)]}' { "log": [ { "commit": "249bb070f5e821503c1118e1e87c0ccb1432d191", "tree": "8736af4ab6dfa5e95001194603225b96def849bb", "parents": [ "863b18f4b5e7d9e6903b353328cf6fa084dbb619" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Nov 04 18:56:13 2005 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Nov 10 16:09:17 2005 -0800" }, "message": "[PATCH] PCI: removed unneeded .owner field from struct pci_driver\n\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "78b9c0f91cf908616b8f9f356e1d1220e727ea88", "tree": "84a581820bff0fa9830f18138da02d929e4edcb9", "parents": [ "6b482c6779daaa893b277fc9b70767a7c2e7c5eb", "94382f3562e350ed7c8f7dcd6fc968bdece31328" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Nov 10 13:27:06 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Nov 10 13:27:06 2005 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband\n" }, { "commit": "94382f3562e350ed7c8f7dcd6fc968bdece31328", "tree": "cf4cb4d1d1dc79326f68511addc1391d7d81e0ce", "parents": [ "ae57e24a4006fd46b73d842ee99db9580ef74a02" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:18:23 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:51 2005 -0800" }, "message": "[IB] umad: further ib_unregister_mad_agent() deadlock fixes\n\nThe previous umad deadlock fix left ib_umad_kill_port() still\nvulnerable to deadlocking. This patch fixes that by downgrading our\nlock to a read lock when we might end up trying to reacquire the lock\nfor reading.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ae57e24a4006fd46b73d842ee99db9580ef74a02", "tree": "17115437026be55dcd74641be21561fecf33dcdb", "parents": [ "64044bcf75063cb5a6d42712886a712449df2ce3" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Wed Nov 09 14:59:57 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:51 2005 -0800" }, "message": "[IB] mthca: fix posting long lists of receive work requests\n\nIn Tavor mode, when posting a long list of receive work requests, a\ndoorbell must be rung every 256 requests. Add code to do this when\nrequired.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "64044bcf75063cb5a6d42712886a712449df2ce3", "tree": "984d2fc62c548af3d01450135f33b5b97aecf00b", "parents": [ "62abb8416f1923f4cef50ce9ce841b919275e3fb" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 09 12:23:17 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:51 2005 -0800" }, "message": "[IB] mthca: fix wraparound handling in mthca_cq_clean()\n\nHandle case where prod_index has wrapped around and become less than\ncq-\u003econs_index by checking that their difference as a signed int is\npositive rather than comparing directly.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "62abb8416f1923f4cef50ce9ce841b919275e3fb", "tree": "704990abeb22a51ed2722e92536d22135f60957f", "parents": [ "77369ed31daac51f4827c50d30f233c45480235a" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Wed Nov 09 11:30:14 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] mthca: fix posting of atomic operations\n\nThe size of work requests for atomic operations was computed\nincorrectly in mthca: all sizeofs need to be divided by 16.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "77369ed31daac51f4827c50d30f233c45480235a", "tree": "308dce81364b1cbb563942a1a57146c1808e8911", "parents": [ "ec914c52d6208d8752dfd85b48a9aff304911434" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Wed Nov 09 11:26:07 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] uverbs: have kernel return QP capabilities\n\nMove the computation of QP capabilities (max scatter/gather entries,\nmax inline data, etc) into the kernel, and have the uverbs module\nreturn the values as part of the create QP response. This keeps\nprecise knowledge of device limits in the low-level kernel driver.\n\nThis requires an ABI bump, so while we\u0027re making changes, get rid of\nthe max_sge parameter for the modify SRQ command -- it\u0027s not used and\nshouldn\u0027t be there.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ec914c52d6208d8752dfd85b48a9aff304911434", "tree": "2741f22c820fb664f6958becc4f3d415eea0e61b", "parents": [ "40de2e548c225e3ef859e3c60de9785e37e1b5b1" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 09 09:58:10 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] umad: get rid of unused mr array\n\nNow that ib_umad uses the new MAD sending interface, it no longer\nneeds its own L_Key. So just delete the array of MRs that it keeps.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "40de2e548c225e3ef859e3c60de9785e37e1b5b1", "tree": "e7b9ffe6fca9246f29a0a3cdf6417770f5821cef", "parents": [ "0b4ff2c0e624089ad87dc1604e239b7c3201c53f" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Nov 08 11:10:25 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] Have cq_resize() method take an int, not int*\n\nChange the struct ib_device.resize_cq() method to take a plain integer\nthat holds the new CQ size, rather than a pointer to an integer that\nit uses to return the new size. This makes the interface match the\nexported ib_resize_cq() signature, and allows the low-level driver to\nupdate the CQ size with proper locking if necessary.\n\nNo in-tree drivers are exporting this method yet.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0b4ff2c0e624089ad87dc1604e239b7c3201c53f", "tree": "08d94f59d6f80937db5d87f0bb60eafcedd811d1", "parents": [ "8c608a32e3cd7ff14498ad996ca32d1452245a97" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Nov 07 22:01:02 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] mthca: fix typo in catastrophic error polling\n\nFix a typo in the rearming of the catastrophic error polling timer: we\nshould rearm the timer as long as the stop flag is _not_ set.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8c608a32e3cd7ff14498ad996ca32d1452245a97", "tree": "f96eacca4c5259be3981190bd6d3297afee6fd8c", "parents": [ "2f76e82947b977a1008cfd2868351a701c93c69c" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Nov 07 10:49:38 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IPoIB] no need to set skb-\u003edev right before freeing skb\n\nFor cut-and-paste reasons, the IPoIB driver was setting skb-\u003edev right\nbefore calling dev_kfree_skb_any(). Get rid of this.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2f76e82947b977a1008cfd2868351a701c93c69c", "tree": "e6df3542a0a9dd36a8fa9939cdaeedf90f4884a5", "parents": [ "1732b0ef3b3a02e3df328086fb3018741c5476da" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Nov 07 10:41:29 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:50 2005 -0800" }, "message": "[IB] umad: avoid potential deadlock when unregistering MAD agents\n\nib_unregister_mad_agent() completes all pending MAD sends and waits\nfor the agent\u0027s send_handler routine to return. umad\u0027s send_handler()\ncalls queue_packet(), which does down_read() on the port mutex to look\nup the agent ID. This means that the port mutex cannot be held for\nwriting while calling ib_unregister_mad_agent(), or else it will\ndeadlock. This patch fixes all the calls to ib_unregister_mad_agent()\nin the umad module to avoid this deadlock.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1732b0ef3b3a02e3df328086fb3018741c5476da", "tree": "380f460367321984ab7c7dcd72c37257d95de9a0", "parents": [ "8b37b94721533f2729c79bcb6fa0bb3e2bc2f400" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Nov 07 10:33:11 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 10 10:22:49 2005 -0800" }, "message": "[IPoIB] add path record information in debugfs\n\nAdd ibX_path files to debugfs that contain information about the IPoIB\npath cache. IPoIB ARP only gives GIDs, which the IPoIB driver must\nresolve to real IB paths through the ib_sa module. For debugging,\nwhen the ARP table looks OK but traffic isn\u0027t flowing, it\u0027s useful to\nbe able to see if the resolution from GID to path worked.