4f12e56e2484954b891411b74d" ], "author": { "name": "Tom Talpey", "email": "talpey@netapp.com", "time": "Thu Oct 09 14:59:59 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Oct 10 15:09:24 2008 -0400" }, "message": "RPC/RDMA: add data types and new FRMR memory registration enum.\n\nInternal RPC/RDMA structure updates in preparation for FRMR support.\n\nSigned-off-by: Tom Talpey \u003ctalpey@netapp.com\u003e\nAcked-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "107e0008dfb8bd6366bc8827f5bbbc0c1f795d2d", "tree": "12d6be652e3de5cfcfcec5fe0bc37e4b1d983b44", "parents": [ "2937391385807b3da9cd7a39345259caf550b032", "67080c82361b7510b602c87b83399421aa2d2895" ], "author": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Oct 08 18:22:18 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Oct 08 18:22:18 2008 -0400" }, "message": "Merge branch \u0027from-tomtucker\u0027 into for-2.6.28\n" }, { "commit": "146b6df6a537939570c5772ebd7db826fdbd5d82", "tree": "1acf55f3244719e543fb5f6eed6de11c5dd3110e", "parents": [ "5b180a9a64ca2217a658bd515ef910eafefc5e5a" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Tue Aug 12 15:12:10 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon Oct 06 14:46:01 2008 -0500" }, "message": "svcrdma: Modify the RPC recv path to use FRMR when available\n\nRPCRDMA requests that specify a read-list are fetched with RDMA_READ. Using\nan FRMR to map the data sink improves NFSRDMA security on transports that\nplace the RDMA_READ data sink LKEY on the wire because the valid lifetime\nof the MR is only the duration of the RDMA_READ. The LKEY is invalidated\nwhen the last RDMA_READ WR completes.\n\nMapping the data sink also allows for very large amounts to data to be\nfetched with a single WR, so if the client is also using FRMR, the entire\nRPC read-list can be fetched with a single WR.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "e1183210625cc8e02ce13eec78fb7a246567fc59", "tree": "c01cb7326fb64b899cc375bbadcea0e4579c6043", "parents": [ "3a5c63803d0552a3ad93b85c262f12cd86471443" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Fri Oct 03 15:22:18 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon Oct 06 14:45:49 2008 -0500" }, "message": "svcrdma: Add a service to register a Fast Reg MR with the device\n\nFast Reg MR introduces a new WR type. Add a service to register the\nregion with the adapter and update the completion handling to support\ncompletions with a NULL WR context.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "64be8608c163bd480cf5ec4b34366f11e0f3c87f", "tree": "e0a2499259c799d1ac97935107a25c4cefd7cb0c", "parents": [ "0d3ebb9ae9f9c887518fd4c81a68084111d154d7" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon Oct 06 14:45:18 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon Oct 06 14:45:18 2008 -0500" }, "message": "svcrdma: Add FRMR get/put services\n\nAdd services for the allocating, freeing, and unmapping Fast Reg MR. These\nservices will be used by the transport connection setup, send and receive\nroutines.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "2937391385807b3da9cd7a39345259caf550b032", "tree": "6b75733afd1c5eb9a847c72b0c745ffc24aef559", "parents": [ "26a414092353590ceaa5955bcb53f863d6ea7549" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Fri Oct 03 17:15:38 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Sat Oct 04 17:12:27 2008 -0400" }, "message": "NLM: Remove unused argument from svc_addsock() function\n\nClean up: The svc_addsock() function no longer uses its \"proto\"\nargument, so remove it.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nCc: Neil Brown \u003cneilb@suse.de\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "0d3ebb9ae9f9c887518fd4c81a68084111d154d7", "tree": "dd6e68bb5f9f25c1c212fec567ba68e13d214e58", "parents": [ "bc45eb8950b8c14487385cfd2bda1613ca8d9703" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Tue Sep 30 13:06:13 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Fri Oct 03 15:39:58 2008 -0500" }, "message": "svcrdma: Add Fast Reg MR Data Types\n\nAdd data types to track Fast Reg Memory Regions. The core data type is\nsvc_rdma_fastreg_mr that associates a device MR with a host kva and page\nlist. A field is added to the WR context to keep track of the FRMR\nused to map the local memory for an RPC.\n\nAn FRMR list and spin lock are added to the transport instance to keep\ntrack of all FRMR allocated for the transport. Also added are device\ncapability flags to indicate what the memory registration\ncapabilities are for the underlying device and whether or not fast\nmemory registration is supported.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "d5b337b4877f7c4e1d761434ee04d045b0201e03", "tree": "0c31c4cfb0f2010a87b0f1ae7246cbd50068306e", "parents": [ "97eb89bb0e5d9ab20dbc677cb18fad1421473287" ], "author": { "name": "Benny Halevy", "email": "bhalevy@panasas.com", "time": "Sun Sep 28 09:21:26 2008 +0300" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Sep 29 18:13:40 2008 -0400" }, "message": "nfsd: use nfs client rpc callback program\n\nsince commit ff7d9756b501744540be65e172d27ee321d86103\n\"nfsd: use static memory for callback program and stats\"\ndo_probe_callback uses a static callback program\n(NFS4_CALLBACK) rather than the one set in clp-\u003ecl_callback.cb_prog\nas passed in by the client in setclientid (4.0)\nor create_session (4.1).