)]}' { "log": [ { "commit": "1b90c137cc2a0e9b813a8ae316827c493c664146", "tree": "2ca251f9875e44f02011bbd0b560ba384262398b", "parents": [ "a4e6d5d3817ebae167e78e5957cd9e624be200c7" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Sat Mar 29 03:10:28 2008 +0000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 30 14:20:24 2008 -0700" }, "message": "trivial endianness annotations: infiniband core\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1f71f50342c6fe4fbdebe63b0fd196972a70e281", "tree": "44e39c73f0e194e3a7df6c5a34e667161b2f0dc4", "parents": [ "8c178beeb20ce3801c4851d41342d0ca32ad292c" ], "author": { "name": "Roland Dreier", "email": "rdreier@cisco.com", "time": "Fri Mar 28 10:28:17 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 28 10:45:32 2008 -0700" }, "message": "RDMA/cxgb3: Program hardware IRD with correct value\n\nBecause of a typo in iwch_accept_cr(), the cxgb3 connection handling\ncode programs the hardware IRD (incoming RDMA read queue depth) with\nthe value that is passed in for the ORD (outgoing RDMA read queue\ndepth). In particular this means that if an application passes in IRD\n\u003e 0 and ORD \u003d 0 (which is a completely sane and valid thing to do for\nan app that expects only incoming RDMA read requests), then the\nhardware will end up programmed with IRD \u003d 0 and the app will fail in\na mysterious way.\n\nFix this by using \"ep-\u003eird\" instead of \"ep-\u003eord\" in the intended place.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f2b2b59b93082d02ee5ec8fda4ed94f498551a47", "tree": "de2f5566fc128dccdfe958f4b9b401989e820dfa", "parents": [ "dba92d3bc49c036056a48661d2d8fefe4c78375a" ], "author": { "name": "Chien Tung", "email": "ctung@neteffect.com", "time": "Thu Mar 20 19:55:30 2008 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Mar 21 13:59:28 2008 -0700" }, "message": "RDMA/nes: Fix MSS calculation on RDMA path\n\nFix the calculation of the MSS for RDMA connections: we need to\nallow space in frames for a VLAN tag too.\n\nSigned-off-by: Chien Tung \u003cctung@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "10313cbb92206450b450e14f2b3f6ccde42d9a34", "tree": "6e10b1066fcb3f3a699960c62535f0cbea240738", "parents": [ "4200406b8fbbf309f4fffb339bd16c4553ae0c30" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Mar 12 07:51:03 2008 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Mar 12 07:51:03 2008 -0700" }, "message": "IPoIB: Allocate priv-\u003etx_ring with vmalloc()\n\nCommit 7143740d (\"IPoIB: Add send gather support\") made struct\nipoib_tx_buf significantly larger, since the mapping member changed\nfrom a single u64 to an array with MAX_SKB_FRAGS + 1 entries. This\nmeans that allocating tx_rings with kzalloc() may fail because there\nis not enough contiguous memory for the new, much bigger size. Fix\nthis regression by allocating the rings with vmalloc() instead.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4200406b8fbbf309f4fffb339bd16c4553ae0c30", "tree": "cb734632dc9fce70535cd873c8c2fa244262a9e6", "parents": [ "b3e2749bf32f61e7beb259eb7cfb066d2ec6ad65" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 18:35:20 2008 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 18:35:20 2008 -0700" }, "message": "IPoIB/cm: Set tx_wr.num_sge in connected mode post_send()\n\nCommit 7143740d (\"IPoIB: Add send gather support\") made it possible\nfor tx_wr.num_sge to be !\u003d 1 -- this happens if send gather support is\nenabled. However, the code in the connected mode post_send() function\nassumes the old invariant, namely that tx_wr.num_sge is always 1. Fix\nthis by explicitly setting tx_wr.num_sge to 1 in the CM post_send().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b3e2749bf32f61e7beb259eb7cfb066d2ec6ad65", "tree": "3fb0ce0dd2c470b00a9bb942c163fd77f2a1dc51", "parents": [ "450bb3875f5f5ab3679823c941d6045d16967370" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Tue Mar 11 16:10:02 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 14:12:03 2008 -0700" }, "message": "IPoIB: Don\u0027t drop multicast sends when they can be queued\n\nWhen set_multicast_list() is called the multicast task is restarted\nand the IPOIB_MCAST_STARTED bit is cleared. As a result for some\nwindow of time, multicast packets are not transmitted nor queued but\nrather dropped by ipoib_mcast_send(). These dropped packets are\npainful in two cases:\n\n - bonding fail-over which both calls set_multicast_list() on the new\n active slave and sends Gratuitous ARP through that slave.\n\n - IP_DROP_MEMBERSHIP code which both calls set_multicast_list() on the\n device and issues IGMP leave.\n\nIn both these cases, depending on the scheduling of the IPoIB\nmulticast task, the packets would be dropped. As a result, in the\nbonding case, the failover would not be detected by the peers until\ntheir neighbour is renewed the neighbour (which takes a few tens of\nseconds). In the IGMP case, the IP router doesn\u0027t get an IGMP leave\nand would only learn on that from further probes on the group (also a\ndelay of at least a few tens of seconds).\n\nFix this by allowing transmission (or queuing) depending on the\nIPOIB_FLAG_OPER_UP flag instead of the IPOIB_MCAST_STARTED flag.\n\nSigned-off-by: Olga Shern \u003colgas@voltaire.com\u003e\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "450bb3875f5f5ab3679823c941d6045d16967370", "tree": "8e92a0d4caf8a625ef5ab37a417897f88a5ee420", "parents": [ "2a049e514b890c8b70b965bbd9f4e3c963af69c9" ], "author": { "name": "Patrick Marchand Latifi", "email": "patrick.latifi@qlogic.com", "time": "Wed Feb 20 19:08:10 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 14:04:35 2008 -0700" }, "message": "IB/ipath: Reset the retry counter for RDMA_READ_RESPONSE_MIDDLE packets\n\nReset the retry counter when we get a good RDMA_READ_RESPONSE_MIDDLE\npacket. This fix will prevent the requester from reporting a retry\nexceeded error too early.\n\nSigned-off-by: Patrick Marchand Latifi \u003cpatrick.latifi@qlogic.com\u003e\n" }, { "commit": "2a049e514b890c8b70b965bbd9f4e3c963af69c9", "tree": "ec090051bcddfd780c20ad51a855df45578302d8", "parents": [ "4cd5060cf7c2207c31e2e368f8a6343355362e51" ], "author": { "name": "Patrick Marchand Latifi", "email": "patrick.latifi@qlogic.com", "time": "Thu Jan 31 00:24:37 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 14:03:54 2008 -0700" }, "message": "IB/ipath: Fix error completion put on send CQ instead of recv CQ\n\nA work completion entry could be placed on the wrong completion\nqueue when an RC QP is placed in the error state.\n\nSigned-off-by: Patrick Marchand Latifi \u003cpatrick.latifi@qlogic.com\u003e\nAcked-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4cd5060cf7c2207c31e2e368f8a6343355362e51", "tree": "aa1a5eb987aa39b06c3f582cb0206ba3f4ee5626", "parents": [ "87d5aed85b2d79e4075ad2ca1449e9b98f657a09" ], "author": { "name": "Patrick Marchand Latifi", "email": "patrick.latifi@qlogic.com", "time": "Fri Jan 18 20:10:48 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 14:02:32 2008 -0700" }, "message": "IB/ipath: Fix RC QP initialization\n\nThis patch fixes the initialization of RC QPs, since we would rely on\nthe queue pair type (ibqp-\u003eqp_type) being set, but this field is only\ninitialized when we return from ipath_create_qp (it is initialized by\nthe user-level verbs library).\n\nThe fix is to not depend on this field to initialize the send and\nthe receive state of the RC QP.\n\nSigned-off-by: Patrick Marchand Latifi \u003cpatrick.latifi@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "87d5aed85b2d79e4075ad2ca1449e9b98f657a09", "tree": "fa8c4f5b110148b913c3f9b5663d7a95358d4fd4", "parents": [ "140277e9a710202608914b5b731948d2769399bc" ], "author": { "name": "Patrick Marchand Latifi", "email": "patrick.latifi@qlogic.com", "time": "Mon Jan 07 23:43:04 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 14:01:14 2008 -0700" }, "message": "IB/ipath: Fix potentially wrong RNR retry counter returned in ipath_query_qp()\n\nThere can be a case where the requester\u0027s rnr retry counter\n(s_rnr_retry) is less than the number of rnr retries allowed per QP\n(s_rnr_retry_cnt). This can happen if the s_rnr_retry counter is being\ndecremented and an ipath_query_qp call is issued during that time frame.\nThe fix is to always return the number of rnr retries allowed per QP\ninstead of the requester\u0027s rnr counter.\n\nFound by code review.\n\nSigned-off-by: Patrick Marchand Latifi \u003cpatrick.latifi@qlogic.com\u003e\nAcked-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "140277e9a710202608914b5b731948d2769399bc", "tree": "d30fa34b17ec1f42508c5d046926ed7a717126c4", "parents": [ "baadac8b10c5ac15ce3d26b68fa266c8889b163f" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Dec 14 01:53:56 2007 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Mar 11 13:58:22 2008 -0700" }, "message": "IB/ipath: Fix IB compliance problems with link state vs physical state\n\nSubnet manager SetPortinfo messages distingush between changing the link\nstate (DOWN, ARM, ACTIVE) and the link physical state (POLL, SLEEP,\nDISABLED). These are somewhat independent commands and affect when link\nwidth and speed changes take effect. Without this patch, a link DOWN\nphysical state NOP command was causing the link width and speed settings\nto take effect which should only happen when the link physical state is\ngoes down (either by a SMP or some link physical error like link errors\nexceeding the threshold).