)]}' { "log": [ { "commit": "cf5046323ea254be72535648a9d090b18b8510f3", "tree": "d44894722bd965b2f28a54c4dc4157f22b618c34", "parents": [ "ae937debe178b4327fd67d604ee83a20f22aa0de", "8d34ff34016959d464fd5582ea6a8226fe57ab0e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jun 14 13:53:22 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jun 14 13:53:22 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n mlx4_core: Don\u0027t double-free IRQs when falling back from MSI-X to INTx\n IB/mthca: Don\u0027t double-free IRQs when falling back from MSI-X to INTx\n IB/mlx4: Add strong ordering to local inval and fast reg work requests\n IB/ehca: Remove superfluous bitmasks from QP control block\n RDMA/cxgb3: Limit fast register size based on T3 limitations\n RDMA/cxgb3: Report correct port state and MTU\n mlx4_core: Add module parameter for number of MTTs per segment\n IB/mthca: Add module parameter for number of MTTs per segment\n RDMA/nes: Fix off-by-one bugs in reset_adapter_ne020() and init_serdes()\n infiniband: Remove void casts\n IB/ehca: Increment version number\n IB/ehca: Remove unnecessary memory operations for userspace queue pairs\n IB/ehca: Fall back to vmalloc() for big allocations\n IB/ehca: Replace vmalloc() with kmalloc() for queue allocation\n" }, { "commit": "8d34ff34016959d464fd5582ea6a8226fe57ab0e", "tree": "63523621aa0b1f5811916297b7ea7e505157e4d0", "parents": [ "3026c19a14ba71ccd4dc4925abab9395ea12839c", "25a52393270ca48c7d0848672ad4423313033c3d", "5b891a9332dc4212bf166a4506092fbcd60f2319", "2ac6bf4ddc87c3b6b609f8fa82f6ebbffeac12f4", "d1fdf24b4074a8d962f9a28519c99dcdd66bdee3", "28e43a519b9edb8277fc6b490ad17aa38c45a02b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jun 14 13:31:19 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jun 14 13:31:19 2009 -0700" }, "message": "Merge branches \u0027cxgb3\u0027, \u0027ehca\u0027, \u0027misc\u0027, \u0027mlx4\u0027, \u0027mthca\u0027 and \u0027nes\u0027 into for-linus\n" }, { "commit": "9aa0a489d909af0cc36c41d3061ef956c7442ce2", "tree": "e973b5748c54060325ba1dd90ab0ce8ccdc66195", "parents": [ "c1f67a88bf62fac0f4151c007b361199c2cd1988" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Jun 13 15:14:09 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Jun 13 15:14:09 2009 -0700" }, "message": "IB/mthca: Don\u0027t double-free IRQs when falling back from MSI-X to INTx\n\nWhen both MSI-X and legacy INTx fail to generate an interrupt, the\ndriver frees the MSI-X interrupts twice. Fix this by clearing the\nhave_irq flag for the MSI-X interrupts when they are freed the first\ntime.\n\nReported-by: Yinghai Lu \u003cyhlu.kernel@gmail.com\u003e\nTested-by: Yinghai Lu \u003cyhlu.kernel@gmail.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2ac6bf4ddc87c3b6b609f8fa82f6ebbffeac12f4", "tree": "7de468eac9f948f85faebb2f86efbfd66737d8d9", "parents": [ "ab6bf42e2339580b5d87746d0ff4da4b1578b03e" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Fri Jun 05 10:36:24 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jun 05 10:36:24 2009 -0700" }, "message": "IB/mlx4: Add strong ordering to local inval and fast reg work requests\n\nThe ConnectX Programmer\u0027s Reference Manual states that the \"SO\" bit\nmust be set when posting Fast Register and Local Invalidate send work\nrequests. When this bit is set, the work request will be executed\nonly after all previous work requests on the send queue have been\nexecuted. (If the bit is not set, Fast Register and Local Invalidate\nWQEs may begin execution too early, which violates the defined\nsemantics for these operations)\n\nThis fixes the issue with NFS/RDMA reported in\n\u003chttp://lists.openfabrics.org/pipermail/general/2009-April/059253.html\u003e\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "25a52393270ca48c7d0848672ad4423313033c3d", "tree": "9c2541736a54e059ea2cabc89299dda95c3ca63e", "parents": [ "bde2cfaf8ff5511b4f434078554f89ff6cb677f2" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Wed Jun 03 13:25:42 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Jun 03 13:25:42 2009 -0700" }, "message": "IB/ehca: Remove superfluous bitmasks from QP control block\n\nAll the fields in the control block are nicely right-aligned, so no\nmasking is necessary.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3026c19a14ba71ccd4dc4925abab9395ea12839c", "tree": "9264da66d2ab516b7ffdf310d2cf5ac078cc4415", "parents": [ "7ab1a2b31d4a8b4f519ccff5a84c53a5b87fd1be" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Wed May 27 14:43:39 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 27 14:43:39 2009 -0700" }, "message": "RDMA/cxgb3: Limit fast register size based on T3 limitations\n\nT3 firmware only supports one WRs worth of page list for fast register\nwork requests. The driver currently allows 2 WRs worth, which\ndoesn\u0027t work for T3, so reduce the limit in the driver.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7ab1a2b31d4a8b4f519ccff5a84c53a5b87fd1be", "tree": "d11363c949e8f15b3f933e3839374dacef0429ac", "parents": [ "210af919c949a7d6bd330916ef376cec2907d81e" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Wed May 27 14:42:36 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 27 14:42:36 2009 -0700" }, "message": "RDMA/cxgb3: Report correct port state and MTU\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c1f67a88bf62fac0f4151c007b361199c2cd1988", "tree": "5bb527966e1a6cd2a3c06f6df70199df948e3817", "parents": [ "210af919c949a7d6bd330916ef376cec2907d81e" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Wed May 27 14:36:16 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 27 14:36:16 2009 -0700" }, "message": "IB/mthca: Add module parameter for number of MTTs per segment\n\nThe current MTT allocator uses kmalloc() to allocate a buffer for its\nbuddy allocator, and thus is limited in the amount of MTT segments\nthat it can control. As a result, the size of memory that can be\nregistered is limited too. This patch uses a module parameter to\ncontrol the number of MTT entries that each segment represents,\nallowing more memory to be registered with the same number of\nsegments.\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b3cd5050bf8eb32ceecee129cac7c59e6f1668c4", "tree": "5aa4999dd1c5c7fe83354f3965764f0849afaff3", "parents": [ "1336aed10b8af791378b017f0fa8da4e5b827b8d" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Wed May 13 17:57:49 2009 -0500" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Sat May 23 15:44:13 2009 -0500" }, "message": "[SCSI] libiscsi: add task aborted state\n\nIf a task did not complete normally due to a TMF, libiscsi will\nnow complete the task with the state ISCSI_TASK_ABRT_TMF. Drivers\nlike bnx2i that need to free resources if a command did not complete normally\ncan then check the task state. If a driver does not need to send\na special command if we have dropped the session then they can check\nfor ISCSI_TASK_ABRT_SESS_RECOV.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "10eb0f013c63c71c82ede77945a5f390c10cfda6", "tree": "6f4b738aa1dfc422087a0b05ebde4388cf84a3f0", "parents": [ "5a2537959fa8781012e8c286fc1614e0f6991327" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Wed May 13 17:57:38 2009 -0500" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Sat May 23 15:44:09 2009 -0500" }, "message": "[SCSI] iscsi: pass ep connect shost\n\nWhen we create the tcp/ip connection by calling ep_connect, we currently\njust go by the routing table info.\n\nI think there are two problems with this.\n\n1. Some drivers do not have access to a routing table. Some drivers like\nqla4xxx do not even know about other ports.\n\n2. If you have two initiator ports on the same subnet, the user may have\nset things up so that session1 was supposed to be run through port1. and\nsession2 was supposed to be run through port2. It looks like we could\nend with both sessions going through one of the ports.\n\nFixes for cxgb3i from Karen Xie.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "28e43a519b9edb8277fc6b490ad17aa38c45a02b", "tree": "56fd31d1d4e67a770d4fe55bff0ae570e4a7df3f", "parents": [ "210af919c949a7d6bd330916ef376cec2907d81e" ], "author": { "name": "Roel Kluin", "email": "roel.kluin@gmail.com", "time": "Fri May 15 10:16:45 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri May 15 10:16:45 2009 -0700" }, "message": "RDMA/nes: Fix off-by-one bugs in reset_adapter_ne020() and init_serdes()\n\nWith a postfix increment, i is incremented one past 10K/5K before the\nloop ends, so the error messages will be displayed too soon if the\ntest succeeds on the last iteration. Fix the comparisons to be \u003e\ninstead of \u003e\u003d.\n\nSigned-off-by: Roel Kluin \u003croel.kluin@gmail.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5b891a9332dc4212bf166a4506092fbcd60f2319", "tree": "26a0d3966d0c00bedf9315ddfd968f80b2635970", "parents": [ "210af919c949a7d6bd330916ef376cec2907d81e" ], "author": { "name": "Jack Stone", "email": "jwjstone@fastmail.fm", "time": "Wed May 13 16:53:39 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 16:53:39 2009 -0700" }, "message": "infiniband: Remove void casts\n\nRemove uneeded casts of void *.\n\nSigned-off-by: Jack Stone \u003cjwjstone@fastmail.fm\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bde2cfaf8ff5511b4f434078554f89ff6cb677f2", "tree": "7128fdb84c8753defeddb1107bccf4d6f534bfb8", "parents": [ "1988d1fa1a9d642c5714a6afc9775fba0627f3ed" ], "author": { "name": "Stefan Roscher", "email": "ossrosch@linux.vnet.ibm.com", "time": "Wed May 13 16:52:43 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 16:52:43 2009 -0700" }, "message": "IB/ehca: Increment version number\n\nSigned-off-by: Stefan Roscher \u003cstefan.roscher@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1988d1fa1a9d642c5714a6afc9775fba0627f3ed", "tree": "9eeca20f2831d8ebe32795838c1190f1ecc01017", "parents": [ "c94f156f63c835ffc02b686f9d4238b106f31a5d" ], "author": { "name": "Stefan Roscher", "email": "ossrosch@linux.vnet.ibm.com", "time": "Wed May 13 16:52:43 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 16:52:43 2009 -0700" }, "message": "IB/ehca: Remove unnecessary memory operations for userspace queue pairs\n\nThe queue map for flush completion circumvention is only used for\nkernel space queue pairs. This patch skips the allocation of the\nqueue maps in case the QP is created for userspace. In addition, this\npatch does not iomap the galpas for kernel usage if the queue pair is\nonly used in userspace. These changes will improve the performance of\ncreation of userspace queue pairs.