\n\nThis patches introduces rpc_create_args.prognumber that allows\noverriding program-\u003enumber when creating rpc_clnt.\n\nSigned-off-by: Benny Halevy \u003cbhalevy@panasas.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "5344b12d4f97d4a9a62d806425977a6ff64b6baf", "tree": "48b7200d09b898c66c3e278ffeb12cd4453022bf", "parents": [ "396cb3d003c2ce72b50c8c06fddfbb7516f30eb1" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Wed Aug 27 16:57:46 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Sep 29 18:13:38 2008 -0400" }, "message": "SUNRPC: Make svc_addr\u0027s argument a constant\n\nClean up: Add extra type safety and squelch a few compiler complaints\nin upcoming patches.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "a26cfad6e0a308a2c68df1f1ef50aabd48b17e6d", "tree": "9d200a635720bde73b66fd96e4db9e7ff9be3af7", "parents": [ "7252d575ab0e8771269a3d245c36a05ace5152bd" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Mon Aug 18 19:34:16 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Sep 29 18:13:38 2008 -0400" }, "message": "SUNRPC: Support IPv6 when registering kernel RPC services\n\nIn order to advertise NFS-related services on IPv6 interfaces via\nrpcbind, the kernel RPC server implementation must use\nrpcb_v4_register() instead of rpcb_register().\n\nA new kernel build option allows distributions to use the legacy\nv2 call until they integrate an appropriate user-space rpcbind\ndaemon that can support IPv6 RPC services.\n\nI tried adding some automatic logic to fall back if registering\nwith a v4 protocol request failed, but there are too many corner\ncases. So I just made it a compile-time switch that distributions\ncan throw when they\u0027ve replaced portmapper with rpcbind.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "14aeb2118d6e9fd9ee988324c740a00c80979093", "tree": "ca98af50a2e90fb782605ab6235c3a4b30c5b1a4", "parents": [ "b6632339e3afbcbb438a3c8935190ea22464fc99" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Mon Aug 18 19:34:00 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Sep 29 18:13:37 2008 -0400" }, "message": "SUNRPC: Simplify rpcb_register() API\n\nBruce suggested there\u0027s no need to expose the difference between an error\nsending the PMAP_SET request and an error reply from the portmapper to\nrpcb_register\u0027s callers. The user space equivalent of rpcb_register() is\npmap_set(3), which returns a bool_t : either the PMAP set worked, or it\ndidn\u0027t. Simple.\n\nSo let\u0027s remove the \"*okay\" argument from rpcb_register() and\nrpcb_v4_register(), and simply return an error if any part of the call\ndidn\u0027t work.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "e851db5b05408b89b9a9429a66814b79fabee2a1", "tree": "99bac979ed90efed99c89a49cb77719ab4c0a229", "parents": [ "72d31053f62c4bc464c2783974926969614a8649" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Mon Jun 30 18:45:30 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Sep 29 17:56:56 2008 -0400" }, "message": "SUNRPC: Add address family field to svc_serv data structure\n\nIntroduce and initialize an address family field in the svc_serv structure.\n\nThis field will determine what family to use for the service\u0027s listener\nsockets and what families are advertised via the local rpcbind daemon.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "24b8b44780a2c53ecb738f4a1c08d114f5eda27c", "tree": "f772e59438783c6230158ea65bd5aeb3f84f4202", "parents": [ "30a2f3c60a84092c8084dfe788b710f8d0768cd4" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Aug 13 11:05:41 2008 -0500" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Aug 13 16:57:31 2008 -0400" }, "message": "svcrdma: Fix race between svc_rdma_recvfrom thread and the dto_tasklet\n\nRDMA_READ completions are kept on a separate queue from the general\nI/O request queue. Since a separate lock is used to protect the RDMA_READ\ncompletion queue, a race exists between the dto_tasklet and the\nsvc_rdma_recvfrom thread where the dto_tasklet sets the XPT_DATA\nbit and adds I/O to the read-completion queue. Concurrently, the\nrecvfrom thread checks the generic queue, finds it empty and resets\nthe XPT_DATA bit. A subsequent svc_xprt_enqueue will fail to enqueue\nthe transport for I/O and cause the transport to \"stall\".\n\nThe fix is to protect both lists with the same lock and set the XPT_DATA\nbit with this lock held.