)]}' { "log": [ { "commit": "ec0afc9311adcfb10b90e547c23250f63939f990", "tree": "2093d2668898a8a03f30acbfd5568e65b8c086b9", "parents": [ "804f18536984939622ddca60ab6b25743e0ec68d", "776e58ea3d3735f85678155398241d2513afa67a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 17 18:40:35 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 17 18:40:35 2011 -0700" }, "message": "Merge branch \u0027kvm-updates/2.6.39\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\n* \u0027kvm-updates/2.6.39\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm: (55 commits)\n KVM: unbreak userspace that does not sets tss address\n KVM: MMU: cleanup pte write path\n KVM: MMU: introduce a common function to get no-dirty-logged slot\n KVM: fix rcu usage in init_rmode_* functions\n KVM: fix kvmclock regression due to missing clock update\n KVM: emulator: Fix permission checking in io permission bitmap\n KVM: emulator: Fix io permission checking for 64bit guest\n KVM: SVM: Load %gs earlier if CONFIG_X86_32_LAZY_GS\u003dn\n KVM: x86: Remove useless regs_page pointer from kvm_lapic\n KVM: improve comment on rcu use in irqfd_deassign\n KVM: MMU: remove unused macros\n KVM: MMU: cleanup page alloc and free\n KVM: MMU: do not record gfn in kvm_mmu_pte_write\n KVM: MMU: move mmu pages calculated out of mmu lock\n KVM: MMU: set spte accessed bit properly\n KVM: MMU: fix kvm_mmu_slot_remove_write_access dropping intermediate W bits\n KVM: Start lock documentation\n KVM: better readability of efer_reserved_bits\n KVM: Clear async page fault hash after switching to real mode\n KVM: VMX: Initialize vm86 TSS only once.\n ...\n" }, { "commit": "c55d267de274d308927b60c3e740c1a826832317", "tree": "21b53a8c725d9f9650f60d94b349459d5b8dae10", "parents": [ "61ef46fd45c3c62dc7c880a45dd2aa841b9af8fb", "bc898c97f7ba24def788d9f80786cf028a197122" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 17 17:54:40 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 17 17:54:40 2011 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (170 commits)\n [SCSI] scsi_dh_rdac: Add MD36xxf into device list\n [SCSI] scsi_debug: add consecutive medium errors\n [SCSI] libsas: fix ata list corruption issue\n [SCSI] hpsa: export resettable host attribute\n [SCSI] hpsa: move device attributes to avoid forward declarations\n [SCSI] scsi_debug: Logical Block Provisioning (SBC3r26)\n [SCSI] sd: Logical Block Provisioning update\n [SCSI] Include protection operation in SCSI command trace\n [SCSI] hpsa: fix incorrect PCI IDs and add two new ones (2nd try)\n [SCSI] target: Fix volume size misreporting for volumes \u003e 2TB\n [SCSI] bnx2fc: Broadcom FCoE offload driver\n [SCSI] fcoe: fix broken fcoe interface reset\n [SCSI] fcoe: precedence bug in fcoe_filter_frames()\n [SCSI] libfcoe: Remove stale fcoe-netdev entries\n [SCSI] libfcoe: Move FCOE_MTU definition from fcoe.h to libfcoe.h\n [SCSI] libfc: introduce __fc_fill_fc_hdr that accepts fc_hdr as an argument\n [SCSI] fcoe, libfc: initialize EM anchors list and then update npiv EMs\n [SCSI] Revert \"[SCSI] libfc: fix exchange being deleted when the abort itself is timed out\"\n [SCSI] libfc: Fixing a memory leak when destroying an interface\n [SCSI] megaraid_sas: Version and Changelog update\n ...\n\nFix up trivial conflicts due to whitespace differences in\ndrivers/scsi/libsas/{sas_ata.c,sas_scsi_host.c}\n" }, { "commit": "0014bd990e69063b0fb78940b35439d7980ce3ee", "tree": "56d4576cc07954eb304abaf602aba44a6aa2a4f1", "parents": [ "91c9c3eda4f3066980d13a6907ef84f3a99364bd" ], "author": { "name": "Huang Ying", "email": "ying.huang@intel.com", "time": "Sun Jan 30 11:15:47 2011 +0800" }, "committer": { "name": "Marcelo Tosatti", "email": "mtosatti@redhat.com", "time": "Thu Mar 17 13:08:27 2011 -0300" }, "message": "mm: export __get_user_pages\n\nIn most cases, get_user_pages and get_user_pages_fast should be used\nto pin user pages in memory. But sometimes, some special flags except\nFOLL_GET, FOLL_WRITE and FOLL_FORCE are needed, for example in\nfollowing patch, KVM needs FOLL_HWPOISON. To support these users,\n__get_user_pages is exported directly.\n\nThere are some symbol name conflicts in infiniband driver, fixed them too.\n\nSigned-off-by: Huang Ying \u003cying.huang@intel.com\u003e\nCC: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCC: Michel Lespinasse \u003cwalken@google.com\u003e\nCC: Roland Dreier \u003croland@kernel.org\u003e\nCC: Ralph Campbell \u003cinfinipath@qlogic.com\u003e\nSigned-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n" }, { "commit": "7a6362800cb7d1d618a697a650c7aaed3eb39320", "tree": "087f9bc6c13ef1fad4b392c5cf9325cd28fa8523", "parents": [ "6445ced8670f37cfc2c5e24a9de9b413dbfc788d", "ceda86a108671294052cbf51660097b6534672f5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 16 16:29:25 2011 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 16 16:29:25 2011 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1480 commits)\n bonding: enable netpoll without checking link status\n xfrm: Refcount destination entry on xfrm_lookup\n net: introduce rx_handler results and logic around that\n bonding: get rid of IFF_SLAVE_INACTIVE netdev-\u003epriv_flag\n bonding: wrap slave state work\n net: get rid of multiple bond-related netdevice-\u003epriv_flags\n bonding: register slave pointer for rx_handler\n be2net: Bump up the version number\n be2net: Copyright notice change. Update to Emulex instead of ServerEngines\n e1000e: fix kconfig for crc32 dependency\n netfilter ebtables: fix xt_AUDIT to work with ebtables\n xen network backend driver\n bonding: Improve syslog message at device creation time\n bonding: Call netif_carrier_off after register_netdevice\n bonding: Incorrect TX queue offset\n net_sched: fix ip_tos2prio\n xfrm: fix __xfrm_route_forward()\n be2net: Fix UDP packet detected status in RX compl\n Phonet: fix aligned-mode pipe socket buffer header reserve\n netxen: support for GbE port settings\n ...\n\nFix up conflicts in drivers/staging/brcm80211/brcmsmac/wl_mac80211.c\nwith the staging updates.\n" }, { "commit": "043332cf28368e38fdfcee1a4827025e8bdbc540", "tree": "fb179157040c52ac2c8184f52e28de65342a6267", "parents": [ "a396d43a35fb91f2d4920a4700d25ecc5ec92404", "db5d040d7b2d15539d2c84932f93621d9bd482f7", "2a543904ddcb463db9d56d1efcb2f80884ea55f3", "36b87b419c7616a8aaa9889566d9a9b50689dee1" ], "author": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:58:04 2011 -0700" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:58:04 2011 -0700" }, "message": "Merge branches \u0027cma\u0027, \u0027cxgb4\u0027, \u0027ipath\u0027 and \u0027qib\u0027 into for-next\n" }, { "commit": "a396d43a35fb91f2d4920a4700d25ecc5ec92404", "tree": "5658f67fe3b18c51c51340b777cbb0fd803376ba", "parents": [ "8d8ac86564b616bc17054cbb6e727588da64c86b" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Feb 23 09:05:39 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:57:34 2011 -0700" }, "message": "RDMA/cma: Replace global lock in rdma_destroy_id() with id-specific one\n\nrdma_destroy_id currently uses the global rdma cm \u0027lock\u0027 to test if an\nrdma_cm_id has been bound to a device. This prevents an active\naddress resolution callback handler from assigning a device to the\nrdma_cm_id after rdma_destroy_id checks for one.\n\nInstead, we can replace the use of the global lock around the check to\nthe rdma_cm_id device pointer by setting the id state to destroying,\nthen flushing all active callbacks. The latter is accomplished by\nacquiring and releasing the handler_mutex. Any active handler will\ncomplete first, and any newly scheduled handlers will find the\nrdma_cm_id in an invalid state.\n\nIn addition to optimizing the current locking scheme, the use of the\nrdma_cm_id mutex is a more intuitive synchronization mechanism than\nthat of the global lock. These changes are based on feedback from\nDoug Ledford \u003cdledford@redhat.com\u003e while he was trying to debug a\ncrash in the rdma cm destroy path.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "8d8ac86564b616bc17054cbb6e727588da64c86b", "tree": "f17f09108b2fa639d5248120409e2fe562069545", "parents": [ "29963437a48475036353b95ab142bf199adb909e" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Thu Mar 03 23:31:06 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:56:12 2011 -0700" }, "message": "IB/cm: Cancel pending LAP message when exiting IB_CM_ESTABLISH state\n\nThis problem was reported by Moni Shoua \u003cmonis@mellanox.com\u003e and Amir\nVadai \u003camirv@mellanox.com\u003e:\n\n\tWhen destroying a cm_id from a context of a work queue and if\n\tthe lap_state of this cm_id is IB_CM_LAP_SENT, we need to\n\trelease the reference of this id that was taken upon the send\n\tof the LAP message. Otherwise, if the expected APR message\n\tgets lost, it is only after a long time that the reference\n\twill be released, while during that the work handler thread is\n\tnot available to process other things.\n\nIt turns out that we need to cancel any pending LAP messages whenever\nwe transition out of the IB_CM_ESTABLISH state. This occurs when\ndisconnecting - either sending or receiving a DREQ. It can also\nhappen in a corner case where we receive a REJ message after sending\nan RTU, followed by a LAP. Add checks and cancel any outstanding LAP\nmessages in these three cases.