\n\nSigned-off-by: Stefan Roscher \u003cstefan.roscher@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c94f156f63c835ffc02b686f9d4238b106f31a5d", "tree": "2b9e4a8543259fd077f79624838d93b2587e3f73", "parents": [ "bf31a1a02eb28d9bda0bb74345df7889faeb7335" ], "author": { "name": "Stefan Roscher", "email": "ossrosch@linux.vnet.ibm.com", "time": "Wed May 13 16:52:42 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 16:52:42 2009 -0700" }, "message": "IB/ehca: Fall back to vmalloc() for big allocations\n\nIn case of large queue pairs there is the possibillity of allocation\nfailures due to memory fragmentation when using kmalloc(). To ensure\nthe memory is allocated even if kmalloc() can not find chunks which\nare big enough, we fall back to allocating the memory with vmalloc().\n\nSigned-off-by: Stefan Roscher \u003cstefan.roscher@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bf31a1a02eb28d9bda0bb74345df7889faeb7335", "tree": "ede5fa4f19c993ec6809e94941280f6e3600921e", "parents": [ "210af919c949a7d6bd330916ef376cec2907d81e" ], "author": { "name": "Anton Blanchard", "email": "antonb@au1.ibm.com", "time": "Wed May 13 16:52:40 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 16:52:40 2009 -0700" }, "message": "IB/ehca: Replace vmalloc() with kmalloc() for queue allocation\n\nTo improve performance of driver resource allocation, replace\nvmalloc() calls with kmalloc().\n\nSigned-off-by: Stefan Roscher \u003cstefan.roscher@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c98861f7de20fa982de0b5bbe01628414d7a7fbf", "tree": "e93ad89fbe9eb136867670e7368aeba6ebd1fb2e", "parents": [ "0a40a55fd1fde3875769025ee8a4a7059c7a042d", "8be741b0ac8e8857fb64da269e81da8a131377bb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 13 16:31:12 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 13 16:31:12 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n IB/mlx4: Don\u0027t overwrite fast registration page list when posting work request\n RDMA/cxgb3: Don\u0027t complete flushed send work requests twice\n" }, { "commit": "8be741b0ac8e8857fb64da269e81da8a131377bb", "tree": "90be5cea26a35cf3990cc303e78f1940ae43c9d5", "parents": [ "ec6995ddaa3db2c5f028184ed9b74c0c448c4947", "2b6b7d4be487bada8c727df829c25068c7b5e5a3" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 15:16:17 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed May 13 15:16:17 2009 -0700" }, "message": "Merge branches \u0027cxgb3\u0027 and \u0027mlx4\u0027 into for-linus\n" }, { "commit": "265e771e8197cdb22a1e2556663173fb62c9cd91", "tree": "3672112a8f7010756d5e61ecd0b10d3406d553f2", "parents": [ "c96f58573778ddf96ff67108a635f3f642ea63d3" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed May 06 01:36:04 2009 -0400" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Sat May 09 10:49:40 2009 -0400" }, "message": "Fix deadlock in ipathfs -\u003eget_sb()\n\nforgot to unlock superblock before calling deactivate_super()...\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "2b6b7d4be487bada8c727df829c25068c7b5e5a3", "tree": "d1fa04daa97cafe89f04b9f16c8a58925f364203", "parents": [ "56a50adda49b2020156616c4eb15353e0f9ad7de" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Thu May 07 21:35:13 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu May 07 21:35:13 2009 -0700" }, "message": "IB/mlx4: Don\u0027t overwrite fast registration page list when posting work request\n\nThe low-level mlx4 driver modified the page-list addresses for fast\nregister work requests post send to big-endian, and set a \"present\"\nbit. This caused problems later when the consumer attempted to unmap\nthe pages using the page-list (using the list addresses which were\nassumed to be still in CPU-endian order). Fix the mlx4 driver to\nallocate two buffers and use a private buffer for the hardware-format\nbus addresses.\n\nThis patch fixes \u003chttps://bugs.openfabrics.org/show_bug.cgi?id\u003d1571\u003e,\nan NFS/RDMA server crash. The cause of the crash was found by Vu Pham\nof Mellanox. The fix is along the lines suggested by Steve Wise in\ncomment #21 in bug 1571.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "61bd1e858db743af64f6e363c526f7e433d12e0c", "tree": "a9213c472c1458e9e7686f9d9fa1c4ed1362e3f3", "parents": [ "8c0c3f7ff0daa119f0bb109148f6f0e39573b429", "e9da4d7f731dafc2b93ce7b31aa09c4d935ef978" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 02 16:36:34 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 02 16:36:34 2009 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6: (53 commits)\n [SCSI] libosd: OSD2r05: on-the-wire changes for latest OSD2 revision 5.\n [SCSI] libosd: OSD2r05: OSD_CRYPTO_KEYID_SIZE will grow 20 \u003d\u003e 32 bytes\n [SCSI] libosd: OSD2r05: Prepare for rev5 attribute list changes\n [SCSI] libosd: fix potential ERR_PTR dereference in osd_initiator.c\n [SCSI] mpt2sas : bump driver version to 01.100.02.00\n [SCSI] mpt2sas: fix hotplug event processing\n [SCSI] mpt2sas : release diagnotic buffers prior host reset\n [SCSI] mpt2sas : Broadcast Primative AEN bug fix\n [SCSI] mpt2sas : Identify Dell series-7 adapters at driver load time\n [SCSI] mpt2sas : driver name needs to be in the MPT2IOCINFO ioctl\n [SCSI] mpt2sas : running out of message frames\n [SCSI] mpt2sas : fix oops when firmware sends large sense buffer size\n [SCSI] mpt2sas : the sanity check in base_interrupt needs to be on dword boundary\n [SCSI] mpt2sas : unique ioctl magic number\n [SCSI] fix sign extension with 1.5TB usb-storage LBD\u003dy\n [SCSI] ipr: Fix sleeping function called with interrupts disabled\n [SCSI] fcoe: fip: add multicast filter to receive FIP advertisements.\n [SCSI] libfc: Fix compilation warnings with allmodconfig\n [SCSI] fcoe: fix spelling typos and bad comments\n [SCSI] fcoe: don\u0027t export functions that are internal to fcoe\n ...\n" }, { "commit": "ec6995ddaa3db2c5f028184ed9b74c0c448c4947", "tree": "1d4635d10ed0a9f1e4d6f1893321d9dadc0b4cf6", "parents": [ "56a50adda49b2020156616c4eb15353e0f9ad7de" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Wed Apr 29 15:15:59 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 29 15:15:59 2009 -0700" }, "message": "RDMA/cxgb3: Don\u0027t complete flushed send work requests twice\n\nWhen the SQ is flushed, mark the flushed entries as not signaled so\nthe poll logic doesn\u0027t re-insert the CQ entry thinking its an out of\norder completion.\n\nThe bug can cause the NFS/RDMA server to crash due to processing the\nsame completed work request twice.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9308f96c7901dd851d93a38e1b5b7a68ee8aa74f", "tree": "b6bdd97168ad94d0488fd10081f643768c09cddc", "parents": [ "cde9e2f9307dd6af6bf8c5f02e41a6e27da9682a", "e028cc55cc5c90a1c57eefe560a0cbb4df1fed14", "8531f1f14a85c004d5063a0a78c72d0b686ccb8e", "5bf0da7dd0ae193e072412519cba1d77b6196c61", "26cc5e57bbe770916bc67af169477fdd3ea1be4c" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Apr 28 16:01:31 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Apr 28 16:01:31 2009 -0700" }, "message": "Merge branches \u0027cxgb3\u0027, \u0027ipoib\u0027, \u0027mthca\u0027, \u0027mlx4\u0027 and \u0027nes\u0027 into for-linus\n" }, { "commit": "26cc5e57bbe770916bc67af169477fdd3ea1be4c", "tree": "d7cc6b1d102f7e577f66f2ce8077cd0c1c919d5f", "parents": [ "9256b2513074164b4555617c4a3b82d36abf03e5" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:46:29 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:46:29 2009 -0700" }, "message": "RDMA/nes: Update iw_nes version\n\nUpdate version number to 1.5.0.0\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9256b2513074164b4555617c4a3b82d36abf03e5", "tree": "ffccbec443c56132035336ef80f65759947beb62", "parents": [ "109d67e4f12b828113ca8ccf4a735972dd984f40" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Mon Apr 27 13:45:19 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:45:19 2009 -0700" }, "message": "RDMA/nes: Fix error path in nes_accept()\n\nIf reg_phys_mem() fails, we need to free memory allocated for MPA\nframe with private data before returning the error. Also move\nnes_add_ref() after the reg_phys_mem() is successful.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "109d67e4f12b828113ca8ccf4a735972dd984f40", "tree": "44b21f18ea6cb05ec6411b4ec00028152f4f1c5d", "parents": [ "4e9c390036196f89208cf9574dfd19daae146776" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Mon Apr 27 13:41:06 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:41:06 2009 -0700" }, "message": "RDMA/nes: Fix hang issues for large cluster dynamic connections\n\nRunning large cluster setup, we are hanging after many hours of\ntesting. Fixing this required going over the code and making sure the\nrexmit entry was properly removed based on the cm_node\u0027s state and\npacket received. Also when receiving a FIN packet, check seq# and\nmake sure there were no errors before calling handle_fin().\n\nFollowing are the changes done in nes_cm.c:\n\n* handle_ack_pkt() needs to return error value, so in case of error,\n handle_fin() is not called. Some cleanup done while going over the code.\n\n* handle_rst_pkt(), handling of cm_node\u0027s NES_CM_STATE_LAST_ACK is missing.\n\n* process_packet(), in case of FIN only packet is received, call\n check_seq() before processing.\n\n* in handle_fin_pkt(), we are calling cleanup_retrans_entry() for all\n conditions, even if the packets need to be dropped.