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "14b395e35d1afdd8019d11b92e28041fad591b71", "tree": "cff7ba9bed7a38300b19a5bacc632979d64fd9c8", "parents": [ "734b397cd14f3340394a8dd3266bec97d01f034b", "5108b27651727b5aba0826e8fd7be71b42428701" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 20 21:21:46 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 20 21:21:46 2008 -0700" }, "message": "Merge branch \u0027for-2.6.27\u0027 of git://linux-nfs.org/~bfields/linux\n\n* \u0027for-2.6.27\u0027 of git://linux-nfs.org/~bfields/linux: (51 commits)\n nfsd: nfs4xdr.c do-while is not a compound statement\n nfsd: Use C99 initializers in fs/nfsd/nfs4xdr.c\n lockd: Pass \"struct sockaddr *\" to new failover-by-IP function\n lockd: get host reference in nlmsvc_create_block() instead of callers\n lockd: minor svclock.c style fixes\n lockd: eliminate duplicate nlmsvc_lookup_host call from nlmsvc_lock\n lockd: eliminate duplicate nlmsvc_lookup_host call from nlmsvc_testlock\n lockd: nlm_release_host() checks for NULL, caller needn\u0027t\n file lock: reorder struct file_lock to save space on 64 bit builds\n nfsd: take file and mnt write in nfs4_upgrade_open\n nfsd: document open share bit tracking\n nfsd: tabulate nfs4 xdr encoding functions\n nfsd: dprint operation names\n svcrdma: Change WR context get/put to use the kmem cache\n svcrdma: Create a kmem cache for the WR contexts\n svcrdma: Add flush_scheduled_work to module exit function\n svcrdma: Limit ORD based on client\u0027s advertised IRD\n svcrdma: Remove unused wait q from svcrdma_xprt structure\n svcrdma: Remove unneeded spin locks from __svc_rdma_free\n svcrdma: Add dma map count and WARN_ON\n ...\n" }, { "commit": "49997d75152b3d23c53b0fa730599f2f74c92c65", "tree": "46e93126170d02cfec9505172e545732c1b69656", "parents": [ "a0c80b80e0fb48129e4e9d6a9ede914f9ff1850d", "5b664cb235e97afbf34db9c4d77f08ebd725335e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jul 18 02:39:39 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jul 18 02:39:39 2008 -0700" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6\n\nConflicts:\n\n\tDocumentation/powerpc/booting-without-of.txt\n\tdrivers/atm/Makefile\n\tdrivers/net/fs_enet/fs_enet-main.c\n\tdrivers/pci/pci-acpi.c\n\tnet/8021q/vlan.c\n\tnet/iucv/iucv.c\n" }, { "commit": "c2e1b09ff237c0a3687b9a804cc8bf489743cffc", "tree": "9571ae56e29de0d0b165b136863c2363f2a78569", "parents": [ "babe80eb4994dfdc97d5be19a68b5af66d667585" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Mon Jul 14 16:03:30 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Jul 15 18:08:55 2008 -0400" }, "message": "SUNRPC: Support registering IPv6 interfaces with local rpcbind daemon\n\nIntroduce a new API to register RPC services on IPv6 interfaces to allow\nthe NFS server and lockd to advertise on IPv6 networks.\n\nUnlike rpcb_register(), the new rpcb_v4_register() function uses rpcbind\nprotocol version 4 to contact the local rpcbind daemon. The version 4\nSET/UNSET procedures allow services to register address families besides\nAF_INET, register at specific network interfaces, and register transport\nprotocols besides UDP and TCP. All of this functionality is exposed via\nthe new rpcb_v4_register() kernel API.\n\nA user-space rpcbind daemon implementation that supports version 4 of the\nrpcbind protocol is required in order to make use of this new API.\n\nNote that rpcbind version 3 is sufficient to support the new rpcbind\nfacilities listed above, but most extant implementations use version 4.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "a486aeda9b2b0d944aecce7871b3186379b898de", "tree": "724ee69f012083efcd379733ba783efbe6341ab2", "parents": [ "d25a03cf966f2cf9990dc0bf2a921a554919ea34" ], "author": { "name": "\\\\\\\"J. Bruce Fields\\\\\\", "email": "bfields@citi.umich.edu", "time": "Mon Jun 09 16:51:35 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Jul 09 12:09:14 2008 -0400" }, "message": "rpc: minor cleanup of scheduler callback code\n\nTry to make the comment here a little more clear and concise.\n\nAlso, this macro definition seems unnecessary.\n\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "b6b6152c46861dd914d0e6cea9c27df057d6e235", "tree": "0d63ce529c13c09a1997da06199e27b049bb55aa", "parents": [ "48b605f83c920d8daa50e43fc2c7f718e04c7bfa" ], "author": { "name": "Olga Kornievskaia", "email": "aglo@citi.umich.edu", "time": "Mon Jun 09 16:51:31 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Jul 09 12:09:10 2008 -0400" }, "message": "rpc: bring back cl_chatty\n\nThe cl_chatty flag alows us to control whether a given rpc client leaves\n\n\t\"server X not responding, timed out\"\n\nmessages in the syslog. Such messages make sense for ordinary nfs\nclients (where an unresponsive server means applications on the\nmountpoint are probably hanging), but not for the callback client (which\ncan fail more commonly, with the only result just of disabling some\noptimizations).\n\nPreviously cl_chatty was removed, do to lack of users; reinstate it, and\nuse it for the nfsd\u0027s callback client.