\n\nAlso clean up the formatting of the existing _mcg debugfs files.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "733482e445ca4450cf41381b1c95e2b8c7145114", "tree": "d31a2fdaeb8e439fc348a781c780a035d794266d", "parents": [ "c1a0f5e3c01d28b6782457bee5ae5ace3a9958ec" ], "author": { "name": "Olaf Hering", "email": "olh@suse.de", "time": "Tue Nov 08 21:34:55 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Nov 09 07:55:57 2005 -0800" }, "message": "[PATCH] changing CONFIG_LOCALVERSION rebuilds too much, for no good reason\n\nThis patch removes almost all inclusions of linux/version.h. The 3\n#defines are unused in most of the touched files.\n\nA few drivers use the simple KERNEL_VERSION(a,b,c) macro, which is\nunfortunatly in linux/version.h.\n\nThere are also lots of #ifdef for long obsolete kernels, this was not\ntouched. In a few places, the linux/version.h include was move to where\nthe LINUX_VERSION_CODE was used.\n\nquilt vi `find * -type f -name \"*.[ch]\"|xargs grep -El \u0027(UTS_RELEASE|LINUX_VERSION_CODE|KERNEL_VERSION|linux/version.h)\u0027|grep -Ev \u0027(/(boot|coda|drm)/|~$)\u0027`\n\nsearch pattern:\n/UTS_RELEASE\\|LINUX_VERSION_CODE\\|KERNEL_VERSION\\|linux\\/\\(utsname\\|version\\).h\n\nSigned-off-by: Olaf Hering \u003colh@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "6044ec8882c726e325017bd948aa0cd94ad33abc", "tree": "3bfc5dc93434e8ad556540f6689abcd2699aa45d", "parents": [ "2ea7533060e361810c21b2f5ee02151c4dfb85d8" ], "author": { "name": "Jesper Juhl", "email": "jesper.juhl@gmail.com", "time": "Mon Nov 07 01:01:32 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Nov 07 07:54:05 2005 -0800" }, "message": "[PATCH] kfree cleanup: misc remaining drivers\n\nThis is the remaining misc drivers/ part of the big kfree cleanup patch.\n\nRemove pointless checks for NULL prior to calling kfree() in misc files in\ndrivers/.\n\nSigned-off-by: Jesper Juhl \u003cjesper.juhl@gmail.com\u003e\nAcked-by: Aristeu Sergio Rozanski Filho \u003caris@cathedrallabs.org\u003e\nAcked-by: Roland Dreier \u003crolandd@cisco.com\u003e\nAcked-by: Pierre Ossman \u003cdrzeus@drzeus.cx\u003e\nAcked-by: Jean Delvare \u003ckhali@linux-fr.org\u003e\nAcked-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\nAcked-by: Len Brown \u003clen.brown@intel.com\u003e\nAcked-by: \"Antonino A. Daplas\" \u003cadaplas@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8c65b4a60450590e79a28e9717ceffa9e4debb3f", "tree": "e0e42b5faee0a1c44746a36d9df7a8fbb2a2c24c", "parents": [ "6fdcc2162285a8fc96ab12ff85086c37bceaa494" ], "author": { "name": "Tim Schmielau", "email": "tim@physik3.uni-rostock.de", "time": "Mon Nov 07 00:59:43 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Nov 07 07:53:41 2005 -0800" }, "message": "[PATCH] fix remaining missing includes\n\nFix more include file problems that surfaced since I submitted the previous\nfix-missing-includes.patch. This should now allow not to include sched.h\nfrom module.h, which is done by a followup patch.\n\nSigned-off-by: Tim Schmielau \u003ctim@physik3.uni-rostock.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8b37b94721533f2729c79bcb6fa0bb3e2bc2f400", "tree": "f681c9c9ea858c5b14f593077e7cadf9e93ad255", "parents": [ "0f69ce1e4474e5d5e266457e8a1f4166cf71f6c7" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Sun Nov 06 15:47:02 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Nov 06 15:47:02 2005 -0800" }, "message": "[IB] umad: two small fixes\n\nTwo small fixes for the umad module:\n - set kobject name for issm device properly\n - in ib_umad_add_one(), s is subtracted from the index i when\n initializing ports, so s should be subtracted from the index when\n freeing ports in the error path as well.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0f69ce1e4474e5d5e266457e8a1f4166cf71f6c7", "tree": "ea857a2670e77bd8e8e8538f42504bcaa1a515d5", "parents": [ "7015faa7df829876a0f931cd18aa6d7c24a1b581" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Fri Nov 04 16:03:32 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Nov 04 21:28:21 2005 -0800" }, "message": "[IB] mthca: report page size capability\n\nReport the device\u0027s real page size capability in mthca_query_device().\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "127f2fa31ac624c744f3767363c4919209980956", "tree": "972b7ffbcd2c846bbb34879efb91e324dc0677be", "parents": [ "ba77df570c6710c9c19b31e0e48e4bcdf31cefe8", "aef9ec39c47f0cece886ddd6b53c440321e0b2a6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Nov 04 16:32:36 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Nov 04 16:32:36 2005 -0800" }, "message": "Merge branch \u0027srp\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband\n" }, { "commit": "ba77df570c6710c9c19b31e0e48e4bcdf31cefe8", "tree": "997c210135b5578982df256dd5a0140e04d972d1", "parents": [ "602d4a7e2f4b843d1a67375d4d7104073495b758", "d09e32764176b61c4afee9fd5e7fe04713bfa56f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Nov 04 16:31:54 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Nov 04 16:31:54 2005 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband\n" }, { "commit": "d09e32764176b61c4afee9fd5e7fe04713bfa56f", "tree": "1818a0043dcbc9789dbc12c6affe5a6924d65366", "parents": [ "0c99cb6d5fe77872c5a32cff837c05f70158ce15" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Thu Nov 03 14:58:33 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 03 14:58:33 2005 -0800" }, "message": "[IB] mthca: check P_Key index in modify QP\n\nMake sure that the P_Key index passed into mthca_modify_qp() is\nwithin the device\u0027s P_Key table.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0c99cb6d5fe77872c5a32cff837c05f70158ce15", "tree": "b974a31452cb645f063589262bde09b6c5b05701", "parents": [ "87cfe32375e0b69b999b59bf8287f501df3e43f7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 03 12:01:18 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Nov 03 12:01:18 2005 -0800" }, "message": "[IB] umad: fix hot remove of IB devices\n\nFix hotplug of devices for ib_umad module: when a device goes away,\nkill off all MAD agents for open files associated with that device,\nand make sure that the device is not touched again after ib_umad\nreturns from its remove_one function.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "87cfe32375e0b69b999b59bf8287f501df3e43f7", "tree": "2cbc1b1e7bb230afcf4903b6527e3238f689de89", "parents": [ "8ae5a8a24f7fe797027d481f88c1464b0e47eede" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 22:59:37 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 22:59:37 2005 -0800" }, "message": "[IB] mthca: fix format of FW version\n\nMellanox has decided that the components of the firmware version are\nreally meant to be displayed in decimal, e.g. 0x000400070190 is\nversion 4.7.400. Change the format we use from \"%x.%x.%x\" to\n\"%d.%d.%d\" to match this convention.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8ae5a8a24f7fe797027d481f88c1464b0e47eede", "tree": "389cecdfb0769cdddd0e901c1d60b9549b0a6322", "parents": [ "21a384897d48c116b879924c3dd9e96f6f1e764b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 20:51:01 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 20:51:01 2005 -0800" }, "message": "[IPoIB] don\u0027t compile debug code if debugging isn\u0027t enabled\n\nDon\u0027t build ipoib_mcast_iter_ functions if CONFIG_INFINIBAND_IPOIB_DEBUG\nis not enabled -- their only callers will not be built either.\n\nAlso move the prototype for ipoib_open() to ipoib.h to fix a sparse warning.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "aef9ec39c47f0cece886ddd6b53c440321e0b2a6", "tree": "d9a0f0fb69ef336bec4317ef5f95adc8c062e445", "parents": [ "89fbb69c4f285019ba5e029963dc11cc6beb078a" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 14:07:13 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 14:07:13 2005 -0800" }, "message": "IB: Add SCSI RDMA Protocol (SRP) initiator\n\nAdd an InfiniBand SCSI RDMA Protocol (SRP) initiator. This driver is\nused to talk talk to InfiniBand SRP targets (storage devices).