)]}' { "log": [ { "commit": "53d412fce05e73dd0b25b0ebfa83c7ee94f16451", "tree": "3aca42d6408bfcded0d5e829a9446a1c04af9b21", "parents": [ "51cf22495ae2fe60ba28123e04059cff4ddd9461" ], "author": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Jul 24 14:41:13 2007 +0200" }, "committer": { "name": "Jens Axboe", "email": "jens.axboe@oracle.com", "time": "Tue Oct 16 11:20:59 2007 +0200" }, "message": "infiniband: sg chaining support\n\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n" }, { "commit": "200d1713b47200aa478f27e454e3d957264d49be", "tree": "cc7a1849ebfac45538075af5819aadfeeba4b331", "parents": [ "732a2170f499ce7cf5f0bdd4f9e0b0c8337b67e1" ], "author": { "name": "Moni Shoua", "email": "monis@voltaire.com", "time": "Tue Oct 09 19:43:37 2007 -0700" }, "committer": { "name": "Jeff Garzik", "email": "jeff@garzik.org", "time": "Mon Oct 15 14:20:45 2007 -0400" }, "message": "IB/ipoib: Verify address handle validity on send\n\nWhen the bonding device senses a carrier loss of its active slave it replaces\nthat slave with a new one. In between the times when the carrier of an IPoIB\ndevice goes down and ipoib_neigh is destroyed, it is possible that the\nbonding driver will send a packet on a new slave that uses an old ipoib_neigh.\nThis patch detects and prevents this from happenning.\n\nSigned-off-by: Moni Shoua \u003cmonis at voltaire.com\u003e\nSigned-off-by: Or Gerlitz \u003cogerlitz at voltaire.com\u003e\nAcked-by: Roland Dreier \u003crdreier@cisco.com\u003e\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\n" }, { "commit": "732a2170f499ce7cf5f0bdd4f9e0b0c8337b67e1", "tree": "db71c08db18421709b829aa1ea87803bbaf3e752", "parents": [ "a8a935dae5de65a8f5a5371a58ad2aa73a0ea2da" ], "author": { "name": "Moni Shoua", "email": "monis@voltaire.com", "time": "Tue Oct 09 19:43:36 2007 -0700" }, "committer": { "name": "Jeff Garzik", "email": "jeff@garzik.org", "time": "Mon Oct 15 14:20:45 2007 -0400" }, "message": "IB/ipoib: Bound the net device to the ipoib_neigh structue\n\nIPoIB uses a two layer neighboring scheme, such that for each struct neighbour\nwhose device is an ipoib one, there is a struct ipoib_neigh buddy which is\ncreated on demand at the tx flow by an ipoib_neigh_alloc(skb-\u003edst-\u003eneighbour)\ncall.\n\nWhen using the bonding driver, neighbours are created by the net stack on behalf\nof the bonding (master) device. On the tx flow the bonding code gets an skb such\nthat skb-\u003edev points to the master device, it changes this skb to point on the\nslave device and calls the slave hard_start_xmit function.\n\nUnder this scheme, ipoib_neigh_destructor assumption that for each struct\nneighbour it gets, n-\u003edev is an ipoib device and hence netdev_priv(n-\u003edev)\ncan be casted to struct ipoib_dev_priv is buggy.\n\nTo fix it, this patch adds a dev field to struct ipoib_neigh which is used\ninstead of the struct neighbour dev one, when n-\u003edev-\u003eflags has the\nIFF_MASTER bit set.\n\nSigned-off-by: Moni Shoua \u003cmonis at voltaire.com\u003e\nSigned-off-by: Or Gerlitz \u003cogerlitz at voltaire.com\u003e\nAcked-by: Roland Dreier \u003crdreier@cisco.com\u003e\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\n" }, { "commit": "df3d80f5a5c74168be42788364d13cf6c83c7b9c", "tree": "892a964c2fd28d028f2fb7471e8543d3f4006a58", "parents": [ "3d06f7a5f74a813cee817c4b30b5e6f0398da0be", "c8e91b0a8fc8493e3bf3efcb3c8f866e9453cf1c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Oct 15 08:19:33 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Oct 15 08:19:33 2007 -0700" }, "message": "Merge master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6\n\n* master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (207 commits)\n [SCSI] gdth: fix CONFIG_ISA build failure\n [SCSI] esp_scsi: remove __dev{init,exit}\n [SCSI] gdth: !use_sg cleanup and use of scsi accessors\n [SCSI] gdth: Move members from SCp to gdth_cmndinfo, stage 2\n [SCSI] gdth: Setup proper per-command private data\n [SCSI] gdth: Remove gdth_ctr_tab[]\n [SCSI] gdth: switch to modern scsi host registration\n [SCSI] gdth: gdth_interrupt() gdth_get_status() \u0026 gdth_wait() fixes\n [SCSI] gdth: clean up host private data\n [SCSI] gdth: Remove virt hosts\n [SCSI] gdth: Reorder scsi_host_template intitializers\n [SCSI] gdth: kill gdth_{read,write}[bwl] wrappers\n [SCSI] gdth: Remove 2.4.x support, in-kernel changelog\n [SCSI] gdth: split out pci probing\n [SCSI] gdth: split out eisa probing\n [SCSI] gdth: split out isa probing\n gdth: Make one abuse of scsi_cmnd less obvious\n [SCSI] NCR5380: Use scsi_eh API for REQUEST_SENSE invocation\n [SCSI] usb storage: use scsi_eh API in REQUEST_SENSE execution\n [SCSI] scsi_error: Refactoring scsi_error to facilitate in synchronous REQUEST_SENSE\n ...\n" }, { "commit": "7eff2e7a8b65c25920207324e56611150eb1cd9a", "tree": "02a0eeba9d25d996233e30c18f258dfae0ae2139", "parents": [ "8380770c842faef3001e44662953d64ad9a93663" ], "author": { "name": "Kay Sievers", "email": "kay.sievers@vrfy.org", "time": "Tue Aug 14 15:15:12 2007 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Fri Oct 12 14:51:01 2007 -0700" }, "message": "Driver core: change add_uevent_var to use a struct\n\nThis changes the uevent buffer functions to use a struct instead of a\nlong list of parameters. It does no longer require the caller to do the\nproper buffer termination and size accounting, which is currently wrong\nin some places. It fixes a known bug where parts of the uevent\nenvironment are overwritten because of wrong index calculations.\n\nMany thanks to Mathieu Desnoyers for finding bugs and improving the\nerror handling.\n\nSigned-off-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nCc: Mathieu Desnoyers \u003cmathieu.desnoyers@polymtl.ca\u003e\nCc: Cornelia Huck \u003ccornelia.huck@de.ibm.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n\n\n" }, { "commit": "aebd5e476ecc8ceb53577b20f2a352ff4ceffd8d", "tree": "43103d35db1aa80440515db3fc968b0d96dd1d5e", "parents": [ "3236822b1c9b67ad10745d965515b528818f1120" ], "author": { "name": "FUJITA Tomonori", "email": "tomof@acm.org", "time": "Wed Jul 11 15:08:15 2007 +0900" }, "committer": { "name": "James Bottomley", "email": "jejb@mulgrave.localdomain", "time": "Fri Oct 12 14:37:46 2007 -0400" }, "message": "[SCSI] transport_srp: add rport roles attribute\n\nThis adds a \u0027roles\u0027 attribute to rport like transport_fc. The role can\nbe initiator or target. That is, the initiator driver creates target\nremote ports and the target driver creates initiator remote ports.