\n\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "e86322f611eef95aafaf726fd3965e5b211f1985", "tree": "28547e26df4fc6ae671dc8cc6912a53717e4db08", "parents": [ "b001a1b6aa960949a24c2cdc28257dfcc9428d74", "8948896c9e098c6fd31a6a698a598a7cbd7fa40e" ], "author": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Thu Jul 03 16:24:06 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Thu Jul 03 16:24:06 2008 -0400" }, "message": "Merge branch \u0027for-bfields\u0027 of git://linux-nfs.org/~tomtucker/xprt-switch-2.6 into for-2.6.27\n" }, { "commit": "8948896c9e098c6fd31a6a698a598a7cbd7fa40e", "tree": "465a4d93c8becb0b2e4750cc1286391916f887b5", "parents": [ "bf5927d84e70d522f234ca247b27d27c63878b93" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed May 28 15:14:02 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Jul 02 15:02:02 2008 -0500" }, "message": "svcrdma: Change WR context get/put to use the kmem cache\n\nChange the WR context pool to be shared across mount points. This\nreduces the RDMA transport memory footprint significantly since\nidle mounts don\u0027t consume WR context memory.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "779a48577ba88b6a7e9748a04b0b739f36c5e6f6", "tree": "40e1a912d98c66ccae3bc7c99e617c1453d8262d", "parents": [ "94dba4918d4570bfa98776e54a5fa527c848dc78" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 10:17:09 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Jul 02 15:01:58 2008 -0500" }, "message": "svcrdma: Remove unused wait q from svcrdma_xprt structure\n\nThe sc_read_wait queue head is no longer used. Remove it.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "87295b6c5c7fd7bbc0ce3e7f42d2adbbac7352b9", "tree": "11a273d3fd2c61bbbf102b80c0b45a0f0021d499", "parents": [ "e6ab9143719ff76f0b95f0866c4d0f6c743ad2e0" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed May 28 13:17:44 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Jul 02 15:01:56 2008 -0500" }, "message": "svcrdma: Add dma map count and WARN_ON\n\nAdd a dma map count in order to verify that all DMA mapping resources\nhave been freed when the transport is closed.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "f820c57ebf5493d4602cc00577c8b0fadd27a7b8", "tree": "7a39a4e18e98d95d3a2a9d642ae93509c5c77177", "parents": [ "34d16e42a6ab74a4a4389c061dfa3c6609e08fa0" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Tue May 27 17:03:14 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Jul 02 15:01:55 2008 -0500" }, "message": "svcrdma: Use reply and chunk map for RDMA_READ processing\n\nModify the RDMA_READ processing to use the reply and chunk list mapping data\ntypes. Also add a special purpose \u0027hdr_count\u0027 field in in the context to hold\nthe header page count instead of overloading the SGE length field and\ncorrupting the DMA map length.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "ab96dddbedf4bb8a7a0fe44012efc1d99598c36f", "tree": "bb3fcfe50264d9a0eb63b768aab8154d70185975", "parents": [ "e1441b9a41c33aa9236008a7cfe49a8e723fb397" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed May 28 13:54:04 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Jul 02 15:01:53 2008 -0500" }, "message": "svcrdma: Add a type for keeping NFS RPC mapping\n\nCreate a new data structure to hold the remote client address space\nto local server address space mapping.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "d00953a53e9a2edbe005c1e596f1e96a8a293401", "tree": "3e476deb8cfd5e97a48a725bb21af28dfdea879d", "parents": [ "8837abcab3d16608bd2c7fac051a839d48f2f30c" ], "author": { "name": "Kevin Coffman", "email": "kwc@citi.umich.edu", "time": "Wed Apr 30 12:45:53 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Jun 23 13:47:25 2008 -0400" }, "message": "gss_krb5: create a define for token header size and clean up ptr location\n\ncleanup:\nDocument token header size with a #define instead of open-coding it.\n\nDon\u0027t needlessly increment \"ptr\" past the beginning of the header\nwhich makes the values passed to functions more understandable and\neliminates the need for extra \"krb5_hdr\" pointer.\n\nClean up some intersecting white-space issues flagged by checkpatch.pl.\n\nThis leaves the checksum length hard-coded at 8 for DES. A later patch\ncleans that up.\n\nSigned-off-by: Kevin Coffman \u003ckwc@citi.umich.edu\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "a75c5d01e4235a7dd785548ac756f248b1b40107", "tree": "602c517b868a219d184e42d23cda4e834396c8ca", "parents": [ "9867d76ca16b3f455f9ca83861f4ce5c94a25928" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue Jun 10 08:40:39 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Jun 23 13:02:49 2008 -0400" }, "message": "sunrpc: remove sv_kill_signal field from svc_serv struct\n\nSince we no longer make any distinction between shutdown signals with\nnfsd, then it becomes easier to just standardize on a particular signal\nto use to bring it down (SIGINT, in this case).