\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d7c1fbd6606085dbf95e47068d6bd2db8a180e38", "tree": "0852af87269facc5f07f08bd20387f183976cca2", "parents": [ "d33ed425c6cc14370d8c418b504328d2c3db58b4" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Mar 04 16:44:52 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 10 21:22:22 2008 -0700" }, "message": "RDMA/iwcm: Don\u0027t access a cm_id after dropping reference\n\ncm_work_handler() can access cm_id_priv after it drops its reference\nby calling iwch_deref_id(), which might cause it to be freed. The fix\nis to look at whether IWCM_F_CALLBACK_DESTROY is set _before_ dropping\nthe reference. Then if it was set, free the cm_id on this thread.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d33ed425c6cc14370d8c418b504328d2c3db58b4", "tree": "a026c1189014d6dc8a972c7584a6150e84679f8e", "parents": [ "9a378270c085080b2f38dee6308de4d8413b5141" ], "author": { "name": "Arne Redlich", "email": "arne.redlich@xiranet.com", "time": "Tue Mar 04 14:11:54 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 10 21:17:51 2008 -0700" }, "message": "IB/iser: Handle iser_device allocation error gracefully\n\n\"iser_device\" allocation failure is \"handled\" with a BUG_ON() right\nbefore dereferencing the NULL-pointer - fix this!\n\nSigned-off-by: Arne Redlich \u003carne.redlich@xiranet.com\u003e\nSigned-off-by: Erez Zilber \u003cerezz@voltaire.com\u003e\n" }, { "commit": "9a378270c085080b2f38dee6308de4d8413b5141", "tree": "347ce131d4b06f11430632ded2b6a257950a6193", "parents": [ "4fa45725df0f00c2bf86a0fc2670e88bfe0ceee7" ], "author": { "name": "Arne Redlich", "email": "arne.redlich@xiranet.com", "time": "Tue Mar 04 14:07:22 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Mar 10 21:15:49 2008 -0700" }, "message": "IB/iser: Fix list iteration bug\n\nThe iteration through the list of \"iser_device\"s during device\nlookup/creation is broken -- it might result in an infinite loop if\nmore than one HCA is used with iSER. Fix this by using\nlist_for_each_entry() instead of the open-coded flawed list iteration\ncode.\n\nSigned-off-by: Arne Redlich \u003carne.redlich@xiranet.com\u003e\nSigned-off-by: Erez Zilber \u003cerezz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4fa45725df0f00c2bf86a0fc2670e88bfe0ceee7", "tree": "37a7cf4f16bcc9ca36ebee018fea015b7110a334", "parents": [ "1bab74e691d3c7845df2342d202c0f1c2344c834" ], "author": { "name": "Jon Mason", "email": "jon@opengridcomputing.com", "time": "Sun Mar 09 13:54:12 2008 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Mar 09 13:54:12 2008 -0700" }, "message": "RDMA/cxgb3: Fix iwch_create_cq() off-by-one error\n\nThe cxbg3 driver is unnecessarily decreasing the number of CQ entries by\none when creating a CQ. This will cause the CQ not to have as many\nentries as requested by the user if the user requests a power of 2 size.\n\nSigned-off-by: Jon Mason \u003cjon@opengridcomputing.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1bab74e691d3c7845df2342d202c0f1c2344c834", "tree": "9b99db64e082944f8892df1916f29a4bf1fda9a8", "parents": [ "331552925d17ffa2f5c676e282d4fd37c852d9e3" ], "author": { "name": "Jon Mason", "email": "jon@opengridcomputing.com", "time": "Fri Feb 29 13:53:18 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 29 13:53:18 2008 -0800" }, "message": "RDMA/cxgb3: Return correct max_inline_data when creating a QP\n\nSet cap.max_inline_data to the actual max inline data that the adapter\nsupport, so that userspace apps see the right value returned.\n\nSigned-off-by: Jon Mason \u003cjon@opengridcomputing.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "331552925d17ffa2f5c676e282d4fd37c852d9e3", "tree": "6dba57eb575399cef6e7498110670209ee99f148", "parents": [ "35fb5340e3de5dff86923eb0cded748c3a6e05e7" ], "author": { "name": "Pete Wyckoff", "email": "pw@osc.edu", "time": "Tue Feb 26 13:27:53 2008 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 29 13:31:48 2008 -0800" }, "message": "IB/fmr_pool: Flush all dirty FMRs from ib_fmr_pool_flush()\n\nCommit a3cd7d90 (\"IB/fmr_pool: ib_fmr_pool_flush() should flush all\ndirty FMRs\") caused a regression for iSER and was reverted in\ne5507736.\n\nThis change attempts to redo the original patch so that all used FMR\nentries are flushed when ib_flush_fmr_pool() is called without\naffecting the normal FMR pool cleaning thread. Simply move used\nentries from the clean list onto the dirty list in ib_flush_fmr_pool()\nbefore letting the cleanup thread do its job.\n\nSigned-off-by: Pete Wyckoff \u003cpw@osc.edu\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "35fb5340e3de5dff86923eb0cded748c3a6e05e7", "tree": "49a0c8b339366ae134f02f1eede6a4a0804bd62e", "parents": [ "84ba284cd78c130818e2de53150f39b92504593b" ], "author": { "name": "Pete Wyckoff", "email": "pw@osc.edu", "time": "Tue Feb 26 13:27:31 2008 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 29 13:29:19 2008 -0800" }, "message": "Revert \"IB/fmr_pool: ib_fmr_pool_flush() should flush all dirty FMRs\"\n\nThis reverts commit a3cd7d9070be417a21905c997ee32d756d999b38.\n\nThe original commit breaks iSER reliably, making it complain:\n\n iser: iser_reg_page_vec:ib_fmr_pool_map_phys failed: -11\n\nThe FMR cleanup thread runs ib_fmr_batch_release() as dirty entries\nbuild up. This commit causes clean but used FMR entries also to be\npurged. During that process, another thread can see that there are no\nfree FMRs and fail, even though there should always have been enough\navailable.\n\nSigned-off-by: Pete Wyckoff \u003cpw@osc.edu\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "84ba284cd78c130818e2de53150f39b92504593b", "tree": "28d0af4c0bd5ff24eb3aa988464fff8641d64248", "parents": [ "18b8c8f170ce346b88884ebe4060cd6dbe64e1cc" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Fri Feb 22 10:40:45 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 29 13:27:52 2008 -0800" }, "message": "IB/cm: Flush workqueue when removing device\n\nWhen a CM MAD is received, it is queued to a CM workqueue for\nprocessing. The queued work item references the port and device on\nwhich the MAD was received. If that device is removed from the system\nbefore the work item can execute, the work item will reference freed\nmemory.\n\nTo fix this, flush the workqueue after unregistering to receive MAD,\nand before the device is be freed.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4b1cc7e7ca5715907d17619dcb49144db6efe1c9", "tree": "82e8617c051c3148d902476464e36b418cbe531c", "parents": [ "30da7cff87f0ffa169fe07b766c3d6a5f6d1f6ab" ], "author": { "name": "John Lacombe", "email": "jlacombe@neteffect.com", "time": "Thu Feb 21 08:34:58 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:29 2008 -0800" }, "message": "RDMA/nes: Fix interrupt moderation low threshold\n\nInterrupt moderation low threshold value was incorrectly triggering,\nindicating that the threshold should be lowered.\n\nThe impact was the timer was likely to become 40usecs and get stuck\nthere. The biggest side effect was too many interrupts and nonoptimal\nperformance.\n\nSigned-off-by: John Lacombe \u003cjlacombe@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "30da7cff87f0ffa169fe07b766c3d6a5f6d1f6ab", "tree": "e68206af4e17a9542b978ca97aaca1a373ba8349", "parents": [ "a2e9c384ce76993cd68d6de57eaa81985b4618e3" ], "author": { "name": "Faisal Latif", "email": "flatif@neteffect.com", "time": "Thu Feb 21 08:31:22 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:29 2008 -0800" }, "message": "RDMA/nes: Fix CRC endianness for RDMA connection establishment on big-endian\n\nWith commit ef19454b (\"[LIB] crc32c: Keep intermediate crc state in\ncpu order\"), the behavior of crc32c changes on big-endian platforms.\n\nOur algorithm expects the previous behavior; otherwise we have RDMA\nconnection establishment failure on big-endian platforms like powerpc.\nApply cpu_to_le32() to value returned by crc32c() to get the previous\nbehavior.\n\nSigned-off-by: Faisal Latif \u003cflatif@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a2e9c384ce76993cd68d6de57eaa81985b4618e3", "tree": "adea958874e4289b400635633b4f6829270de8eb", "parents": [ "f84fba6f969065c6622669bbaa955c26fc1461ae" ], "author": { "name": "Faisal Latif", "email": "flatif@neteffect.com", "time": "Thu Feb 21 08:27:32 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:29 2008 -0800" }, "message": "RDMA/nes: Fix use-after-free in mini_cm_dec_refcnt_listen()\n\nFix use-after-free spotted by Coverity checker flagged by Adrian Bunk.\n\nSigned-off-by: Faisal Latif \u003cflatif@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f84fba6f969065c6622669bbaa955c26fc1461ae", "tree": "b562989918c69b271c04b83df7b27df786229d8a", "parents": [ "a4435febd4c0f14b25159dca249ecf91301c7c76" ], "author": { "name": "Glenn Streiff", "email": "gstreiff@neteffect.com", "time": "Thu Feb 21 08:17:54 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:29 2008 -0800" }, "message": "RDMA/nes: Fix use-after-free in nes_create_cq()\n\nJust delete the debugging statement so we don\u0027t use cqp_request after\nfreeing it. Adrian Bunk flagged this use-after-free issue spotted by\nthe Coverity checker.\n\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a4435febd4c0f14b25159dca249ecf91301c7c76", "tree": "c75976729afd4f9754517249b831abfd3aeb5ee1", "parents": [ "ed0ba33d64fb933f5fd985aa8f641984efd9d658" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Thu Feb 21 08:13:47 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:29 2008 -0800" }, "message": "RDMA/nes: Fix a check-after-use in nes_probe()\n\nFix a check-after-use spotted by the Coverity checker.