\n\nCanceling the LAP when sending a DREQ fixes the destroy problem\nreported by Moni. When a cm_id is destroyed in the IB_CM_ESTABLISHED\nstate, it sends a DREQ to the remote side to notify the peer that the\nconnection is going away.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "29963437a48475036353b95ab142bf199adb909e", "tree": "23564281f4d60b7a9cdbaab5e1babe3f67cc8ca8", "parents": [ "25ae21a10112875763c18b385624df713a288a05" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Feb 23 08:17:40 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:56:12 2011 -0700" }, "message": "IB/cm: Bump reference count on cm_id before invoking callback\n\nWhen processing a SIDR REQ, the ib_cm allocates a new cm_id. The\nrefcount of the cm_id is initialized to 1. However, cm_process_work\nwill decrement the refcount after invoking all callbacks. The result\nis that the cm_id will end up with refcount set to 0 by the end of the\nsidr req handler.\n\nIf a user tries to destroy the cm_id, the destruction will proceed,\nunder the incorrect assumption that no other threads are referencing\nthe cm_id. This can lead to a crash when the cm callback thread tries\nto access the cm_id.\n\nThis problem was noticed as part of a larger investigation with kernel\ncrashes in the rdma_cm when running on a real time OS.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nAcked-by: Doug Ledford \u003cdledford@redhat.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "25ae21a10112875763c18b385624df713a288a05", "tree": "e05138aae6bd98f055ce2be0437922cee8e9f17e", "parents": [ "a5bbef0b2deb7b943f095181309ecc9e1fc91c0f" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Feb 23 08:11:32 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Mar 15 10:00:28 2011 -0700" }, "message": "RDMA/cma: Fix crash in request handlers\n\nDoug Ledford and Red Hat reported a crash when running the rdma_cm on\na real-time OS. The crash has the following call trace:\n\n cm_process_work\n cma_req_handler\n cma_disable_callback\n rdma_create_id\n kzalloc\n init_completion\n cma_get_net_info\n cma_save_net_info\n cma_any_addr\n cma_zero_addr\n rdma_translate_ip\n rdma_copy_addr\n cma_acquire_dev\n rdma_addr_get_sgid\n ib_find_cached_gid\n cma_attach_to_dev\n ucma_event_handler\n kzalloc\n ib_copy_ah_attr_to_user\n cma_comp\n\n[ preempted ]\n\n cma_write\n copy_from_user\n ucma_destroy_id\n copy_from_user\n _ucma_find_context\n ucma_put_ctx\n ucma_free_ctx\n rdma_destroy_id\n cma_exch\n cma_cancel_operation\n rdma_node_get_transport\n\n rt_mutex_slowunlock\n bad_area_nosemaphore\n oops_enter\n\nThey were able to reproduce the crash multiple times with the\nfollowing details:\n\n Crash seems to always happen on the:\n mutex_unlock(\u0026conn_id-\u003ehandler_mutex);\n as conn_id looks to have been freed during this code path.\n\nAn examination of the code shows that a race exists in the request\nhandlers. When a new connection request is received, the rdma_cm\nallocates a new connection identifier. This identifier has a single\nreference count on it. If a user calls rdma_destroy_id() from another\nthread after receiving a callback, rdma_destroy_id will proceed to\ndestroy the id and free the associated memory. However, the request\nhandlers may still be in the process of running. When control returns\nto the request handlers, they can attempt to access the newly created\nidentifiers.\n\nFix this by holding a reference on the newly created rdma_cm_id until\nthe request handler is through accessing it.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nAcked-by: Doug Ledford \u003cdledford@redhat.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "2a543904ddcb463db9d56d1efcb2f80884ea55f3", "tree": "57cdb9098e4299cc47bb21e8061b671b24d70219", "parents": [ "a5bbef0b2deb7b943f095181309ecc9e1fc91c0f" ], "author": { "name": "Nicolas Kaiser", "email": "nikai@nikai.net", "time": "Tue Oct 26 15:47:52 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 14:25:59 2011 -0700" }, "message": "IB/ipath: Don\u0027t reset disabled devices\n\nThe comment some lines above states that disabled devices must not reset.\n\nSigned-off-by: Nicolas Kaiser \u003cnikai@nikai.net\u003e\n" }, { "commit": "36b87b419c7616a8aaa9889566d9a9b50689dee1", "tree": "a1509da10d80a58e2803420804ed2f1d8c780f26", "parents": [ "4634b7945cf0d6a66036ad10c3d658ae4eb39fa0" ], "author": { "name": "Mitko Haralanov", "email": "mitko@qlogic.com", "time": "Fri Mar 04 18:53:17 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:11:51 2011 -0700" }, "message": "IB/qib: Fix M_Key field in SubnGet and SubnGetResp MADs\n\nSet the M_Key field in SubnGet and SugnGetResp MADs based on correctly\ninterpreting the protection level specified in the M_KeyProtBits field.\n\nSigned-off-by: Mitko Haralanov \u003cmitko@qlogic.com\u003e\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "4634b7945cf0d6a66036ad10c3d658ae4eb39fa0", "tree": "55e10d1d51b6c40d5dff669bde13832c55e8a977", "parents": [ "cc7fb05946fb1cd2fd0582f9e39f759e20dfeefa" ], "author": { "name": "Mitko Haralanov", "email": "mitko@qlogic.com", "time": "Mon Feb 28 13:39:49 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:10:34 2011 -0700" }, "message": "IB/qib: Set default LE2 value for active cables to 0\n\nFor active and far-EQ cables use an LE2 value of 0 for improved SI.\n\nSigned-off-by: Mitko Haralanov \u003cmitko@qlogic.com\u003e\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "db5d040d7b2d15539d2c84932f93621d9bd482f7", "tree": "115c8e629b40106cc6d2dc6211a97d9c77d4bfb6", "parents": [ "767fbe8151d1a7cc8a69e52e354e4220a5e804fb" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:29:50 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:14 2011 -0700" }, "message": "RDMA/cxgb4: Debugfs dump_qp() updates\n\n- Show whether the SQ is in onchip memory or not.\n- Dump both SQ and RQ QIDs.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "767fbe8151d1a7cc8a69e52e354e4220a5e804fb", "tree": "805495187c3d9ffe78b23bccd82fc21ffc955c86", "parents": [ "b48f3b9c10d731160f0af5c3028ad57d9c66673b" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:53 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:13 2011 -0700" }, "message": "RDMA/cxgb4: Dispatch FATAL event on EEH errors\n\nThis at least kicks the user mode applications that are watching for\ndevice events.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "b48f3b9c10d731160f0af5c3028ad57d9c66673b", "tree": "505fbe19fa360d0a18f08af3ccdb727e801fa115", "parents": [ "a9c7719800ac513b2df14e267d062ec84dc9313e" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:21 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:12 2011 -0700" }, "message": "RDMA/cxgb4: Use ULP_MODE_TCPDDP\n\nSet the ULP mode for initial RDMA connection setup to the proper DDP\nmode. This avoids wasting some HW resources while in streaming mode.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "a9c7719800ac513b2df14e267d062ec84dc9313e", "tree": "44ed2a27f269a126a767cde283424417f7081f20", "parents": [ "ffc3f7487ff0b32500b319c770475383f6f6efab" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:11 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:12 2011 -0700" }, "message": "RDMA/cxgb4: Enable on-chip SQ support by default\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "ffc3f7487ff0b32500b319c770475383f6f6efab", "tree": "64beb59d6b0b843316a47bdae156a24893254380", "parents": [ "294281373999e7fff393c04eb16092a8f00ad5aa" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:42 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:11 2011 -0700" }, "message": "RDMA/cxgb4: Do CIDX_INC updates every 1/16 CQ depth CQE reaps\n\nThis avoids the CIDX_INC overflow issue with T4A2 when running\nkernel RDMA applications.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "294281373999e7fff393c04eb16092a8f00ad5aa", "tree": "3608fd01c734f99670f43f4cfac691da63c7a5d2", "parents": [ "b52fe09e3309c3d7069cd0e5a3bdb5b4ba45e01f" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:32 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:10 2011 -0700" }, "message": "RDMA/cxgb4: Remove db_drop_task\n\nUnloading iw_cxgb4 can crash due to the unload code trying to use\ndb_drop_task, which is uninitialized. So remove this dead code.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "b52fe09e3309c3d7069cd0e5a3bdb5b4ba45e01f", "tree": "a8b5beeed11f6c7bbe0ee09f1607619e10a54627", "parents": [ "a5bbef0b2deb7b943f095181309ecc9e1fc91c0f" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Mar 11 22:30:01 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Mon Mar 14 12:09:09 2011 -0700" }, "message": "RDMA/cxgb4: Turn on delayed ACK\n\nSet the default to on.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "4c9483b2fb5d2548c3cc1fe03cdd4484ceeb5d1c", "tree": "c29c8070012cffb38fe249cf528589a675f622b1", "parents": [ "9cce96df5b76691712dba22e83ff5efe900361e1" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 16:22:43 2011 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 15:08:54 2011 -0800" }, "message": "ipv6: Convert to use flowi6 where applicable.