\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "9867d76ca16b3f455f9ca83861f4ce5c94a25928", "tree": "1715f69545bf6e766e6e18b33c39270ea7bd15ce", "parents": [ "e096bbc6488d3e49d476bf986d33752709361277" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Tue Jun 10 08:40:38 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Mon Jun 23 13:02:49 2008 -0400" }, "message": "knfsd: convert knfsd to kthread API\n\nThis patch is rather large, but I couldn\u0027t figure out a way to break it\nup that would remain bisectable. It does several things:\n\n- change svc_thread_fn typedef to better match what kthread_create expects\n- change svc_pool_map_set_cpumask to be more kthread friendly. Make it\n take a task arg and and get rid of the \"oldmask\"\n- have svc_set_num_threads call kthread_create directly\n- eliminate __svc_create_thread\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "0b040829952d84bf2a62526f0e24b624e0699447", "tree": "b2e836e7e64343f19b949e50099bbe2ea554ec30", "parents": [ "573bf470e693f73a6ac437b17a64a10902ba54bf" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Jun 10 22:46:50 2008 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Jun 11 21:00:38 2008 -0700" }, "message": "net: remove CVS keywords\n\nThis patch removes CVS keywords that weren\u0027t updated for a long time\nfrom comments.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "008fdbc57164b0ac237ad6ee2766944f02ac9c28", "tree": "e1637f39b6454a6ae94a00e90bd047d47bdfdf7e", "parents": [ "af261af4db14230fb35bcdc0ba9ef78ed6cf7bc1" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed May 07 15:47:42 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 07:34:01 2008 -0500" }, "message": "svcrdma: Change svc_rdma_send_error return type to void\n\nThe svc_rdma_send_error function is called when an RPCRDMA protocol\nerror is detected. This function attempts to post an error reply message.\nSince an error posting to a transport in error is ignored, change\nthe return type to void.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "8da91ea8de873ee8be82377ff18637d05e882058", "tree": "9cfb692f27dde588a360a0f612b8ea14b04a76a4", "parents": [ "47698e083e40bbd3ef87f5561390ae33abb13cd0" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Apr 30 22:00:46 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 07:33:54 2008 -0500" }, "message": "svcrdma: Move destroy to kernel thread\n\nSome providers may wait while destroying adapter resources.\nSince it is possible that the last reference is put on the\ndto_tasklet, the actual destroy must be scheduled as a work item.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "8740767376b32a7772607e1b2b07cde0c24120cc", "tree": "62e9831548e0cdd2e70951a9ab9df4b93ca440aa", "parents": [ "02e7452de74d308ca642f54f7e5ef801ced60a92" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Apr 30 20:44:39 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 07:33:52 2008 -0500" }, "message": "svcrdma: Use standard Linux lists for context cache\n\nReplace the one-off linked list implementation used to implement the\ncontext cache with the standard Linux list_head lists. Add a context\ncounter to catch resource leaks. A WARN_ON will be added later to\nensure that we\u0027ve freed all contexts.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "02e7452de74d308ca642f54f7e5ef801ced60a92", "tree": "9300b5d12be31a6d48689a4260abb1b81e8a5551", "parents": [ "10a38c33f46d128d11e299acba744bc325cde420" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Apr 30 19:50:56 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 07:33:51 2008 -0500" }, "message": "svcrdma: Simplify RDMA_READ deferral buffer management\n\nAn NFS_WRITE requires a set of RDMA_READ requests to fetch the write\ndata from the client. There are two principal pieces of data that\nneed to be tracked: the list of pages that comprise the completed RPC\nand the SGE of dma mapped pages to refer to this list of pages. Previously\nthis whole bit was managed as a linked list of contexts with the\ncontext containing the page list buried in this list. This patch\nsimplifies this processing by not keeping a linked list, but rather only\na pionter from the last submitted RDMA_READ\u0027s context to the context\nthat maps the set of pages that describe the RPC. This significantly\nsimplifies this code path. SGE contexts are cleaned up inline in the DTO\npath instead of at read completion time.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "10a38c33f46d128d11e299acba744bc325cde420", "tree": "3454e7f32f823ad169f09f8e886c7df007c2ce43", "parents": [ "d16d40093a95f2b31007d7a7abefc50e6b27e236" ], "author": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Wed Apr 30 17:32:17 2008 -0500" }, "committer": { "name": "Tom Tucker", "email": "tom@opengridcomputing.com", "time": "Mon May 19 07:33:50 2008 -0500" }, "message": "svcrdma: Remove unused READ_DONE context flags bit\n\nThe RDMACTXT_F_READ_DONE bit is not longer used. Remove it.\n\nSigned-off-by: Tom Tucker \u003ctom@opengridcomputing.com\u003e\n" }, { "commit": "563307b2fa15d687abc54bd980b9847ebf0e3231", "tree": "7f16be30217a2e66360ede97aa27d07a4ebd7e55", "parents": [ "10c993a6b5418cb1026775765ba4c70ffb70853d", "233607dbbc823caf685e778cabc49fb7f679900b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 24 11:46:16 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 24 11:46:16 2008 -0700" }, "message": "Merge git://git.linux-nfs.org/projects/trondmy/nfs-2.6\n\n* git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (80 commits)\n SUNRPC: Invalidate the RPCSEC_GSS session if the server dropped the request\n make nfs_automount_list static\n NFS: remove duplicate flags assignment from nfs_validate_mount_data\n NFS - fix potential NULL pointer dereference