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ed0ba33d64fb933f5fd985aa8f641984efd9d658", "tree": "12c45202489de5470f732080aa6ec0be540345a9", "parents": [ "65b07ec29354b345ff93914d064c2467aef4c862" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Thu Feb 21 08:12:06 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 26 16:24:27 2008 -0800" }, "message": "RDMA/nes: Fix a memory leak in schedule_nes_timer()\n\nFix a memory leak spotted by the Coverity checker.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "65b07ec29354b345ff93914d064c2467aef4c862", "tree": "a82fe79c2b17a54e0ee5ca8419d5810b340caa74", "parents": [ "9300c0c06788a409a97d54bbe3360d2f385fc621" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Thu Feb 21 08:01:18 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 25 16:00:30 2008 -0800" }, "message": "RDMA/nes: Fix off-by-one\n\nFix an off-by-one spotted by the Coverity checker.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9300c0c06788a409a97d54bbe3360d2f385fc621", "tree": "177ad00f8cbabaf84f561e3c98fca6b84f2dac1f", "parents": [ "82d416fffb5e8e39e899be7075dbeeac5fb8f0c2" ], "author": { "name": "Chien Tung", "email": "ctung@neteffect.com", "time": "Thu Feb 21 07:51:17 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 25 16:00:30 2008 -0800" }, "message": "RDMA/nes: Resurrect error path dead code\n\nAdrian Bunk pointed out that a Coverity scan found some apparently\ndead code in nes_verbs.c that really shouldn\u0027t have been dead.\n\nThe function nes_create_cq() was missing the assignment\n\n\terr \u003d 1;\n\njust prior to an iteration that conditionally set err \u003d 0 if a PBL was\nfound for a given virtual CQ. I also noticed we should have been\nreturning -EFAULT on a couple related error paths.\n\nSigned-off-by: Chien Tung \u003cctung@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "82d416fffb5e8e39e899be7075dbeeac5fb8f0c2", "tree": "6b09634809c71b5bbd1141413d5a324e0f4de7c4", "parents": [ "bfa274e2436fc7ef72ef51c878083647f1cfd429" ], "author": { "name": "Bryan Rosenburg", "email": "rosnbrg@us.ibm.com", "time": "Wed Feb 20 17:31:48 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 25 16:00:29 2008 -0800" }, "message": "RDMA/cxgb3: Fix shift calc in build_phys_page_list() for 1-entry page lists\n\nA single entry (addr 0x10001000, size 0x2000) will get converted to\npage address 0x10000000 with a page size of 0x4000. The code as it\nstands doesn\u0027t address the single buffer case, but in fact it allows\nthe subsequent single-buffer special case to be eliminated entirely.\nBecause the mask now includes the (page adjusted) starting and ending\naddresses, the general case works for the single buffer case as well.\n\nSigned-off-by: Bryan Rosenburg \u003crosnbrg@us.ibm.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b7f9c112a5e7f68d77a31c79f5942a3272ae9a3b", "tree": "23839936fc223e7076eb4a482e8720b81801f107", "parents": [ "ec229e5e81b3cf757e5e8b6a8bd0b4f32fe52f8c" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 19 10:42:50 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 19 10:42:50 2008 -0800" }, "message": "IB/mthca: Free correct MPT on error exit from mthca_fmr_alloc()\n\nWhen mthca_fmr_alloc() returns an error, it should free the MPT at the\nindex key, not mr-\u003eibmr.lkey, since the lkey has been mangled by\nhw_index_to_key() and no longer is the real index. This bug causes\ncorruption of the MPT table free bitmap when mthca_fmr_alloc() fails.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ec229e5e81b3cf757e5e8b6a8bd0b4f32fe52f8c", "tree": "0a257aa06672c5212110390baafc09752f4ca425", "parents": [ "9a4c8546f3e7c893888bccc2b3416d6214f2664a" ], "author": { "name": "Pradeep Satyanarayana", "email": "pradeeps@linux.vnet.ibm.com", "time": "Tue Feb 12 15:00:59 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 19 10:25:11 2008 -0800" }, "message": "IPoIB/cm: Fix ipoib_cm_dev_stop() cleanup when drain times out\n\nCommit efcd9971 (\"IPoIB/cm: Factor out ipoib_cm_free_rx_reap_list()\")\nintroduced a bug in ipoib_cm_dev_stop() when the receive drain times\nout. In that case, the function moves all the pending rx stuff into a\nprivate list but then calls ipoib_cm_free_rx_reap_list(), which\nhandles a different list.\n\nFix this by moving everything to the rx_reap_list that will actually\nget freed up.\n\nThis fixes \u003chttps://bugs.openfabrics.org/show_bug.cgi?id\u003d906\u003e.\n\nSigned-off-by: Pradeep Satyanarayana \u003cpradeeps@linux.vnet.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "51af33e8e45b845d8ee85446f58e31bc4c118048", "tree": "5d82fbb684c0adc0a01f2277f93fab7da2ac3810", "parents": [ "edd2fd643c500c812cae5b0d314ab9db9f959898" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 18 10:33:59 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 18 10:33:59 2008 -0800" }, "message": "RDMA/nes: Fix possible array overrun\n\nIn nes_create_qp(), the test\n\n\tif (nesqp-\u003emmap_sq_db_index \u003e NES_MAX_USER_WQ_REGIONS) {\n\nis used to error out if the db_index is too large; however, if the\ntest doesn\u0027t trigger, then the index is used as\n\n\tnes_ucontext-\u003emmap_nesqp[nesqp-\u003emmap_sq_db_index] \u003d nesqp;\n\nand mmap_nesqp is declared as\n\n\tstruct nes_qp *mmap_nesqp[NES_MAX_USER_WQ_REGIONS];\n\nwhich leads to an array overrun if the index is exactly equal to\nNES_MAX_USER_WQ_REGIONS. Fix this by bailing out if the index is\ngreater than or equal to NES_MAX_USER_WQ_REGIONS.\n\nThis was spotted by the Coverity checker (CID 2162).\n\nAcked-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "edd2fd643c500c812cae5b0d314ab9db9f959898", "tree": "46e7fb390f08cc58ca521c714742651cadf89098", "parents": [ "11e0704b7ec3abfeaee7e3a56f11c82024d1ae01" ], "author": { "name": "Chien Tung", "email": "ctung@neteffect.com", "time": "Fri Feb 15 21:00:36 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Feb 16 21:16:33 2008 -0800" }, "message": "RDMA/nes: Fix VLAN support\n\nWe need to account for the VLAN header size in nes_netdev_change_mtu()\nand nes_netdev_init(). Also, add spin lock/unlock during VLAN RX\nregistration so only one process can assign VLAN group for a given\ninterface at a time.\n\nSigned-off-by: Chien Tung \u003cctung@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "11e0704b7ec3abfeaee7e3a56f11c82024d1ae01", "tree": "b4929f77a34339c47ee4863fed47d8bca717c538", "parents": [ "c7482b81c8b524193d736488c149adbe27a7eb7f" ], "author": { "name": "Glenn Streiff", "email": "gstreiff@neteffect.com", "time": "Fri Feb 15 11:41:27 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 15 15:05:05 2008 -0800" }, "message": "RDMA/nes: Fix MAC interrupt erroneously masked on ifdown\n\nOnly mask out MAC interrupt if necessary and re-enable on ifup. There\ncould be multiple netdevs going through the same MAC. MAC interrupts\nshould not be masked off until the last netdev is downed.\n\nSigned-off-by: Chien Tung \u003cctung@neteffect.com\u003e\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c7482b81c8b524193d736488c149adbe27a7eb7f", "tree": "a9b68e294a9565335463fdceee8fc0543974d28d", "parents": [ "101142c37be8e5af9b847860219217e6b958c739" ], "author": { "name": "Li Zefan", "email": "lizf@cn.fujitsu.com", "time": "Fri Feb 15 10:24:49 2008 +0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 15 15:05:05 2008 -0800" }, "message": "IB: Fix return value in ib_device_register_sysfs()\n\nIf kobject_create_and_add() fails and returns NULL, the current code\nin ib_device_register_sysfs() does not set ret and hence returns 0.\nSet ret to -ENOMEM for this failure, so that the caller knows that\nib_device_register_sysfs() actually failed.\n\nSigned-off-by: Li Zefan \u003clizf@cn.fujitsu.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ead595aeb0974171eddd012df115424752413c26", "tree": "6505d5bc351667a8a3b35fd1025f123081a0bba4", "parents": [ "11e75a7455a7bc73e752c0c985986c2b1f8c930a" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Feb 13 14:33:53 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Feb 14 15:30:41 2008 -0800" }, "message": "RDMA/cma: Do not issue MRA if user rejects connection request\n\nThere\u0027s an undesirable interaction with issuing MRA requests to\nincrease connection timeouts and the listen backlog.\n\nWhen the rdma_cm receives a connection request, it queues an MRA with\nthe ib_cm. (The ib_cm will send an MRA if it receives a duplicate\nREQ.) The rdma_cm will then create a new rdma_cm_id and give that to\nthe user, which in this case is the rdma_user_cm.\n\nIf the listen backlog maintained in the rdma_user_cm is full, it\ndestroys the rdma_cm_id, which in turns destroys the ib_cm_id. The\nib_cm_id generates a REJ because the state of the ib_cm_id has changed\nto MRA sent, versus REQ received. When the backlog is full, we just\nwant to drop the REQ so that it is retried later.\n\nFix this by deferring queuing the MRA until after the user of the\nrdma_cm has examined the connection request.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e6028c0e004d334bb9ed75d4c918f4c763af1b9f", "tree": "9b83da3ad6221f4935f22459ddfaf6a1702588c2", "parents": [ "a9d1884925c80b96a621939a4fef5d74de58debe" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Thu Feb 14 10:39:36 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Feb 14 10:39:36 2008 -0800" }, "message": "IB/mlx4: mlx4_ib_fmr_alloc() should call mlx4_fmr_enable()\n\nCurrently mlx4_ib_fmr_alloc() calls mlx4_mr_enable() instead of\nmlx4_fmr_enable(). The two functions are equivalent at the moment, but \nthis is not really correct (and the change is needed to fix a bug).