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1d28f42c1bd4bb2363d88df74d0128b4da135b4a", "tree": "cb2e652fe79a2bc307e871bc2d3fa51cc8051e45", "parents": [ "ca116922afa8cc5ad46b00c0a637b1cde5ca478a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 00:29:39 2011 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 15:08:44 2011 -0800" }, "message": "net: Put flowi_* prefix on AF independent members of struct flowi\n\nI intend to turn struct flowi into a union of AF specific flowi\nstructs. There will be a common structure that each variant includes\nfirst, much like struct sock_common.\n\nThis is the first step to move in that direction.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "78fbfd8a653ca972afe479517a40661bfff6d8c3", "tree": "9dccc5c16bf269d53d8499064ec95a998e84c646", "parents": [ "1561747ddf9d28185548687b11aae7074d6129c4" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 00:00:52 2011 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Mar 12 15:08:42 2011 -0800" }, "message": "ipv4: Create and use route lookup helpers.\n\nThe idea here is this minimizes the number of places one has to edit\nin order to make changes to how flows are defined and used.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b23dd4fe42b455af5c6e20966b7d6959fa8352ea", "tree": "bf97323eae9a8d084170e573ff2c0c40bc72c3cd", "parents": [ "452edd598f60522c11f7f88fdbab27eb36509d1a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 02 14:31:35 2011 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 02 14:31:35 2011 -0800" }, "message": "ipv4: Make output route lookup return rtable directly.\n\nInstead of on the stack.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "273447b352e69c327efdecfd6e1d6fe3edbdcd14", "tree": "dae1e0778ca973c25b74fd3dc9728616d6e65b73", "parents": [ "5df65e5567a497a28067019b8ff08f98fb026629" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 01 14:27:04 2011 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 01 14:27:04 2011 -0800" }, "message": "ipv4: Kill can_sleep arg to ip_route_output_flow()\n\nThis boolean state is now available in the flow flags.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "420d44daa7aa1cc847e9e527f0a27a9ce61768ca", "tree": "a5aab8c6b925ba3da1079b7262f7d6c504406eb8", "parents": [ "abdf7e7239da270e68262728f125ea94b9b7d42d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 01 14:19:23 2011 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Mar 01 14:19:23 2011 -0800" }, "message": "ipv4: Make final arg to ip_route_output_flow to be boolean \"can_sleep\"\n\nSince that is what the current vague \"flags\" argument means.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c53c6f89d7a6487986c51cd73ae9a9be338a8f4", "tree": "b1762d88f6906093be6cb80b9c8a4607d973e764", "parents": [ "d8585bcd7da071f4278710f1c39e18dfe7cb0280" ], "author": { "name": "Mike Christie", "email": "michaelc@cs.wisc.edu", "time": "Wed Feb 16 15:04:40 2011 -0600" }, "committer": { "name": "James Bottomley", "email": "James.Bottomley@suse.de", "time": "Thu Feb 24 12:41:23 2011 -0500" }, "message": "[SCSI] iser: export addr and port\n\nThis pactch has iser export the address and port\nof the endpoint.\n\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@suse.de\u003e\n" }, { "commit": "cc7fb05946fb1cd2fd0582f9e39f759e20dfeefa", "tree": "5b6bc0af27d0312f558f70b52ff82e17c1694e2b", "parents": [ "a5bbef0b2deb7b943f095181309ecc9e1fc91c0f" ], "author": { "name": "Mitko Haralanov", "email": "mitko@qlogic.com", "time": "Tue Feb 22 16:56:37 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Tue Feb 22 16:56:37 2011 -0800" }, "message": "IB/qib: Return correct MAD when setting link width to 255\n\nFix a bug which causes the driver to return incorrect MADs as a\nresponse to Set(PortInfo) which sets the link width to 0xFF or link\nspeed to 0xF.\n\nSigned-off-by: Mitko Haralanov \u003cmitko@qlogic.com\u003e\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "da935c66bacb3ed9ada984b053297f87c2dff63a", "tree": "46278da2b312c73f1375b830d7e5912bf23abd78", "parents": [ "9435eb1cf0b76b323019cebf8d16762a50a12a19", "2205a6ea93fea76f88b43727fea53f3ce3790d6f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Feb 19 19:17:35 2011 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Feb 19 19:17:35 2011 -0800" }, "message": "Merge branch \u0027master\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6\n\nConflicts:\n\tDocumentation/feature-removal-schedule.txt\n\tdrivers/net/e1000e/netdev.c\n\tnet/xfrm/xfrm_policy.c\n" }, { "commit": "814b0a61204d24f9fba6f7c575e6450d15ce2cf1", "tree": "978d0265d1387036e8285685bc0692cade005f90", "parents": [ "25a54a6bb87dc966f6a3fc1f2ac6e88db1f5614c", "c0af2c057d7ce3f0b260f9380d187a82bb5cab28" ], "author": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Thu Feb 17 14:04:59 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Thu Feb 17 14:04:59 2011 -0800" }, "message": "Merge branches \u0027nes\u0027 and \u0027qib\u0027 into for-next\n" }, { "commit": "c0af2c057d7ce3f0b260f9380d187a82bb5cab28", "tree": "30b59bf396145825d5b005dcc8f6fa1abf5c2b1f", "parents": [ "414ed90cee32486c50f91b28990443e0dc21c868" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Wed Feb 16 15:48:25 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Thu Feb 17 14:04:50 2011 -0800" }, "message": "IB/qib: Prevent double completions after a timeout or RNR error\n\nThere is a double completion associated with error handling for RC QPs.\n\nThe sequence is:\n\n - The do_rc_ack() routine fields an RNR nack and there are 0\n rnr_retries configured on the QP.\n - qib_error_qp() stops the pending timer\n - qib_rc_send_complete() is called from sdma_complete()\n - qib_rc_send_complete() starts the timer because the msb of the psn\n just completed says an ack is needed.\n - a bunch of flushes occur as ipoib posts WQEs to an error\u0027ed QP\n - rc_timeout() calls qib_restart_rc()\n - qib_restart_rc() calls qib_send_complete() with a\n IB_WC_RETRY_EXC_ERR on a wqe that has already been completed in the\n past\n\nThe fix avoids starting the timer since another packet will never\narrive.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "1765a575334f1a232c1478accdee5c7d19f4b3e3", "tree": "6613f53ce9095b799048d891f0b4b6e34a479fe0", "parents": [ "d59cfde2fb960b5970ccb5a38cea25d38b37a8e8" ], "author": { "name": "Jiri Pirko", "email": "jpirko@redhat.com", "time": "Sat Feb 12 06:48:36 2011 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Feb 13 10:42:07 2011 -0800" }, "message": "net: make dev-\u003emaster general\n\ndev-\u003emaster is now tightly connected to bonding driver. This patch makes\nthis pointer more general and ready to be used by others.\n\n - netdev_set_master() - bond specifics moved to new function\n netdev_set_bond_master()\n - introduced netif_is_bond_slave() to check if device is a bonding slave\n\nSigned-off-by: Jiri Pirko \u003cjpirko@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "414ed90cee32486c50f91b28990443e0dc21c868", "tree": "bcb6fc2551b3bb230c6aa9ebf18502728f573228", "parents": [ "831d52bc153971b70e64eccfbed2b232394f22f8" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Thu Feb 10 14:11:28 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Thu Feb 10 11:24:08 2011 -0800" }, "message": "IB/qib: Fix double add_timer()\n\nThe following panic BUG_ON occurs during qib testing:\n\n Kernel BUG at include/linux/timer.h:82\n\n RIP [\u003cffffffff881f7109\u003e] :ib_qib:start_timer+0x73/0x89\n RSP \u003cffffffff80425bd0\u003e\n \u003c0\u003eKernel panic - not syncing: Fatal exception\n \u003c0\u003eDumping qib trace buffer from panic\n qib_set_lid INFO: IB0:1 got a lid: 0xf8\n Done dumping qib trace buffer\n BUG: warning at kernel/panic.c:137/panic() (Tainted: G\n\nThe flaw is due to a missing state test when processing responses that\nresults in an add_timer() call when the same timer is already queued.\nThis code was executing in parallel with a QP destroy on another CPU\nthat had changed the state to reset, but the missing test caused to\nresponse handling code to run on into the panic.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "25a54a6bb87dc966f6a3fc1f2ac6e88db1f5614c", "tree": "5ac11304c2b46557439fd22d688e2d93bd70fce3", "parents": [ "831d52bc153971b70e64eccfbed2b232394f22f8" ], "author": { "name": "Maciej Sosnowski", "email": "maciej.sosnowski@intel.com", "time": "Thu Feb 03 15:55:26 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Thu Feb 03 15:55:26 2011 -0800" }, "message": "RDMA/nes: Don\u0027t generate async events for unregistered devices\n\nnes_port_ibevent() should not be called when the nes RDMA device is not\nregistered with the RDMA core. Add missing checks of of_device_registered flag.