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4e9c390036196f89208cf9574dfd19daae146776", "tree": "c6f80bd2fdebbbfe87ba6f1addf84ed0e6d1af58", "parents": [ "c11470f9f4d4490cd7e9563f604c4c7868caf6de" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Mon Apr 27 13:39:36 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:39:36 2009 -0700" }, "message": "RDMA/nes: Increase rexmit timeout interval\n\nUnder heavy load with large cluster testing, it may take longer to\nreceive a response to MPA requests. Change the driver to wait longer\nafter each rexmit to max time value.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c11470f9f4d4490cd7e9563f604c4c7868caf6de", "tree": "3fec0b3fac5499d1c6812a5bcd4ae19cd57d4831", "parents": [ "53094c388f11d79f742eaf743c9fd740a881f2c0" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Mon Apr 27 13:38:31 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:38:31 2009 -0700" }, "message": "RDMA/nes: Check for sequence number wrap-around\n\ncheck_seq() was not checking if the seq#s have wrapped. Fix it.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "53094c388f11d79f742eaf743c9fd740a881f2c0", "tree": "fc6532504678f4b5e52ec89fb1be26af31ac9435", "parents": [ "1f0dba1e51cfc93bf4545811839a84c879086fd4" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Mon Apr 27 13:37:34 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:37:34 2009 -0700" }, "message": "RDMA/nes: Do not set apbvt entry for loopback\n\nWhen a connect request comes, apbvt should only be set for\nnon-loopback connections.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1f0dba1e51cfc93bf4545811839a84c879086fd4", "tree": "24cd91ae82c72dbb6c931d5a582edfa6f2d3b4fa", "parents": [ "0e4562da9e533188108d00022cf3650cb9e29aae" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:36:03 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:36:03 2009 -0700" }, "message": "RDMA/nes: Fix unused variable compile warning when INFINIBAND_NES_DEBUG\u003dn\n\nRemove the NES_DEBUG that is causing the compile warning about an\nunused variable when INFINIBAND_NES_DEBUG is not enabled.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0e4562da9e533188108d00022cf3650cb9e29aae", "tree": "126b9e4750c21f7c906c8a15f365d95f434f27a4", "parents": [ "923223776b53013443d062a87e0a3d57d6513f04" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:33:48 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:33:48 2009 -0700" }, "message": "RDMA/nes: Fix fw_ver in /sys\n\n/sys/class/infiniband/nes?/fw_ver is not displaying firmware version\nproperly (it shows 0.0.0 with the current code). Fill in the correct\nfirmware version number.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "923223776b53013443d062a87e0a3d57d6513f04", "tree": "d1823410e4a78a6d5cf68b3969afe914b3f22adf", "parents": [ "e998c25bc29f2b409b39fa63dad3df499982a887" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:30:35 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:30:35 2009 -0700" }, "message": "RDMA/nes: Set trace length to 1 inch for SFP_D\n\nWith updated PHY firmware for SFP_D, setting the trace length to 1\ninch for SFP_D provides a more stable link.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e998c25bc29f2b409b39fa63dad3df499982a887", "tree": "cecd8abc9fc7920d19fe78997ffdfe687c5e8d1f", "parents": [ "366835e24977f4590ef353bdc70f0dda278c2a84" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:29:42 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:29:42 2009 -0700" }, "message": "RDMA/nes: Enable repause timer for port 1\n\nEnable repause timer for port 1. Without this setting, under stress,\nthe chip may misbehave.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "366835e24977f4590ef353bdc70f0dda278c2a84", "tree": "a48c57e07bc0227281f184d48ae4343414d11392", "parents": [ "010db4d127d1ae7324d5e00035fe4362e27f0508" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:28:41 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:28:41 2009 -0700" }, "message": "RDMA/nes: Correct CDR loop filter setting for port 1\n\nIn commit 1b949324 (\"RDMA/nes: Fix SFP+ PHY initialization\") there is\na mistake in the clean up code that removed port 1 CDR loop filter\nsettings for 10G cards other than CX4. Put the correct setting back\nfor appropriate PHY types.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "010db4d127d1ae7324d5e00035fe4362e27f0508", "tree": "141275610a4858ccc4092d1d78c67297aee291c0", "parents": [ "5d1af5c83232c5a02b9dc0fe43053b4ddc005224" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 27 13:27:21 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 27 13:27:21 2009 -0700" }, "message": "RDMA/nes: Modify thermo mitigation to flip SerDes1 ref clk to internal\n\nChange thermo mitigation code to flip the SerDes1 reference clock to\ninternal, to match the change in commit a4849fc1 (\"RDMA/nes: Add\nwide_ppm_offset parm for switch compatibility\").\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6b5d6c443a9b4fd71b633cef66b5db4de8a85787", "tree": "6b0ca6eaba58d51d3a0715ba9be40bf5449b40ed", "parents": [ "9a6510eb3f030cedba32664498a610dc6d084d46" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Tue Apr 21 15:32:32 2009 -0500" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Mon Apr 27 10:09:54 2009 -0500" }, "message": "[SCSI] cxgb3i, iser, iscsi_tcp: set target can queue\n\nSet target can queue limit to the number of preallocated\nsession tasks we have.\n\nThis along with the cxgb3i can_queue patch will fix a throughput\nproblem where it could only queue one LU worth of data at a time.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "5d1af5c83232c5a02b9dc0fe43053b4ddc005224", "tree": "705fa258b696855d355fa8ebf2af84ced5f3fa29", "parents": [ "cc005fa20c5229c283bea4958869da1e3c8a3720" ], "author": { "name": "Miroslaw Walukiewicz", "email": "miroslaw.walukiewicz@intel.com", "time": "Tue Apr 21 16:16:48 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Apr 21 16:16:48 2009 -0700" }, "message": "RDMA/nes: Fix resource issues in nes_create_cq() and nes_destroy_cq()\n\nIn error paths where a CQ is not created, pbl is not freeed properly.\n\nIn nes_destroy_cq(), add the corresponding check for nescq-\u003emcrqf to\nnot call nes_free_resource() when it is already done in nes_create_cq().\n\nSigned-off-by: Miroslaw Walukiewicz \u003cmiroslaw.walukiewicz@intel.com\u003e\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "cc005fa20c5229c283bea4958869da1e3c8a3720", "tree": "b196ea46b9e1eb80ec1dd4436db8d6e1e4b579d5", "parents": [ "3f32eb1185170524a81dadff2e67521585943a53" ], "author": { "name": "Matt Kraai", "email": "kraai@ftbfs.org", "time": "Tue Apr 21 10:43:21 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Apr 21 10:43:21 2009 -0700" }, "message": "RDMA/nes: Remove root_256()\u0027s unused pbl_count_256 parameter\n\nSigned-off-by: Matt Kraai \u003ckraai@ftbfs.org\u003e\nAcked-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8531f1f14a85c004d5063a0a78c72d0b686ccb8e", "tree": "0cc55a1673d1966096828c8e057977f7784a9e5a", "parents": [ "a939b96cccdb65df80a52447ec8e4a6d79c56dbb" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Mon Apr 20 21:12:25 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 21:12:25 2009 -0700" }, "message": "IB/mthca: Fix timeout for INIT_HCA and a few other commands\n\nCommands INIT_HCA, CLOSE_HCA, SYS_EN, SYS_DIS, and CLOSE_IB all have 1\nsecond timeouts. For INIT_HCA this causes problems when had more than\n2^18 are QPs configured, since the command takes more than 1 second to\ncomplete.\n\nAll other commands have 60-second timeouts. This patch makes the\nabove commands consistent with the rest of the commands (and with the\nchip documentation).\n\nThis patch is an expansion of a patch from Arthur Kepner\n\u003cakepner@sgi.com\u003e fixing just the INIT_HCA timeout.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "cde9e2f9307dd6af6bf8c5f02e41a6e27da9682a", "tree": "c0503da1e1ee368d5cbd5d37f6309f1798be49e7", "parents": [ "96ac7e88922da6ab33efea87c6b560ba5ab11e75" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Apr 20 17:00:53 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 17:00:53 2009 -0700" }, "message": "RDMA/cxgb3: Don\u0027t zero QP attrs when moving to IDLE\n\nQP attributes must stay initialized when moving back to IDLE. Zeroing\nthem will crash the system in _flush_qp() if the QP is subsequently\nmoved to ERROR and back to IDLE.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3f32eb1185170524a81dadff2e67521585943a53", "tree": "3c44461bed14392563c06b611ccff7e78f80a1cb", "parents": [ "1af9222b5223a7b04102dc8c403fa55fa15d5184" ], "author": { "name": "Don Wood", "email": "donald.e.wood@intel.com", "time": "Mon Apr 20 14:53:00 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 14:53:00 2009 -0700" }, "message": "RDMA/nes: Fix bugs in nes_reg_phys_mr()\n\nThe code incorrectly failed memory registration if the buffer was not\npage aligned. Also, the length field is mangled causing the hardware\nto think the registration is much larger than it really is.\n\nThe fix is to remove the page alignment restriction as well the\nincorrect length adjustment. Also make sure that all buffers after\nthe first start at a page boundary, and all buffers except the last\nend on a page boundary.