\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nSigned-off-by: Mike Christie \u003cmichaelc@cs.wisc.edu\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@SteelEye.com\u003e\n" }, { "commit": "3236822b1c9b67ad10745d965515b528818f1120", "tree": "afcffc2cd7a15f8ef362e4a114965a3e2c9d7cff", "parents": [ "4d68041907b150b07640b607c6c626391cf3fe8b" ], "author": { "name": "FUJITA Tomonori", "email": "fujita.tomonori@lab.ntt.co.jp", "time": "Wed Jun 27 16:33:12 2007 +0900" }, "committer": { "name": "James Bottomley", "email": "jejb@mulgrave.localdomain", "time": "Fri Oct 12 14:37:42 2007 -0400" }, "message": "[SCSI] ib_srp: convert to use the srp transport class\n\nThis converts ib_srp to use the srp transport class.\n\nI don\u0027t have ib hardware so I\u0027ve not tested this patch.\n\nSigned-off-by: FUJITA Tomonori \u003cfujita.tomonori@lab.ntt.co.jp\u003e\nCc: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: James Bottomley \u003cJames.Bottomley@SteelEye.com\u003e\n" }, { "commit": "ce9d3c9a6a9aef61525be07fe6ba27d937236aa2", "tree": "1b29bcb8f60fc6b59fa0d7b833cc733b8ebe17c9", "parents": [ "038a5008b2f395c85e6e71d6ddf3c684e7c405b0", "3d73c2884f45f9a297cbc956cea101405a9703f2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Thu Oct 11 19:43:13 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Thu Oct 11 19:43:13 2007 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (87 commits)\n mlx4_core: Fix section mismatches\n IPoIB: Allow setting policy to ignore multicast groups\n IB/mthca: Mark error paths as unlikely() in post_srq_recv functions\n IB/ipath: Minor fix to ordering of freeing and zeroing of tid pages.\n IB/ipath: Remove redundant link state checks\n IB/ipath: Fix IB_EVENT_PORT_ERR event\n IB/ipath: Better handling of unexpected GPIO interrupts\n IB/ipath: Maintain active time on all chips\n IB/ipath: Fix QHT7040 serial number check\n IB/ipath: Indicate a couple of chip bugs to userspace\n IB/ipath: iba6110 rev4 no longer needs recv header overrun workaround\n IB/ipath: Use counters in ipath_poll and cleanup interrupts in ipath_close\n IB/ipath: Remove duplicate copy of LMC\n IB/ipath: Add ability to set the LMC via the sysfs debugging interface\n IB/ipath: Optimize completion queue entry insertion and polling\n IB/ipath: Implement IB_EVENT_QP_LAST_WQE_REACHED\n IB/ipath: Generate flush CQE when QP is in error state\n IB/ipath: Remove redundant code\n IB/ipath: Future proof eeprom checksum code (contents reading)\n IB/ipath: UC RDMA WRITE with IMMEDIATE doesn\u0027t send the immediate\n ...\n" }, { "commit": "227b60f5102cda4e4ab792b526a59c8cb20cd9f8", "tree": "2c9e372601ba794894833b0618bc531a9f5d57c4", "parents": [ "06393009000779b00a558fd2f280882cc7dc2008" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@linux-foundation.org", "time": "Wed Oct 10 17:30:46 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 10 17:30:46 2007 -0700" }, "message": "[INET]: local port range robustness\n\nExpansion of original idea from Denis V. Lunev \u003cden@openvz.org\u003e\n\nAdd robustness and locking to the local_port_range sysctl.\n1. Enforce that low \u003c high when setting.\n2. Use seqlock to ensure atomic update.\n\nThe locking might seem like overkill, but there are\ncases where sysadmin might want to change value in the\nmiddle of a DoS attack.\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9153f66a5b8e63c61374df4e6a4cbd0056e45178", "tree": "e0a0ae6fbbee67c25a8c414a4e9ae81226e3723a", "parents": [ "f2cade13371f3ca80c5479c284e6455c681c8a0c" ], "author": { "name": "Roland Dreier", "email": "rdreier@cisco.com", "time": "Tue Oct 09 15:46:13 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:55:30 2007 -0700" }, "message": "IPoIB: Fix unused variable warning\n\nThe conversion to use netdevice internal stats left an unused variable\nin ipoib_neigh_free(), since there\u0027s no longer any reason to get\nnetdev_priv() in order to increment dropped packets. Delete the\nunused priv variable.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\n" }, { "commit": "de90351219a1f1fd3cb45cf6fcc4e9d6407fd2c9", "tree": "5fd6193eb233e900452f719fd56d7065b2fb939e", "parents": [ "587117414909e9c52f50e3c9d1f85b3dc1815d75" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Sep 28 15:33:51 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:53:41 2007 -0700" }, "message": "[IPoIB]: Convert to netdevice internal stats\n\nUse the stats member of struct netdevice in IPoIB, so we can save\nmemory by deleting the stats member of struct ipoib_dev_priv, and save\ncode by deleting ipoib_get_stats().\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3b04ddde02cf1b6f14f2697da5c20eca5715017f", "tree": "9da1341a5a399a507b5ea6bf5a3047506b8d8f8f", "parents": [ "b95cce3576813ac3f86bafa6b5daaaaf7574b0fe" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@linux-foundation.org", "time": "Tue Oct 09 01:40:57 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:52:52 2007 -0700" }, "message": "[NET]: Move hardware header operations out of netdevice.\n\nSince hardware header operations are part of the protocol class\nnot the device instance, make them into a separate object and\nsave memory.\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "10d024c1b2fd58af8362670d7d6e5ae52fc33353", "tree": "dbfb03c539986e2c1270385eb0083aaf0dfca8ab", "parents": [ "596c5c97431eab8465739c169401ea611127b9ad" ], "author": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Mon Sep 17 13:11:17 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:51:13 2007 -0700" }, "message": "[NET]: Nuke SET_MODULE_OWNER macro.\n\nIt\u0027s been a useless no-op for long enough in 2.6 so I figured it\u0027s time to\nremove it. The number of people that could object because they\u0027re\nmaintaining unified 2.4 and 2.6 drivers is probably rather small.\n\n[ Handled drivers added by netdev tree and some missed IRDA cases... -DaveM ]\n\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "881d966b48b035ab3f3aeaae0f3d3f9b584f45b2", "tree": "c579d59a4107cbbe9e2b85939bc0d496b815c887", "parents": [ "b4b510290b056b86611757ce1175a230f1080f53" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Mon Sep 17 11:56:21 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:49:10 2007 -0700" }, "message": "[NET]: Make the device list and device lookups per namespace.\n\nThis patch makes most of the generic device layer network\nnamespace safe. This patch makes dev_base_head a\nnetwork namespace variable, and then it picks up\na few associated variables. The functions:\ndev_getbyhwaddr\ndev_getfirsthwbytype\ndev_get_by_flags\ndev_get_by_name\n__dev_get_by_name\ndev_get_by_index\n__dev_get_by_index\ndev_ioctl\ndev_ethtool\ndev_load\nwireless_process_ioctl\n\nwere modified to take a network namespace argument, and\ndeal with it.