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "21a384897d48c116b879924c3dd9e96f6f1e764b", "tree": "3179960b8e0f3ccb4feff19eb5582298d48324a0", "parents": [ "de6eb66b56d9df5ce6bd254994f05e065214e8cd" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 10:07:59 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 10:07:59 2005 -0800" }, "message": "[IPoIB] remove unneeded initializations to 0\n\nShrink our source and .text a little by removing a few assignments of\nNULL and 0 to memory that is already cleared as part of the allocation.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "de6eb66b56d9df5ce6bd254994f05e065214e8cd", "tree": "7463446a05b5e9a5d2fc400da0be8d4a6c2ff6f1", "parents": [ "7b28b0d000eeb62d77add636f5d6eb0da04e48aa" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 07:23:14 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Nov 02 07:23:14 2005 -0800" }, "message": "[IB] kzalloc() conversions\n\nReplace kmalloc()+memset(,0,) with kzalloc(), for a net savings of 35\nsource lines and about 500 bytes of text.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3bc12e75b23c0499cc2c0873a5f77494be173761", "tree": "ecb02f68e1055343bb45fc38350a8e33c827efc9", "parents": [ "e5b251a24a9cd34a7ef98e361eb94e7ab122a554" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Oct 30 13:20:09 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 31 07:10:33 2005 -0800" }, "message": "[IPoIB] cleanups: fix comment, remove useless variables\n\nMinor cleanups: fix a misleading comment, and get rid of attr_mask\nvariables that are only used to hold constants (just use the constants\ndirectly).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e5b251a24a9cd34a7ef98e361eb94e7ab122a554", "tree": "c29760bafd7107252389712965ad7e4ed0791a82", "parents": [ "7162a3e0db34e914a8bc5bf74bbae0b386310cf8" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Oct 30 13:07:03 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 31 07:10:32 2005 -0800" }, "message": "[IB] mthca: Avoid SRQ free WQE list corruption\n\nFix wqe_to_link() to use a structure field that we know is definitely\nalways unused for receive work requests, so that it really avoids the\nfree list corruption bug that the comment claims it does.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7162a3e0db34e914a8bc5bf74bbae0b386310cf8", "tree": "ffd7eba03f29dd2932dd32ac4adc2921bde7644b", "parents": [ "a20583a7c2e35d80b1dfc1f60c9729498838725e" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Oct 30 09:50:04 2005 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 31 07:10:32 2005 -0800" }, "message": "[IB] uverbs: Avoid NULL pointer deref on CQ async event\n\nUserspace CQs that have no completion event channel attached end up\nwith their cq_context set to NULL. However, asynchronous events like\n\"CQ overrun\" can still occur on such CQs, so add a uverbs_file member\nto struct ib_ucq_object that we can follow to deliver these events.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4e57b6817880946a3a78d5d8cad1ace363f7e449", "tree": "b6b5f3f9e8e52cc55d98239a4992e72e983c8fa4", "parents": [ "b0423a0d9cc836b2c3d796623cd19236bfedfe63" ], "author": { "name": "Tim Schmielau", "email": "tim@physik3.uni-rostock.de", "time": "Sun Oct 30 15:03:48 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 30 17:37:32 2005 -0800" }, "message": "[PATCH] fix missing includes\n\nI recently picked up my older work to remove unnecessary #includes of\nsched.h, starting from a patch by Dave Jones to not include sched.h\nfrom module.h. This reduces the number of indirect includes of sched.h\nby ~300. Another ~400 pointless direct includes can be removed after\nthis disentangling (patch to follow later).\nHowever, quite a few indirect includes need to be fixed up for this.\n\nIn order to feed the patches through -mm with as little disturbance as\npossible, I\u0027ve split out the fixes I accumulated up to now (complete for\ni386 and x86_64, more archs to follow later) and post them before the real\npatch. This way this large part of the patch is kept simple with only\nadding #includes, and all hunks are independent of each other. So if any\nhunk rejects or gets in the way of other patches, just drop it. My scripts\nwill pick it up again in the next round.\n\nSigned-off-by: Tim Schmielau \u003ctim@physik3.uni-rostock.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "a20583a7c2e35d80b1dfc1f60c9729498838725e", "tree": "e7fbd856e7522b65d309e9dfd425541d8f45a0bd", "parents": [ "affcd50546d4788b7849e2b2e2ec7bc50d64c5f8" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Oct 29 13:54:40 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Oct 29 13:54:40 2005 -0700" }, "message": "[IPoIB] use spin_trylock_irqsave()\n\nUse spin_trylock_irqsave() in ipoib_start_xmit() instead of\nreinventing it out of local_irq_save(), spin_trylock() and\nlocal_irq_restore().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2d3c0b7bedfd7a8e1870a046a434bd467e7c5349", "tree": "1073d1e05f604faeebbde344a0d4e624fc116d65", "parents": [ "4b4a5eaedfc098d825d4c34cad1f1115802512b7" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Sat Oct 29 06:46:03 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 29 10:35:07 2005 -0700" }, "message": "[PATCH] missing include in infiniband\n\nuse of IS_ERR/PTR_ERR in infiniband/core/agent.c, without a portable\nchain of includes pulling err.h (breaks on a bunch of platforms).\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nAcked-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "affcd50546d4788b7849e2b2e2ec7bc50d64c5f8", "tree": "e4e6a0f5f2203569b6ada4c101a146c3a4f24c28", "parents": [ "89fbb69c4f285019ba5e029963dc11cc6beb078a" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Sat Oct 29 07:39:42 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Oct 29 07:39:42 2005 -0700" }, "message": "[IB] mthca: report asynchronous CQ events\n\nImplement reporting asynchronous CQ events in Mellanox HCA driver.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4cce3390c998600f6648e075e475cf8f6dd8cebe", "tree": "2b800e2e2d890e98db792ca67590e65d22384926", "parents": [ "a4da0628efd788eb199dd9df225e296035ec2539" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 16:38:15 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 16:38:15 2005 -0700" }, "message": "[IB] fix up class_device_create() calls\n\nFix class_device_create() calls to match the new prototype which\ntakes a parent device pointer.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a4da0628efd788eb199dd9df225e296035ec2539", "tree": "1e611030c993e1480dc2f07f26976d2d33002b0d", "parents": [ "70a30e16a8a9d22396a4d1e96af86e43594df584", "20731945ae743034353a88c307920d1f16cf8ac8" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:50:51 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:50:51 2005 -0700" }, "message": "Merge master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6\n" }, { "commit": "70a30e16a8a9d22396a4d1e96af86e43594df584", "tree": "0d8582567c62f7929620ec03128c79f9d8c65a58", "parents": [ "a74968f8c3b1166cfe0942901b56165f06ab6f60" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:38:26 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:38:26 2005 -0700" }, "message": "[IB] uverbs: Fix device lifetime problems\n\nMove ib_uverbs module to using cdev_alloc() and class_device_create()\nso that we can handle device lifetime properly. Now we can make sure\nwe keep all of our data structures around until the last way to reach\nthem is gone.