v2\n SUNRPC: Don\u0027t change the RPCSEC_GSS context on a credential that is in use\n SUNRPC: Fix a race in gss_refresh_upcall()\n SUNRPC: Don\u0027t disconnect more than once if retransmitting NFSv4 requests\n SUNRPC: Remove the unused export of xprt_force_disconnect\n SUNRPC: remove XS_SENDMSG_RETRY\n SUNRPC: Protect creds against early garbage collection\n NFSv4: Attempt to use machine credentials in SETCLIENTID calls\n NFSv4: Reintroduce machine creds\n NFSv4: Don\u0027t use cred-\u003ecr_ops-\u003ecr_name in nfs4_proc_setclientid()\n nfs: fix printout of multiword bitfields\n nfs: return negative error value from nfs{,4}_stat_to_errno\n NLM/lockd: Ensure client locking calls use correct credentials\n NFS: Remove the buggy lock-if-signalled case from do_setlk()\n NLM/lockd: Fix a race when cancelling a blocking lock\n NLM/lockd: Ensure that nlmclnt_cancel() returns results of the CANCEL call\n NLM: Remove the signal masking in nlmclnt_proc/nlmclnt_cancel\n ...\n" }, { "commit": "c0401ea008fb7c785a93428752d69dccafb127ec", "tree": "c24c8c6e8871ecba8906098c0360bf3980fedf73", "parents": [ "b7872fe86db78cc96c85a13338ea6e3fe1aef610" ], "author": { "name": "Chuck Lever", "email": "chuck.lever@oracle.com", "time": "Mon Apr 14 12:27:30 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:43 2008 -0400" }, "message": "SUNRPC: Update RPC server\u0027s TCP record marker decoder\n\nClean up: Update the RPC server\u0027s TCP record marker decoder to match the\nconstructs used by the RPC client\u0027s TCP socket transport.\n\nSigned-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "8774282c4cef82695ccca8bd09976de5d6e49610", "tree": "d5061bde10232534900729aedb3ee63a78183ae5", "parents": [ "e1ba1ab76e68de9f4a93fae8406627924efaed99" ], "author": { "name": "Jeff Layton", "email": "jlayton@redhat.com", "time": "Mon Apr 07 16:45:37 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:42 2008 -0400" }, "message": "SUNRPC: remove svc_create_thread()\n\nNow that the nfs4 callback thread uses the kthread API, there are no\nmore users of svc_create_thread(). Remove it.\n\nSigned-off-by: Jeff Layton \u003cjlayton@redhat.com\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "5743d65c2f77d5145fb4c4262c4dd70c3f078776", "tree": "aad96f19d1087e577aff8a88dd08c444c09f5068", "parents": [ "1a448fdb3c5495405bc44d77ea676150f9195444" ], "author": { "name": "Kevin Coffman", "email": "kwc@citi.umich.edu", "time": "Mon Mar 31 10:31:33 2008 -0400" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:41 2008 -0400" }, "message": "gss_krb5: consistently use unsigned for seqnum\n\nConsistently use unsigned (u32 vs. s32) for seqnum.\n\nIn get_mic function, send the local copy of seq_send,\nrather than the context version.\n\nSigned-off-by: Kevin Coffman \u003ckwc@citi.umich.edu\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "30aef3166ab27f7bcb14c5e809205af8126fa10b", "tree": "0c42969032c9ec9a2a0c71406a0eed7b174b43b7", "parents": [ "3d4a6886786f839976c36e62303507692bf87d8d" ], "author": { "name": "Kevin Coffman", "email": "kwc@citi.umich.edu", "time": "Thu Feb 21 13:44:27 2008 -0500" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:40 2008 -0400" }, "message": "Remove define for KRB5_CKSUM_LENGTH, which will become enctype-dependent\n\ncleanup: When adding new encryption types, the checksum length\ncan be different for each enctype. Face the fact that the\ncurrent code only supports DES which has a checksum length of 8.\n\nSigned-off-by: Kevin Coffman \u003ckwc@citi.umich.edu\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "67eb6ff610d50da231a37beb634d6dea4b5025ab", "tree": "5a14b4e8ca8ed716f541bb681d5fdc0f0b07f767", "parents": [ "6a85fa3adddd3a74bd5b94c4b72668d307b88377" ], "author": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Thu Jan 31 16:14:54 2008 -0500" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:37 2008 -0400" }, "message": "svcrpc: move unused field from cache_deferred_req\n\nThis field is set once and never used; probably some artifact of an\nearlier implementation idea.\n\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "f15364bd4cf8799a7677b6daeed7b67d9139d974", "tree": "bef48016533d184ea171d4b64336a5ad65ebc18e", "parents": [ "d751a7cd0695554498f25d3026ca6710dbb3698f" ], "author": { "name": "Aurélien Charbon", "email": "aurelien.charbon@ext.bull.net", "time": "Fri Jan 18 15:50:56 2008 +0100" }, "committer": { "name": "J. Bruce Fields", "email": "bfields@citi.umich.edu", "time": "Wed Apr 23 16:13:36 2008 -0400" }, "message": "IPv6 support for NFS server export caches\n\nThis adds IPv6 support to the interfaces that are used to express nfsd\nexports. All addressed are stored internally as IPv6; backwards\ncompatibility is maintained using mapped addresses.