\n\nvlan_ioctl_set and brioctl_set were modified so their\nhooks will receive a network namespace argument.\n\nSo basically anthing in the core of the network stack that was\naffected to by the change of dev_base was modified to handle\nmultiple network namespaces. The rest of the network stack was\nsimply modified to explicitly use \u0026init_net the initial network\nnamespace. This can be fixed when those components of the network\nstack are modified to handle multiple network namespaces.\n\nFor now the ifindex generator is left global.\n\nFundametally ifindex numbers are per namespace, or else\nwe will have corner case problems with migration when\nwe get that far.\n\nAt the same time there are assumptions in the network stack\nthat the ifindex of a network device won\u0027t change. Making\nthe ifindex number global seems a good compromise until\nthe network stack can cope with ifindex changes when\nyou change namespaces, and the like.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bea3348eef27e6044b6161fd04c3152215f96411", "tree": "f0990b263e5ce42505d290a4c346fe990bcd4c33", "parents": [ "dde4e47e8fe333a5649a3fa0e7db1fa7c08d6158" ], "author": { "name": "Stephen Hemminger", "email": "shemminger@linux-foundation.org", "time": "Wed Oct 03 16:41:36 2007 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Oct 10 16:47:45 2007 -0700" }, "message": "[NET]: Make NAPI polling independent of struct net_device objects.\n\nSeveral devices have multiple independant RX queues per net\ndevice, and some have a single interrupt doorbell for several\nqueues.\n\nIn either case, it\u0027s easier to support layouts like that if the\nstructure representing the poll is independant from the net\ndevice itself.\n\nThe signature of the -\u003epoll() call back goes from:\n\n\tint foo_poll(struct net_device *dev, int *budget)\n\nto\n\n\tint foo_poll(struct napi_struct *napi, int budget)\n\nThe caller is returned the number of RX packets processed (or\nthe number of \"NAPI credits\" consumed if you want to get\nabstract). The callee no longer messes around bumping\ndev-\u003equota, *budget, etc. because that is all handled in the\ncaller upon return.\n\nThe napi_struct is to be embedded in the device driver private data\nstructures.\n\nFurthermore, it is the driver\u0027s responsibility to disable all NAPI\ninstances in it\u0027s -\u003estop() device close handler. Since the\nnapi_struct is privatized into the driver\u0027s private data structures,\nonly the driver knows how to get at all of the napi_struct instances\nit may have per-device.\n\nWith lots of help and suggestions from Rusty Russell, Roland Dreier,\nMichael Chan, Jeff Garzik, and Jamal Hadi Salim.\n\nBug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra,\nJoseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan.\n\n[ Ported to current tree and all drivers converted. Integrated\n Stephen\u0027s follow-on kerneldoc additions, and restored poll_list\n handling to the old style to fix mutual exclusion issues. -DaveM ]\n\nSigned-off-by: Stephen Hemminger \u003cshemminger@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "335a64a5a958002bc238c90de695e120c3c8c120", "tree": "e75f6aba0b89516e4100b5b9d6e77be86ce4d79d", "parents": [ "55a98e955caab78a5959933a4a3a0136e2491d6c" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@voltaire.com", "time": "Mon Oct 08 10:13:00 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Oct 10 13:02:30 2007 -0700" }, "message": "IPoIB: Allow setting policy to ignore multicast groups\n\nThe kernel IB stack allows (through the RDMA CM) userspace\napplications to join and use multicast groups from the IPoIB MGID\nrange. This allows multicast traffic to be handled directly from\nuserspace QPs, without going through the kernel stack, which gives\nbetter performance for some applications.\n\nHowever, to fully interoperate with IP multicast, such userspace\napplications need to participate in IGMP reports and queries, or else\nrouters may not forward the multicast traffic to the system where the\napplication is running. The simplest way to do this is to share the\nkernel IGMP implementation by using the IP_ADD_MEMBERSHIP option to\njoin multicast groups that are being handled directly in userspace.\n\nHowever, in such cases, the actual multicast traffic should not also\nbe handled by the IPoIB interface, because that would burn resources\nhandling multicast packets that will just be discarded in the kernel.\n\nTo handle this, this patch adds lookup on the database used for IB\nmulticast group reference counting when IPoIB is joining multicast\ngroups, and if a multicast group is already handled by user space,\nthen the IPoIB kernel driver ignores the group. This is controlled by\na per-interface policy flag. When the flag is set, IPoIB will not\njoin and attach its QP to a multicast group which already has an entry\nin the database; when the flag is cleared, IPoIB will behave as before\nthis change.\n\nFor each IPoIB interface, the /sys/class/net/$intf/umcast attribute\ncontrols the policy flag. The default value is off/0.\n\nSigned-off-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "55a98e955caab78a5959933a4a3a0136e2491d6c", "tree": "e1e69c6b548b48a9d8d762a4fa9423f9e3acf549", "parents": [ "3ac8c70f74ca67111c570f4ba828cc4b6fc229f4" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Wed Oct 10 17:55:37 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Oct 10 11:24:33 2007 -0700" }, "message": "IB/mthca: Mark error paths as unlikely() in post_srq_recv functions\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3ac8c70f74ca67111c570f4ba828cc4b6fc229f4", "tree": "d8e99d579fc2c7cead30215c68d5c006badc7abd", "parents": [ "bda94e32b39c0e60d43b34a175363601b6f12ca4" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Wed Oct 03 12:47:00 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 21:03:02 2007 -0700" }, "message": "IB/ipath: Minor fix to ordering of freeing and zeroing of tid pages.\n\nFixed to be the same as everywhere else. copy and then zero the page *\nin the array first, and then pass the copy to the VM routines.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "bda94e32b39c0e60d43b34a175363601b6f12ca4", "tree": "449de15b91f5188d498a2e5174b90af45e179b77", "parents": [ "49739b3e24a10d819d3167a1c5b319d0b1186245" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Oct 05 16:03:21 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 21:02:08 2007 -0700" }, "message": "IB/ipath: Remove redundant link state checks\n\nThis patch removes some redundant checks when the SMA changes the link\nstate since the same checks are made in the lower level function that\nsets the state.