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a74968f8c3b1166cfe0942901b56165f06ab6f60", "tree": "8b6bfdad7c5a56186bb84a8f282270998f64db0f", "parents": [ "1993d683f39f77ddb46a662d7146247877d50b8f" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:37:23 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:37:23 2005 -0700" }, "message": "[IB] umad: Fix device lifetime problems\n\nMove ib_umad module to using cdev_alloc() and class_device_create() so\nthat we can handle device lifetime properly. Now we can make sure we\nkeep all of our data structures around until the last way to reach\nthem is gone.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1993d683f39f77ddb46a662d7146247877d50b8f", "tree": "608391d9daa25e13b70cc11d6b99f800a7079050", "parents": [ "cb0f0910f4b41772a6771bdb4fb2d419b27bcd77" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:30:34 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 28 15:30:34 2005 -0700" }, "message": "[IPoIB] Drop RX packets when out of memory\n\nChange the way IPoIB handles RX packets when it can\u0027t allocate a new\nreceive skbuff. If the allocation of a new receive skb fails, we now\ndrop the packet we just received and repost the original receive skb.\nThis means that the receive ring always stays full and we don\u0027t have\nto monkey around with trying to schedule a refill task for later.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6fbfddcb52d8d9fa2cd209f5ac2a1c87497d55b5", "tree": "c0414e89678fcef7ce3493e048d855bde781ae8d", "parents": [ "1a222bca26ca691e83be1b08f5e96ae96d0d8cae", "27d1097d39509494706eaa2620ef3b1e780a3224" ], "author": { "name": "Greg KH", "email": "greg@press.(none)", "time": "Fri Oct 28 10:13:16 2005 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Oct 28 10:13:16 2005 -0700" }, "message": "Merge ../bleed-2.6\n" }, { "commit": "53f4654272df7c51064825024340554b39c9efba", "tree": "e3e7b82a6bb0040ffbd267b250be2720704b98f2", "parents": [ "51d172d5f3a193e4b8f76179b2e55d7a36b94117" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Thu Oct 27 22:25:43 2005 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Oct 28 09:52:52 2005 -0700" }, "message": "[PATCH] Driver Core: fix up all callers of class_device_create()\n\nThe previous patch adding the ability to nest struct class_device\nchanged the paramaters to the call class_device_create(). This patch\nfixes up all in-kernel users of the function.\n\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "87b750dc4b7109aa744e7d331dc93df3fe5c1c28", "tree": "a76309b353fc5c04eca8f6b5753b65e68f00c4d3", "parents": [ "c53033f6b0bd7cc133b7f433083f0394cf29ac70" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Oct 21 03:22:13 2005 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Oct 28 08:16:50 2005 -0700" }, "message": "[PATCH] gfp_t: drivers/infiniband\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "cb0f0910f4b41772a6771bdb4fb2d419b27bcd77", "tree": "fe34505915c31b3b77296e3828cd7675f4616ba9", "parents": [ "089a1bedd84be16a4f49a319e7ccb4a128da5ce9" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Thu Oct 27 20:48:11 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 27 20:48:11 2005 -0700" }, "message": "[IB] ib_umad: various cleanups\n\nSimplify user_mad.c code in a few places, and convert from kmalloc() +\nmemset() to kzalloc(). This also fixes a theoretical race window by\nnot accessing packet-\u003elength after posting the send buffer (the send\ncould complete and packet could be freed before we get to the return\nstatement at the end of ib_umad_write()).\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "089a1bedd84be16a4f49a319e7ccb4a128da5ce9", "tree": "d2bad46f16a76769b1f8d87aad369d50b4966cb6", "parents": [ "3d155f8cd0d077938d271225d26ee52f8eb26082" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 27 20:33:43 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 27 20:33:43 2005 -0700" }, "message": "[IB] ib_umad: fix crash when freeing send buffers\n\nThe conversion of user_mad.c to the new MAD send API was slightly off:\nin a few places, we used packet-\u003emsg instead of packet-\u003emsg-\u003emad when\nreferring to the actual data buffer, which ended up corrupting the\nunderlying data structure and crashing when we free an invalid pointer.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3d155f8cd0d077938d271225d26ee52f8eb26082", "tree": "28f65ed2dc5e9f12884daf2a97f50367f145d8a0", "parents": [ "7cc656efb560cda66b5ed48444cad7556ea4fe99" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 27 11:03:38 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 27 11:03:38 2005 -0700" }, "message": "[IB] mthca: first pass at catastrophic error reporting\n\nAdd some initial support for detecting and reporting catastrophic\nerrors reported by Mellanox HCAs. We start a periodic timer which\npolls the catastrophic error reporting buffer in device memory. If an\nerror is detected, we dump the contents of the buffer for port-mortem\ndebugging, and report a fatal asynchronous error to higher levels.\n\nIn the future we can try to recover from these errors by resetting the\ndevice, but this will require some work in higher-level code as well.\nLet\u0027s get this in now, so that we at least get catastrophic errors\nreported in logs.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7cc656efb560cda66b5ed48444cad7556ea4fe99", "tree": "bfc4d1e256e755623c9782b1a0883fc337d88e6e", "parents": [ "547e3090738b04be650770b64265835dbb6ddf92" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 25 15:13:54 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 25 15:13:54 2005 -0700" }, "message": "[IB] simplify mad_rmpp.c:alloc_response_msg()\n\nChange alloc_response_msg() in mad_rmpp.c to return the struct\nit allocates directly (or an error code a la ERR_PTR), rather than\nreturning a status and passing the struct back in a pointer param.\nThis simplifies the code and gets rid of warnings like\n\n drivers/infiniband/core/mad_rmpp.c: In function nack_recv:\n drivers/infiniband/core/mad_rmpp.c:192: warning: msg may be used uninitialized in this function\n\nwith newer versions of gcc.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "547e3090738b04be650770b64265835dbb6ddf92", "tree": "9329491379623349a7ef3a42be9bad302eea011d", "parents": [ "34816ad98efe4d47ffd858a0345321f9d85d9420" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 25 10:57:32 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 25 10:57:32 2005 -0700" }, "message": "[IB] mthca: correct modify QP attribute masks for UC\n\nThe UC transport does not support RDMA reads or atomic operations, so\nwe shouldn\u0027t require or even allow the consumer to set attributes\nrelating to these operations for UC QPs.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "34816ad98efe4d47ffd858a0345321f9d85d9420", "tree": "8a5ed6a9b80e667c4c02d9993711ced06d158555", "parents": [ "ae7971a7706384ca373fb7e212fe195698e6c5a1" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Tue Oct 25 10:51:39 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 25 10:51:39 2005 -0700" }, "message": "[IB] Fix MAD layer DMA mappings to avoid touching data buffer once mapped\n\nThe MAD layer was violating the DMA API by touching data buffers used\nfor sends after the DMA mapping was done. This causes problems on\nnon-cache-coherent architectures, because the device doing DMA won\u0027t\nsee updates to the payload buffers that exist only in the CPU cache.\n\nFix this by having all MAD consumers use ib_create_send_mad() to\nallocate their send buffers, and moving the DMA mapping into the MAD\nlayer so it can be done just before calling send (and after any\nmodifications of the send buffer by the MAD layer).\n\nTested on a non-cache-coherent PowerPC 440SPe system.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ae7971a7706384ca373fb7e212fe195698e6c5a1", "tree": "5c0d0fecf3faa6c899f8438c4589fa8dede9da72", "parents": [ "ec329a135974f1c400214dab02f09584d4beeba9" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Mon Oct 24 12:33:56 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 24 12:33:56 2005 -0700" }, "message": "[IB] CM: Fix initialization of QP attributes for UC QPs.