\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a9d1884925c80b96a621939a4fef5d74de58debe", "tree": "89541ba3ecfbbf21a3d15f6e1905ca68563f34ae", "parents": [ "167c42655cca188657aa9bb4e06d1194af3c73a5" ], "author": { "name": "Eli Cohen", "email": "eli@dev.mellanox.co.il", "time": "Thu Feb 14 13:15:28 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Feb 14 10:30:50 2008 -0800" }, "message": "IPoIB: Remove unused struct ipoib_cm_tx.ibwc member\n\nstruct ipoib_cm_tx.ibwc is unused since commit 1b524963 (\"IPoIB/cm:\nUse common CQ for CM send completions\"), so remove it.\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\n" }, { "commit": "167c42655cca188657aa9bb4e06d1194af3c73a5", "tree": "52e500e426a6d4a77acdc68333c6dcae0bc342c8", "parents": [ "5163dc1a645bc9ed7984fa484f1a77378c166d23" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Wed Feb 13 16:23:50 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Feb 14 10:15:06 2008 -0800" }, "message": "IPoIB: On P_Key change event, reset state properly\n\nIn P_Key event handling, if the old P_Key is no longer available, the\ndriver must call ipoib_ib_dev_stop() -- just as it does when the P_Key\nis still available (see procedure __ipoib_ib_dev_flush()).\n\nWhen a P_Key becomes available, the driver will perform ipoib_open(),\nwhich assumes that the QP is in RESET, the cm_id has been\ndestroyed/deleted, etc. If ipoib_ib_dev_stop() is not called as\ndescribed above, then these assumptions will be false, and the attempt\nto bring the interface up will fail.\n\nFound by Mellanox QA.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5163dc1a645bc9ed7984fa484f1a77378c166d23", "tree": "986189b025cacec7e736dae41de4737e05b76df4", "parents": [ "8704e9a8790cc9e394198663c1c9150c899fb9a2" ], "author": { "name": "Marcin Slusarz", "email": "marcin.slusarz@gmail.com", "time": "Wed Feb 13 00:06:08 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 13 07:47:47 2008 -0800" }, "message": "IB/mthca: Convert to use be16_add_cpu()\n\nreplace:\n\n\tbig_endian_variable \u003d cpu_to_beX(beX_to_cpu(big_endian_variable) +\n\t\t\t\t\t\texpression_in_cpu_byteorder);\n\nwith:\n\n\tbeX_add_cpu(\u0026big_endian_variable, expression_in_cpu_byteorder);\n\nGenerated with a semantic patch.\n\nSigned-off-by: Marcin Slusarz \u003cmarcin.slusarz@gmail.com\u003e\nCc: Sean Hefty \u003csean.hefty@intel.com\u003e\nCc: Hal Rosenstock \u003chal.rosenstock@gmail.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8704e9a8790cc9e394198663c1c9150c899fb9a2", "tree": "68f6ba71227f445d2fb81d5385f22d36fc1ba895", "parents": [ "7c7a9bccd2ba9f17e4b588461f140578a0a7b073" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Feb 12 16:09:29 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 13 07:47:42 2008 -0800" }, "message": "RDMA/cxgb3: Fail loopback connections\n\nThe cxgb3 HW and driver don\u0027t support loopback RDMA connections. So\nfail any connection attempt where the destination address is local.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7c7a9bccd2ba9f17e4b588461f140578a0a7b073", "tree": "97a8a5e88339a5defa770c19466190b8a20c0943", "parents": [ "ab64b960673250518e748f8b4f1545447136b68b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:27 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:27 2008 -0800" }, "message": "IB/cm: Fix infiniband_cm class kobject ref counting\n\nCommit 9af57b7a (\"IB/cm: Add basic performance counters\") introduced a\nbug in how the reference count for cm_class.subsys.kobj was handled:\nthe path that released a device did a kobject_put() on that kobject, but\nthere was no kobject_get() in the path the handles adding a device. So\nthe reference count ended up too low, which leads to bad things. Fix up\nand simplify the reference counting to avoid this.\n\n(Actually, I introduced the bug when fixing the patch up to match some\nof Greg\u0027s kobject changes, but who\u0027s counting)\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ab64b960673250518e748f8b4f1545447136b68b", "tree": "5932c3f64013a2e4c9ffa352344e939672aadb32", "parents": [ "fe174357eb2deb184c93269846c92adf5743115b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:27 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:27 2008 -0800" }, "message": "IB/cm: Remove debug printk()s that snuck upstream\n\nPesky little devils, sneaking around...\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "fe174357eb2deb184c93269846c92adf5743115b", "tree": "f034fd33459be36d5d2f47d176fee8ed79b7c7af", "parents": [ "96b5a46e2a72dc1829370c87053e0cd558d58bc0" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:22 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Feb 12 14:38:22 2008 -0800" }, "message": "IB/mthca: Add missing sg_init_table() in mthca_map_user_db()\n\nUsually harmless, since the scatterlist is always hard-coded to a length\nof 1, but it triggers a BUG() if CONFIG_DEBUG_SG\u003dy, so we better fix it.\nThis fixes \u003chttp://bugzilla.kernel.org/show_bug.cgi?id\u003d9934\u003e.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7143740d26098aca84ecc7376ccfe2c58fd0412e", "tree": "c29029045ebdbe197f94ead3e1d6610f498f1880", "parents": [ "eb14032f9eb595621270f3269f40094adb3144e8" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Wed Jan 30 18:30:53 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 08 14:32:37 2008 -0800" }, "message": "IPoIB: Add send gather support\n\nThis patch acts as a preparation for using checksum offload for IB\ndevices capable of inserting/verifying checksum in IP packets. The\npatch does not actaully turn on NETIF_F_SG - we defer that to the\npatches adding checksum offload capabilities.\n\nWe only add support for send gathers for datagram mode, since existing\nHW does not support checksum offload on connected QPs.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "eb14032f9eb595621270f3269f40094adb3144e8", "tree": "88c837fe13a5415e26e2744fd309a01e034832e2", "parents": [ "ea54b10c7773007e173da31fe7adcc049da33331" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Wed Jan 30 18:30:46 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 08 13:39:26 2008 -0800" }, "message": "IPoIB: Add high DMA feature flag\n\nAll current InfiniBand devices can handle all DMA addresses, and it\u0027s\nhard to imagine anyone would be silly enough to build a new device\nthat couldn\u0027t. Therefore, enable the NETIF_F_HIGHDMA feature for IPoIB.\n\nThis has no effect for no, but is needed when we enable gather/scatter\nsupport and checksum stateless offloads.\n\nSigned-off-by: Eli Cohen \u003celi@mellnaox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ea54b10c7773007e173da31fe7adcc049da33331", "tree": "b13b77fb3807071a5b93ece7b564f6748d962bbc", "parents": [ "b57aacfa7a95328f469d0360e49289b023c47e9e" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Mon Jan 28 10:40:59 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Feb 08 13:30:02 2008 -0800" }, "message": "IB/mlx4: Use multiple WQ blocks to post smaller send WQEs\n\nConnectX HCA supports shrinking WQEs, so that a single work request\ncan be made of multiple units of wqe_shift. This way, WRs can differ\nin size, and do not have to be a power of 2 in size, saving memory and\nspeeding up send WR posting. Unfortunately, if we do this then the\nwqe_index field in CQEs can\u0027t be used to look up the WR ID anymore, so\nour implementation does this only if selective signaling is off.\n\nFurther, on 32-bit platforms, we can\u0027t use vmap() to make the QP\nbuffer virtually contigious. Thus we have to use constant-sized WRs to\nmake sure a WR is always fully within a single page-sized chunk.\n\nFinally, we use WRs with the NOP opcode to avoid wrapping around the\nqueue buffer in the middle of posting a WR, and we set the\nNoErrorCompletion bit to avoid getting completions with error for NOP\nWRs. However, NEC is only supported starting with firmware 2.2.232,\nso we use constant-sized WRs for older firmware. And, since MLX QPs\nonly support SEND, we use constant-sized WRs in this case.\n\nWhen stamping during NOP posting, do stamping following setting of the\nNOP WQE valid bit.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1c69fc2a9012e160c8d459f63df74a6b01db8322", "tree": "a32a2be6b2c8b9f5c4e006bacebc1eeb723e2ff5", "parents": [ "21511abd0a248a3f225d3b611cfabb93124605a7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 06 21:07:54 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Feb 06 21:07:54 2008 -0800" }, "message": "IB/mlx4: Consolidate code to get an entry from a struct mlx4_buf\n\nWe use struct mlx4_buf for kernel QP, CQ and SRQ buffers, and the code\nto look up an entry is duplicated in get_cqe_from_buf() and the QP and\nSRQ versions of get_wqe(). Factor this out into mlx4_buf_offset().\n\nThis will also make it easier to switch over to using vmap() for buffers.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3c2d774cad5bf4fad576363da77870e9e6530b7a", "tree": "344ae8167730ea361ff4bd4bf098ba64bfeef5e5", "parents": [ "2c78853472a36c7cf51a84a34edc370e21c93ce4" ], "author": { "name": "Glenn Streiff", "email": "gstreiff@neteffect.com", "time": "Mon Feb 04 20:20:45 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:45 2008 -0800" }, "message": "RDMA/nes: Add a driver for NetEffect RNICs\n\nAdd a standard NIC and RDMA/iWARP driver for NetEffect 1/10Gb ethernet adapters.\n\nSigned-off-by: Glenn Streiff \u003cgstreiff@neteffect.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2c78853472a36c7cf51a84a34edc370e21c93ce4", "tree": "a89ab43cc596415f80e434470595fcbb4f8e7544", "parents": [ "f33afc26dc03e6e0513e2e300f2aa0ad5463c2d2" ], "author": { "name": "Olaf Kirch", "email": "olaf.kirch@oracle.