\n\nSigned-off-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "9118626a30f8a3f58674623bebd3c34961e558af", "tree": "9dac3d8f95f82b84c39695a648ee3a718daba676", "parents": [ "aba99437f5af7a419a82438f7fab16bb3ddd9b44", "e51c7b1ab05d4a6fe4d153b2769290d37e077479" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 03 11:19:26 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 03 11:19:26 2011 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n RDMA: Update missed conversion of flush_scheduled_work()\n RDMA/ucma: Copy iWARP route information on queries\n RDMA/amso1100: Fix compile warnings\n RDMA/cxgb4: Set the correct device physical function for iWARP connections\n RDMA/cxgb4: Limit MAXBURST EQ context field to 256B\n IB/qib: Hold link for TX SERDES settings\n mlx4_core: Add ConnectX-3 device IDs\n" }, { "commit": "e51c7b1ab05d4a6fe4d153b2769290d37e077479", "tree": "97b3fe2f6cb216a47237b2f57f5e58ae79837b85", "parents": [ "f9a4f6dcdd3f9b6d938484c2c394f39007c14f38", "e86f8b06f5fa884a84c22b2dcd0df37ed0a75827", "94788657c94169171971968c9d4b6222c5e704aa", "96e61fa55e86fbd10c526567aacae5b484c4b63c", "31dd272e8cbb32ef31a411492cc642c363bb54b9", "d70585f7de4b4c3725062b7ce7d492f4903e4833" ], "author": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Sat Jan 29 20:45:04 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Sat Jan 29 20:45:04 2011 -0800" }, "message": "Merge branches \u0027amso1100\u0027, \u0027cma\u0027, \u0027cxgb4\u0027, \u0027misc\u0027, \u0027mlx4\u0027 and \u0027qib\u0027 into for-next\n" }, { "commit": "96e61fa55e86fbd10c526567aacae5b484c4b63c", "tree": "ae95defe8cc2597352baa30e720ccb7ba21bd587", "parents": [ "a25cecce88194b2edf38b8c3b1665e9318eb2d22" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Mon Jan 24 11:06:54 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 16:39:08 2011 -0800" }, "message": "RDMA: Update missed conversion of flush_scheduled_work()\n\nCommit f06267104dd9 (\"RDMA: Update workqueue usage\") introduced ib_wq\nand removed the use of flush_scheduled_work(); however, during the merge\nprocess one chunk was lost in ib_sa_remove_one(). Fix it.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "e86f8b06f5fa884a84c22b2dcd0df37ed0a75827", "tree": "4aaf6a46755752417f6eff901dcaf5afe0fbec15", "parents": [ "a25cecce88194b2edf38b8c3b1665e9318eb2d22" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Jan 21 03:40:46 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 16:34:05 2011 -0800" }, "message": "RDMA/ucma: Copy iWARP route information on queries\n\nFor iWARP rdma_cm ids, the \"route\" information is the L2 src and\nnext hop addresses.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "f9a4f6dcdd3f9b6d938484c2c394f39007c14f38", "tree": "2a9cbd728a8376a4d1b6f643d60c4cbda7892a91", "parents": [ "a25cecce88194b2edf38b8c3b1665e9318eb2d22" ], "author": { "name": "Ralf Thielow", "email": "ralf.thielow@googlemail.com", "time": "Sun Jan 23 16:30:03 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 15:40:25 2011 -0800" }, "message": "RDMA/amso1100: Fix compile warnings\n\nFix compile warnings on 32-bit by using \"0\" instead of \"(u64) NULL\" to\nassign to \"c2_vq_req-\u003ereply_msg\".\n\nSigned-off-by: Ralf Thielow \u003cralf.thielow@googlemail.com\u003e\n\n[ Change from \"(unsigned long) NULL\" to plain old \"0\" as suggested by\n Bart Van Assche \u003cbvanassche@acm.org\u003e. - Roland ]\n\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "94788657c94169171971968c9d4b6222c5e704aa", "tree": "8aef275617eda6e0b9e7b582d7b1dabc88e36ff9", "parents": [ "6a09a9d6946dd516d243d072bee83fae3c683471" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Jan 21 17:00:34 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 15:34:28 2011 -0800" }, "message": "RDMA/cxgb4: Set the correct device physical function for iWARP connections\n\nThe PF passed to FW was 0, causing PCI failures in an SR-IOV environment.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "6a09a9d6946dd516d243d072bee83fae3c683471", "tree": "776e0b9bf5ed1e5fd541b0661d9bf124298269cf", "parents": [ "a25cecce88194b2edf38b8c3b1665e9318eb2d22" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Fri Jan 21 17:00:29 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 15:34:24 2011 -0800" }, "message": "RDMA/cxgb4: Limit MAXBURST EQ context field to 256B\n\nMAXBURST cannot exceed 256B for on-chip queues. With a 512B MAXBURST,\nwe can lock up the chip.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "d70585f7de4b4c3725062b7ce7d492f4903e4833", "tree": "3c5a65c8f8350a794369444a7d2219395a4904f4", "parents": [ "a25cecce88194b2edf38b8c3b1665e9318eb2d22" ], "author": { "name": "Mitko Haralanov", "email": "mitko@qlogic.com", "time": "Fri Jan 21 13:45:17 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "roland@purestorage.com", "time": "Fri Jan 28 15:30:02 2011 -0800" }, "message": "IB/qib: Hold link for TX SERDES settings\n\nHold the IB link at DISABLED until we get the correct TX settings\non mezz boards.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003croland@purestorage.com\u003e\n" }, { "commit": "6a108a14fa356ef607be308b68337939e56ea94e", "tree": "1bf260572bd8f95ed867307a2bcf5d881c8ae4a6", "parents": [ "12fcdba1b7ae8b25696433f420b775aeb556d89b" ], "author": { "name": "David Rientjes", "email": "rientjes@google.com", "time": "Thu Jan 20 14:44:16 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 20 17:02:05 2011 -0800" }, "message": "kconfig: rename CONFIG_EMBEDDED to CONFIG_EXPERT\n\nThe meaning of CONFIG_EMBEDDED has long since been obsoleted; the option\nis used to configure any non-standard kernel with a much larger scope than\nonly small devices.\n\nThis patch renames the option to CONFIG_EXPERT in init/Kconfig and fixes\nreferences to the option throughout the kernel. A new CONFIG_EMBEDDED\noption is added that automatically selects CONFIG_EXPERT when enabled and\ncan be used in the future to isolate options that should only be\nconsidered for embedded systems (RISC architectures, SLOB, etc).\n\nCalling the option \"EXPERT\" more accurately represents its intention: only\nexpert users who understand the impact of the configuration changes they\nare making should enable it.\n\nReviewed-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nAcked-by: David Woodhouse \u003cdavid.woodhouse@intel.com\u003e\nSigned-off-by: David Rientjes \u003crientjes@google.com\u003e\nCc: Greg KH \u003cgregkh@suse.de\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Robin Holt \u003cholt@sgi.com\u003e\nCc: \u003clinux-arch@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6845a44a314c0c626549de373131bf108f9cc1f1", "tree": "9152a5c557434515dd7306d288e8f55ecc64652e", "parents": [ "eee2a817df7c5a6e569f353f8be78cc1b3604bb6", "4790f4dc5f4326dab5d81ed8fb8c9473e620bdbb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jan 17 14:45:48 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jan 17 14:45:48 2011 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n RDMA: Update workqueue usage\n RDMA/nes: Fix incorrect SFP+ link status detection on driver init\n RDMA/nes: Fix SFP+ link down detection issue with switch port disable\n RDMA/nes: Generate IB_EVENT_PORT_ERR/PORT_ACTIVE events\n RDMA/nes: Fix bonding on iw_nes\n IB/srp: Test only once whether iu allocation succeeded\n IB/mlx4: Handle protocol field in multicast table\n RDMA: Use vzalloc() to replace vmalloc()+memset(0)\n mlx4_{core, ib, en}: Fix driver when sizeof (phys_addr_t) \u003e sizeof (long)\n IB/mthca: Fix driver when sizeof (phys_addr_t) \u003e sizeof (long)\n" }, { "commit": "4790f4dc5f4326dab5d81ed8fb8c9473e620bdbb", "tree": "72d5bbcdc8aa72c4e48a81ce33be5ed797093224", "parents": [ "f06267104dd9112f11586830d22501d0e26245ea", "da995a8aee044bc5d0847e19e351cd48a2cb8bcc", "eb4a7cbf27082bea34764bab3bc85595683f967b", "843276ad985cb59212ceb70d989474521ff516f6", "695b83495e2fba9d3a883193cfc9d5eefa96a911" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 21:22:41 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 21:22:41 2011 -0800" }, "message": "Merge branches \u0027misc\u0027, \u0027mlx4\u0027, \u0027mthca\u0027, \u0027nes\u0027 and \u0027srp\u0027 into for-next\n" }, { "commit": "f06267104dd9112f11586830d22501d0e26245ea", "tree": "8f7c364abc84a5f69269974eaa2b955b24d8f421", "parents": [ "948579cd8c6ea7c8c98c52b79f4470952e182ebd" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Tue Oct 19 15:24:36 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 21:16:31 2011 -0800" }, "message": "RDMA: Update workqueue usage\n\n* ib_wq is added, which is used as the common workqueue for infiniband\n instead of the system workqueue. All system workqueue usages\n including flush_scheduled_work() callers are converted to use and\n flush ib_wq.\n\n* cancel_delayed_work() + flush_scheduled_work() converted to\n cancel_delayed_work_sync().\n\n* qib_wq is removed and ib_wq is used instead.\n\nThis is to prepare for deprecation of flush_scheduled_work().\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "843276ad985cb59212ceb70d989474521ff516f6", "tree": "d8c80eb0e35533735e219dc94d5ea43bcde6cd6a", "parents": [ "5f61b2c6939bb6d26393df15765bc3cb260db063" ], "author": { "name": "Maciej Sosnowski", "email": "maciej.sosnowski@intel.com", "time": "Wed Nov 24 17:29:54 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 13:23:35 2011 -0800" }, "message": "RDMA/nes: Fix incorrect SFP+ link status detection on driver init\n\nDuring iw_nes initialization the link status for SFP+ PHY is always\ndetected as \"up\" regardless of real state (cable either connected or\ndisconnected). Add SFP+ PHY specific link status detection to the\niw_nes initialization procedure. Use link status recheck for\nnetdev_open to detect delayed state updates.