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "49739b3e24a10d819d3167a1c5b319d0b1186245", "tree": "487d4df80efa73d31ab414564021503d289b6b3f", "parents": [ "6a733cdc71b7aa8107caa57f2a16629aa731242a" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Wed Sep 19 16:47:31 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 21:01:38 2007 -0700" }, "message": "IB/ipath: Fix IB_EVENT_PORT_ERR event\n\nThe link state event calls were being generated when the SM told the SMA\nto change link states. This works for IB_EVENT_PORT_ACTIVE but not if\nthe link goes down and stays down. The fix is to generate event calls\nfrom the interrupt handler when the HW link state changes.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6a733cdc71b7aa8107caa57f2a16629aa731242a", "tree": "0e367cd9105c89e6f21baf91ba0c13dc5c6921a2", "parents": [ "192594d5230f447ef2df8de9d7902ac90d11c118" ], "author": { "name": "Michael Albaugh", "email": "Michael.Albaugh@Qlogic.com", "time": "Wed Oct 03 10:47:38 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 21:00:47 2007 -0700" }, "message": "IB/ipath: Better handling of unexpected GPIO interrupts\n\nThe General Purpose I/O pins can be configured to cause interrupts. At\nthe end of the interrupt code dealing with all known causes, a message\nis output if any bits remain un-handled. Since this is a \"can\u0027t happen\"\nscenario, it should only be triggered by bugs elsewhere. It is harmless,\nand potentially beneficial, to limit the damage by masking any such\nunexpected interrupts.\n\nThis patch adds disabling of interrupts from any pins that should\nnot have been allowed to interrupt, in addition to emitting a message.\n\nSigned-off-by: Michael Albaugh \u003cMichael.Albaugh@Qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "192594d5230f447ef2df8de9d7902ac90d11c118", "tree": "8867ed0ba446ac2a9f1c6549cff18fb7a70f0bfc", "parents": [ "aa7c79abd154ed9aba4c19b861d439ef6af35d3a" ], "author": { "name": "Michael Albaugh", "email": "Michael.Albaugh@Qlogic.com", "time": "Tue Oct 02 13:26:45 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 21:00:08 2007 -0700" }, "message": "IB/ipath: Maintain active time on all chips\n\nThere is a count of \"active hours\" maintained in EEPROM, to aid\ntroubleshooting. The definition of \"active\" is based on traffic\nexceeding a threshold in any given 5-second polling interval. As\noriginally written, the check was inadvertently bypassed for chips whose\ncounters were 64-bits wide, and only applied to chips with 32-bit wide\ncounters.\n\nThis patch moves the test for amount of traffic \"out\" to a more common\nlocation, rather than depending on a side-effect of the software\nemulation of 64-bit counts on chips whose hardware is only 32-bits wide.\n\nSigned-off-by: Michael Albaugh \u003cMichael.Albaugh@Qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "aa7c79abd154ed9aba4c19b861d439ef6af35d3a", "tree": "717bac066316893e3fb059ab97c5157dfa60794d", "parents": [ "20bed343142bfaf08765e35aaefa56dc5cc287db" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Thu Aug 16 18:10:43 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:58:49 2007 -0700" }, "message": "IB/ipath: Fix QHT7040 serial number check\n\nRemove all the OEM and bringup boards, and complain and fail\ninitialization if one is found. QHT7040 with GPIO rework (128ywwuuuu)\nis OK, older 112ywwuuuu is no longer supported). The check that had been\nadded was failing both the 112 and 128 series.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "20bed343142bfaf08765e35aaefa56dc5cc287db", "tree": "b27c69854e783b41c87a4ed3bbc8c2eecd95b900", "parents": [ "4bec0b9155d6757847b754e21b55ecafcecef839" ], "author": { "name": "Arthur Jones", "email": "arthur.jones@qlogic.com", "time": "Tue Sep 18 14:44:45 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:57:54 2007 -0700" }, "message": "IB/ipath: Indicate a couple of chip bugs to userspace\n\nA couple of chip bugs in the iba6110 and in the iba6120 are not in more\nrecent chips. This first bug swaps two of the pioavail register\nlocations. In the second bug, the chip can sometimes forget to dma the\npio avail register to memory. We indicate the presence of these bugs\nwith runtime flags and we indicate the presence of the flags by bumping\nthe SWMINOR.\n\nSigned-off-by: Arthur Jones \u003carthur.jones@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4bec0b9155d6757847b754e21b55ecafcecef839", "tree": "8df789c5b0ef303d883c25321547589c92ddb3d6", "parents": [ "70c51da2c4f84317bb13a2b564600afdcebd686f" ], "author": { "name": "Arthur Jones", "email": "arthur.jones@qlogic.com", "time": "Tue Sep 18 14:24:23 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:56:59 2007 -0700" }, "message": "IB/ipath: iba6110 rev4 no longer needs recv header overrun workaround\n\niba6110 rev3 and earlier had a chip bug where the chip could overrun the\nrecv header queue. rev4 fixed this chip bug so userspace no longer needs\nto workaround it. Now we only set the workaround flag for older chip\nversions.\n\nSigned-off-by: Arthur Jones \u003carthur.jones@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "70c51da2c4f84317bb13a2b564600afdcebd686f", "tree": "d7c39e3efaf6c90482cfe9d0e85012c3c9c511a7", "parents": [ "542869a17eee2edf389273f40f757aa4e662b3da" ], "author": { "name": "Arthur Jones", "email": "arthur.jones@qlogic.com", "time": "Fri Sep 14 12:22:49 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:56:23 2007 -0700" }, "message": "IB/ipath: Use counters in ipath_poll and cleanup interrupts in ipath_close\n\nipath_poll() suffered from a couple subtle bugs. Under the right\nconditions we could leave recv interrupts enabled on an ipath user\ncontext on close, thereby taking potentially unwanted interrupts on the\nnext open -- this is fixed by unconditionally turning off recv\ninterrupts on close. Also, we now use counters rather than set/clear\nbits which allows us to make sure we catch all interrupts at the cost of\nchanging the semantics slightly (it\u0027s now give me all events since the\nlast time I called poll() rather than give me all events since I called\n_this_ poll routine). We also added some memory barriers which may help\nensure we get all notifications in a timely manner.