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB/mthca: Return proper error codes from mthca_fmr_alloc()\n\nIf the allocation of the MTT or the mailbox failed, mthca_fmr_alloc()\nwould return 0 (success) no matter what. This leads to crashes a\nlittle down the road, when we try to dereference eg mr-\u003emtt, which was\nreally ERR_PTR(-Ewhatever).\n\nSigned-off-by: Olaf Kirch \u003colaf.kirch@oracle.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f33afc26dc03e6e0513e2e300f2aa0ad5463c2d2", "tree": "8d4645dd7c20c466a1c421a2ed05d480497f629f", "parents": [ "68f3948dab39249d178eb007c071f87fb6481fc6" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB: Avoid marking __devinitdata as const\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "68f3948dab39249d178eb007c071f87fb6481fc6", "tree": "c67659a811309d0e63d8570a64ce934c9febc2fd", "parents": [ "1d368c546566e249da8181e933c53788093965cf" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB/mlx4: Actually print out the driver version\n\nThe string mlx4_ib_version was defined, but never used. Print out the\nversion once when the first device is initialized.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1d368c546566e249da8181e933c53788093965cf", "tree": "08ea2e0280f125f2981ab7ada0d335b5928cc670", "parents": [ "1203c42e7be1aa0be641b701f42b6d38c2d94b39" ], "author": { "name": "Eli Cohen", "email": "eli at mellanox.co.il", "time": "Thu Jan 24 06:38:06 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB/ib_mthca: Pre-link receive WQEs in Tavor mode\n\nWe have recently discovered that Tavor mode requires each WQE in a\nposted list of receive WQEs to have a valid NDA field at all times.\nThis requirement holds true for regular QPs as well as for SRQs. This\npatch prelinks the receive queue in a regular QP and keeps the free\nlist in SRQ always properly linked.\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nReviewed-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1203c42e7be1aa0be641b701f42b6d38c2d94b39", "tree": "3f84fb72b8fb61d460f3207d21ddf06b4a78700f", "parents": [ "1d96354e617990799b1cb5d7ff8f7c467b8767c8" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB/mthca: Remove checks for srq-\u003efirst_free \u003c 0\n\nThe SRQ receive posting functions make sure that srq-\u003efirst_free never\nbecomes negative, so we can remove tests of whether it is negative.\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1d96354e617990799b1cb5d7ff8f7c467b8767c8", "tree": "7d106bf836c8cd99ee844845e71216e83c004248", "parents": [ "9fe4bcf45ece0b0081031edaaa41581c85ef7049" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Tue Jan 29 12:56:18 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:44 2008 -0800" }, "message": "IB/fmr_pool: Allocate page list for pool FMRs only when caching enabled\n\nAllocate memory for the page_list field of struct ib_pool_fmr only\nwhen caching is enabled for the FMR pool, since the field is not used\notherwise. This can save significant amounts of memory for large\npools with caching turned off.\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9fe4bcf45ece0b0081031edaaa41581c85ef7049", "tree": "b4d7cd3aff19fe8802d71d031d79340962164434", "parents": [ "893da75956ab48545e8732b46e1cf4350bd25f9c" ], "author": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Tue Jan 08 17:08:52 2008 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:43 2008 -0800" }, "message": "IB/srp: Retry stale connections\n\nWhen a host just goes away (crash, power loss, etc.) without tearing\ndown its IB connections, it can get stale connection errors when it\ntries to reconnect to targets upon rebooting. Retrying the connection\na few times will prevent sysadmins from playing the \"which disk(s)\nwent missing?\" game.\n\nThis would have made things slightly quicker when tracking down some\nof the recent bugs, but it also helps quite a bit when you\u0027ve got a\nlarge number of targets hanging off a wedged server.\n\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "893da75956ab48545e8732b46e1cf4350bd25f9c", "tree": "fe5125ffd8ec7d84f3b4937f6ade9ee65f7a4913", "parents": [ "6ccef1de2c1718729dd1c7ee8bd98473519eb3b3" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Sun Jan 27 18:13:25 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:43 2008 -0800" }, "message": "mlx4_core: Don\u0027t read reserved fields in mlx4_QUERY_ADAPTER()\n\nThe firmware QUERY_ADAPTER command does not return vendor_id,\ndevice_id, and revision_id; eliminate these fields from the query.\n\nInitialize the rev_id field of the mlx4 device via init_node_data (MAD\nIFC query), as is done in the query_device verb implementation.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6ccef1de2c1718729dd1c7ee8bd98473519eb3b3", "tree": "7685f0b85ce22ae65ea95a4af9771e03ec6206ad", "parents": [ "7bc531dd883b955e6198c8e202161f22d2e8c472" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Sun Jan 27 18:13:20 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:43 2008 -0800" }, "message": "IB/mthca: Don\u0027t read reserved fields in mthca_QUERY_ADAPTER()\n\nFor memfree devices, the firmware QUERY_ADAPTER command does not\nreturn vendor_id, device_id, and revision_id; do not return these\nfields in the QUERY_ADAPTER function for memfree devices.\n\nInstead, for memfree devices, initialize the rev_id field of the mthca\ndevice via init_node_data (MAD IFC query), as is done in the\nquery_device verb implementation.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7bc531dd883b955e6198c8e202161f22d2e8c472", "tree": "509aaa21f88d024c422937339783fb7c41c94e44", "parents": [ "bafff9741704959e99fb65a7327c017251019a19" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Tue Jan 29 12:57:56 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:43 2008 -0800" }, "message": "IPoIB: Remove a misleading debug print\n\nCommit 732a2170 (\"IB/ipoib: Bound the net device to the ipoib_neigh\nstructue\") left a misleading debug print (n-\u003edev would be a bond\ndevice only if boding is used). Clean it up.\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bafff9741704959e99fb65a7327c017251019a19", "tree": "571a380b5be9b35d46d76003d2b362a377eb7bbb", "parents": [ "0d89fe2c0ca12ad2ee4e35a0661319746af6e94a" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Thu Jan 17 17:03:45 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:43 2008 -0800" }, "message": "IPoIB: Handle bonding failover race for connected neighbours too\n\nMove up the code that checks for a situation where the remote GID\nstored in the ipoib_neigh is different than the one present in the\nneighbour (handle gratuitous ARP) or that a bonding fail over has\nhappened but the neighbour still has a pointer to an ipoib_neigh\ncreated by a different device than the current slave. This will cause\nthe driver to apply the check also for connected mode neighbours.\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0d89fe2c0ca12ad2ee4e35a0661319746af6e94a", "tree": "fcc5d6598121a4863740437baf10217b701ca153", "parents": [ "2b5e6b120e58d44cace68e6c7204b541a8b0b43f" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "message": "IB/mthca: Fix and simplify page size calculation in mthca_reg_phys_mr()\n \nIn mthca_reg_phys_mr(), we calculate the page size for the HCA\nhardware to use to map the buffer list passed in by the consumer.\nFor example, if the consumer passes in\n\n [0] addr 0x1000, size 0x1000\n [1] addr 0x2000, size 0x1000\n\nthen the algorithm would come up with a page size of 0x2000 and a list\nof two pages, at 0x0000 and 0x2000. Usually, this would work fine\nsince the memory region would start at an offset of 0x1000 and have a\nlength of 0x2000.\n\nHowever, the old code did not take into account the alignment of the\nIO virtual address passed in. For example, if the consumer passed in\na virtual address of 0x6000 for the above, then the offset of 0x1000\nwould not be used correctly because the page mask of 0x1fff would\nresult in an offset of 0.\n\nWe can fix this quite neatly by making sure that the page shift we use\nis no bigger than the first bit where the start of the first buffer\nand the IO virtual address differ. Also, we can further simplify the\ncode by removing the special case for a single buffer by noticing that\nit doesn\u0027t matter if we use a page size that is too big. This allows\nthe loop to compute the page shift to be replaced with __ffs().\n\nThanks to Bryan S Rosenburg \u003crosnbrg@us.ibm.com\u003e for pointing out the\noriginal bug and suggesting several ways to improve this patch.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2b5e6b120e58d44cace68e6c7204b541a8b0b43f", "tree": "fa2cb052b146b22ec285928a000f133da3e7d73c", "parents": [ "528b03f73247c30750b740dcad16ad1914e56e89" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@de.ibm.com", "time": "Fri Jan 25 21:18:27 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "message": "IB/ehca: Add PMA support\n\nThis patch enables ehca to redirect any PMA queries to the\nactual PMA QP.\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nReviewed-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nReviewed-by: Christoph Raisch \u003craisch@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "528b03f73247c30750b740dcad16ad1914e56e89", "tree": "414d7724b9f111fc80909975f53f65f675b431c4", "parents": [ "2b7274c39228d7a8c81a411dc3969763e9069c56" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Fri Jan 25 21:12:39 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "message": "IB/ehca: Update sma_attr also in case of disruptive config change\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2b7274c39228d7a8c81a411dc3969763e9069c56", "tree": "596f69786ab9957a69a1d1e5da82d2fc900602c1", "parents": [ "3971c9f6dbf26f077b929dbe14ced60a697ebcf0" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Thu Jan 24 17:59:08 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "message": "IB/ehca: Prevent sending UD packets to QP0\n\nThe IB spec doesn\u0027t allow packets to QP0 sent on any other VL than VL15.