\n\nSigned-off-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5f61b2c6939bb6d26393df15765bc3cb260db063", "tree": "0a243de60e121a3bfc06c86d7dc76ca3540fe8c4", "parents": [ "ea623455b736d82f476460647e8b5fe5dc36f4f2" ], "author": { "name": "Maciej Sosnowski", "email": "maciej.sosnowski@intel.com", "time": "Wed Nov 24 17:29:46 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 13:23:34 2011 -0800" }, "message": "RDMA/nes: Fix SFP+ link down detection issue with switch port disable\n\nIn case of SFP+ PHY, link status check at interrupt processing can\ngive false results. For proper link status change detection a delayed\nrecheck is needed to give nes registers time to settle. Add a\nperiodic link status recheck scheduled at interrupt to detect\npotential delayed registers state changes.\n\nAddresses: http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id\u003d2117\nSigned-off-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ea623455b736d82f476460647e8b5fe5dc36f4f2", "tree": "c1e2343adc9b2e8f7e11d58dc561370905d60bb8", "parents": [ "2a4c97ead4b375a64063523210939b87ad225b85" ], "author": { "name": "Maciej Sosnowski", "email": "maciej.sosnowski@intel.com", "time": "Wed Nov 24 17:29:38 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 13:23:34 2011 -0800" }, "message": "RDMA/nes: Generate IB_EVENT_PORT_ERR/PORT_ACTIVE events\n\nDepending on link state change, IB_EVENT_PORT_ERR or\nIB_EVENT_PORT_ACTIVE should be generated when handling MAC interrupts.\n\nPlugging in a cable happens to result in series of interrupts changing\ndriver\u0027s link state a number of times before finally staying at link\nup (e.g. link up, link down, link up, link down, ..., link up). To\nprevent sending series of redundant IB_EVENT_PORT_ACTIVE and\nIB_EVENT_PORT_ERR events, we use a timer to debounce them in\nnes_port_ibevent().\n\nSigned-off-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2a4c97ead4b375a64063523210939b87ad225b85", "tree": "be408f5b7e31d24e0fe3dc98612c2a1afec4e260", "parents": [ "4162cf64973df51fc885825bc9ca4d055891c49f" ], "author": { "name": "Maciej Sosnowski", "email": "maciej.sosnowski@intel.com", "time": "Wed Nov 24 17:29:30 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Jan 16 13:23:33 2011 -0800" }, "message": "RDMA/nes: Fix bonding on iw_nes\n\nEnable configuring bonds on nes devices by adding missing support for\nmaster net_device to the driver.\n\nSigned-off-by: Maciej Sosnowski \u003cmaciej.sosnowski@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "695b83495e2fba9d3a883193cfc9d5eefa96a911", "tree": "23c03ec4fc840d6df5c038fe7eb49e224123f081", "parents": [ "4162cf64973df51fc885825bc9ca4d055891c49f" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Thu Jan 13 19:02:25 2011 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Thu Jan 13 14:00:43 2011 -0800" }, "message": "IB/srp: Test only once whether iu allocation succeeded\n\nMerge the two tests in srp_queuecommand() of whether information unit\nallocation succeeded into one. An intended side effect of this change\nis that we fix the warning:\n\n drivers/infiniband/ulp/srp/ib_srp.c: In function \u0027srp_queuecommand\u0027:\n drivers/infiniband/ulp/srp/ib_srp.c:1116: warning: \u0027req\u0027 may be used uninitialized in this function\n\n(seen with CONFIG_CC_OPTIMIZE_FOR_SIZE\u003dy at least with gcc 4.4.4)\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\nAcked-by: David Dillow \u003cdillowda@ornl.gov\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "008d23e4852d78bb2618f2035f8b2110b6a6b968", "tree": "81c88f744f6f3fc84132527c1ddc0b4da410c5e2", "parents": [ "8f685fbda43deccd130d192c9fcef1444649eaca", "bfc672dcf323877228682aff79dff8ecd9f30ff8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 10:05:56 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jan 13 10:05:56 2011 -0800" }, "message": "Merge branch \u0027for-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial\n\n* \u0027for-next\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (43 commits)\n Documentation/trace/events.txt: Remove obsolete sched_signal_send.\n writeback: fix global_dirty_limits comment runtime -\u003e real-time\n ppc: fix comment typo singal -\u003e signal\n drivers: fix comment typo diable -\u003e disable.\n m68k: fix comment typo diable -\u003e disable.\n wireless: comment typo fix diable -\u003e disable.\n media: comment typo fix diable -\u003e disable.\n remove doc for obsolete dynamic-printk kernel-parameter\n remove extraneous \u0027is\u0027 from Documentation/iostats.txt\n Fix spelling milisec -\u003e ms in snd_ps3 module parameter description\n Fix spelling mistakes in comments\n Revert conflicting V4L changes\n i7core_edac: fix typos in comments\n mm/rmap.c: fix comment\n sound, ca0106: Fix assignment to \u0027channel\u0027.\n hrtimer: fix a typo in comment\n init/Kconfig: fix typo\n anon_inodes: fix wrong function name in comment\n fix comment typos concerning \"consistent\"\n poll: fix a typo in comment\n ...\n\nFix up trivial conflicts in:\n - drivers/net/wireless/iwlwifi/iwl-core.c (moved to iwl-legacy.c)\n - fs/ext4/ext4.h\n\nAlso fix missed \u0027diabled\u0027 typo in drivers/net/bnx2x/bnx2x.h while at it.\n" }, { "commit": "da995a8aee044bc5d0847e19e351cd48a2cb8bcc", "tree": "000cb11c5e49e1bc91a2212ed5a2299f279c990b", "parents": [ "4979d18fe105297f8f065743f31f8f735da8df2d" ], "author": { "name": "Aleksey Senin", "email": "alex@senin.name", "time": "Thu Dec 02 11:44:49 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Jan 12 14:49:17 2011 -0800" }, "message": "IB/mlx4: Handle protocol field in multicast table\n\nThe newest device firmware stores IB vs. Ethernet protocol in two bits\nin members_count field of multicast group table (0: Infiniband, 1:\nEthernet). When changing the QP members count for a multicast group,\nit important not to reset this information. When calling multicast\nattach first time, the protocol type should be specified. In this\npatch we always set it IB, but in the future we will handle Ethernet\ntoo. When looking for a QP, the protocol type shoud be checked too.\n\nSigned-off-by: Aleksey Senin \u003calekseys@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "948579cd8c6ea7c8c98c52b79f4470952e182ebd", "tree": "77e85adbdd07be8394fa60d08d3f1dbda3c67393", "parents": [ "4162cf64973df51fc885825bc9ca4d055891c49f" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Fri Nov 05 03:07:36 2010 +0000" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Jan 12 11:11:58 2011 -0800" }, "message": "RDMA: Use vzalloc() to replace vmalloc()+memset(0)\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4979d18fe105297f8f065743f31f8f735da8df2d", "tree": "36776f85b2bee47c5a671db498939305589d23aa", "parents": [ "4162cf64973df51fc885825bc9ca4d055891c49f" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Jan 12 09:50:36 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Jan 12 09:50:36 2011 -0800" }, "message": "mlx4_{core, ib, en}: Fix driver when sizeof (phys_addr_t) \u003e sizeof (long)\n\nSome systems have PCI addresses that don\u0027t fit in unsigned long (eg some\n32-bit PowerPC 440 systems have 36-bit bus addresses). Fix up mlx4 drivers\nby using phys_addr_t where appropriate, so we don\u0027t truncate any PCI\nresource addresses before ioremapping them.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "eb4a7cbf27082bea34764bab3bc85595683f967b", "tree": "4d0d799a04c7d62d9ceb35e73f01112a49dc5d83", "parents": [ "4162cf64973df51fc885825bc9ca4d055891c49f" ], "author": { "name": "John L. Burr", "email": "jlburr@cadence.com", "time": "Tue Jan 11 20:39:46 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Jan 11 20:39:47 2011 -0800" }, "message": "IB/mthca: Fix driver when sizeof (phys_addr_t) \u003e sizeof (long)\n\nSome systems have PCI addresses that don\u0027t fit in unsigned long (eg some\n32-bit PowerPC 440 systems have 36-bit bus addresses). Fix up the driver\nby using phys_addr_t where appropriate, so we don\u0027t truncate any PCI\nresource addresses before ioremapping them.\n\nSigned-off-by: John L. Burr \u003cjlburr@cadence.com\u003e\n\n[ Update to apply to current driver source. - Roland ]\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n\n" }, { "commit": "f1d6d6cd9029daa7e7d4a0b14347b5392320f22a", "tree": "673e1940d385cfc625ed7583b54117ca97216734", "parents": [ "b9d919a4ac6cf031b8e065f82ad8f1b0c9ed74b1", "2b76c05794e66655e10633d2d78287854c991f63" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 11 16:30:08 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jan 11 16:30:08 2011 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (42 commits)\n IB/qib: Fix refcount leak in lkey/rkey validation\n IB/qib: Improve SERDES tunning on QMH boards\n IB/qib: Unnecessary delayed completions on RC connection\n IB/qib: Issue pre-emptive NAKs on eager buffer overflow\n IB/qib: RDMA lkey/rkey validation is inefficient for large MRs\n IB/qib: Change QPN increment\n IB/qib: Add fix missing from earlier patch\n IB/qib: Change receive queue/QPN selection\n IB/qib: Fix interrupt mitigation\n IB/qib: Avoid duplicate writes to the rcv head register\n IB/qib: Add a few new SERDES tunings\n IB/qib: Reset packet list after freeing\n IB/qib: New SERDES init routine and improvements to SI quality\n IB/qib: Clear WAIT_SEND flags when setting QP to error state\n IB/qib: Fix context allocation with multiple HCAs\n IB/qib: Fix multi-Florida HCA host panic on reboot\n IB/qib: Handle transitions from ACTIVE_DEFERRED to ACTIVE better\n IB/qib: UD send with immediate receive completion has wrong size\n IB/qib: Set port physical state even if other fields are invalid\n IB/qib: Generate completion callback on errors\n ...