\n\nSigned-off-by: Arthur Jones \u003carthur.jones@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "542869a17eee2edf389273f40f757aa4e662b3da", "tree": "720a0099a4e91e325671d0bddd506a5a2b51471c", "parents": [ "15cba26f42c13ca30cbb4388f132ac0ddf4df538" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Thu Sep 13 11:42:52 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:55:06 2007 -0700" }, "message": "IB/ipath: Remove duplicate copy of LMC\n\nThe LMC value was being saved by the SMA in two places. This patch\ncleans it up so only one copy is kept.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "15cba26f42c13ca30cbb4388f132ac0ddf4df538", "tree": "e1b0d4efd1078987c7416642aeaddb694bf4aa90", "parents": [ "6cff2faaf135b602c914710f3414630c3fc1ee83" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Wed Sep 12 15:00:58 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:53:50 2007 -0700" }, "message": "IB/ipath: Add ability to set the LMC via the sysfs debugging interface\n\nThis patch adds the ability to set the LMC via a sysfs file as if the SM\nsent a SubnSet(PortInfo) MAD. It is useful for debugging when no SM is\nrunning.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6cff2faaf135b602c914710f3414630c3fc1ee83", "tree": "a63266c7640b5739a6ab948194da0c7283229f96", "parents": [ "d42b01b584b6f55f70c56f6a3dabc26f4982d30d" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Sep 07 16:54:01 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:52:23 2007 -0700" }, "message": "IB/ipath: Optimize completion queue entry insertion and polling\n\nThe code to add an entry to the completion queue stored the QPN which is\nneeded for the user level verbs view of the completion queue entry but\nthe kernel struct ib_wc contains a pointer to the QP instead of a QPN.\nWhen the kernel polled for a completion queue entry, the QPN was lookup\nup and the QP pointer recovered. This patch stores the CQE differently\nbased on whether the CQ is a kernel CQ or a user CQ thus avoiding the\nQPN to QP lookup overhead.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d42b01b584b6f55f70c56f6a3dabc26f4982d30d", "tree": "7ad73c8795a5b594b2af67af280b58e0b3644580", "parents": [ "c9cf7db2bca9180f5888eebc23dc607666a9685b" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Sat Aug 25 16:45:03 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:51:20 2007 -0700" }, "message": "IB/ipath: Implement IB_EVENT_QP_LAST_WQE_REACHED\n\nThis patch implements the IB_EVENT_QP_LAST_WQE_REACHED event which is\nneeded by ib_ipoib to destroy the QP when used in connected mode.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c9cf7db2bca9180f5888eebc23dc607666a9685b", "tree": "5503c082c930bf69d1f0b0b05d131e4e03347b23", "parents": [ "036be09ca55ee8512c05742f4f6b88911d012a90" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Sat Aug 25 16:48:29 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:50:29 2007 -0700" }, "message": "IB/ipath: Generate flush CQE when QP is in error state\n\nFollow the IB spec. (C10-96) for post send which states that a flushed \ncompletion event should be generated for work requests posted when a QP\nis in the error state.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "036be09ca55ee8512c05742f4f6b88911d012a90", "tree": "a41e06e1fe9797cf1f5b394bbfcbd67a12fd77ba", "parents": [ "d29cc6efb9731a415957b8d0ff16e31729ed6837" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Mon Aug 20 16:52:35 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:46:23 2007 -0700" }, "message": "IB/ipath: Remove redundant code\n\nThis patch removes some redundant initialization code.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d29cc6efb9731a415957b8d0ff16e31729ed6837", "tree": "aff221b626e734db9b0ad9598dfafa6af41fe54f", "parents": [ "55046698faf6357ff7c53593dbfd43a9a3f681a7" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Fri Aug 17 14:42:14 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:45:57 2007 -0700" }, "message": "IB/ipath: Future proof eeprom checksum code (contents reading)\n\nIn an earlier change, the amount of data read from the flash was\nmistakenly limited to the size known to the current driver. This causes\nproblems when the length is increased, and written with the new longer\nversion; the checksum would fail because not enough data was read.\nAlways read the full 128 byte length to prevent this.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "55046698faf6357ff7c53593dbfd43a9a3f681a7", "tree": "3635486eaad2407d924fdbdce196fb7a020441e0", "parents": [ "9ef8617af77136e743e5dd4b081a61797888a977" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Fri Aug 17 11:28:48 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:44:56 2007 -0700" }, "message": "IB/ipath: UC RDMA WRITE with IMMEDIATE doesn\u0027t send the immediate\n\nThis patch fixes a bug in the receive processing for UC RDMA WRITE with\nimmediate which caused the last packet to be dropped.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9ef8617af77136e743e5dd4b081a61797888a977", "tree": "93999550ddf98ff3c743fd27a7bd942d8cbfdf09", "parents": [ "1793b4771d258c93ed86a6356c95cac418781fdd" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Thu Aug 09 15:18:48 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:44:20 2007 -0700" }, "message": "IB/ipath: Correctly describe workaround for TID write chip bug\n\nThis is a comment change, only, correcting the comment to match the\nimplemented workaround, rather than the original workaround, and\nclarifying why it\u0027s needed.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1793b4771d258c93ed86a6356c95cac418781fdd", "tree": "4aaccda9ed17c33fd50e328b8c92e27d7fe906eb", "parents": [ "9bec3992312b8bb3aee71bd3b57d106a0a649479" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Tue Aug 07 18:09:34 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:43:17 2007 -0700" }, "message": "IB/ipath: Remove unneeded code for ipathfs\n\nThe ipathfs file system is used to export binary data verses ASCII data\nsuch as through /sys. This patch removes some unneeded files since the\ndata is available through other /sys files.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9bec3992312b8bb3aee71bd3b57d106a0a649479", "tree": "7ef51e3e5eb5cd84682e9d2742679f86b9227fbb", "parents": [ "4ee97180ac76deb5a715ac45b7d7516e6ee82ae7" ], "author": { "name": "Dave Olson", "email": "dave.olson@qlogic.com", "time": "Fri Jun 01 13:01:47 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:20:15 2007 -0700" }, "message": "IB/ipath: Verify host bus bandwidth to chip will not limit performance\n\nThere have been a number of issues where host bandwidth via HT or PCIe\nto the InfiniPath chip has been limited in some fashion (BIOS,\nconfiguration, etc.), resulting in user confusion. This check gives a\nclear warning that something is wrong and needs to be resolved.