\nHardware doesn\u0027t filter those packets on the send side, so we need to do\nthis in the driver and firmware.\n\nAs eHCA doesn\u0027t support QP0, we can just filter out all traffic going to\nQP0, regardless of SL or VL.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3971c9f6dbf26f077b929dbe14ced60a697ebcf0", "tree": "005053ac7b0b42f43b296cfc0384bf11567673e3", "parents": [ "e8f9b2ed9882874ca96716597bd8c7113289e77b" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Mon Dec 10 15:53:25 2007 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Feb 04 20:20:42 2008 -0800" }, "message": "IB/cm: Add interim support for routed paths\n\nPaths with hop_limit \u003e 1 indicate that the connection will be routed\nbetween IB subnets. Update the subnet local field in the CM REQ based\non the hop_limit value. In addition, if the path is routed, then set\nthe LIDs in the REQ to the permissive LIDs. This is used to indicate\nto the passive side that it should use the LIDs in the received local\nroute header (LRH) associated with the REQ when programming the QP.\n\nThis is a temporary work-around to the IB CM to support IB router\ndevelopment until the IB router specification is completed. It is not\nanticipated that this work-around will cause any interoperability\nissues with existing stacks or future stacks that will properly\nsupport IB routers when defined.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d3f46f39b7092594b498abc12f0c73b0b9913bde", "tree": "6d595bdecbdd41a958e05e425664209f5d71ecf1", "parents": [ "b8de16318410f6f8611a879678a531237e4aadc9" ], "author": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Tue Jan 15 11:11:46 2008 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@HansenPartnership.com", "time": "Wed Jan 30 13:14:02 2008 -0600" }, "message": "[SCSI] remove use_sg_chaining\n\nWith the sg table code, every SCSI driver is now either chain capable\nor broken (or has sg_tablesize set so chaining is never activated), so\nthere\u0027s no need to have a check in the host template.\n\nAlso tidy up the code by moving the scatterlist size defines into the\nSCSI includes and permit the last entry of the scatterlist pools not\nto be a power of two.\nSigned-off-by: James Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\n" }, { "commit": "0ba6c33bcddc64a54b5f1c25a696c4767dc76292", "tree": "62e616f97a4762d8e75bf732e4827af2d15d52c5", "parents": [ "21af0297c7e56024a5ccc4d8ad2a590f9ec371ba", "85040bcb4643cba578839e953f25e2d1965d83d0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 29 22:54:01 2008 +1100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 29 22:54:01 2008 +1100" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6.25\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6.25: (1470 commits)\n [IPV6] ADDRLABEL: Fix double free on label deletion.\n [PPP]: Sparse warning fixes.\n [IPV4] fib_trie: remove unneeded NULL check\n [IPV4] fib_trie: More whitespace cleanup.\n [NET_SCHED]: Use nla_policy for attribute validation in ematches\n [NET_SCHED]: Use nla_policy for attribute validation in actions\n [NET_SCHED]: Use nla_policy for attribute validation in classifiers\n [NET_SCHED]: Use nla_policy for attribute validation in packet schedulers\n [NET_SCHED]: sch_api: introduce constant for rate table size\n [NET_SCHED]: Use typeful attribute parsing helpers\n [NET_SCHED]: Use typeful attribute construction helpers\n [NET_SCHED]: Use NLA_PUT_STRING for string dumping\n [NET_SCHED]: Use nla_nest_start/nla_nest_end\n [NET_SCHED]: Propagate nla_parse return value\n [NET_SCHED]: act_api: use PTR_ERR in tcf_action_init/tcf_action_get\n [NET_SCHED]: act_api: use nlmsg_parse\n [NET_SCHED]: act_api: fix netlink API conversion bug\n [NET_SCHED]: sch_netem: use nla_parse_nested_compat\n [NET_SCHED]: sch_atm: fix format string warning\n [NETNS]: Add namespace for ICMP replying code.\n ...\n" }, { "commit": "f206351a50ea86250fabea96b9af8d8f8fc02603", "tree": "014148d8b45db1995d98374ec3b30e635f243197", "parents": [ "f1b050bf7a88910f9f00c9c8989c1bf5a67dd140" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Tue Jan 22 22:07:34 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 28 15:11:07 2008 -0800" }, "message": "[NETNS]: Add namespace parameter to ip_route_output_key.\n\nNeeded to propagate it down to the ip_route_output_flow.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f1b050bf7a88910f9f00c9c8989c1bf5a67dd140", "tree": "37e8dce4268cb657a75e5074a38db48264f1e42f", "parents": [ "611c183ebcb5af384df3a4ddb391034a1b6ac255" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Tue Jan 22 22:07:10 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 28 15:11:06 2008 -0800" }, "message": "[NETNS]: Add namespace parameter to ip_route_output_flow.\n\nNeeded to propagate it down to the __ip_route_output_key.\n\nSigned_off_by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1ab352768fc73838b062776ca5d1add3876a019f", "tree": "b30bd7744e30e90b6db01b71a5c8b86140538255", "parents": [ "010278ec4cdf404aefc0bbd5e7406674fec95286" ], "author": { "name": "Denis V. Lunev", "email": "den@openvz.org", "time": "Tue Jan 22 22:04:30 2008 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 28 15:11:04 2008 -0800" }, "message": "[NETNS]: Add namespace parameter to ip_dev_find.\n\nin_dev_find() need a namespace to pass it to fib_get_table(), so add\nan argument.\n\nSigned-off-by: Denis V. Lunev \u003cden@openvz.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6360a02af1599e46b023ccbb85545ed97c6f662c", "tree": "55f2cfdf0a34c98fcf47e7534558d1eef7d8f153", "parents": [ "b5cb2bbc4c6cb489941be881e8adfe136ee45b8e" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Sun Dec 16 13:47:33 2007 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Jan 28 14:58:17 2008 -0800" }, "message": "[IPV4] drivers/infiniband: Use ipv4_is_\u003ctype\u003e\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "aff5905778d2fb1826ec623e4b1da3a7ff048ca4", "tree": "81af31da314140408653c95b53401d7405eceb91", "parents": [ "42d71c44d5fc09ae47783b7e15fe0d6b60711230" ], "author": { "name": "WANG Cong", "email": "xiyou.wangcong@gmail.com", "time": "Tue Jan 01 21:41:08 2008 +0800" }, "committer": { "name": "Sam Ravnborg", "email": "sam@ravnborg.org", "time": "Mon Jan 28 23:14:37 2008 +0100" }, "message": "INFINIBAND: Remove \u0027TOPDIR\u0027 from Makefiles\n\nThis patch removes TOPDIR from infiniband Makefile and delete\none include statement pointing to a non-existing directory\n\nCc: Roland Dreier \u003crolandd@cisco.com\u003e\nCc: Sean Hefty \u003cmshefty@ichips.intel.com\u003e\nCc: Hal Rosenstock \u003chal.rosenstock@gmail.com\u003e\nSigned-off-by: WANG Cong \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n" }, { "commit": "9b73e76f3cf63379dcf45fcd4f112f5812418d0a", "tree": "4e6bef87cd0cd6d848fc39a5ae25b981dbbe035b", "parents": [ "50d9a126240f9961cfdd063336bbeb91f77a7dce", "23c3e290fb9ce38cabc2822b47583fc8702411bf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 25 17:19:08 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 25 17:19:08 2008 -0800" }, "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: (200 commits)\n [SCSI] usbstorage: use last_sector_bug flag universally\n [SCSI] libsas: abstract STP task status into a function\n [SCSI] ultrastor: clean up inline asm warnings\n [SCSI] aic7xxx: fix firmware build\n [SCSI] aacraid: fib context lock for management ioctls\n [SCSI] ch: remove forward declarations\n [SCSI] ch: fix device minor number management bug\n [SCSI] ch: handle class_device_create failure properly\n [SCSI] NCR5380: fix section mismatch\n [SCSI] sg: fix /proc/scsi/sg/devices when no SCSI devices\n [SCSI] IB/iSER: add logical unit reset support\n [SCSI] don\u0027t use __GFP_DMA for sense buffers if not required\n [SCSI] use dynamically allocated sense buffer\n [SCSI] scsi.h: add macro for enclosure bit of inquiry data\n [SCSI] sd: add fix for devices with last sector access problems\n [SCSI] fix pcmcia compile problem\n [SCSI] aacraid: add Voodoo Lite class of cards.\n [SCSI] aacraid: add new driver features flags\n [SCSI] qla2xxx: Update version number to 8.02.00-k7.\n [SCSI] qla2xxx: Issue correct MBC_INITIALIZE_FIRMWARE command.\n ...\n" }, { "commit": "8176d297c73a06e6076c9c31f6404047567f6324", "tree": "d3ea63287941d458f853dc06e1e28115025cab19", "parents": [ "f7fca1e8a8eeb6660396327b48995e3d9a7b6dc2" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Thu Jan 24 16:30:16 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:47 2008 -0800" }, "message": "RDMA/cxgb3: Fix the T3A workaround checks\n\nCorrectly work around T3A issues by checking \"hwtype !\u003d T3A\" instead of\n\"hwtype \u003d\u003d T3B\". This will be needed for new hardware types.