\n" }, { "commit": "2b76c05794e66655e10633d2d78287854c991f63", "tree": "f6782ce9fc2f8e975189c7f6e47e1ac6ab75ea61", "parents": [ "db8b10167126d72829653690f57b9c7ca53c4d54", "1eba27e87a85f6ed68905055bc9a7dbfb024c255", "8ae31e5b1fc73751d800d551fb30340caa53c7dd", "1397490938aa0aca39001c3fd5a9fc9387110d86", "d0444f1527f22d193c209d4fdad0dbabbf2be2ee", "601d87b0795c5affe5b73318c1088edbf10ce72f", "4db62d4786e946e6fc8c2bb1f9201508f7f46c41", "9af762719e8f8fa282de02997dced593030eb238" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:43:30 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:43:30 2011 -0800" }, "message": "Merge branches \u0027cxgb4\u0027, \u0027ipath\u0027, \u0027ipoib\u0027, \u0027mlx4\u0027, \u0027mthca\u0027, \u0027nes\u0027, \u0027qib\u0027 and \u0027srp\u0027 into for-next\n" }, { "commit": "4db62d4786e946e6fc8c2bb1f9201508f7f46c41", "tree": "dbd1bf62fa05e53d225f54d24e6db28eab7bb6cc", "parents": [ "f2d255a0787119f7f4dc0e6093a0bd2700a49402" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.org", "time": "Mon Jan 10 17:42:23 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:23 2011 -0800" }, "message": "IB/qib: Fix refcount leak in lkey/rkey validation\n\nThe mr optimization introduced a reference count leak on an exception\ntest. The lock/refcount manipulation is moved down and the problematic\nexception test now calls bail to insure that the lock is released.\n\nAdditional fixes as suggested by Ralph Campbell \u003cralph.campbell@qlogic.org\u003e:\n- reduce lock scope of dma regions\n- use explicit values on returns vs. automatic ret value\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f2d255a0787119f7f4dc0e6093a0bd2700a49402", "tree": "68de18e409adc3a48fd03d0704806fdfc854f5e3", "parents": [ "dd04e43d46ad7a4e625a9ff3b270dc0db9abe81d" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "message": "IB/qib: Improve SERDES tunning on QMH boards\n\nImprove the QMH SERDES tunning on initial driver load by having the\ndriver go through a link state change.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "dd04e43d46ad7a4e625a9ff3b270dc0db9abe81d", "tree": "9cb5e6c86c1ee179cb03b7335bf1ecb2c58ebb7d", "parents": [ "994bcd28a36af1413381dfe0aac065e2cbc2af40" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "message": "IB/qib: Unnecessary delayed completions on RC connection\n\nCurrently on receipt of a response message (ACKs, RDMA Response,\nAtomic Responses etc.) if the SDMA completion counter is not advanced\nthe driver delays the completion of the WQE. In most cases this is\noverly pessimistic as the response (ACK) to a previously transmitted\nsend implies that the send is complete. Ensure that SDMA queue is\nprogressed appropriately before determining if a send has delayed\ncompletions.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "994bcd28a36af1413381dfe0aac065e2cbc2af40", "tree": "26f4685a39d94dbfbab346862061ef6f577653f2", "parents": [ "2a600f14d25fda341b5633c75cc50a7574fc1007" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "message": "IB/qib: Issue pre-emptive NAKs on eager buffer overflow\n\nUnder congestion resulting in eager buffer overflow attempt to send\npre-emptive NAKs if header queue entries with TID errors are generated\nand a valid header is present. This prevents long timeouts and flow\nrestarts if a trailing set of packets are dropped due to eager\noverflows. Pre-emptive NAKs are currently only supported for RDMA\nwrites.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2a600f14d25fda341b5633c75cc50a7574fc1007", "tree": "02d0af0c74dc1e7ef99d2798d4b5e0a375b4deb8", "parents": [ "7c3edd3ff3098093e594dbcbc9dbeeae09b1b4a0" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "message": "IB/qib: RDMA lkey/rkey validation is inefficient for large MRs\n\nThe current code loops during rkey/lkey validiation to isolate the MR\nfor the RDMA, which is expensive when the current operation is inside\na very large memory region.\n\nThis fix optimizes rkey/lkey validation routines for user memory\nregions and fast memory regions. The MR entry can be isolated by\nshifts/mods instead of looping. The existing loop is preserved for\nphys memory regions for now.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7c3edd3ff3098093e594dbcbc9dbeeae09b1b4a0", "tree": "8cb8652beb2d2b1261949ed07c7a2b83ebf54067", "parents": [ "057ae62faceccb50b0c2387af60dbbbcd40aab84" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:22 2011 -0800" }, "message": "IB/qib: Change QPN increment\n\nChanging from +1 to +2 allows for better QP distribution across\nreceive contexts.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "057ae62faceccb50b0c2387af60dbbbcd40aab84", "tree": "e606a9f0c458306e7c4b6645aa1f3ba8a548145c", "parents": [ "2528ea60f94ef9e1e1cd82066d55f62a1d19fde1" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Add fix missing from earlier patch\n\nThe upstream code was missing part of a receive/error race fix from\nthe internal tree. Add the missing part, which makes future merges\npossible.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2528ea60f94ef9e1e1cd82066d55f62a1d19fde1", "tree": "f2e3494c3b1ac698efb1ba3ba5675d1f21d03d06", "parents": [ "19ede2e422496b2a064b9b22823c6afb66ff927b" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Change receive queue/QPN selection\n\nThe basic idea is that on SusieQ, the difficult part of mapping QPN to\ncontext is handled by the mapping registers so the generic QPN\nallocation doesn\u0027t need to worry about chip specifics. For Monty and\nLinda, there is no mapping table so the qpt-\u003emask (same as\ndd-\u003eqpn_mask), is used to see if the QPN to context falls within\n[zero..dd-\u003en_krcv_queues).\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "19ede2e422496b2a064b9b22823c6afb66ff927b", "tree": "5d440053d4662acd8f9a5dac227156878bd99db2", "parents": [ "aa7374ac19ca08715693b0a2d9f88f479af3ea7c" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Fix interrupt mitigation\n\nFor SusieQ we need to write to the interrupt timer register before\nupdating the header queue head with interrupt count. This is to\nensure that the timer is enabled properly and a receive available\ninterrupt is delivered. Otherwise this interrupt can be lost if the\nreceiver header/eager queues are full before the timer is enabled.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "aa7374ac19ca08715693b0a2d9f88f479af3ea7c", "tree": "dcdc8913026be84b1cc9d072d7a6c9103b7740f1", "parents": [ "e706203c7c1cff8c27f9ce6d58911014a6bd826c" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Avoid duplicate writes to the rcv head register\n\nAvoid duplicate writes to the head register as this can lead to lost\ninterrupts if the context goes full before the second write is done.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e706203c7c1cff8c27f9ce6d58911014a6bd826c", "tree": "d6dbf93ed8cc4f777b1857b3de8495bad922569d", "parents": [ "f73df408b299e4bc5ed66cc50058585af64a8f9e" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Add a few new SERDES tunings\n\nAdd new SERDES tuning to aid manufacturing.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f73df408b299e4bc5ed66cc50058585af64a8f9e", "tree": "6c62aace6c68102baae77473575d9a07978d2b42", "parents": [ "a0a234d47dcacfdb0a8dfcb861e0bd8300702674" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:21 2011 -0800" }, "message": "IB/qib: Reset packet list after freeing\n\nReset the list pointers after freeing the SDMA packet list. This is\ndone to any potential double-free cases.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a0a234d47dcacfdb0a8dfcb861e0bd8300702674", "tree": "2be8f11631c7f1c61a78b38c74348c73e7dd891c", "parents": [ "16028f27778cb6439516c36c0a72446d29805691" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: New SERDES init routine and improvements to SI quality\n\nImplement new SERDES initialization routine and improvements to signal\nintegrity -- disable LE1 adaptation, disable LOS after link-up, set\nbetter SERDES parameters.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "16028f27778cb6439516c36c0a72446d29805691", "tree": "1e291c492bb5f3009f770fff116d47c9ca4f4c2a", "parents": [ "6676b3f746ff164a4a367c9aa5ae4b5b1145083c" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: Clear WAIT_SEND flags when setting QP to error state\n\nIf these flags are set when the QP is transitioned to the error state,\nit will wait until the flags are cleared, which may never happen if\nthe error transition is due to a link going down.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6676b3f746ff164a4a367c9aa5ae4b5b1145083c", "tree": "4a9d0ce7dc03a56cae8f2c95da9a55d670933697", "parents": [ "5dbbcb97cc55dd6e0c34cf06f9e1ee181247d054" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: Fix context allocation with multiple HCAs\n\nThe driver was incorrectly choosing HCAs on which to allocate new user\ncontexts based on overall count of usable ports regardless whether the\nusable port was on the currently selected HCA.