\n\nThanks to Bruce Fields, Brian Haley, Neil Brown and Hideaki Joshifuji\nfor comments\n\nSigned-off-by: Aurelien Charbon \u003caurelien.charbon@bull.net\u003e\nCc: Neil Brown \u003cneilb@suse.de\u003e\nCc: Brian Haley \u003cbrian.haley@hp.com\u003e\nCc: YOSHIFUJI Hideaki / 吉藤英明 \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: J. Bruce Fields \u003cbfields@citi.umich.edu\u003e\n" }, { "commit": "7c1d71cf56feebfb5b98219b9d11dfc3a2feca62", "tree": "5a073d987ec79900b2aff217a226c8c78e94b579", "parents": [ "636ac43318ce6939c1698fb67e714d421314ed71" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Apr 17 16:52:57 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Apr 19 16:55:12 2008 -0400" }, "message": "SUNRPC: Don\u0027t disconnect more than once if retransmitting NFSv4 requests\n\nNFSv4 requires us to ensure that we break the TCP connection before we\u0027re\nallowed to retransmit a request. However in the case where we\u0027re\nretransmitting several requests that have been sent on the same\nconnection, we need to ensure that we don\u0027t interfere with the attempt to\nreconnect and/or break the connection again once it has been established.\n\nWe therefore introduce a \u0027connection\u0027 cookie that is bumped every time a\nconnection is broken. This allows requests to track if they need to force a\ndisconnection.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "7c67db3a8a98045744f06fcd6d8f476d9df0ba5c", "tree": "4d0041249b48f9ac66560550e0e96c9c86c37891", "parents": [ "78ea323be6380a9313e87fe241809e912e8ae401" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Apr 07 20:50:11 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Apr 19 16:54:56 2008 -0400" }, "message": "NFSv4: Reintroduce machine creds\n\nWe need to try to ensure that we always use the same credentials whenever\nwe re-establish the clientid on the server. If not, the server won\u0027t\nrecognise that we\u0027re the same client, and so may not allow us to recover\nstate.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "78ea323be6380a9313e87fe241809e912e8ae401", "tree": "6ea9a772ab677012c52a08c6b32e366d7bf129d4", "parents": [ "441092415770ddec648800701895913c4bfd60c1" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Apr 07 20:49:28 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Apr 19 16:54:53 2008 -0400" }, "message": "NFSv4: Don\u0027t use cred-\u003ecr_ops-\u003ecr_name in nfs4_proc_setclientid()\n\nWith the recent change to generic creds, we can no longer use\ncred-\u003ecr_ops-\u003ecr_name to distinguish between RPCSEC_GSS principals and\nAUTH_SYS/AUTH_NULL identities. Replace it with the rpc_authops-\u003eau_name\ninstead...\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "b6ddf64ffe9d59577a9176856bb6fe69a539f573", "tree": "dd12c31555cf9449dd6094428d377186b3c4c373", "parents": [ "24b74bf0c9e08cbda74d3c64af69ad402ed54e04" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Apr 17 18:52:19 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Sat Apr 19 16:52:44 2008 -0400" }, "message": "SUNRPC: Fix up xprt_write_space()\n\nThe rest of the networking layer uses SOCK_ASYNC_NOSPACE to signal whether\nor not we have someone waiting for buffer memory. Convert the SUNRPC layer\nto use the same idiom.\nRemove the unlikely()s in xs_udp_write_space and xs_tcp_write_space. In\nfact, the most common case will be that there is nobody waiting for buffer\nspace.\n\nSOCK_NOSPACE is there to tell the TCP layer whether or not the cwnd was\nlimited by the application window. Ensure that we follow the same idiom as\nthe rest of the networking layer here too.\n\nFinally, ensure that we clear SOCK_ASYNC_NOSPACE once we wake up, so that\nwrite_space() doesn\u0027t keep waking things up on xprt-\u003epending.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "5c691044ecbca04dd558fca4c754121689fe1b34", "tree": "7a56b63c44766fcca0d7ac245ce0e19077b6f118", "parents": [ "9a559efd4199c9812d339e23cc1b6055366b224f" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 12 16:21:07 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Mar 14 13:42:41 2008 -0400" }, "message": "SUNRPC: Add an rpc_credop callback for binding a credential to an rpc_task\n\nWe need the ability to treat \u0027generic\u0027 creds specially, since they want to\nbind instances of the auth cred instead of binding themselves.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "9a559efd4199c9812d339e23cc1b6055366b224f", "tree": "44ae54434a6bd368693ea76b210448496e222e46", "parents": [ "4ccda2cdd8d156b6f49440653d5d6997e0facf97" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 12 12:24:49 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Mar 14 13:42:38 2008 -0400" }, "message": "SUNRPC: Add a generic RPC credential\n\nAdd an rpc credential that is not tied to any particular auth mechanism,\nbut that can be cached by NFS, and later used to look up a cred for\nwhichever auth mechanism that turns out to be valid when the RPC call is\nbeing made.