\n\nFix cm_init_qp_init_attr(), cm_init_qp_rtr_attr() and cm_init_qp_rts_attr()\nso that they correctly handle the differences between UC and RC QPs. This\nfixes problems with setting up UC QPs through the CM.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ec329a135974f1c400214dab02f09584d4beeba9", "tree": "235be754f76f4e7d51975d2434f226cfc75cdd08", "parents": [ "bbf207860931b6a033d0fbcd170ae2332c0d8216", "5d7edb3c1a01310725d86f0d83fb3be45685dc82" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 24 10:55:29 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 24 10:55:29 2005 -0700" }, "message": "Manual merge of for-linus to upstream (fix conflicts in drivers/infiniband/core/ucm.c)\n" }, { "commit": "5d7edb3c1a01310725d86f0d83fb3be45685dc82", "tree": "bbe21779313fea41eb644e0e0f06d5949cd4ed49", "parents": [ "e7d311da9cba1e113f82176130d3af1be77dc3ee" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 24 10:53:25 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 24 10:53:25 2005 -0700" }, "message": "[IB] Add idr_destroy() calls on module unload\n\nAdd idr_destroy() calls to the module_exit() functions of the four IB\ndriver modules that use idrs, so we don\u0027t leak idr_layer_cache objects\nwhen these modules are unloaded.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "75eeec2f3fd9e8a16777219ebf1bf8395845faa7", "tree": "347883a070001786aafdbf284d8d6971ab44dbff", "parents": [ "8d3b35914aa54232b27e6a2b57d84092aadc5e86" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Oct 23 12:57:19 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 23 16:38:39 2005 -0700" }, "message": "[PATCH] ib: mthca: Always re-arm EQs in mthca_tavor_interrupt()\n\nWe should always re-arm an event queue\u0027s interrupt in\nmthca_tavor_interrupt() if the corresponding bit is set in the event cause\nregister (ECR), even if we didn\u0027t find any entries in the EQ. If we don\u0027t,\nthen there\u0027s a window where we miss an EQ entry and then get stuck because\nwe don\u0027t get another EQ event.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "c8e0ca683dfea7242ff29814561dfe761945e5b3", "tree": "9c7be6ca5f815f72c6be614576ad6ef6c3110f8c", "parents": [ "2cc78eb52bc1ae89f0a4fa5a00eb998dffde4a9f" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Oct 22 09:43:29 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sat Oct 22 09:43:29 2005 -0700" }, "message": "[IB] mthca: Always re-arm EQs in mthca_tavor_interrupt()\n\nWe should always re-arm an event queue\u0027s interrupt in\nmthca_tavor_interrupt() if the corresponding bit is set in the event\ncause register (ECR), even if we didn\u0027t find any entries in the EQ.\nIf we don\u0027t, then there\u0027s a window where we miss an EQ entry and then\nget stuck because we don\u0027t get another EQ event.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bbf207860931b6a033d0fbcd170ae2332c0d8216", "tree": "9ffa1c4d1b045e1b5bfbe23a8e8f07973d170c9a", "parents": [ "2e0c512aff978a7040464e81fc9c0dfdf4639c23" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:54:01 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:54:01 2005 -0700" }, "message": "[IB] user_mad: Use class_device.devt\n\nUse devt member of struct class_device so that we don\u0027t have to create\nour own \"dev\" file in sysfs.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2e0c512aff978a7040464e81fc9c0dfdf4639c23", "tree": "0b278a811609e159bcaa134099af1692cdabd3ee", "parents": [ "3910f44d79a8a4a26fcfefbf16c5b8e1a7743175" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:30:16 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:30:16 2005 -0700" }, "message": "[IB] user_mad: trivial coding style fixes\n\nAdd spaces after \"sizeof\" operator to match the rest of file.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3910f44d79a8a4a26fcfefbf16c5b8e1a7743175", "tree": "5dd548fbe2fdabecebde7e6cca0665730f3c767b", "parents": [ "7150bf8a98f14f1ba67e090ef2778004c746f465" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:29:36 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 20 12:29:36 2005 -0700" }, "message": "[IB] cm: Add missing break in switch\n\nAdd missing \"break\" in switch statement. Without the break, the\nCM ended up always falling through and setting every connection\nrequest to use RC transport, which meant that UC connections\ndidn\u0027t work.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7150bf8a98f14f1ba67e090ef2778004c746f465", "tree": "6fd181880b24748b91029fddc629d1dd6f47d1c3", "parents": [ "ba8e9310244180bbc820e865d2e4378809b4a443" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Tue Oct 18 14:46:38 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 14:46:38 2005 -0700" }, "message": "[IB] mthca: Don\u0027t enter QP into MCG more than once.\n\nAvoid entering a QP as member of a multicast group multiple times.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ba8e9310244180bbc820e865d2e4378809b4a443", "tree": "407dc58bb932a7cd09df5c5b0c13f28cfd7f1c61", "parents": [ "d476306f1cc9156ce4b42586aacce70a356ef656" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 14:14:56 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 14:14:56 2005 -0700" }, "message": "[IB] Fail sysfs queries after device is unregistered\n\nWe keep IB device structures around until the last sysfs reference is\ngone, but we shouldn\u0027t ask the low-level driver to do anything after\nthe LLD unregisters the device. To handle this, check the reg_state\nfield and just fail sysfs show() requests if the device has already\nbeen unregistered.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d476306f1cc9156ce4b42586aacce70a356ef656", "tree": "045da3b196f276bbe83cd25f08956578d5e918f6", "parents": [ "c6f5cb7be0ab478e0618e3c2c6ada27f56d1e7fb" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 14:07:07 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 14:07:07 2005 -0700" }, "message": "[IB] mthca: Add struct pci_driver.owner field\n\nSet mthca_driver.owner to THIS_MODULE.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c6f5cb7be0ab478e0618e3c2c6ada27f56d1e7fb", "tree": "37446873b146ad0e8a4d3c5c6378e4400e797544", "parents": [ "4b2d319b53810ab00ef3d8fdfc1c1ab0647ab0a7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 13:22:16 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 13:22:16 2005 -0700" }, "message": "[IB] mthca: Use enum in mthca_alloc_db() prototype\n\nMake the type parameter of mthca_alloc_db() be an enum mthca_db_type\ninstead of an int. This doesn\u0027t have any practical effect but\ndocuments the functions a little better.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4b2d319b53810ab00ef3d8fdfc1c1ab0647ab0a7", "tree": "e54ac6da33e561bb38cf31d2d1ba43b913983cab", "parents": [ "762a03e21ed08daa5170d9cd98e83539861e7692" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 12:20:06 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 18 12:20:06 2005 -0700" }, "message": "[IPoIB] Improve ipoib_timeout() output\n\nUse jiffies_to_msecs() so we print a human-readable time so\nwe don\u0027t have to worry about what HZ is configured to, and\nprint out a few values to make post-mortem analysis easier.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "762a03e21ed08daa5170d9cd98e83539861e7692", "tree": "43c626e5100387e9e66b8cc152194f51701a08db", "parents": [ "07d357d0cbf89d9980b1769d5444a3c70f000e00" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:38:50 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:38:50 2005 -0700" }, "message": "[IB] ucm: quiet sparse warnings\n\nMake ctx_id_mutex and ctx_id_table static to quiet sparse warnings.