\n\nSigned-off-by: Don Wood \u003cdonald.e.wood@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1af9222b5223a7b04102dc8c403fa55fa15d5184", "tree": "67a94d7fa273da026dc3601cc15f95e7a74660f4", "parents": [ "a939b96cccdb65df80a52447ec8e4a6d79c56dbb" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Mon Apr 20 14:50:36 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 14:50:36 2009 -0700" }, "message": "RDMA/nes: Fix compiler warning at nes_verbs.c:1955\n\nInitialize pbl_count_256 to 0 to get rid of the warning:\n\n drivers/infiniband/hw/nes/nes_verbs.c: In function \u0027nes_reg_mr\u0027:\n drivers/infiniband/hw/nes/nes_verbs.c:1955: warning: \u0027pbl_count_256\u0027 may be used uninitialized in this function\n\nReported-by: Roland Dreier \u003crdreier@cisco.com\u003e\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e028cc55cc5c90a1c57eefe560a0cbb4df1fed14", "tree": "b1ef203c107a4589536a089737ae550e0b18451b", "parents": [ "a939b96cccdb65df80a52447ec8e4a6d79c56dbb" ], "author": { "name": "Yossi Etigin", "email": "yosefe@voltaire.com", "time": "Mon Apr 20 13:58:08 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 13:58:08 2009 -0700" }, "message": "IPoIB: Disable NAPI while CQ is being drained\n\nIf NAPI is enabled while IPoIB\u0027s CQ is being drained, it creates a\nrace on priv-\u003eibwc between ipoib_poll() and ipoib_drain_cq(), leading\nto memory corruption.\n\nThe solution is to enable/disable NAPI in ipoib_ib_dev_{open/stop}()\ninstead of in ipoib_{open/stop}(), and sync NAPI on the INITIALIZED\nflag instead on the ADMIN_UP flag. This way NAPI will be disabled when\nipoib_drain_cq() is called.\n\nThis fixes \u003chttps://bugs.openfabrics.org/show_bug.cgi?id\u003d1587\u003e.\n\nSigned-off-by: Yossi Etigin \u003cyosefe@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "96ac7e88922da6ab33efea87c6b560ba5ab11e75", "tree": "da84d3d855280ba138108249b0dfa22cae5ac8bb", "parents": [ "a939b96cccdb65df80a52447ec8e4a6d79c56dbb" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Apr 20 13:53:15 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Apr 20 13:53:15 2009 -0700" }, "message": "RDMA/cxgb3: Adjust ORD/IRD (if needed) for peer2peer connections\n\nNFS/RDMA currently fails to set up connections if peer2peer is on.\nThis is due to the fact that the NFS/RDMA client sets its ORD to 0.\n\nIf peer2peer is set, make sure the active side ORD is \u003e\u003d 1 and the\npassive side IRD is \u003e\u003d1.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0534c8cb5c8a8a954751fa01eef7831a475a9ec5", "tree": "1ba3037634378966a78a35bbdec66b384fbff20d", "parents": [ "54f93b74cf8c575bb89f9bf32a40a39b83ace1e9", "07306c0b98a9a0399bff1b91229b4c3b55912189" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 09 16:42:26 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 09 16:42:26 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n RDMA/nes: Add support for new SFP+ PHY\n RDMA/nes: Add wide_ppm_offset parm for switch compatibility\n RDMA/nes: Fix SFP+ PHY initialization\n RDMA/nes: Fix nes_nic_cm_xmit() error handling\n RDMA/nes: Fix error handling issues\n RDMA/nes: Fix incorrect casts on 32-bit architectures\n IPoIB: Document newish features\n RDMA/cma: Create cm id even when IB port is down\n RDMA/cma: Use rate from IPoIB broadcast when joining IPoIB multicast groups\n IPoIB: Avoid free_netdev() BUG when destroying a child interface\n mlx4_core: Don\u0027t leak mailbox for SET_PORT on Ethernet ports\n RDMA/cxgb3: Release dependent resources only when endpoint memory is freed.\n RDMA/cxgb3: Handle EEH events\n IB/mlx4: Use pgprot_writecombine() for BlueFlame pages\n" }, { "commit": "07306c0b98a9a0399bff1b91229b4c3b55912189", "tree": "95251e31a7a67f9b894c9cb10be56f1ec7383e7e", "parents": [ "d2ca39f262806aa2f035f680a14aa55ff9e3d889", "874d8df5ed6e36fed07b524c266f6a96dd6d10d9", "6a3335b43342b42dd6c69b4bbbde15d622cb49ca", "352b09edd7fa8145bfc9e5db0cc0fed971b69440", "4303565df4eb425851ddd22136fec69bdfeede61" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:28:21 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:28:21 2009 -0700" }, "message": "Merge branches \u0027cma\u0027, \u0027cxgb3\u0027, \u0027ipoib\u0027, \u0027mlx4\u0027 and \u0027nes\u0027 into for-next\n" }, { "commit": "4303565df4eb425851ddd22136fec69bdfeede61", "tree": "c6ec136b3cfc7cff41c5e91735954394cfd0a3a5", "parents": [ "a4849fc157cdbe4fb68cfe37e7222697f003deb5" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Wed Apr 08 14:27:56 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:27:56 2009 -0700" }, "message": "RDMA/nes: Add support for new SFP+ PHY\n\nAdd new register settings for new SFP+ PHY/firmware.\nAdd new PHY to to nes_netdev_get/set_settings.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a4849fc157cdbe4fb68cfe37e7222697f003deb5", "tree": "315f1f55d5167b8006cf40d175b6eaae6914d43c", "parents": [ "1b9493248cf5e9f1ecc045488100cbf3ccd91be1" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Wed Apr 08 14:27:18 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:27:18 2009 -0700" }, "message": "RDMA/nes: Add wide_ppm_offset parm for switch compatibility\n\nWe have observed unstable link with a new BNT switch.\n\nAdd wide_ppm_offset parameter to allow the user to control the clock\nppm offset on the CX4 interface for better compatibility. Default is\n100ppm, setting it to 1 will increase it to 300ppm. Change default\nSerDes1 reference clock to external source.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1b9493248cf5e9f1ecc045488100cbf3ccd91be1", "tree": "b52f5aa52452130048dba841b0302ad7dee89f47", "parents": [ "5962c2c8036b4dcf10ec6c481be656ae4700b664" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Wed Apr 08 14:27:09 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:27:09 2009 -0700" }, "message": "RDMA/nes: Fix SFP+ PHY initialization\n\nSFP+ PHY initialization has very long delays, incorrect settings for\ndirect attach copper cables, and inconsistent link detection.\n\nAdjust delays to the minimum required by the PHY. Worst case is now\nless than 4 seconds. Add new register settings for direct attach\ncables. Change link detection logic to use two new registers for more\nconsistent link state detection. Reorganize code to shorten line\nlength.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5962c2c8036b4dcf10ec6c481be656ae4700b664", "tree": "a20bfcbb93e52f7a1dc161d7c6333ceadc5ba046", "parents": [ "79fc3d7410c861c8ced5b81a5c3759f6bbf891dc" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Wed Apr 08 14:23:55 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:23:55 2009 -0700" }, "message": "RDMA/nes: Fix nes_nic_cm_xmit() error handling\n\nWe are getting crash or hung situation when we are running network\ncable pull tests during RDMA traffic.\n\nIn schedule_nes_timer(), we return an error if nes_nic_cm_xmit()\nreturns failure. This is changed to success as skb is being put on\nthe timer routines to be processed later. In send_syn() case, we are\nindicating connect failure once from nes_connect() and the other when\nthe rexmit retries expires.\n\nThe other issue is skb-\u003eusers which we are incrementing before calling\nnes_nic_cm_xmit() which calls dev_queue_xmit() but in case of failure\nwe are decrementing the skb-\u003eusers at the same time putting the skb on\nthe rexmit path. Even if dev_queue_xmit() fails, the skb-\u003eusers is\ndecremented already. We are removing the decrement of skb-\u003eusers in\ncase of failure from both schedule_nes_timer() as well as from\nnes_cm_timer_tick().\n\nThere is also extra check in nes_cm_timer_tick() for rexmit failure\nwhich does a break from the loop is removed. This causes problem as\nthe other nodes have their cm_node-\u003eref_count incremented and are not\nprocessed.\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "79fc3d7410c861c8ced5b81a5c3759f6bbf891dc", "tree": "99c16570cd52c43d65bc9ca4e60f3c4e34d4ae7e", "parents": [ "7a5efb62f6ae366cefac6be475434906c5061e15" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Wed Apr 08 14:22:20 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:22:20 2009 -0700" }, "message": "RDMA/nes: Fix error handling issues\n\nFix issues found by static code analysis:\n\n(1) Check if cm_node was successfully created for loopback connection.\n\n(2) schedule_nes_timer() does not free up allocated memory after\n encountering an error. There is a WARN_ON() for this condition.\n\n(3) there is a cm_node-\u003efreed flag which is set but not used.\n\nReported-by: Dan Carpenter \u003cerror27@gmail.com\u003e\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7a5efb62f6ae366cefac6be475434906c5061e15", "tree": "834615c1a5928d196aec846cfe2742cea168a78c", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a" ], "author": { "name": "Don Wood", "email": "donald.e.wood@intel.com", "time": "Wed Apr 08 14:21:02 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 14:21:02 2009 -0700" }, "message": "RDMA/nes: Fix incorrect casts on 32-bit architectures\n\nThe were some incorrect casts to unsigned long that caused 64-bit values\nto be truncated on 32-bit architectures and made the driver pass invalid\nadresses and lengths to the hardware. The problems were primarily seen\nwith kernels with highmem configured but some could show up in\nnon-highmem kernels, too.\n\nSigned-off-by: Don Wood \u003cdonald.e.wood@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d2ca39f262806aa2f035f680a14aa55ff9e3d889", "tree": "e58c94b8c6e02b72565daff9c19d156ac0f02655", "parents": [ "84adeee9aaa0d81712de1e0ea74caed3398e4a1d" ], "author": { "name": "Yossi Etigin", "email": "root@voltaire.com", "time": "Wed Apr 08 13:42:33 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 08 13:42:33 2009 -0700" }, "message": "RDMA/cma: Create cm id even when IB port is down\n\nWhen doing rdma_resolve_addr(), if the relevant IB port is down, the\nfunction fails and the cm_id is not bound to the correct device.\nTherefore, application does not have a device handle and cannot wait\nfor the port to become active. The function fails because the\nunderlying IPoIB interface is not joined to the broadcast group and\ntherefore the SA does not have a multicast record to take a Q_Key\nfrom.\n\nThe fix is to use lazy Q_Key resolution - cma_set_qkey() will set\nid_priv-\u003eqkey if it was not set, and will be called just before the\nQ_Key is really required.