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5dbbcb97cc55dd6e0c34cf06f9e1ee181247d054", "tree": "14eb055d1e816e494d29db014d4c3b0c0fec6de2", "parents": [ "b3d5cb2f2067b30da53aa67e42fdd733030fb411" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: Fix multi-Florida HCA host panic on reboot\n\nAdd check when setting configured contexts that the value does not\nexceed the number of contexts allocated for the card. If the value\nexceeds the already allocated count, set it to what is already\nallocated.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b3d5cb2f2067b30da53aa67e42fdd733030fb411", "tree": "6092769d355fbe7a1b76677d39de7521cdb2284c", "parents": [ "c7665e5a6988450a6cc19c2dc1dff4d290077614" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: Handle transitions from ACTIVE_DEFERRED to ACTIVE better\n\nWhen the link transitions from ACTIVE_DEFERRED to ACTIVE, the driver\nonly sees the ACTIVE state. With this change, it will check whether\nthe state was already ACTIVE and if so, it will not generated IB\nevents and will not clear symbol error counts.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c7665e5a6988450a6cc19c2dc1dff4d290077614", "tree": "09dfcc1f57b33ce7d6199044d2947e915fd412f5", "parents": [ "3c9e5f4d657170c2fbc3d382d2daae3820713a6c" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:20 2011 -0800" }, "message": "IB/qib: UD send with immediate receive completion has wrong size\n\nThe code to generate receive completion entries for UD send with\nimmediate contains the wrong payload length. This is because when the\ncode to compute the payload size was moved, the value of hdrsize\ndidn\u0027t get moved too. The fix is to update tlen directly.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3c9e5f4d657170c2fbc3d382d2daae3820713a6c", "tree": "5d3417d0d69c1d6610d1c6ab5ea597ed2242e4cb", "parents": [ "a377acd15188cf11e47a67af5feea098cd8b25d2" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "message": "IB/qib: Set port physical state even if other fields are invalid\n\nThe IBTA vol. 1 release 1.2.1 spec. says:\nC14-24.2.1: If PortInfo:Portstate\u003dDown, then a SubnSet(PortInfo) shall\nmake any changes it specifies to PortInfo:PortPhysicalState; any other\nresult is vendor-dependent.\n\nThe patch changes the error handling so that the reply says there are\ninvalid fields but still attempts to set fields that are in range\nincluding PortInfo:PortPhysicalState.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a377acd15188cf11e47a67af5feea098cd8b25d2", "tree": "07b03137da672a9616b0a75e7296db621bf831e0", "parents": [ "f509f9c14d3f70834f964189293bed3e0e1fc839" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "message": "IB/qib: Generate completion callback on errors\n\nAccording to IBTA vol. 1, C11-30.1.1, a notification callback is\ninvoked if the CQ is armed for the next solicited completion event or\nan error completion. The error case wasn\u0027t being generated correctly.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "f509f9c14d3f70834f964189293bed3e0e1fc839", "tree": "554de15dc05c9db57369104c4c05b59b313bde1c", "parents": [ "0a43e117221702b08a023d6aa1a31ac30e40866b" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "message": "IB/qib: Add support for the new QME7362 card\n\nAdd support to recognize another board variation named QME7362.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0a43e117221702b08a023d6aa1a31ac30e40866b", "tree": "e5977969e7035117ec6c19fac4adfcf69a2ff675", "parents": [ "9d5b243f24212966d6d06058f96b2b1a22482f59" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "message": "IB/qib: Add receive header queue size module parameters\n\nThe receive header queue sizes need to modified for performance\ntuning. Three module parameters are added to support this.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9d5b243f24212966d6d06058f96b2b1a22482f59", "tree": "51b528e81624bc7d31472539e46fad9cd4c93d5d", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Mike Marciniszyn", "email": "mike.marciniszyn@qlogic.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:19 2011 -0800" }, "message": "IB/qib: Remove IB latency turnoff\n\nThis is required for hardware testing.\n\nSigned-off-by: Mike Marciniszyn \u003cmike.marciniszyn@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "601d87b0795c5affe5b73318c1088edbf10ce72f", "tree": "465f6a3b6b39c9b4163e925d16ada2315e0b8856", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Mon Jan 10 17:42:14 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:14 2011 -0800" }, "message": "RDMA/nes: Fix string continuation line\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n\n" }, { "commit": "d0444f1527f22d193c209d4fdad0dbabbf2be2ee", "tree": "d2f17364cd4455470ab5b8ef62151d48aa77507f", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Dan Carpenter", "email": "error27@gmail.com", "time": "Mon Jan 10 17:42:10 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:10 2011 -0800" }, "message": "IB/mthca: Handle -ENOMEM in forward_trap()\n\nib_create_send_mad() can return ERR_PTR(-ENOMEM) here.\n\nSigned-off-by: Dan Carpenter \u003cerror27@gmail.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1397490938aa0aca39001c3fd5a9fc9387110d86", "tree": "8f4baeec7d2349f49face4ba9a90903afe4dca3f", "parents": [ "030b4b3309e29c6c857d8521c3076743663c259e" ], "author": { "name": "Dan Carpenter", "email": "error27@gmail.com", "time": "Mon Jan 10 17:42:06 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:06 2011 -0800" }, "message": "IB/mlx4: Handle -ENOMEM in forward_trap()\n\nib_create_send_mad() can return ERR_PTR(-ENOMEM) here.\n\nSigned-off-by: Dan Carpenter \u003cerror27@gmail.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3afa9f19e5bd16abed998b7bf1b178206403286f", "tree": "2bd80f9204eaa867bbd7d4aee3ca51c11f3c5f75", "parents": [ "f5a49539a64ae5e3833fddec54a3c2524c7fe333" ], "author": { "name": "Vladimir Sokolovsky", "email": "vlad@dev.mellanox.co.il", "time": "Mon Jan 10 17:42:06 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:42:06 2011 -0800" }, "message": "IB/mlx4: Don\u0027t call dma_free_coherent() with irqs disabled\n\nmlx4_ib_free_cq_buf() should not be called under spin_lock_irq() since\nit calls dma_free_coherent(), which needs irqs enabled. Fix this by\ndeferring the free to outside the locked region.\n\nThis was found due to the\n\n\tWARN_ON(irqs_disabled());\n\nin swiotlb_free_coherent().\n\nSigned-off-by: Vladimir Sokolovsky \u003cvlad@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8ae31e5b1fc73751d800d551fb30340caa53c7dd", "tree": "d615fd6f91f150427bf143936fb686f38609cf06", "parents": [ "19e364f6801e38972673278adedaab1abf6f854c" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Mon Jan 10 17:41:55 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:41:55 2011 -0800" }, "message": "IPoIB: Add GRO support\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "19e364f6801e38972673278adedaab1abf6f854c", "tree": "ba2d39f4f449e9b1e911e6d5cbc39e08e54410c3", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Mon Jan 10 17:41:54 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:41:54 2011 -0800" }, "message": "IPoIB: Remove LRO support\n\nAs a first step in moving from LRO to GRO, revert commit af40da894e9\n(\"IPoIB: add LRO support\"). Also eliminate the ethtool set_flags\ncallback which isn\u0027t needed anymore. Finally, we need to include\n\u003clinux/sched.h\u003e directly to get the declaration of restart_syscall()\n(which used to be included implicitly through \u003clinux/inet_lro.h\u003e).\n\nCc: Ben Hutchings \u003cbhutchings@solarflare.com\u003e\nCc: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Vladimir Sokolovsky \u003cvlad@mellanox.co.il\u003e\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1eba27e87a85f6ed68905055bc9a7dbfb024c255", "tree": "aecc39e079df0ab704799a98df7e75604178477e", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Mon Jan 10 17:41:50 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:41:50 2011 -0800" }, "message": "IB/ipath: Use printf extension %pR for struct resource\n\nUsing %pR standardizes the struct resource output.\n\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "db8b10167126d72829653690f57b9c7ca53c4d54", "tree": "3d99ca03dee2fe076932821087a6558572f47094", "parents": [ "c94310916390ac162e70c8c134bf70680c944a02" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Mon Jan 10 17:41:43 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:41:43 2011 -0800" }, "message": "RDMA/cxgb4: Don\u0027t re-init wait object in init/fini paths\n\nRe-initializing the wait object in rdma_init()/rdma_fini() causes a\ntiming window which can lead to a deadlock during close. Once this\ndeadlock hits, all RDMA activity over the T4 device will be stuck.\n\nThere\u0027s no need to re-init the wait object, so remove it.