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "07d357d0cbf89d9980b1769d5444a3c70f000e00", "tree": "cc7bcdee52b4e79e2115295e763f2e3d49c68a86", "parents": [ "595e726a1f28420c5fc7970b1a87cbce77a1cd45" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Mon Oct 17 15:37:43 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:37:43 2005 -0700" }, "message": "[IB] CM: bind IDs to a specific device\n\nBind communication identifiers to a device to support device removal.\nExport per HCA CM devices to userspace.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\n" }, { "commit": "595e726a1f28420c5fc7970b1a87cbce77a1cd45", "tree": "743f39eb3f33c9f0a58a96cf4c7213ca06af50a9", "parents": [ "67cdb40ca444c09853ab4d8a41cf547ac26a4de4" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Mon Oct 17 15:33:47 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:33:47 2005 -0700" }, "message": "[IB] merge ucm.h into ucm.c\n\nEliminate ucm.h. Replace ucm_dbg with direct call to printk KERN_ERR.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\n" }, { "commit": "67cdb40ca444c09853ab4d8a41cf547ac26a4de4", "tree": "fc0ecdf4e2c39db11849c79c594711c9e8e76a13", "parents": [ "91ecd4ae178bed83da4f6a94ced7992e4d7711eb" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 14 15:26:04 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:31 2005 -0700" }, "message": "[IB] uverbs: Implement more commands\n\nAdd kernel support for userspace calling poll CQ, request CQ\nnotification, post send, post receive, post SRQ receive, create AH and\ndestroy AH commands. These commands allow us to support userspace\nverbs for devices that can\u0027t perform these operations directly from\nuserspace (eg the PathScale HCA).\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "883a99c7024c5763d6d4f22d9239c133893e8d74", "tree": "659a7982a51d0ba187dac826a5b1e5ef319692fd", "parents": [ "56c202d6e4f542468fe79b8a735cf128898b87fb" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Oct 14 14:00:58 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:30 2005 -0700" }, "message": "[IB] uverbs: Add a mask of device methods allowed for userspace\n\nGive each device a uverbs_cmd_mask, so that a low-level driver can\ncontrol which methods may be called on behalf of userspace.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "56c202d6e4f542468fe79b8a735cf128898b87fb", "tree": "3080b34f6080d71b88ae26f665b5e23d7f485c69", "parents": [ "305a7e87055dde2466a855c24ef51a606915d149" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 13 10:45:02 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:29 2005 -0700" }, "message": "[IB] fail SA queries if device initialization failed\n\nIf the SA query module\u0027s initialization fails for a device, then that\ndevice won\u0027t have a struct ib_sa_device associated. We should fail SA\nqueries in that case, rather than blindly dereferencing the NULL\npointer we get back from ib_get_client_data().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "305a7e87055dde2466a855c24ef51a606915d149", "tree": "7cfe3df3901ad99677dc1794d684894db8bf5a38", "parents": [ "5b6810e048435de508ef66aebd6b78db13d651b8" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 11 15:39:38 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:29 2005 -0700" }, "message": "[IB] uverbs: unlock correctly in error paths\n\nA couple of functions were missing spin_unlock calls in error paths.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5b6810e048435de508ef66aebd6b78db13d651b8", "tree": "b54df96d5821a6ed41d778bf1ecfb54dbc7a4999", "parents": [ "efaae8f71f3088cc73c9e5ceabbd314aa82ac768" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 11 11:08:24 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:29 2005 -0700" }, "message": "[IPoIB] Rename ipoib_create_qp() -\u003e ipoib_init_qp() and fix error cleanup\n\nipoib_create_qp() no longer creates IPoIB\u0027s QP, so it shouldn\u0027t\ndestroy the QP on failure -- that unwinding happens elsewhere, so the\ncurrent code can cause a double free. While we\u0027re at it, the\nfunction\u0027s name should match what it actually does, so rename it to\nipoib_init_qp().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "efaae8f71f3088cc73c9e5ceabbd314aa82ac768", "tree": "b19084ae0a9dc6837cb2837cbd21386c9f1b2e0e", "parents": [ "4ab6fb7e5b3d34b65a1c3473d80d9d1a462d3a49" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Mon Oct 10 13:48:07 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:29 2005 -0700" }, "message": "[IB] mthca: Better limit checking and reporting\n\nCheck the sizes of CQs, QPs and SRQs when creating objects, and fail\ninstead of creating too-big queues. Also return real limits instead\nof just plausible-sounding values from mthca_query_device().\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4ab6fb7e5b3d34b65a1c3473d80d9d1a462d3a49", "tree": "3a78a17d4499bf95cc23be72ffb88bd49c8e7852", "parents": [ "e23d6d2b090658007732770720a44375cba23200" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 06 13:28:16 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:28 2005 -0700" }, "message": "[IB] Fix leak on MAD initialization failure\n\nThere is a bug in ib_mad_init_device(): if ib_agent_port_open() fails\nfor a given port, then the current code doesn\u0027t call ib_mad_port_close()\nfor that port.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e23d6d2b090658007732770720a44375cba23200", "tree": "277afdc20477a3ab8ca9a1ff26c478e047a3a581", "parents": [ "90f104da22bbf2e2b8a2136c12fb4e013fccf504" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 06 13:25:16 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:28 2005 -0700" }, "message": "[IB] mthca: detect SRQ overflow\n\nThe hardware relies on us keeping one extra work request that never\ngets used in SRQs. Add checks to the SRQ work request posting\nfunctions so that they fail when someone is about to use up that extra\nwork request, rather than when someone uses the very last work request.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "90f104da22bbf2e2b8a2136c12fb4e013fccf504", "tree": "d720f4b226cd72903ee878096707578ccc7abc0c", "parents": [ "116c0074ecfd6f061570856bec52b691d54dbd3c" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Oct 06 13:15:56 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:28 2005 -0700" }, "message": "[IB] mthca: SRQ limit reached events\n\nOur hardware supports generating an event when the number of receives\nposted to a shared receive queue (SRQ) falls below a user-specified\nlimit. Implement mthca_modify_srq() to arm the limit, and add code to\nhandle dispatching SRQ events when they occur.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "116c0074ecfd6f061570856bec52b691d54dbd3c", "tree": "0f9524538accf39d965d270d63728ec43a5f26b8", "parents": [ "f575394f1de9b4afa4b474f1882c7e2d3b8e51d7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 03 09:32:33 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:27 2005 -0700" }, "message": "[IB] Check port number in ib_query_port()/ib_modify_port()\n\nCheck port number before passing query_port or modify_port operations\non to device driver.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f575394f1de9b4afa4b474f1882c7e2d3b8e51d7", "tree": "b9ffc5983cbba3acbb8d7233d26101c38f4d7e61", "parents": [ "9825051e8c81cbd4400333e7cdc04be77a0bab75" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 03 09:18:02 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:27 2005 -0700" }, "message": "[IB] uverbs: reject invalid memory registration permission flags\n\nReject userspace memory registrations with invalid permission flags:\n\"local write\" is required if \"remote write\" or \"remote atomic\" is also\nrequested.\n\nPointed out by Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9825051e8c81cbd4400333e7cdc04be77a0bab75", "tree": "0b14750f603b1feece6c147cb95f31619bfb67c2", "parents": [ "274c0891637c44ac71f3ac40be91b43c2318883a" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Sat Oct 01 13:09:44 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:26 2005 -0700" }, "message": "[IB] mthca: Fill in more fields in query_port method\n\nAdd code to fill in the bad_pkey_cntr, max_mtu, active_mtu and\nsubnet_timeout fields in mthca_query_port().