\n\nSigned-off-by: Yossi Etigin \u003cyosefe@voltaire.com\u003e\nAcked-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "284901a90a9e0b812ca3f5f852cbbfb60d10249d", "tree": "06c1b5a0f83c90cfb662f756e7781977ce739ce8", "parents": [ "6afd142fd0dfba497246d0fab236c20a7b4bf778" ], "author": { "name": "Yang Hongyang", "email": "yanghy@cn.fujitsu.com", "time": "Mon Apr 06 19:01:15 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 07 08:31:11 2009 -0700" }, "message": "dma-mapping: replace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32)\n\nReplace all DMA_32BIT_MASK macro with DMA_BIT_MASK(32)\n\nSigned-off-by: Yang Hongyang\u003cyanghy@cn.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6a35528a8346f6e6fd32ed7e51f04d1fa4ca2c01", "tree": "9caaf8645b573687bbcf3a16b5aa7dd233fed46e", "parents": [ "8a59f5d2526593c6bc1a0754c3a16ccc9ed41ce3" ], "author": { "name": "Yang Hongyang", "email": "yanghy@cn.fujitsu.com", "time": "Mon Apr 06 19:01:13 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 07 08:31:10 2009 -0700" }, "message": "dma-mapping: replace all DMA_64BIT_MASK macro with DMA_BIT_MASK(64)\n\nReplace all DMA_64BIT_MASK macro with DMA_BIT_MASK(64)\n\nSigned-off-by: Yang Hongyang\u003cyanghy@cn.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "84adeee9aaa0d81712de1e0ea74caed3398e4a1d", "tree": "2493682afc1efb9053c60883e68a4e30ded2eb38", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a" ], "author": { "name": "Yossi Etigin", "email": "yosefe@Voltaire.COM", "time": "Wed Apr 01 13:55:32 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Apr 01 13:55:32 2009 -0700" }, "message": "RDMA/cma: Use rate from IPoIB broadcast when joining IPoIB multicast groups\n\nWhen joining an IPoIB multicast group, use the same rate as in the\nbroadcast group. Otherwise, if the RDMA CM creates this group before\nIPoIB does, it might get a different rate. This will cause IPoIB to\nfail joining to the same group later on, because IPoIB uses strict\nrate selection.\n\nSigned-off-by: Yossi Etigin \u003cyosefe@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "edb5abb1e2a84fd8802a3577d95eac84fe1405ab", "tree": "a1337e312058581acada61502ccee9ff99c28f96", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 31 10:22:32 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 31 10:22:32 2009 -0700" }, "message": "IPoIB: Avoid free_netdev() BUG when destroying a child interface\n\nWe have to release the RTNL before calling free_netdev() so that the\ndevice state has a chance to become NETREG_UNREGISTERED. Otherwise\nwhen removing a child interface, we hit the BUG() that tests the\ndevice state in free_netdev().\n\nReported-by: Yossi Etigin \u003cyosefe@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "874d8df5ed6e36fed07b524c266f6a96dd6d10d9", "tree": "22fe735a0a00fda013e8514bdc00a5a31e612b77", "parents": [ "04b5d028f50ff05a8f9ae049ee71f8fdfcf1f5de" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Mar 30 08:37:59 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 30 08:37:59 2009 -0700" }, "message": "RDMA/cxgb3: Release dependent resources only when endpoint memory is freed.\n\nThe cxgb3 l2t entry, hwtid, and dst entry were being released before\nall the iwch_ep references were released. This can cause a crash in\nt3_l2t_send_slow() and other places where the l2t entry is used.\n\nThe fix is to defer releasing these resources until all endpoint\nreferences are gone.\n\nDetails:\n\n- move flags field to the iwch_ep_common struct.\n- add a flag indicating resources are to be released.\n- release resources at endpoint free time instead of close/abort time.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "04b5d028f50ff05a8f9ae049ee71f8fdfcf1f5de", "tree": "a6363d175d07995e0c2b32e035fb16beeca0772a", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Mar 30 08:37:56 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 30 08:37:56 2009 -0700" }, "message": "RDMA/cxgb3: Handle EEH events\n\n- wrap calls into cxgb3 and fail them if we\u0027re in the middle\n of a PCI EEH event.\n\n- correctly unwind and release endpoint and other resources when\n we are in an EEH event.\n\n- dispatch IB_EVENT_DEVICE_FATAL event when cxgb3 notifies iw_cxgb3 of\n a fatal error.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e1d60ec6699f19b760df8261e922ae236ea7bb31", "tree": "6a28185f550d78bf36cf5b7e7d411f204d0a2ad1", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a" ], "author": { "name": "Roland Dreier", "email": "rdreier@cisco.com", "time": "Mon Mar 30 08:31:05 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 30 08:31:05 2009 -0700" }, "message": "IB/mlx4: Use pgprot_writecombine() for BlueFlame pages\n\nThe PAT work on x86 has finally made pgprot_writecombine() a usable API\nfor modular drivers. As the comment indicates, this is exactly what we\nwant to use in mlx4_ib to map BlueFlame pages up to userspace, since\nusing WC for these pages improves small message latency significantly.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d54b3538b0bfb31351d02d1669d4a978d2abfc5f", "tree": "5ce539ecba525b30bbfb1c46c55487099264947e", "parents": [ "5d80f8e5a9dc9c9a94d4aeaa567e219a808b8a4a", "af50bb993dfa673cf21ab812efe620d7e0c36319" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 28 13:30:43 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 28 13:30:43 2009 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (119 commits)\n [SCSI] scsi_dh_rdac: Retry for NOT_READY check condition\n [SCSI] mpt2sas: make global symbols unique\n [SCSI] sd: Make revalidate less chatty\n [SCSI] sd: Try READ CAPACITY 16 first for SBC-2 devices\n [SCSI] sd: Refactor sd_read_capacity()\n [SCSI] mpt2sas v00.100.11.15\n [SCSI] mpt2sas: add MPT2SAS_MINOR(221) to miscdevice.h\n [SCSI] ch: Add scsi type modalias\n [SCSI] 3w-9xxx: add power management support\n [SCSI] bsg: add linux/types.h include to bsg.h\n [SCSI] cxgb3i: fix function descriptions\n [SCSI] libiscsi: fix possbile null ptr session command cleanup\n [SCSI] iscsi class: remove host no argument from session creation callout\n [SCSI] libiscsi: pass session failure a session struct\n [SCSI] iscsi lib: remove qdepth param from iscsi host allocation\n [SCSI] iscsi lib: have lib create work queue for transmitting IO\n [SCSI] iscsi class: fix lock dep warning on logout\n [SCSI] libiscsi: don\u0027t cap queue depth in iscsi modules\n [SCSI] iscsi_tcp: replace scsi_debug/tcp_debug logging with iscsi conn logging\n [SCSI] libiscsi_tcp: replace tcp_debug/scsi_debug logging with session/conn logging\n ...\n" }, { "commit": "7c757eb9f804782fb39d0ae2c1a88ffb9309138e", "tree": "99caa29ce033e4ba8b14687022d5da4de3b76232", "parents": [ "e2ab3dff9d515ef69ac7c245b5ad1e348f2106be" ], "author": { "name": "Roland Dreier", "email": "rdreier@cisco.com", "time": "Thu Mar 26 16:25:49 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 17:00:25 2009 -0700" }, "message": "RDMA/nes: Fix mis-merge\n\nWhen net-next and infiniband were merged upstream, each branch deleted\none of a pair of adjacent lines from nes_nic.c, but when Linus fixed the\nconflict up, he brought back both of the lines. Fix up to the intended\nfinal tree state.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nAcked-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6671de344cd661453bbee3cfde5574a974332436", "tree": "826873ae0180eea53b8fcc6e41938b67c656ba3f", "parents": [ "831576fe40f4175e0767623cffa4aeb28157943a", "7c526e1fef8d604a9ec022d9145bba5dbfe40a11" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 16:05:42 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 16:05:42 2009 -0700" }, "message": "Merge branch \u0027timers-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip\n\n* \u0027timers-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (26 commits)\n posix timers: fix RLIMIT_CPU \u0026\u0026 fork()\n time: ntp: fix bug in ntp_update_offset() \u0026 do_adjtimex(), fix\n time: ntp: clean up second_overflow()\n time: ntp: simplify ntp_tick_adj calculations\n time: ntp: make 64-bit constants more robust\n time: ntp: refactor do_adjtimex() some more\n time: ntp: refactor do_adjtimex()\n time: ntp: fix bug in ntp_update_offset() \u0026 do_adjtimex()\n time: ntp: micro-optimize ntp_update_offset()\n time: ntp: simplify ntp_update_offset_fll()\n time: ntp: refactor and clean up ntp_update_offset()\n time: ntp: refactor up ntp_update_frequency()\n time: ntp: clean up ntp_update_frequency()\n time: ntp: simplify the MAX_TICKADJ_SCALED definition\n time: ntp: simplify the second_overflow() code flow\n time: ntp: clean up kernel/time/ntp.c\n x86: hpet: stop HPET_COUNTER when programming periodic mode\n x86: hpet: provide separate functions to stop and start the counter\n x86: hpet: print HPET registers during setup (if hpet\u003dverbose is used)\n time: apply NTP frequency/tick changes immediately\n ...\n" }, { "commit": "13220a94d35708d5378114e96ffcc88d0a74fe99", "tree": "be6530677d5f9536c7211e05ba012923e4c0b307", "parents": [ "8690d8a9f6c2d5728a9c9f68231f1bb4de109e3a", "08abe18af1f78ee80c3c3a5ac47c3e0ae0beadf6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 15:54:36 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 15:54:36 2009 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1750 commits)\n ixgbe: Allow Priority Flow Control settings to survive a device reset\n net: core: remove unneeded include in net/core/utils.c.\n e1000e: update version number\n e1000e: fix close interrupt race\n e1000e: fix loss of multicast packets\n e1000e: commonize tx cleanup routine to match e1000 \u0026 igb\n netfilter: fix nf_logger name in ebt_ulog.\n netfilter: fix warning in ebt_ulog init function.\n netfilter: fix warning about invalid const usage\n e1000: fix close race with interrupt\n e1000: cleanup clean_tx_irq routine so that it completely cleans ring\n e1000: fix tx hang detect logic and address dma mapping issues\n bridge: bad error handling when adding invalid ether address\n bonding: select current active slave when enslaving device for mode tlb and alb\n gianfar: reallocate skb when headroom is not enough for fcb\n Bump release date to 25Mar2009 and version to 0.22\n r6040: Fix second PHY address\n qeth: fix wait_event_timeout handling\n qeth: check for completion of a running recovery\n qeth: unregister MAC addresses during recovery.