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "4ccda2cdd8d156b6f49440653d5d6997e0facf97", "tree": "fd75203f002edfd1657c9ca7bcb5494a7b783cf1", "parents": [ "af093835774931de898a9baf7b4041fa0d100f77" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 12 16:20:55 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Mar 14 13:42:35 2008 -0400" }, "message": "SUNRPC: Clean up rpcauth_bindcred()\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "af093835774931de898a9baf7b4041fa0d100f77", "tree": "f7617eb4556ead8615f6c26d8b02fcec8f64c6b3", "parents": [ "25337fdc85951dfeac944f16cb565904c619077a" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 12 12:12:16 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Mar 14 13:42:32 2008 -0400" }, "message": "SUNRPC: Fix RPCAUTH_LOOKUP_ROOTCREDS\n\nThe current RPCAUTH_LOOKUP_ROOTCREDS flag only works for AUTH_SYS\nauthentication, and then only as a special case in the code. This patch\nremoves the auth_sys special casing, and replaces it with generic code.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "25337fdc85951dfeac944f16cb565904c619077a", "tree": "8c0f37eca64775825ea84ade40460862a2ba377e", "parents": [ "9446389ef612096704fdf18fa79bab423d4110f0" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Mar 12 14:40:14 2008 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Mar 14 13:42:29 2008 -0400" }, "message": "SUNRPC: Fix a bug in rpcauth_lookup_credcache()\n\nThe hash bucket is for some reason always being set to zero.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "cdd0972945dbcb8ea24db365d9b0e100af2a27bb", "tree": "d41204db20ccc81cecf99a332e2fa764075e8637", "parents": [ "5e4424af9a1f062c6451681dff24a26e27741cc6", "cbc20059259edee4ea24c923e6627c394830c972" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:48:05 2008 -0800" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:48:05 2008 -0800" }, "message": "Merge branch \u0027cleanups\u0027 into next\n" }, { "commit": "5e4424af9a1f062c6451681dff24a26e27741cc6", "tree": "fcba54a8fc2d94c1d691ab5b8b956ac71b5cb8d4", "parents": [ "ff2d7db848f8db7cade39e55f78f86d77e0de01a" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Feb 25 21:53:49 2008 -0800" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:26:28 2008 -0800" }, "message": "SUNRPC: Remove now-redundant RCU-safe rpc_task free path\n\nNow that we\u0027ve tightened up the locking rules for RPC queue wakeups, we can\nremove the RCU-safe kfree calls...\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "f5fb7b06e4e4ab18326f067f4317b2016ce18af2", "tree": "c1a2c972d373d503b432f592b52d4fd2a59a1b46", "parents": [ "eb276c0e10187702928aeaa133e1d3dbaf3eafc7" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Feb 25 21:40:50 2008 -0800" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:26:25 2008 -0800" }, "message": "SUNRPC: Eliminate the now-redundant rpc_start_wakeup()\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "eb276c0e10187702928aeaa133e1d3dbaf3eafc7", "tree": "f7cfb639d97747f32573dc72970aa9bf589fb0da", "parents": [ "36df9aae3158ce8fc4ede241169dc94ac910d884" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Feb 22 17:27:59 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:26:21 2008 -0800" }, "message": "SUNRPC: Switch tasks to using the rpc_waitqueue\u0027s timer function\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "36df9aae3158ce8fc4ede241169dc94ac910d884", "tree": "5f6caa22ce9414850fd34eefb54150b120a8f4e7", "parents": [ "f6a1cc89309f0ae847a9b6fe418d1c4215e5bc55" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Wed Jul 18 16:18:52 2007 -0400" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:21:59 2008 -0800" }, "message": "SUNRPC: Add a timer function to wait queues.\n\nThis is designed to replace the timeout timer in the individual rpc_tasks.\nBy putting the timer function in the wait queue, we will eventually be able\nto reduce the total number of timers in use by the RPC subsystem.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "f6a1cc89309f0ae847a9b6fe418d1c4215e5bc55", "tree": "4f5d585dd901718d66d9c354625f9a0bc85c234d", "parents": [ "5d00837b90340af9106dcd93af75fd664c8eb87f" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Feb 22 17:06:55 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Thu Feb 28 23:17:27 2008 -0800" }, "message": "SUNRPC: Add a (empty for the moment) destructor for rpc_wait_queues\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "5d00837b90340af9106dcd93af75fd664c8eb87f", "tree": "f537dc84421cf150d66b630e56ea8107078c07a8", "parents": [ "fda1393938035559b417dd5b26b9cc293a7aee00" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Feb 22 16:34:17 2008 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Mon Feb 25 21:40:44 2008 -0800" }, "message": "SUNRPC: Run rpc timeout functions as callbacks instead of in softirqs\n\nAn audit of the current RPC timeout functions shows that they don\u0027t really\never need to run in the softirq context. As long as the softirq is\nable to signal that the wakeup is due to a timeout (which it can do by\nsetting task-\u003etk_status to -ETIMEDOUT) then the callback functions can just\nrun as standard task-\u003etk_callback functions (in the rpciod/process\ncontext).\n\nThe only possible border-line case would be xprt_timer() for the case of\nUDP, when the callback is used to reduce the size of the transport\ncongestion window. In testing, however, the effect of moving that update\nto a callback would appear to be minor.\n\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" } ], "next": "fda1393938035559b417dd5b26b9cc293a7aee00" }