\n\nSigned-off-by: Dave Olson \u003cdave.olson@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "4ee97180ac76deb5a715ac45b7d7516e6ee82ae7", "tree": "6683d1c34d3f36271a9d8275a645ce67222ffc56", "parents": [ "210d6ca3db058cd1d6e6fd235ee3e25d6ac221cd" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Wed Jul 25 11:08:28 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:05:49 2007 -0700" }, "message": "IB/ipath: Change UD to queue work requests like RC \u0026 UC\n\nThe code to post UD sends tried to process work requests at the time\nib_post_send() is called without using a WQE queue. This was fine as\nlong as HW resources were available for sending a packet. This patch\nchanges UD to be handled more like RC and UC and shares more code.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "210d6ca3db058cd1d6e6fd235ee3e25d6ac221cd", "tree": "3fb104498d7b4144a1da0ca9d4a2241560a96ef4", "parents": [ "327a338d4fd018d33e7cacde46c0d82622b4bda8" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Tue Jul 24 13:55:39 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:04:14 2007 -0700" }, "message": "IB/ipath: Performance optimization for CPU differences\n\nDifferent processors have different ordering restrictions for write\ncombining. By taking advantage of this, we can eliminate some write\nbarriers when writing to the send buffers.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "327a338d4fd018d33e7cacde46c0d82622b4bda8", "tree": "813d439197f63ff8b9512325afd8462c7853739e", "parents": [ "76dea3bc2644e99cce1d98d0bbd3124314e5b50a" ], "author": { "name": "Arthur Jones", "email": "arthur.jones@qlogic.com", "time": "Thu Aug 02 14:46:29 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 20:02:46 2007 -0700" }, "message": "IB/ipath: iba6110 rev4 GPIO counters support\n\nOn iba6110 rev4, support for three more IB counters were added. The\nLocalLinkIntegrityError counter, the ExcessiveBufferOverrunErrors\ncounter and support for error counting of flow control packets on an\ninvalid VL. These counters trigger GPIO interrupts and the sw keeps\ntrack of the counts. Since we also use GPIO interrupts to signal packet\nreception, we need to turn off the fast interrupts, or we risk losing a\nGPIO interrupt.\n\nSigned-off-by: Arthur Jones \u003carthur.jones@qlogic.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "76dea3bc2644e99cce1d98d0bbd3124314e5b50a", "tree": "14ffcb341294948c429274922cbe6a4bcedcf5bb", "parents": [ "ede6bc04f3a07a9c93f02c92cdc281d254398321" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:18 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:18 2007 -0700" }, "message": "IB/ehca: Fix clipping of device limits to INT_MAX\n\nDoing min_t(int, foo, INT_MAX) doesn\u0027t work correctly, because if foo\nis bigger than INT_MAX, then when treated as a signed integer, it will\nbecome negative and hence such an expression is just an elaborate NOP.\n\nFix such cases in ehca to do min_t(unsigned, foo, INT_MAX) instead.\nThis fixes negative reported values for max_cqe, max_pd and max_ah:\n\nBefore:\n\n max_cqe: -64\n max_pd: -1\n max_ah: -1\n\nAfter:\n max_cqe: 2147483647\n max_pd: 2147483647\n max_ah: 2147483647\n\nBased on a bug report and fix from Anton Blanchard \u003canton@samba.org\u003e.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ede6bc04f3a07a9c93f02c92cdc281d254398321", "tree": "28dee50188606384bb90d8977224b88a255e0bf7", "parents": [ "2e61c646edfa013203e3428762f8d6a72e10bdea" ], "author": { "name": "Dotan Barak", "email": "dotanb@dev.mellanox.co.il", "time": "Sun Oct 07 09:30:48 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:18 2007 -0700" }, "message": "IPoIB/cm: Clean up initialization of QP attr in ipoib_cm_create_tx_qp()\n\nMake the way QP is being created in ipoib_cm_create_tx_qp()\nconsistent with ipoib_cm_create_rx_qp().\n\nSigned-off-by: Dotan Barak \u003cdotanb@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "76d7cc0345a037e8eea426f8abc710abd22946dd", "tree": "49936369d48985749a4beda048b6f3b5e5f31c2e", "parents": [ "dcb3f974da827c964cb8d419fbb4350cdc08a559" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "message": "IB/mthca: Use mmiowb() to avoid firmware commands getting jumbled up\n\nFirmware commands are sent to the HCA by writing multiple words to a\ncommand register block. Access to this block of registers is\nserialized with a mutex. However, on large SGI systems, problems were\nseen with multiple CPUs issuing FW commands at the same time, because\nthe writes to the register block may be reordered within the system\ninterconnect and reach the HCA in a different order than they were\nissued (even with the mutex). Fix this by adding an mmiowb() before\ndropping the mutex.\n\nTested-by: Arthur Kepner \u003cakepner@sgi.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "dcb3f974da827c964cb8d419fbb4350cdc08a559", "tree": "367f8e517387037a8b79407f3a0a673a762f7a7c", "parents": [ "de98b693e9857e183679cd2f49b3c30d2bc57629" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 01 14:47:16 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "message": "RDMA/cma: Queue IB CM MRAs to avoid unnecessary remote retries\n\nAutomatically queue MRA message to decrease the number of retries sent\nby the remote side during connection establishment. This also has the\neffect of increasing the overall connection timeout without using a\nlonger retry time in the case of dropped packets.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "de98b693e9857e183679cd2f49b3c30d2bc57629", "tree": "bfb31d9109f5057926b8849671e2937df22b37fd", "parents": [ "1a1eb6a646f52dc62e3f9ceac4aab9c27e781186" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 01 13:49:53 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "message": "IB/cm: Modify interface to send MRAs in response to duplicate messages\n\nThe IB CM provides a message received acknowledged (MRA) message that\ncan be sent to indicate that a REQ or REP message has been received, but\nwill require more time to process than the timeout specified by those\nmessages. In many cases, the application may not know how long it will\ntake to respond to a CM message, but the majority of the time, it will\nusually respond before a retry has been sent. Rather than sending an\nMRA in response to all messages just to handle the case where a longer\ntimeout is needed, it is more efficient to queue the MRA for sending in\ncase a duplicate message is received.\n\nThis avoids sending an MRA when it is not needed, but limits the number\nof times that a REQ or REP will be resent. It also provides for a\nsimpler implementation than generating the MRA based on a timer event.