\n ...\n\nManually fixed up conflicts in:\n\tdrivers/infiniband/hw/cxgb3/cxio_hal.h\n\tdrivers/infiniband/hw/nes/nes_nic.c\n" }, { "commit": "39b566eedbe9e35d38502cc5e62ef7abf1aff9c9", "tree": "114d963f3eb4e8aff401ed74ae0429aefc55c9fd", "parents": [ "39f15003c7b268e4199d5ddce60a6944a74a14b7", "09f98bafea792644f2dea39eb080aa57d854f5b3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 15:47:08 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 26 15:47:08 2009 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (30 commits)\n RDMA/cxgb3: Enforce required firmware\n IB/mlx4: Unregister IB device prior to CLOSE PORT command\n mlx4_core: Add link type autosensing\n mlx4_core: Don\u0027t perform SET_PORT command for Ethernet ports\n RDMA/nes: Handle MPA Reject message properly\n RDMA/nes: Improve use of PBLs\n RDMA/nes: Remove LLTX\n RDMA/nes: Inform hardware that asynchronous event has been handled\n RDMA/nes: Fix tmp_addr compilation warning\n RDMA/nes: Report correct vendor_id and vendor_part_id\n RDMA/nes: Update copyright to new legal entity and year\n RDMA/nes: Account for freed PBL after HW operation\n IB: Remove useless ibdev_is_alive() tests from sysfs code\n IB/sa_query: Fix AH leak due to update_sm_ah() race\n IB/mad: Fix ib_post_send_mad() returning 0 with no generate send comp\n IB/mad: initialize mad_agent_priv before putting on lists\n IB/mad: Fix null pointer dereference in local_completions()\n IB/mad: Fix RMPP header RRespTime manipulation\n IB/iser: Remove hard setting of path MTU\n mlx4_core: Add device IDs for MT25458 10GigE devices\n ...\n" }, { "commit": "08abe18af1f78ee80c3c3a5ac47c3e0ae0beadf6", "tree": "2be39bf8942edca1bcec735145e144a682ca9cd3", "parents": [ "f0de70f8bb56952f6e016a65a8a8d006918f5bf6", "0384e2959127a56d0640505d004d8dd92f9c29f5" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 26 15:23:24 2009 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Mar 26 15:23:24 2009 -0700" }, "message": "Merge branch \u0027master\u0027 of /home/davem/src/GIT/linux-2.6/\n\nConflicts:\n\tdrivers/net/wimax/i2400m/usb-notif.c\n" }, { "commit": "7c526e1fef8d604a9ec022d9145bba5dbfe40a11", "tree": "0234bf6c439cb0b91c69ac0eed28c526d05cbf2f", "parents": [ "e8684605adf87d8e65f64ce7b0274d60a05085bd", "74019224ac34b044b44a31dd89a54e3477db4896", "a2a5ac8650b570bea3cb3614f77739dcd07d6632", "37bebc70d7ad4144c571d74500db3bb26ec0c0eb" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Mar 26 15:45:52 2009 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Thu Mar 26 15:45:52 2009 +0100" }, "message": "Merge branches \u0027timers/new-apis\u0027, \u0027timers/ntp\u0027 and \u0027timers/urgent\u0027 into timers/core\n" }, { "commit": "09f98bafea792644f2dea39eb080aa57d854f5b3", "tree": "cb2261414a0206429e2c27008c45f50335d9d119", "parents": [ "d1fbe04eee32ed2642cff139b8592866f1d43f41", "9c3da0991754d480328eeaa2b90cb231a1cea9b6", "e538052746d570c874650a24eed89fca6e4c93dc", "71d98b4628ee869d62814f6d8607d76cab4b9ec5", "1aedb7721f05461f777fdee25b50d8a168c425ed", "4780c1953f9bef07365b13af01ae4e8238ecd3de", "6b708b3dde0ab3a10a0eea7774c1d6482f32f587", "a6a47771b113be8e694aedd80f66ea94d05bd8df", "270b8b85134c299799dddec624ceeb5671330131", "9d5ab13325d0bb855cf856946c140a68ceb20e32", "6432f366842c78d17b2df974c68f39cb5794f2c2" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 24 20:44:41 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 24 20:44:41 2009 -0700" }, "message": "Merge branches \u0027cxgb3\u0027, \u0027endian\u0027, \u0027ipath\u0027, \u0027ipoib\u0027, \u0027iser\u0027, \u0027mad\u0027, \u0027misc\u0027, \u0027mlx4\u0027, \u0027mthca\u0027, \u0027nes\u0027 and \u0027sysfs\u0027 into for-next\n" }, { "commit": "d1fbe04eee32ed2642cff139b8592866f1d43f41", "tree": "b3a934fd0415f794fe7acca9bfbf9905e5967f38", "parents": [ "426328963078f644c7194403a308588cf684d4c6" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Mar 24 20:44:18 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 24 20:44:18 2009 -0700" }, "message": "RDMA/cxgb3: Enforce required firmware\n\nThe cxgb3 NIC driver can handle more firmware versions than iw_cxgb3,\nand since commit 8207befa (\"cxgb3: untie strict FW matching\") cxgb3\nwill load with firmware versions that iw_cxgb3 can\u0027t handle. The FW\nmajor number indicates a specific interface between the FW and\niw_cxgb3. Thus if the major number of the running firmware does not\nmatch the required version compiled into iw_cxgb3, then iw_cxgb3 must\nnot register that device.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "fe8114e8e1d15ba07ddcaebc4741957a1546f307", "tree": "ad557d7d4183497ff2537bf80f41cbd3b1e9c275", "parents": [ "d0929553bebcac828b612e7d6d239559e08feaf4" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@vyatta.com", "time": "Fri Mar 20 19:35:32 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 21 19:19:14 2009 -0700" }, "message": "infiniband: convert ipoib to net_device_ops\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d0929553bebcac828b612e7d6d239559e08feaf4", "tree": "e6d2e4561d26455c81b84c093487b72b9e0721d2", "parents": [ "687c75dcf342f71329bd193af553e96a29581238" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@vyatta.com", "time": "Fri Mar 20 19:35:31 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 21 19:19:13 2009 -0700" }, "message": "infiniband: convert nes driver to net_device_ops\n\nAlso, removed unnecessary memset() since alloc_netdev returns\nzeroed memory.\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "687c75dcf342f71329bd193af553e96a29581238", "tree": "44486e047227112d61c55e8597cc1e45b2e51d85", "parents": [ "dde09758557120cb71fb760cfeaed1b8e27209ef" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@vyatta.com", "time": "Fri Mar 20 19:35:30 2009 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 21 19:19:13 2009 -0700" }, "message": "infiniband: convert c2 to net_device_ops\n\nConvert this driver to new net_device_ops infrastructure.\nAlso use default net_device get-stats infrastructure\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nReviewed-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a6a47771b113be8e694aedd80f66ea94d05bd8df", "tree": "cd301081f57a9813e0eae4e7b58c9351e13241b2", "parents": [ "27bf91d6a0d5a9c7224e8687754249bba67dd4cf" ], "author": { "name": "Yevgeny Petrilin", "email": "yevgenyp@mellanox.co.il", "time": "Wed Mar 18 19:49:54 2009 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Mar 18 19:49:54 2009 -0700" }, "message": "IB/mlx4: Unregister IB device prior to CLOSE PORT command\n\nAccording to the ConnectX programmer\u0027s reference manual, all\noperations should be stopped, all QPs should be torn down and all WQEs\nflushed before the CLOSE_PORT command is invoked. In some cases\nreversing the order of operations (as implemented now) could cause\na loss of completions.\n\nSigned-off-by: Yevgeny Petrilin \u003cyevgenyp@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5e7facb77ff4b6961d936773fb1f175f7abf76b7", "tree": "d94349cb57a1ef07528e23ac043ca0e1a3eeceb3", "parents": [ "40a06e755d8524cd0b24f795e8bdce5ad19fc41b" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Thu Mar 05 14:46:06 2009 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Mar 13 15:29:39 2009 -0500" }, "message": "[SCSI] iscsi class: remove host no argument from session creation callout\n\nWe do not need to have llds set the host no for the session\u0027s\nparent, because we know the session\u0027s parent is going to be\nthe host. This removes it from the session creation callback\nand converts the drivers.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "4d1083509a69a36cc1394f188b7b8956e5526a16", "tree": "7afdc7d1e1ae88ee5199d57d3fc8952f985620b7", "parents": [ "32ae763e3fce4192cd008956a340353a2e5c3192" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Thu Mar 05 14:46:04 2009 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Mar 13 15:28:55 2009 -0500" }, "message": "[SCSI] iscsi lib: remove qdepth param from iscsi host allocation\n\nThe qdepth setting was useful when we needed libiscsi to verify\nthe setting. Now we just need to make sure if older tools\npassed in zero then we need to set some default.\n\nSo this patch just has us use the sht-\u003ecmd_per_lun or if\nfor LLD does a host per session then we can set it on per\nhost basis.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "32ae763e3fce4192cd008956a340353a2e5c3192", "tree": "af13b359c4e54fe60c8eb76c8927b41468dcbe57", "parents": [ "06d25af4edb60f9e9c7e74d342a6963a32e3392f" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Thu Mar 05 14:46:03 2009 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Mar 13 15:28:37 2009 -0500" }, "message": "[SCSI] iscsi lib: have lib create work queue for transmitting IO\n\nWe were using the shost work queue which ended up being\na little akward since all iscsi hosts need a thread for\nscanning, but only drivers hooked into libiscsi need\na workqueue for transmitting. So this patch moves the\nxmit workqueue to the lib.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "e28f3d5b51ed07d822f135cd941b01e2d485270e", "tree": "e64637e1e4f3bf6921f91729b6f299165929d6df", "parents": [ "c93f87c727ad4e6a5d94cfab219b1492ccc5ca5e" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Thu Mar 05 14:46:01 2009 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Mar 13 15:28:06 2009 -0500" }, "message": "[SCSI] libiscsi: don\u0027t cap queue depth in iscsi modules\n\nThere is no need to cap the queue depth in the modules. We set\nthis in userspace and can do that there. For performance testing\nwith ram based targets, this is helpful since we can have very\nhigh queue depths.