\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "274c0891637c44ac71f3ac40be91b43c2318883a", "tree": "fc7d9f07ab7e663a8acb771069cb9f1053ca2302", "parents": [ "eb9d3cd5ce2828fbb99ed7270089ea152c8f64b3" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Sep 29 14:17:48 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:26 2005 -0700" }, "message": "[IB] uverbs: Add device-specific ABI version attribute\n\nAdd abi_version attribute to uverbs class devices to allow for\nABI versioning of device-specific interfaces.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "eb9d3cd5ce2828fbb99ed7270089ea152c8f64b3", "tree": "2d0de42bc6c7777c66ee76d5817312bcfd161718", "parents": [ "6b73597e7062118c0549c2702bfb7d273518c906" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 27 15:07:25 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:25 2005 -0700" }, "message": "[IB] uverbs: Fix up resource creation error paths\n\nBy waiting to add resources to our lists until after the last\noperation that can fail, we don\u0027t have to remove them from their lists\nin the error path. Also, we should hold the idr mutex until we know\nwhether resource creation has succeed or failed, to avoid someone\nfinding a resource in our table before we\u0027re ready.\n\nLoosely based on work by Robert Walsh \u003crjwalsh@pathscale.com\u003e.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6b73597e7062118c0549c2702bfb7d273518c906", "tree": "6034aae7493b32d75d1c8818a801b09b77979acd", "parents": [ "33033b797225553e48ca68d8d8dc5e64ec22e02b" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 13:53:25 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:25 2005 -0700" }, "message": "[IB] uverbs: ABI-breaking fixes for userspace verbs\n\nIntroduce new userspace verbs ABI version 3. This eliminates some\nunneeded commands, and adds support for user-created completion\nchannels. This cleans up problems with file leaks on error paths, and\nalso makes sure that file descriptors are always installed into the\ncorrect process.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "33033b797225553e48ca68d8d8dc5e64ec22e02b", "tree": "f76970269a0a16eb6b301f3d048af7670232d652", "parents": [ "0c33aeedb2cf99d877ad9adc7c3df07870f60293" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Mon Sep 26 12:30:02 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:24 2005 -0700" }, "message": "[IB] mthca: Report correct atomic capability\n\nReturn correct atomic capability flag from mthca query function.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0c33aeedb2cf99d877ad9adc7c3df07870f60293", "tree": "7f6b23172b327ebd8b5d4f5d40aa216ba48600a2", "parents": [ "2cc78eb52bc1ae89f0a4fa5a00eb998dffde4a9f" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Mon Sep 26 11:47:53 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Oct 17 15:20:24 2005 -0700" }, "message": "[IB] Add checks to multicast attach and detach\n\nAdd checks so that we only allow multicast attach/detach with\na valid multicast GID and the correct QP type.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "dd0fc66fb33cd610bc1a5db8a5e232d34879b4d7", "tree": "51f96a9db96293b352e358f66032e1f4ff79fafb", "parents": [ "3b0e77bd144203a507eb191f7117d2c5004ea1de" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Fri Oct 07 07:46:04 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 08 15:00:57 2005 -0700" }, "message": "[PATCH] gfp flags annotations - part 1\n\n - added typedef unsigned int __nocast gfp_t;\n\n - replaced __nocast uses for gfp flags with gfp_t - it gives exactly\n the same warnings as far as sparse is concerned, doesn\u0027t change\n generated code (from gcc point of view we replaced unsigned int with\n typedef) and documents what\u0027s going on far better.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "aba7a22f291c13448177b28e0e3d01260ed04fbe", "tree": "9fb794e8e53edcef79d075e13363805a9eb23a96", "parents": [ "d70ed6075f15bdbb0548d162394bf10332769c88" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Fri Sep 30 13:55:50 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Sep 30 13:55:50 2005 -0700" }, "message": "[IB] mthca: Fix memory leak on device close\n\nRemember to free the multicast group context memory table.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d70ed6075f15bdbb0548d162394bf10332769c88", "tree": "f34b72480922d614a3210898b1bb990b61b91e40", "parents": [ "a1c337afaf4ec4d4eabc75a5e1170d03161de4e1" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Sep 28 19:56:57 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Sep 28 19:56:57 2005 -0700" }, "message": "[IPoIB] Rename IPoIB\u0027s path_lookup() to avoid name clashes\n\nRename IPoIB driver\u0027s path_lookup() to ipoib_path_lookup() to avoid a\nclashes with the kernel global path_lookup(). We don\u0027t hit this with\nthe current kernel source, but some external patches seem to trigger\nthis, and it\u0027s cleaner to avoid clashing with global names anyway.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nrefs/heads/for-linus\n" }, { "commit": "a1c337afaf4ec4d4eabc75a5e1170d03161de4e1", "tree": "a0d90a032233b63c93f3053ee2e9f9ea86e3ee9c", "parents": [ "f02b16bea2d8411b21a531fc381e066985895387" ], "author": { "name": "Jack Morgenstein", "email": "jackm@mellanox.co.il", "time": "Tue Sep 27 13:54:44 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 27 13:54:44 2005 -0700" }, "message": "[IB] mthca: fix hw_ver value returned from mthca_query_device\n\nThe IB spec defines the field to be 32 bits, not 16 bits.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f02b16bea2d8411b21a531fc381e066985895387", "tree": "e129e595b978238b2f2c9966acab0cc8a761e0ad", "parents": [ "63c47c286d062d93e0501d60797274c84a587e97" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Mon Sep 26 21:12:26 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 21:12:26 2005 -0700" }, "message": "[IB] mthca: Round up number of slots in HCA context memory table\n\nWhen allocating a table for mem-free HCA context, don\u0027t assume that\nobj_size * nobj is an even multiple of MTHCA_TABLE_CHUNK_SIZE. In\nparticular, make sure we allocate at least one slot even if the table\nis smaller than MTHCA_TABLE_CHUNK_SIZE.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "63c47c286d062d93e0501d60797274c84a587e97", "tree": "7be7c3402e456ed857c5041cd5ee6a699f772701", "parents": [ "44dd823b00fa64bf01e53557d28555011f122a88" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 13:01:03 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 13:01:03 2005 -0700" }, "message": "[IB] uverbs: Close some exploitable races\n\nAl Viro pointed out that the current IB userspace verbs interface\nallows userspace to cause mischief by closing file descriptors before\nwe\u0027re ready, or issuing the same command twice at the same time. This\npatch closes those races, and fixes other obvious problems such as a\nmodule reference leak.\n\nSome other interface bogosities will require an ABI change to fix\nproperly, so I\u0027m deferring those fixes until 2.6.15.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "44dd823b00fa64bf01e53557d28555011f122a88", "tree": "7245a298b11123ef81ded5afb92f1c711d643fed", "parents": [ "f7ed3a5971da98acdc506bdbdef25cfe51c334a2" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Mon Sep 26 09:42:09 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 09:42:09 2005 -0700" }, "message": "[IB] mthca: Fix off by one bug in mthca_map_cmd\n\nThe loop in mthca_map_cmd() would fill one entry past the end of the\nmailbox buffer before calling the firmware command.