\n(That is, trying to send the MRA after receiving the first REQ or REP if\na response has not been generated, so that it is received at the remote\nside before a duplicate REQ or REP has been received)\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1a1eb6a646f52dc62e3f9ceac4aab9c27e781186", "tree": "b3102cea7b2a142531e89876ab453dcb651b83d6", "parents": [ "e57ac0c297b9e1074dcb6b6ee5cd1d6407a7df14" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:17 2007 -0700" }, "message": "IB/mthca: Increase max number of QPs per multicast group to 56\n\nIncrease the number of QPs allowed per multicast group from 8 to 56.\nThis allows for one QP per core on 16-core systems, which are now\nquite common, and allows some space for future growth.\n\nThis is basically the same patch that Jack Morgenstein\n\u003cjackm@dev.mellanox.co.il\u003e just supplied for mlx4.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8ad11fb6b0739e704953e2b0aed453bf7d75d4f6", "tree": "671838bc9293987efba214f2d645d7e40befc5a8", "parents": [ "d7bb58fb1c0e7264a7261c7d0304121ef9402e94" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Wed Aug 01 12:29:05 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:16 2007 -0700" }, "message": "IB/mlx4: Implement FMRs\n\nImplement FMRs for mlx4. This is an adaptation of code from mthca.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d7bb58fb1c0e7264a7261c7d0304121ef9402e94", "tree": "aefbae7ee4b4b84022a7646372ddfcdb0a99e72f", "parents": [ "121964ec38d3e17b5ea6183b3c0571df2f4b0eb6" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Wed Aug 01 12:28:53 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:16 2007 -0700" }, "message": "mlx4_core: Write MTTs from CPU instead with of WRITE_MTT FW command\n\nWrite MTT entries directly to ICM from the driver (eliminating use of\nWRITE_MTT command). This reduces the number of FW commands needed to\nregister an MR by at least a factor of 2 and speeds up memory\nregistration significantly. This code will also be used to implement\nFMRs.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "04d29b0ede242000b24cfc34cc78fbd164c47e1a", "tree": "d507a61d43b04326523a3425afcc2e76d68012a9", "parents": [ "a394f83bdfec10b09d8cb111e622556b2e6fd0de" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "message": "IB/uverbs: Make ib_uverbs_release_event_file() static\n\nib_uverbs_release_event_file() is only used in uverbs_main.c, so make it\nstatic to that file. Also move the definition before the first use, so\na forward declaration is not needed.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a394f83bdfec10b09d8cb111e622556b2e6fd0de", "tree": "bc9735ed3ccda810634173f01528741cefc71a6c", "parents": [ "2be8e3ee8efd6f99ce454115c29d09750915021a" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "message": "IB/umad: Fix bit ordering and 32-on-64 problems on big endian systems\n\nThe declaration of struct ib_user_mad_reg_req.method_mask[] exported\nto userspace was an array of __u32, but the kernel internally treated\nit as a bitmap made up of longs. This makes a difference for 64-bit\nbig-endian kernels, where numbering the bits in an array of__u32 gives:\n\n |31.....0|63....31|95....64|127...96|\n\nwhile numbering the bits in an array of longs gives:\n\n |63..............0|127............64|\n\n64-bit userspace can handle this by just treating method_mask[] as an\narray of longs, but 32-bit userspace is really stuck: the meaning of\nthe bits in method_mask[] depends on whether the kernel is 32-bit or\n64-bit, and there\u0027s no sane way for userspace to know that.\n\nFix this by updating \u003crdma/ib_user_mad.h\u003e to make it clear that\nmethod_mask[] is an array of longs, and using a compat_ioctl method to\nconvert to an array of 64-bit longs to handle the 32-on-64 problem.\nThis fixes the interface description to match existing behavior (so\nworking binaries continue to work) in almost all situations, and gives\nconsistent semantics in the case of 32-bit userspace that can run on\neither a 32-bit or 64-bit kernel, so that the same binary can work for\nboth 32-on-32 and 32-on-64 systems.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2be8e3ee8efd6f99ce454115c29d09750915021a", "tree": "00e21e061a18c7bd1339ebbb637de9701863312d", "parents": [ "c01759cee91379cc3cb551bfd7c76f1b51f91ca2" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "message": "IB/umad: Add P_Key index support\n\nAdd support for setting the P_Key index of sent MADs and getting the\nP_Key index of received MADs. This requires a change to the layout of\nthe ABI structure struct ib_user_mad_hdr, so to avoid breaking\ncompatibility, we default to the old (unchanged) ABI and add a new\nioctl IB_USER_MAD_ENABLE_PKEY that allows applications that are aware\nof the new ABI to opt into using it.\n\nWe plan on switching to the new ABI by default in a year or so, and\nthis patch adds a warning that is printed when an application uses the\nold ABI, to push people towards converting to the new ABI.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\nReviewed-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nReviewed-by: Hal Rosenstock \u003chal@xsigo.com\u003e\n" }, { "commit": "c01759cee91379cc3cb551bfd7c76f1b51f91ca2", "tree": "36f13a3b72678ba469ae6a9f4c5991ef90d7e15d", "parents": [ "a66072237500f31cec19fa688210150de9c9f957" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Fri Sep 28 17:20:05 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:15 2007 -0700" }, "message": "IB/ehca: Return srq_attr-\u003emax_sge in ehca_query_srq()\n\nTotally forgot this.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a66072237500f31cec19fa688210150de9c9f957", "tree": "23608a5e4dbfea3d6895c342a2dc2be9aa31ba96", "parents": [ "03f72a51cb1a0ba530e3308e3de84399a75b41ec" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@linux.vnet.ibm.com", "time": "Fri Sep 28 17:18:47 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:14 2007 -0700" }, "message": "IB/ehca: Adjust 64-bit alignment of create QP response for userspace\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "03f72a51cb1a0ba530e3308e3de84399a75b41ec", "tree": "a9930c35fb8d597455685af389912f462f24f6a7", "parents": [ "cd9281d873c91a01af0cb96ff0f75e9905e54403" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@linux.vnet.ibm.com", "time": "Fri Sep 28 17:16:27 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:14 2007 -0700" }, "message": "IB/ehca: Fix mem leak of firmware ctrlblock in ehca_create_srq()\n\nSigned-off-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "cd9281d873c91a01af0cb96ff0f75e9905e54403", "tree": "a639777f03924198a0c91c8884406e8ba8e40efd", "parents": [ "57cb61d587e990d556385d367589ff61f6c2c0f2" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Tue Sep 18 09:14:18 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:14 2007 -0700" }, "message": "IB/mlx4: Display misc device information under /sys/class/infiniband/\n\ndisplay the following device information under /sys/class/infiniband/mlx4_X:\nboard_id, fw_ver, hw_rev, hca_type.