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "48a237a26db0a31404c83a88e984b37a30ddcf5a", "tree": "37ca02f54e1df67694d0f115b2592ff25955d2af", "parents": [ "091e6dbec966d0727ae7fb2d5a8a2b8ace09a02e" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Thu Mar 05 14:45:57 2009 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Fri Mar 13 15:26:51 2009 -0500" }, "message": "[SCSI] iser: have iser use its own logging\n\niser has its own logging inrfastrucutre. Convert it to use\nit instead of libiscsi.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "c12e56ef6951f4fce1afe9ef6aab9243ea9a9b04", "tree": "677319ae34074efbf98a87d9496a7d5cb47c0524", "parents": [ "7ef0d7377cb287e08f3ae94cebc919448e1f5dff" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Thu Mar 12 14:34:59 2009 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 12 16:21:41 2009 -0700" }, "message": "RDMA/nes: Don\u0027t allow userspace QPs to use STag zero\n\nSTag zero is a special STag that allows consumers to access any bus\naddress without registering memory. The nes driver unfortunately\nallows STag zero to be used even with QPs created by unprivileged\nuserspace consumers, which means that any process with direct verbs\naccess to the nes device can read and write any memory accessible to\nthe underlying PCI device (usually any memory in the system). Such\naccess is usually given for cluster software such as MPI to use, so\nthis is a local privilege escalation bug on most systems running this\ndriver.\n\nThe driver was using STag zero to receive the last streaming mode\ndata; to allow STag zero to be disabled for unprivileged QPs, the\ndriver now registers a special MR for this data.\n\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9d5ab13325d0bb855cf856946c140a68ceb20e32", "tree": "c226cba67009db77052aceb1c7bf885da63fa4b8", "parents": [ "0145f341a951b998d6d0fa38992a42d2a90b5bab" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Fri Mar 06 15:15:01 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:15:01 2009 -0800" }, "message": "RDMA/nes: Handle MPA Reject message properly\n\nWhile doing testing, there are failures as MPA Reject call is not\nhandled. To handle MPA Reject call, following changes are done:\n\n*Handle inbound/outbound MPA Reject response message.\n\tWhen nes_reject() is called for pending MPA request reply,\n\tsend the MPA Reject message to its peer (active\n\tside)cm_node. The peer cm_node (active side) will indicate\n\tReject message event for the pending Connect Request.\n\n*Handle MPA Reject response message for loopback connections and listener.\n\tWhen MPA Request is rejected, check if it is a loopback\n\tconnection and if it is then it will send Reject message event\n\tto its peer loopback node. Also when destroying listener,\n\tcheck if the cm_nodes for that listener are loopback or not.\n\n*Add gracefull connection close with the MPA Reject response message.\n\tSend gracefull close (FIN, FIN ACK..) to terminate the cm_nodes.\n\n*Some code re-org while making the above changes.\n\tRemoved recv_list and recv_list_lock from the cm_node\n\tstructure as there can be only one receive close entry on the\n\ttimer. Also implemented handle_recv_entry() as receive close\n\tentry is processed from both nes_rem_ref_cm_node() as well as\n\tnes_cm_timer_tick().\n\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0145f341a951b998d6d0fa38992a42d2a90b5bab", "tree": "e53b65e3badbaec787ca048aeedd64b6c5199f03", "parents": [ "2869975cfbd58dc6591d8c3ba1f171e7f758be28" ], "author": { "name": "Don Wood", "email": "donald.e.wood@intel.com", "time": "Fri Mar 06 15:15:00 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:15:00 2009 -0800" }, "message": "RDMA/nes: Improve use of PBLs\n\nTwo level 256 byte PBLs was not implemented so the driver could report\nout of memory when in fact there were PBLs still available.\n\nThis solution prefers to use 4KB PBLs over two level 256B PBLs until\nthe number of 4KB PBLs falls below a threshold. At this point the 4KB\nPBL structure is converted to use 256B PBLs which prevents the driver\nfrom running out of 4KB PBLs too quickly.\n\nSigned-off-by: Don Wood \u003cdonald.e.wood@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2869975cfbd58dc6591d8c3ba1f171e7f758be28", "tree": "5a664309bcef41dbc79fd12eb87c63f06369b181", "parents": [ "fd87778cb99429f5e2e041213a5c9c564bbe7b78" ], "author": { "name": "Faisal Latif", "email": "faisal.latif@intel.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "message": "RDMA/nes: Remove LLTX\n\nNETIF_F_LLTX is deprecated. Remove private TX locking from the driver\nand remove the NETIF_F_LLTX feature flag. This also fixes a warning\nin some configs that comes from doing skb_linearize() call in the\nhard_start_xmit method with IRQs disabled (if HIGHMEM is enabled,\nskb_linearize() may end up enabling BHs, which is a no-no if hard IRQs\nare disabled in that context). By getting rid of LLTX, we do not\ndisable IRQs when skb_linearize() is called.\n\nRemove the sq_lock as it is not needed for non-LLTX. Fix ethtool not\nto show the counter for sq_lock.\n\nReported-by: aluno3@poczta.onet.pl\nSigned-off-by: Faisal Latif \u003cfaisal.latif@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "fd87778cb99429f5e2e041213a5c9c564bbe7b78", "tree": "23d5d0e4517a9fe0678989f1506f141d5cfb0277", "parents": [ "7b14ab0b438f20c7d6599985b036bb2864fd2524" ], "author": { "name": "Don Wood", "email": "donald.e.wood@intel.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "message": "RDMA/nes: Inform hardware that asynchronous event has been handled\n\nWhen asynchronous events are processed by software, it is necessary\nto let the hardware know that software has handled the event. This\nfrees up the entry in the asynchronous event queue.\n\nSigned-off-by: Don Wood \u003cdonald.e.wood@intel.com\u003e\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7b14ab0b438f20c7d6599985b036bb2864fd2524", "tree": "97355681c6c8d1b8cd1adfb2abad6a681fc1f786", "parents": [ "b9c367e7e694502085f21d9d2686645798080aa3" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:11 2009 -0800" }, "message": "RDMA/nes: Fix tmp_addr compilation warning\n\nIn find_node(), tmp_addr causes an \"unused variable\" warning when\nINFINIBAND_NES_DEBUG is not defined. It\u0027s only used in a nes_debug()\nand the print does not make sense. So take out the whole thing.\n\nReported-by: Manish Katiyar \u003cmkatiyar@gmail.com\u003e\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b9c367e7e694502085f21d9d2686645798080aa3", "tree": "b64f83ffda2f70f9450f0a80fffcc5234b6aa792", "parents": [ "cd6853d3eb453aee2574521b7ce2cd5a45492a59" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Fri Mar 06 15:12:10 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:10 2009 -0800" }, "message": "RDMA/nes: Report correct vendor_id and vendor_part_id\n\nibv_devinfo displays 0 for vendor_id and vendor_part_id. Fill in OUI\nand device_id for those two fields.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "cd6853d3eb453aee2574521b7ce2cd5a45492a59", "tree": "4e46fd5e81e16e612ea7997824e7e5c9eee5877d", "parents": [ "dae5d13a7e6efcf6e0e00c0febb530b894fa13a3" ], "author": { "name": "Chien Tung", "email": "chien.tin.tung@intel.com", "time": "Fri Mar 06 15:12:10 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:10 2009 -0800" }, "message": "RDMA/nes: Update copyright to new legal entity and year\n\nUpdate copyright to the new legal entity, Intel-NE, Inc., an Intel\ncompany. Update copyright for the new year.\n\nSigned-off-by: Chien Tung \u003cchien.tin.tung@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "dae5d13a7e6efcf6e0e00c0febb530b894fa13a3", "tree": "30a99560bc31bf1f8dcad716024ac503b844c625", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Don Wood", "email": "donald.e.wood@intel.com", "time": "Fri Mar 06 15:12:09 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 06 15:12:09 2009 -0800" }, "message": "RDMA/nes: Account for freed PBL after HW operation\n\nFix occurrences where the software PBL counts were changed before the\nhardware was updated. This bug allowed another thread to overallocate\nthe hardware resources.\n\nAdd proper PBL accounting in case nes_reg_mr() fails.\n\nSigned-off-by: Don Wood \u003cdonald.e.wood@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6432f366842c78d17b2df974c68f39cb5794f2c2", "tree": "1eae97fdf35e8344face61d4eea8ffcc237c7c3d", "parents": [ "9206dff15705267c19f8fed391c4fb95975540a3" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Mar 04 15:22:39 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Mar 04 15:22:39 2009 -0800" }, "message": "IB: Remove useless ibdev_is_alive() tests from sysfs code\n\nSome attribute show functions test ibdev_is_alive() to make sure that\nit\u0027s OK to access device state. However, the sysfs attributes will\nnot be registered until the device is fully initialized, and they\u0027ll\nbe unregistered before anything is torn down, so ibdev_is_alive()\ndoesn\u0027t do anything useful. Remove it.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6b708b3dde0ab3a10a0eea7774c1d6482f32f587", "tree": "fb3f7451940c0d9569bcc10810bbc993c3366535", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Tue Mar 03 14:30:01 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 03 14:30:01 2009 -0800" }, "message": "IB/sa_query: Fix AH leak due to update_sm_ah() race\n\nOur testing uncovered a race condition in ib_sa_event():\n\n\tspin_lock_irqsave(\u0026port-\u003eah_lock, flags);\n\tif (port-\u003esm_ah)\n\t\tkref_put(\u0026port-\u003esm_ah-\u003eref, free_sm_ah);\n\tport-\u003esm_ah \u003d NULL;\n\tspin_unlock_irqrestore(\u0026port-\u003eah_lock, flags);\n\n\tschedule_work(\u0026sa_dev-\u003eport[event-\u003eelement.port_num -\n\t\t\t\t sa_dev-\u003estart_port].update_task);\n\nIf two events occur back-to-back (e.g., client-reregister and LID\nchange), both may pass the spinlock-protected code above before the\nscheduled work updates the port-\u003esm_ah handle. Then if the scheduled\nwork ends up running twice, the second operation will then find a\nnon-NULL port-\u003esm_ah, and will simply overwrite it in update_sm_ah --\nresulting in an AH leak.