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f7ed3a5971da98acdc506bdbdef25cfe51c334a2", "tree": "1e2cff00ff81b1a20be326d5b58e672846663102", "parents": [ "018771f435388f22f388eb8658c652086fb3633e" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Mon Sep 26 09:29:33 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 09:38:34 2005 -0700" }, "message": "[IB] mthca: fix off by one in clr_int calculation\n\nWe should use the first word of the clear interrupt register if\nthe bit we\u0027re after is \u003c 32, not \u003c 31.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "018771f435388f22f388eb8658c652086fb3633e", "tree": "d18f3929acf5acff74a786f233ba8421bfe95c84", "parents": [ "8ddec7460d2f5db3ac35812c03676b1473d1d668" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Sep 21 21:40:12 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 26 09:38:26 2005 -0700" }, "message": "[IB] mthca: Fix doorbell record resource leak\n\nIf we allocate a bunch of doorbell records and then free them, we\u0027ll\nend up with completely empty pages, which we then free. However, when\nwe come back to allocate more doorbell pages, we have to reallocate\nthose empty pages rather than always trying to take a slot that we\u0027ve\nnever used. If we don\u0027t, we eventually use up every slot and fail to\nallocate a doorbell record, even though we have plenty of free space.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f2065e4242e5078d53c521b973c5cd4eae91eca2", "tree": "08cd44403ac326f41715e918f781d077367cb70b", "parents": [ "972d512a17c1bb7c4b784a9da2ca75745fcc6989" ], "author": { "name": "Hal Rosenstock", "email": "halr@voltaire.com", "time": "Wed Sep 21 12:33:09 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Sep 21 12:33:09 2005 -0700" }, "message": "[IB] Fix RMPP receive length calculation\n\nBased on simplification idea from Sean Hefty \u003csean.hefty@intel.com\u003e\n\nSigned-off-by: Hal Rosenstock \u003chalr@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "972d512a17c1bb7c4b784a9da2ca75745fcc6989", "tree": "91eb1a1a0c1affe19e23b15069b6a5fd86d3ca72", "parents": [ "2fe9f798ba3cf7c939e638b78f46975e79039978" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Sep 21 12:31:26 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Sep 21 12:31:26 2005 -0700" }, "message": "[IB] Add MAD data field size definitions\n\nClean up code by using enums instead of hard-coded magic numbers.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "71eea47d853bb0ce0c6befe11b3e08111263170f", "tree": "ff675d57b8d3dc6688a2016b41b0d41b54611231", "parents": [ "8d2cae0651502028bf64844508ab18528bbd65c2" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@mellanox.co.il", "time": "Tue Sep 20 10:54:48 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 20 10:54:48 2005 -0700" }, "message": "[PATCH] IB/mthca: Fix device removal memory leak\n\nClean up QP table array on device removal.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8d2cae0651502028bf64844508ab18528bbd65c2", "tree": "6e166a94e5f581c683d2b29e792665082b21afd5", "parents": [ "eff4c654b1a4a5e5493fbdc3affa6dd48765c085" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 20 10:52:04 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 20 10:52:04 2005 -0700" }, "message": "[PATCH] IPoIB: Don\u0027t flush workqueue from within workqueue\n\nipoib_mcast_restart_task() is always called from within the\nsingle-threaded IPoIB workqueue, so flushing the workqueue from within\nthe function can lead to a recursion overflow. But since we\u0027re\nrunning in a single-threaded workqueue, we\u0027re already synchronized\nagainst other items in the workqueue, so just get rid of the flush in\nipoib_mcast_restart_task().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "eff4c654b1a4a5e5493fbdc3affa6dd48765c085", "tree": "1ddbbdaa978b310f5ac11b9778ac8e5f775d0783", "parents": [ "6577ae51cf52f5fb0e4a85e673dd7bf2d0074e3e" ], "author": { "name": "Hal Rosenstock", "email": "halr@voltaire.com", "time": "Mon Sep 19 13:51:01 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 19 13:51:01 2005 -0700" }, "message": "[PATCH] IB: Fix data length for RMPP SA sends\n\nWe need to subtract off the header length from our payload\nlength when sending multi-packet SA messages.\n\nSigned-off-by: Hal Rosenstock \u003chalr@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6577ae51cf52f5fb0e4a85e673dd7bf2d0074e3e", "tree": "3c3be701f7954d52f850c116903deb2698bb8038", "parents": [ "3853194c2e174cee4da093c67bd54cbf9a38559a" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 19 09:17:56 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Sep 19 09:17:56 2005 -0700" }, "message": "[PATCH] IB/mthca: Don\u0027t try to set srq-\u003elast for userspace SRQs\n\nUserspace SRQs don\u0027t have a buffer allocated for them in the kernel, so\nit doesn\u0027t make sense to set srq-\u003elast during initialization. In fact,\nthis can crash trying to follow a nonexistent buffer pointer.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3853194c2e174cee4da093c67bd54cbf9a38559a", "tree": "3939280d0bf49b4b24068215f30aefdc38aaf0bc", "parents": [ "c915033fc62d7186d243d89f88782d6be33fd8f6" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 14:00:17 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 22:02:38 2005 -0700" }, "message": "[PATCH] IB/mthca: Fix posting work requests to shared receive queues\n\nThe error handling paths in mthca_tavor_post_srq_recv() and\nmthca_arbel_post_srq_recv() are quite bogus, the result of a\nscrewed up merge. Fix them so they work as intended.\n\nPointed out by Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c915033fc62d7186d243d89f88782d6be33fd8f6", "tree": "e509ef63ef6dba42f31807b5c9f6c20b48be3a9a", "parents": [ "ce5b65cc9626feac0d4ffb96f798407e50c45575" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 13:52:06 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 22:02:38 2005 -0700" }, "message": "[PATCH] IB/mthca: Initialize eq-\u003enent before we use it\n\nIn mthca_create_eq(), we call get_eqe() before setting eq-\u003enent. This\nis wrong, because get_eqe() uses eq-\u003enent. Fix this, and clean up the\ncode a little while we\u0027re at it. (We got lucky with the current code,\nbecause eq-\u003enent was cleared to 0, which get_eqe() made happen to do\nthe right thing)\n\nPointed out by Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ce5b65cc9626feac0d4ffb96f798407e50c45575", "tree": "42803ff0ecbc1f0d47adb628d88378232d03b1d6", "parents": [ "d6cff021e24515255b296d399ec517a68bf2ed08" ], "author": { "name": "Hal Rosenstock", "email": "halr@voltaire.com", "time": "Sun Sep 18 13:47:53 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 22:02:38 2005 -0700" }, "message": "[PATCH] IPoIB: Fix SA client retransmission strategy\n\nWe got a little mixed up with what the backoff member holds in the\nIPoIB multicast group structure: sometimes it was used as a number of\nseconds, and sometimes it was used as a number of jiffies. Fix the\ncode so that backoff is always in seconds.\n\nSigned-off-by: Hal Rosenstock \u003chalr@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d6cff021e24515255b296d399ec517a68bf2ed08", "tree": "f3d21aca027cb7065522629c00f65fc3a8fb1d96", "parents": [ "bb4a7f0da75ce9f3e933880428d39b2aa1f16961" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Sep 13 10:41:03 2005 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 18 22:02:37 2005 -0700" }, "message": "[PATCH] IB/mthca: fix posting of first work request\n\nFix posting first WQE for mem-free HCAs: we need to link to previous\nWQE even in that case. While we\u0027re at it, simplify code for\nTavor-mode HCAs. We don\u0027t really need the conditional test there\neither; we can similarly always link to the previous WQE.\n\nBased on Michael S. Tsirkin\u0027s analogous fix for userspace libmthca.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" } ], "next": "bb4a7f0da75ce9f3e933880428d39b2aa1f16961" }