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f7fca1e8a8eeb6660396327b48995e3d9a7b6dc2", "tree": "4c174030d1cd5a47660d1cb9d45396b74b49daef", "parents": [ "1cf18d5aab5144866cb5221250905623b03a32a9" ], "author": { "name": "Jan Engelhardt", "email": "jengelh@computergmbh.de", "time": "Tue Jan 22 20:45:30 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:46 2008 -0800" }, "message": "IB/ipath: Remove unnecessary cast\n\nSigned-off-by: Jan Engelhardt \u003cjengelh@computergmbh.de\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1cf18d5aab5144866cb5221250905623b03a32a9", "tree": "a4acb4c2039485572e81810bee76b916671454ab", "parents": [ "c6b5b5047417c1becb5e48e748e3acb3801079a1" ], "author": { "name": "Jan Engelhardt", "email": "jengelh@computergmbh.de", "time": "Tue Jan 22 20:45:30 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:46 2008 -0800" }, "message": "IPoIB: Constify seq_operations function pointer tables\n\nSigned-off-by: Jan Engelhardt \u003cjengelh@computergmbh.de\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c6b5b5047417c1becb5e48e748e3acb3801079a1", "tree": "8a4efd832544869ec93646be2f5d409119b9b309", "parents": [ "d08ca26ceec4dfbcfdbada4ad728db742ccaecd1" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Jan 21 14:42:13 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:45 2008 -0800" }, "message": "RDMA/cxgb3: Mark QP as privileged based on user capabilities\n\nThis is needed to support zero-stag properly.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d08ca26ceec4dfbcfdbada4ad728db742ccaecd1", "tree": "15c79eb89ce17fcb2b96f1497e52c352d433b0ff", "parents": [ "856b5925047d73a85557203d124d62c5eea1fbd3" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Jan 21 14:42:11 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:45 2008 -0800" }, "message": "RDMA/cxgb3: Fix page shift calculation in build_phys_page_list()\n\nThe existing logic incorrectly maps this buffer list:\n\n 0: addr 0x10001000, size 0x1000\n 1: addr 0x10002000, size 0x1000\n\nTo this bogus page list:\n\n 0: 0x10000000\n 1: 0x10002000\n\nThe shift calculation must also take into account the address of the\nfirst entry masked by the page_mask as well as the last address+size\nrounded up to the next page size.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "856b5925047d73a85557203d124d62c5eea1fbd3", "tree": "a2cd5eaff8a758e41d4870c7f62c2c7cf8aca861", "parents": [ "4e1e93a4189a98cfb0e24865f7f44470ae5f805e" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Jan 21 14:42:09 2008 -0600" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:45 2008 -0800" }, "message": "RDMA/cxgb3: Flush the receive queue when closing\n\n- for kernel mode cqs, call event notification handler when flushing.\n- flush QP when moving from RTS -\u003e CLOSING.\n- fix logic to identify a kernel mode qp.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4e1e93a4189a98cfb0e24865f7f44470ae5f805e", "tree": "b81f51543d0af51edcbac668d851bf053b84d271", "parents": [ "950529e5c6efda8a6d3090a8fef9b904c833397a" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "message": "IB/ipath: Trivial simplification of ipath_make_ud_req()\n\nMove the increment of s_hdrwords into the existing if block that tests \nif we\u0027re doing a send with immediate, to save one test of the opcode.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "950529e5c6efda8a6d3090a8fef9b904c833397a", "tree": "92061c141f77626d14d9bf16c91c6a489fb73ee7", "parents": [ "48fe5e594c979177b7f20affd027be56e8ea2767" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "message": "IB/mthca: Update latest \"native Arbel\" firmware revision\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "48fe5e594c979177b7f20affd027be56e8ea2767", "tree": "7e2885c5f3ac8cd1abc4366b228088589f1ec917", "parents": [ "3d68ea32611095c7e09409ac1b2a56da22fd5eb7" ], "author": { "name": "Krishna Kumar", "email": "krkumar2@in.ibm.com", "time": "Thu Nov 15 10:35:37 2007 +0530" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "message": "IPoIB: Remove redundant check of netif_queue_stopped() in xmit handler\n\nqdisc_run() now tests for queue_stopped() before calling\n__qdisc_run(), and the same check is done in every iteration of\n__qdisc_run(), so another check is not required in the driver xmit.\nThis means that ipoib_start_xmit() no longer needs to test\nnetif_queue_stopped(); the test was added to fix earlier kernels,\nwhere the networking stack did not guarantee that the xmit method of\nan LLTX driver would not be called after the queue was stopped, but\ncurrent kernels do provide this guarantee.\n\nTo validate, I put a debug in the TX_BUSY path which never hit with 64\nthreads running overnight exercising this code a few 100 million\ntimes.\n\nSigned-off-by: Krishna Kumar \u003ckrkumar2@in.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3d68ea32611095c7e09409ac1b2a56da22fd5eb7", "tree": "6713954de767902d74c6b8d9ac432efa3b86b6b9", "parents": [ "6ac50727bda29e961385e4c40318dadbb5730193" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Thu Jan 10 00:50:41 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:44 2008 -0800" }, "message": "IB/ipath: Add mappings from HW register to PortInfo port physical state\n\nAdd new mappings from port physical state (a HW register value) to the\nIB SubnGet(PortInfo) port physical state.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6ac50727bda29e961385e4c40318dadbb5730193", "tree": "522ff1a220c817d4ed00e1311e4cf23b24952b50", "parents": [ "ddb70c83a5ce439271f1699e52a97785a8b45b81" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Thu Aug 09 03:11:38 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:43 2008 -0800" }, "message": "IB/ipath: Changes to support PIO bandwidth check on IBA7220\n\nThe IBA7220 uses a count-based triggering mechanism, and therefore\ncan\u0027t use the same bandwidth verification mechanism as older chips.\n\nTo support the 7220, allow enabling and disabling armlaunch errors on\napplication request. Minor robustness improvements as well.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ddb70c83a5ce439271f1699e52a97785a8b45b81", "tree": "dec9a6a75c595a5348c7970cadd40465b59a4fef", "parents": [ "359193ef433061515fe24d57e5bd5a1318d71bc3" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Tue Jan 08 01:51:46 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:43 2008 -0800" }, "message": "IB/ipath: Minor cleanup of unused fields and chip-specific errors\n\nClean up some unused header fields, minor related cleanup.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "359193ef433061515fe24d57e5bd5a1318d71bc3", "tree": "270f18f7a3752bf1ac9f98f4a8e88eb9b95f8c71", "parents": [ "c4bce8032ef4368063c84d665b19804878d63e7c" ], "author": { "name": "Michael Albaugh", "email": "michael.albaugh@qlogic.com", "time": "Tue Jan 08 00:37:34 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:17:32 2008 -0800" }, "message": "IB/ipath: New sysfs entries to control 7220 features\n\nIBA7220 includes many more configurable IB settings. Getting/setting\nthese is now grouped into a pair of chip specific functions accessed via\nfunction pointers. Provide sysfs access to these settings.\n\nSigned-off-by: Michael Albaugh \u003cmichael.albaugh@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c4bce8032ef4368063c84d665b19804878d63e7c", "tree": "29b7e0b7a97747982237a98f634a97d2111d1d8a", "parents": [ "7387273307139ebf8d7f7fb3bb79d1ca48bd71d6" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Tue Jan 08 02:36:46 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:45 2008 -0800" }, "message": "IB/ipath: Add new chip-specific functions to older chips, consistent init\n\nThis adds the new (sometimes empty) chip-specific functions to the older\nchips, and makes the initialization and related functions consistent across\nall 3 chips.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7387273307139ebf8d7f7fb3bb79d1ca48bd71d6", "tree": "9d65581f68a0dd14589053aaafffa155193e5bf1", "parents": [ "2ec8e662416cc9a171cdfe3d75e1ff00ba757859" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Tue Jan 08 23:16:17 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:44 2008 -0800" }, "message": "IB/ipath: Remove unused MDIO interface code\n\nThis code has been unused for some time, but still had leftovers\nfrom when it was used.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2ec8e662416cc9a171cdfe3d75e1ff00ba757859", "tree": "dbc881205c3177db9a7816385b4fdda762bdb04c", "parents": [ "bbdd267ef2796e96b461b8447b2026ce06e6ec4b" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Thu Jan 17 15:07:24 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:44 2008 -0800" }, "message": "IB/ehca: Prevent RDMA-related connection failures on some eHCA2 hardware\n\nSome HW revisions of eHCA2 may cause an RC connection to break if they\nreceived RDMA Reads over that connection before. This can be\nprevented by assuring that, after the first RDMA Read, the QP receives\na new RDMA Read every few million link packets.\n\nInclude code into the driver that inserts an empty (size 0) RDMA Read\ninto the message stream every now and then if the consumer doesn\u0027t\npost them frequently enough.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bbdd267ef2796e96b461b8447b2026ce06e6ec4b", "tree": "45cea3f33acc271da83fcccca0c5626c0281da4e", "parents": [ "b8b50e353b85bc3c791dd2b99370ac300ebcd186" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@de.ibm.com", "time": "Thu Jan 17 15:05:45 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:44 2008 -0800" }, "message": "IB/ehca: Add \"port connection autodetect mode\"\n\nThis patch enhances ehca with a capability to \"autodetect\" the ports\nbeing connected physically. In order to utilize that function the\nmodule option nr_ports must be set to -1 (default is 2 - two\nports). This feature is experimental and will made the default later.\n\nMore detail:\n\nIf the user connects only one port to the switch, current code requires\n 1) port one to be connected and\n 2) module option nr_ports\u003d1 to be given.\n\nIf autodetect is enabled, ehca will not wait at creation of the GSI QP\nfor the respective port to become active. Since firmware does not\naccept modify_qp() while the port is down at initialization, we need\nto cache all calls to modify_qp() for the SMI/GSI QP and just return a\ngood return code.