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4780c1953f9bef07365b13af01ae4e8238ecd3de", "tree": "ce68e77fc5ad410b9c320cf80ad7f76faa1bc7f1", "parents": [ "d9620a4c82c61a91c9313f80ba951c902573c028" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Tue Mar 03 14:22:17 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 03 14:22:17 2009 -0800" }, "message": "IB/mad: Fix ib_post_send_mad() returning 0 with no generate send comp\n\nIf ib_post_send_mad() returns 0, the API guarantees that there will be\na callback to send_buf-\u003emad_agent-\u003esend_handler() so that the sender\ncan call ib_free_send_mad(). Otherwise, the ib_mad_send_buf will be\nleaked and the mad_agent reference count will never go to zero and the\nIB device module cannot be unloaded. The above can happen without\nthis patch if process_mad() returns (IB_MAD_RESULT_SUCCESS |\nIB_MAD_RESULT_CONSUMED).\n\nIf process_mad() returns IB_MAD_RESULT_SUCCESS and there is no agent\nregistered to receive the mad being sent, handle_outgoing_dr_smp()\nreturns zero which causes a MAD packet which is at the end of the\ndirected route to be incorrectly sent on the wire but doesn\u0027t cause a\nhang since the HCA generates a send completion.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d9620a4c82c61a91c9313f80ba951c902573c028", "tree": "d49b92e161f7fc8176e04084b3d981d94ff9e2bd", "parents": [ "1d9bc6d648ece77ffb41c5a577eab81fac5ad4de" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Feb 27 14:44:32 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 27 14:44:32 2009 -0800" }, "message": "IB/mad: initialize mad_agent_priv before putting on lists\n\nThere is a potential race in ib_register_mad_agent() where the struct\nib_mad_agent_private is not fully initialized before it is added to\nthe list of agents per IB port. This means the ib_mad_agent_private\ncould be seen before the refcount, spin locks, and linked lists are\ninitialized. The fix is to initialize the structure earlier.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1d9bc6d648ece77ffb41c5a577eab81fac5ad4de", "tree": "f4694fe940cfd4cef0c1a96fd436984c1b2c6ea5", "parents": [ "7020cb0fe216fdcec246cdc2412614a3190fbb2f" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Feb 27 10:34:30 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 27 10:34:30 2009 -0800" }, "message": "IB/mad: Fix null pointer dereference in local_completions()\n\nhandle_outgoing_dr_smp() can queue a struct ib_mad_local_private\n*local on the mad_agent_priv-\u003elocal_work work queue with\nlocal-\u003emad_priv \u003d\u003d NULL if device-\u003eprocess_mad() returns\nIB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY and\n(!ib_response_mad(\u0026mad_priv-\u003emad.mad) ||\n!mad_agent_priv-\u003eagent.recv_handler).\n\nIn this case, local_completions() will be called with local-\u003emad_priv\n\u003d\u003d NULL. The code does check for this case and skips calling\nrecv_mad_agent-\u003eagent.recv_handler() but recv \u003d\u003d 0 so\nkmem_cache_free() is called with a NULL pointer.\n\nAlso, since recv isn\u0027t reinitialized each time through the loop, it\ncan cause a memory leak if recv should have been zero.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\n" }, { "commit": "1aedb7721f05461f777fdee25b50d8a168c425ed", "tree": "cb2a6de3a29efa68ca229a8281b0d85bb7d95e96", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Fri Feb 27 10:30:46 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 27 10:30:46 2009 -0800" }, "message": "IB/iser: Remove hard setting of path MTU\n\nRemove hard setting of the IB MTU used by iSER\u0027s RC queue-pair to 1K,\nas this was done due to inter-op issues with an old iser target which\nis not used any more.\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9206dff15705267c19f8fed391c4fb95975540a3", "tree": "fc9722069f10675ed77f2dd90e52daef5230be33", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 25 13:27:46 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 25 13:27:46 2009 -0800" }, "message": "IB: Remove sysfs files before unregistering device\n\nMove the ib_device_unregister_sysfs() call from ib_dealloc_device() to\nib_unregister_device(). The old code allows device unregister to\nproceed even if some sysfs files are open, which leaves a window where\nuserspace can open a file before a device is removed but then end up\nreading the file after the device is removed, which leads to various\nkernel crashes either because the device data structure is freed or\nbecause the low-level driver code is gone after module removal.\n\nBy not returning from ib_unregister_device() until after all sysfs\nentries are removed, we make sure that data structures and/or module\ncode is not freed until after all sysfs access is done.\n\nReported-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e538052746d570c874650a24eed89fca6e4c93dc", "tree": "308b1a9829002328cdd91ef913a486b31b6342c2", "parents": [ "71c4512201575c7cc008b364e2e2c75cc7085c26" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Feb 22 20:14:37 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Feb 22 20:14:37 2009 -0800" }, "message": "IB/ipath: Really run work in ipath_release_user_pages_on_close()\n\nipath_release_user_pages_on_close() just allocated a structure to\nschedule work with but just returned (leaking the structure) rather than \nactually doing schedule_work(). Fix the logic to what was intended.\n\nThis was spotted by the Coverity checker (CID 2700).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "71c4512201575c7cc008b364e2e2c75cc7085c26", "tree": "0a8d76992a500aa82e8943c3dd4743b3d3e3d6e0", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Feb 22 20:04:34 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Feb 22 20:04:34 2009 -0800" }, "message": "IB/ipath: Fix memory leak in init_shadow_tids() error path\n\nIf the second vmalloc() fails, the wrong pointer is pased to vfree(), so\nthe first vmalloc() ends up getting leaked.\n\nThis was spotted by the Coverity checker (CID 2709).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "74019224ac34b044b44a31dd89a54e3477db4896", "tree": "bf7b04319c0c4329eb6cef0788737c14d2fa9030", "parents": [ "5955c7a2cfb6a35429adea5dc480002b15ca8cfc" ], "author": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 18 12:23:29 2009 +0100" }, "committer": { "name": "Ingo Molnar", "email": "mingo@elte.hu", "time": "Wed Feb 18 19:26:33 2009 +0100" }, "message": "timers: add mod_timer_pending()\n\nImpact: new timer API\n\nBased on an idea from Martin Josefsson with the help of\nPatrick McHardy and Stephen Hemminger:\n\nintroduce the mod_timer_pending() API which is a mod_timer()\noffspring that is an invariant on already removed timers.\n\n(regular mod_timer() re-activates non-pending timers.)\n\nThis is useful for the networking code in that it can\nallow unserialized mod_timer_pending() timer-forwarding\ncalls, but a single del_timer*() will stop the timer\nfrom being reactivated again.\n\nAlso while at it:\n\n- optimize the regular mod_timer() path some more, the\n timer-stat and a debug check was needlessly duplicated\n in __mod_timer().\n\n- make the exports come straight after the function, as\n most other exports in timer.c already did.\n\n- eliminate __mod_timer() as an external API, change the\n users to mod_timer().\n\nThe regular mod_timer() code path is not impacted\nsignificantly, due to inlining optimizations and due to\nthe simplifications.\n\nBased-on-patch-from: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nAcked-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Patrick McHardy \u003ckaber@trash.net\u003e\nCc: netdev@vger.kernel.org\nCc: Oleg Nesterov \u003coleg@redhat.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n" }, { "commit": "71d98b4628ee869d62814f6d8607d76cab4b9ec5", "tree": "10b4d085aaf23020c052d5e340330d757730d138", "parents": [ "f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Tue Feb 17 14:51:47 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 17 14:51:47 2009 -0800" }, "message": "IPoIB: In unicast_arp_send(), only free newly-created paths\n\nIf path_rec_start() returns error, call path_free() only if the path\nwas newly-created. If we free an existing path whose valid flag was zero,\n(but do not detach it from the list) we cause corruption of the\npath list (of which it is a member), and get a kernel crash.\n\nThe simplest solution is to not free an existing path -- just leave it\nin the list as-is (i.e., with its valid flag cleared).\n\nThanks to Yossi Etigin of Voltaire for identifying the problem flow\nwhich caused the kernel crash.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Moni Shua \u003cmonis@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "426328963078f644c7194403a308588cf684d4c6", "tree": "4f4d22db8b998b0275800c69075cf1471bdc7b9c", "parents": [ "42fb61f02f9bdc476c7a76d3cce0400d989f44c5" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Feb 16 21:23:32 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 16 21:23:32 2009 -0800" }, "message": "RDMA/cxgb3: Remove modulo math from build_rdma_recv()\n\nRemove modulo usage to avoid a divide in the fast path (not all\ngcc versions do strength reduction here).\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "42fb61f02f9bdc476c7a76d3cce0400d989f44c5", "tree": "1b65628292881e77e81f6e1bdc6aed64b38f0646", "parents": [ "900f4c16c338f742b80f3aa500e12ceb017e86af" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Feb 10 16:38:57 2009 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 10 16:38:57 2009 -0800" }, "message": "RDMA/cxgb3: Connection termination fixes\n\nThe poll and flush code needs to handle all send opcodes: SEND,\nSEND_WITH_SE, SEND_WITH_INV, and SEND_WITH_SE_INV.\n\nIgnore TERM indications if the connection already gone.\n\nIgnore HW receive completions if the RQ is empty.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" } ], "next": "900f4c16c338f742b80f3aa500e12ceb017e86af" }