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c94310916390ac162e70c8c134bf70680c944a02", "tree": "a83115ae8412f2117f4041cbb31a395c113a611e", "parents": [ "3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@vyatta.com", "time": "Mon Jan 10 17:41:43 2011 -0800" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Mon Jan 10 17:41:43 2011 -0800" }, "message": "RDMA/cxgb3,cxgb4: Remove dead code\n\nThis removes unused code found by running \u0027make namespacecheck\u0027;\ncompile tested only.\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@vyatta.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9af762719e8f8fa282de02997dced593030eb238", "tree": "707ebec104ed79499935fe6367357121fbb6906e", "parents": [ "e9684678221441f886b4d7c74f8770bb0981737a" ], "author": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Fri Nov 26 15:34:46 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Mon Jan 10 15:44:51 2011 -0500" }, "message": "IB/srp: consolidate hot-path variables into cache lines\n\nPut the variables accessed together in the hot-path into common\ncachelines, and separate them by RW vs RO to avoid false dirtying.\nWe keep a local copy of the lkey and rkey in the target to avoid\ntraversing pointers (and associated cache lines) to find them.\n\nReviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "e9684678221441f886b4d7c74f8770bb0981737a", "tree": "76ed83fb7dd41c84c2b8e08b785a802e99c6e790", "parents": [ "94a9174c630c8465ed9e97ecd242993429930c05" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 15:08:38 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Mon Jan 10 15:44:50 2011 -0500" }, "message": "IB/srp: stop sharing the host lock with SCSI\n\nWe don\u0027t need protection against the SCSI stack, so use our own lock to\nallow parallel progress on separate CPUs.\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out and small cleanups by David Dillow ]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "94a9174c630c8465ed9e97ecd242993429930c05", "tree": "73ddbe0024d54bad543c90144576148b70922eea", "parents": [ "76c75b258f1fe6abac6af2356989ad4d6518886e" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 14:50:09 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Mon Jan 10 15:44:50 2011 -0500" }, "message": "IB/srp: reduce lock coverage of command completion\n\nWe only need the lock to cover list and credit manipulations, so push\nthose into srp_remove_req() and update the call chains.\n\nWe reorder the request removal and command completion in\nsrp_process_rsp() to avoid the SCSI mid-layer sending another command\nbefore we\u0027ve released our request and added any credits returned by the\ntarget. This prevents us from returning HOST_BUSY unneccesarily.\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out, small cleanups, and modified to avoid potential extraneous\n HOST_BUSY returns by David Dillow ]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "76c75b258f1fe6abac6af2356989ad4d6518886e", "tree": "754db154ad76e44b5ecc6da71097a6f1c096ec6f", "parents": [ "536ae14e7588e85203d4b4147c041309be5b3efb" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 14:37:47 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Mon Jan 10 15:44:49 2011 -0500" }, "message": "IB/srp: reduce local coverage for command submission and EH\n\nWe only need locks to protect our lists and number of credits available.\nBy pre-consuming the credit for the request, we can reduce our lock\ncoverage to just those areas. If we don\u0027t actually send the request,\nwe\u0027ll need to put the credit back into the pool.\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out and small cleanups by David Dillow ]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "536ae14e7588e85203d4b4147c041309be5b3efb", "tree": "9a7e459a914e138d0b02487c7ab40ae9ffe6c05c", "parents": [ "dcb4cb85f4b7caac9769bce464fef16306a4758c" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 13:58:27 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Mon Jan 10 15:44:42 2011 -0500" }, "message": "IB/srp: don\u0027t move active requests to their own list\n\nWe use req-\u003escmnd !\u003d NULL to indicate an active request, so there\u0027s no\nneed to keep a separate list for them. We can afford the array iteration\nduring error handling, and dropping it gives us one less item that needs\nlock protection.\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out and small cleanups by David Dillow ]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "b4a45f5fe8078bfc10837dbd5b98735058bc4698", "tree": "df6f13a27610a3ec7eb4a661448cd779a8f84c79", "parents": [ "01539ba2a706ab7d35fc0667dff919ade7f87d63", "b3e19d924b6eaf2ca7d22cba99a517c5171007b6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 07 08:56:33 2011 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jan 07 08:56:33 2011 -0800" }, "message": "Merge branch \u0027vfs-scale-working\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/npiggin/linux-npiggin\n\n* \u0027vfs-scale-working\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/npiggin/linux-npiggin: (57 commits)\n fs: scale mntget/mntput\n fs: rename vfsmount counter helpers\n fs: implement faster dentry memcmp\n fs: prefetch inode data in dcache lookup\n fs: improve scalability of pseudo filesystems\n fs: dcache per-inode inode alias locking\n fs: dcache per-bucket dcache hash locking\n bit_spinlock: add required includes\n kernel: add bl_list\n xfs: provide simple rcu-walk ACL implementation\n btrfs: provide simple rcu-walk ACL implementation\n ext2,3,4: provide simple rcu-walk ACL implementation\n fs: provide simple rcu-walk generic_check_acl implementation\n fs: provide rcu-walk aware permission i_ops\n fs: rcu-walk aware d_revalidate method\n fs: cache optimise dentry and inode for rcu-walk\n fs: dcache reduce branches in lookup path\n fs: dcache remove d_mounted\n fs: fs_struct use seqlock\n fs: rcu-walk for path lookup\n ...\n" }, { "commit": "dc0474be3e27463d4d4a2793f82366eed906f223", "tree": "41f75e638442cb343bacdcfbabb17ffc3bd5b4ce", "parents": [ "357f8e658bba8a085c4a5d4331e30894be8096b8" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:49:43 2011 +1100" }, "committer": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:50:24 2011 +1100" }, "message": "fs: dcache rationalise dget variants\n\ndget_locked was a shortcut to avoid the lazy lru manipulation when we already\nheld dcache_lock (lru manipulation was relatively cheap at that point).\nHowever, how that the lru lock is an innermost one, we never hold it at any\ncaller, so the lock cost can now be avoided. We already have well working lazy\ndcache LRU, so it should be fine to defer LRU manipulations to scan time.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "b5c84bf6f6fa3a7dfdcb556023a62953574b60ee", "tree": "7a2c299a180713e21d5cb653cb933121adf53c31", "parents": [ "949854d02455080d20cd3e1db28a3a18daf7599d" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:49:38 2011 +1100" }, "committer": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:50:23 2011 +1100" }, "message": "fs: dcache remove dcache_lock\n\ndcache_lock no longer protects anything. remove it.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "b7ab39f631f505edc2bbdb86620d5493f995c9da", "tree": "62be97ebc7fc69ceb601f23312d335ebb8038ee7", "parents": [ "2304450783dfde7b0b94ae234edd0dbffa865073" ], "author": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:49:32 2011 +1100" }, "committer": { "name": "Nick Piggin", "email": "npiggin@kernel.dk", "time": "Fri Jan 07 17:50:21 2011 +1100" }, "message": "fs: dcache scale dentry refcount\n\nMake d_count non-atomic and protect it with d_lock. This allows us to ensure a\n0 refcount dentry remains 0 without dcache_lock. It is also fairly natural when\nwe start protecting many other dentry members with d_lock.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@kernel.dk\u003e\n" }, { "commit": "dcb4cb85f4b7caac9769bce464fef16306a4758c", "tree": "1c21889c8b778415f9a8d46fcf28853eb42ede18", "parents": [ "9709f0e05b827049733f439de82a4a1688b37b86" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 13:22:48 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Wed Jan 05 15:24:25 2011 -0500" }, "message": "IB/srp: allow lockless work posting\n\nOnly one CPU at a time will own an RX IU, so using the address of the IU\nas the work request cookie allows us to avoid taking a lock. We can\nsimilarly prepare the TX path for lockless posting by moving the free TX\nIUs to a list. This also removes the requirement that the queue sizes be\na power of 2.\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out, small cleanups, and modified to avoid needing an extra field\n in the IU by David Dillow]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" }, { "commit": "9709f0e05b827049733f439de82a4a1688b37b86", "tree": "62bf3010092a46c6971eaef148719eee59d85074", "parents": [ "f8b6e31e4e46bf514c27fce38783ed5615cca01d" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Fri Nov 26 13:13:06 2010 -0500" }, "committer": { "name": "David Dillow", "email": "dillowda@ornl.gov", "time": "Wed Jan 05 15:24:25 2011 -0500" }, "message": "IB/srp: consolidate state change code\n\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n[ broken out and small cleanups by David Dillow ]\nSigned-off-by: David Dillow \u003cdillowda@ornl.gov\u003e\n" } ], "next": "f8b6e31e4e46bf514c27fce38783ed5615cca01d" }