\n\nWhen a port is activated and we get a PORT_ACTIVE event, we replay the\ncached modify-qp() parms and re-trigger any posted recv WRs. Only then\ndo we forward the PORT_ACTIVE event to registered clients.\n\nThe result of this autodetect patch is that all ports will be\naccessible by the users. Depending on their respective cabling only\nthose ports that are connected properly will become operable. If a\nuser tries to modify a regular QP of a non-connected port, modify_qp()\nwill fail. Furthermore, ibv_devinfo should show the port state\naccordingly.\n\nNote that this patch primarily improves the loading behaviour of\nehca. If the cable is removed while the driver is operating and\nplugged in again, firmware will handle that properly by sending an\nappropriate async event.\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b8b50e353b85bc3c791dd2b99370ac300ebcd186", "tree": "80b6706a9601b3ce61c1d9caa049eb06520ed87a", "parents": [ "0c86e280fe8a08d4ae30b77e46a1e7da28d756c9" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen at de.ibm.com", "time": "Thu Jan 17 15:04:32 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:44 2008 -0800" }, "message": "IB/ehca: Define array to store SMI/GSI QPs\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0c86e280fe8a08d4ae30b77e46a1e7da28d756c9", "tree": "422504ced2506b3b84a13bb2924f8256c6d3ad90", "parents": [ "6410627eb9804e541b83d220c8e914ce64475b31" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen at de.ibm.com", "time": "Thu Jan 17 15:03:55 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:43 2008 -0800" }, "message": "IB/ehca: Remove CQ-QP-link before destroying QP in error path of create_qp()\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6410627eb9804e541b83d220c8e914ce64475b31", "tree": "8ce0d4be69366fb19c2e71772d578ba065cdfd74", "parents": [ "a4ef1451dfba92f51934e8331f634497b9ed3393" ], "author": { "name": "Erez Zilber", "email": "erezz@voltaire.com", "time": "Thu Jan 17 11:53:17 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:43 2008 -0800" }, "message": "IB/iser: Add change_queue_depth method\n\nAdd a .change_queue_depth handler to the scsi_host_template in the\niSER driver. iscsi_change_queue_depth was added to iscsi_tcp in order\nto solve the problem of queue depth which was too high for some\ntargets. It is also applicable for iSER.\n\nSigned-off-by: Erez Zilber \u003cerezz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a4ef1451dfba92f51934e8331f634497b9ed3393", "tree": "049cdae2c051fd16d8c3bc2570cc078c4afe195e", "parents": [ "a3cd7d9070be417a21905c997ee32d756d999b38" ], "author": { "name": "Erez Zilber", "email": "erezz@voltaire.com", "time": "Thu Jan 17 11:51:58 2008 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:43 2008 -0800" }, "message": "IB/iser: Print information about unhandled RDMA CM events\n\nSome RDMA CM events are not supported or not handled in iSER.\nThis patch adds some info (printk) for the user about them.\n\nSigned-off-by: Erez Zilber \u003cerezz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a3cd7d9070be417a21905c997ee32d756d999b38", "tree": "1576c829374700bb774ed6e560aa9be318240526", "parents": [ "a656eb758fc6e6a42659ecf5ba34a5c5a2aeec17" ], "author": { "name": "Olaf Kirch", "email": "olaf.kirch@oracle.com", "time": "Wed Jan 16 18:37:16 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:43 2008 -0800" }, "message": "IB/fmr_pool: ib_fmr_pool_flush() should flush all dirty FMRs\n\nWhen a FMR is released via ib_fmr_pool_unmap(), the FMR usually ends\nup on the free_list rather than the dirty_list (because we allow a\ncertain number of remappings before actually requiring a flush).\n\nHowever, ib_fmr_batch_release() only looks at dirty_list when flushing\nout old mappings. This means that when ib_fmr_pool_flush() is used to\nforce a flush of the FMR pool, some dirty FMRs that have not reached\ntheir maximum remap count will not actually be flushed.\n\nFix this by flushing all FMRs that have been used at least once in\nib_fmr_batch_release().\n\nSigned-off-by: Olaf Kirch \u003colaf.kirch@oracle.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a656eb758fc6e6a42659ecf5ba34a5c5a2aeec17", "tree": "779bc9dea9396aa1907528472108a90ef3f2bce8", "parents": [ "2fe7e6f7c9f55eac24c5b3cdf56af29ab9b0ca81" ], "author": { "name": "Olaf Kirch", "email": "olaf.kirch@oracle.com", "time": "Wed Jan 16 18:36:27 2008 +0100" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "message": "IB/fmr_pool: Flush serial numbers can get out of sync\n\nNormally, the serial numbers for flush requests and flushes executed\nfor an FMR pool should be in sync.\n\nHowever, if the FMR pool flushes dirty FMRs because the\ndirty_watermark was reached, we wake up the cleanup thread and let it\ndo its stuff. As a side effect, the cleanup thread increments\npool-\u003eflush_ser, which leaves it one higher than pool-\u003ereq_ser. The\nnext time the user calls ib_flush_fmr_pool(), the cleanup thread will\nbe woken up, but ib_flush_fmr_pool() won\u0027t wait for the flush to\ncomplete because flush_ser is already past req_ser. This means the\nFMRs that the user expects to be flushed may not have all been flushed\nwhen the function returns.\n\nFix this by telling the cleanup thread to do work exclusively by\nincrementing req_ser, and by moving the comparison of dirty_len and\ndirty_watermark into ib_fmr_pool_unmap().\n\nSigned-off-by: Olaf Kirch \u003colaf.kirch@oracle.com\u003e\n" }, { "commit": "2fe7e6f7c9f55eac24c5b3cdf56af29ab9b0ca81", "tree": "0e3712ae9033d133e839a22ac28144215a7ece4a", "parents": [ "cf9542aa923982428fbf6a6f815c32ae2c3da8c7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "message": "IB/umad: Simplify and fix locking\n\nIn addition to being overly complex, the locking in user_mad.c is\nbroken: there were multiple reports of deadlocks and lockdep warnings.\nIn particular it seems that a single thread may end up trying to take\nthe same rwsem for reading more than once, which is explicitly\nforbidden in the comments in \u003clinux/rwsem.h\u003e.\n\nTo solve this, we change the locking to use plain mutexes instead of\nrwsems. There is one mutex per open file, which protects the contents\nof the struct ib_umad_file, including the array of agents and list of\nqueued packets; and there is one mutex per struct ib_umad_port, which\nprotects the contents, including the list of open files. We never\nhold the file mutex across calls to functions like ib_unregister_mad_agent(),\nwhich can call back into other ib_umad code to queue a packet, and we\nalways hold the port mutex as long as we need to make sure that a\ndevice is not hot-unplugged from under us.\n\nThis even makes things nicer for users of the -rt patch, since we\nremove calls to downgrade_write() (which is not implemented in -rt).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "cf9542aa923982428fbf6a6f815c32ae2c3da8c7", "tree": "9e10e3b4920f1a98df8853fa59a52a20ac7ce8ec", "parents": [ "1d6e658e8e3247fcf0fa90c40fdfebf2e85a610e" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "message": "IB/ipath: Fix some sparse warnings about shadowed symbols\n\nThere are a few places in the ipath driver where a variable is\nre-declared within a block where it is already in scope. Most of these\nextra declarations can simply be removed, since the variable from the\nouter scope is used in a way so that it does not need to keep its\nvariable across the block with the re-declaration.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1d6e658e8e3247fcf0fa90c40fdfebf2e85a610e", "tree": "fed64880f5886d4991d9e6910985054faefee4f6", "parents": [ "1a7d2dce4169ed42310926a5675fffd0986caa26" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:42 2008 -0800" }, "message": "RDMA/cxgb3: Endianness annotation for irs field\n\nt3_rdma_init_wr.irs is a big-endian field, so declare it as __be32.\nThis fixes one sparse warning.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1a7d2dce4169ed42310926a5675fffd0986caa26", "tree": "ea3dcb6f1f3d114b89603fafb3f39f77292fab71", "parents": [ "5851bb893e5bb87150817c180ccddcf4e78db1b6" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Mon Oct 15 00:49:07 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:41 2008 -0800" }, "message": "IB/ehca: Use round_jiffies() for EQ polling timer\n\nUse round_jiffies() to align ehca\u0027s 1-second timer with other timers\nand potentially save power by sleeping cores for longer.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nAcked-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5851bb893e5bb87150817c180ccddcf4e78db1b6", "tree": "ce0a5e2e7b0888df4379ac3be7e75f505349f5c4", "parents": [ "1f813ca830f51ed404195750bdba86c41a8050ba" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Fri Jan 04 10:47:12 2008 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Jan 25 14:15:41 2008 -0800" }, "message": "RDMA/cma: Override default responder_resources with user value\n\nBy default, the responder_resources parameter is set to that received\nin a connection request. The passive side may override this value\nwhen accepting the connection. Use the value provided by the passive\nside when transitioning the QP to RTR state, rather than the value\ngiven in the connect request. Without this change, the RTR transition\nmay fail if the passive side supports fewer responder_resources than\nthat in the request.\n\nFor code consistency and to protect against QP destruction, restructure\noverriding initiator_depth to match how responder_resources is set.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" } ], "next": "1f813ca830f51ed404195750bdba86c41a8050ba" }