\n\nThis patch makes this information available to userspace utilities\nsuch as ibstat and ibv_devinfo.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "57cb61d587e990d556385d367589ff61f6c2c0f2", "tree": "f9ecc12b569d9971500a5f27f48fe7f059a19b2a", "parents": [ "9faa559c01311281f26544291322252327b65922" ], "author": { "name": "Ralph Campbell", "email": "ralph.campbell@qlogic.com", "time": "Thu Sep 20 16:33:44 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:14 2007 -0700" }, "message": "IB/core: Fix handling of multicast response failures\n\nI was looking at the code for multicast.c and noticed that\nib_sa_join_multicast() calls queue_join() which puts the\nrequest at the front of the group-\u003epending_list. If this\nis a second request, it seems like it would interfere with\nprocess_join_error() since group-\u003elast_join won\u0027t point\nto the member at the head of the pending_list. The sequence\nwould thus be:\n\n1. ib_sa_join_multicast()\n puts member1 on head of pending_list and starts work thread\n2. mcast_work_handler()\n calls send_join() which sets group-\u003elast_join to member1\n3. ib_sa_join_multicast()\n puts member2 on head of pending_list\n4. join operation for member1 receives failures response from SA.\n5. join_handler() is called with error status\n6. process_join_error() fails to process member1 since\n it doesn\u0027t match the first entry in the group-\u003epending_list.\n\nThe impact is that the failed join request is tossed. The second\nrequest is processed, and after it completes, the original request ends\nup being retried.\n\nThis change also results in join requests being processed in FIFO\norder.\n\nSigned-off-by: Ralph Campbell \u003cralph.campbell@qlogic.com\u003e\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "9faa559c01311281f26544291322252327b65922", "tree": "581afde6756cf617baffdbdf3b459a2c2e8ecbcd", "parents": [ "ea98054fefa28b646c740baffe83c2962f8228f6" ], "author": { "name": "Satyam Sharma", "email": "satyam@infradead.org", "time": "Thu Aug 23 04:58:30 2007 +0530" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:14 2007 -0700" }, "message": "IB/ehca: Misc cpuinit section annotations and #ifdef cleanups\n\n* Replace {un}register_cpu_notifier with {un}register_hotcpu_notifier\n thereby losing a couple of #ifdef HOTPLUG_CPU pairs.\n* Move comp_pool_callback_nb declaration to below that of callback\n function so that initialization of .notifier_call and .priority can\n occur at build time itself and not runtime.\n* Mark the notifier_block (and callback function, and another static\n function used by it) as __cpuinit{data} for the sake of consistency\n and remove enclosing #ifdef. (This may increase size for modular\n build of this module, however, because these are no longer dropped\n unconditionally now.)\n\nSigned-off-by: Satyam Sharma \u003csatyam@infradead.org\u003e\nAcked-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ec2a1344ad348a789b1d9d9b32cccbef33161574", "tree": "22d4cfeb3f2ec7330abf5736e91111a5b5fbb263", "parents": [ "935ef2d7a2910ca810fce57511f4d06a0cf00b5e" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:13 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:13 2007 -0700" }, "message": "IB/iser: Remove unnecessary includes\n\n\u003casm/scatterlist.h\u003e is not needed because everyplace it appears,\n\u003clinux/scatterlist.h\u003e also appears. \u003casm/io.h\u003e is not needed because\nnothing seems to be using device IO anyway.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "935ef2d7a2910ca810fce57511f4d06a0cf00b5e", "tree": "b0b3c6d6dd646e31100ca1d11fa5f18bc9014c38", "parents": [ "3a31c41901b6bd3937ec36e0e4a930849e270df6" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Wed Sep 12 05:00:25 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:13 2007 -0700" }, "message": "RDMA/cma: Use neigh_event_send() to start neighbour discovery\n\nCalling arp_send() to initiate neighbour discovery (ND) doesn\u0027t do the\nfull ND protocol. Namely, it doesn\u0027t handle retransmitting the arp\nrequest if it is dropped. The function neigh_event_send() does all\nthis. Without doing full ND, RDMA address resolution fails in the\npresence of dropped ARP broadcast packets.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nAcked-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3a31c41901b6bd3937ec36e0e4a930849e270df6", "tree": "3ec571ad11f0fe161c38ec2f94840394be7382c1", "parents": [ "c8d8beea0383e47c9d65d45f0ca95626ec435fcd" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Thu Sep 13 18:16:20 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:13 2007 -0700" }, "message": "IB/ehca: Only use MR large pages for hugetlb regions\n\n...because, on virtualized hardware like System p, we can\u0027t be sure\nthat the physical pages behind them are contiguous otherwise.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "c8d8beea0383e47c9d65d45f0ca95626ec435fcd", "tree": "00f409205683f3489344eb0cf1b86f8091354fc7", "parents": [ "247e020ee5e2a7bf46f2d7a3d4490a670a712a40" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Thu Sep 13 18:15:28 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:13 2007 -0700" }, "message": "IB/umem: Add hugetlb flag to struct ib_umem\n\nDuring ib_umem_get(), determine whether all pages from the memory\nregion are hugetlb pages and report this in the \"hugetlb\" member.\nLow-level drivers can use this information if they need it.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "247e020ee5e2a7bf46f2d7a3d4490a670a712a40", "tree": "453ce7099947d49023c20d19b7ee6885cdbfd91c", "parents": [ "7ce86409adcd2fda652b628173a66e905950ece1" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 08 15:51:18 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:12 2007 -0700" }, "message": "IB/srp: Add QoS support through service ID\n\nProvide the target service ID when performing a path record query to\nsupport optional QoS capability. QoS requires support from the SA.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "7ce86409adcd2fda652b628173a66e905950ece1", "tree": "37ec32b859f358c77267c465a6e04a642d7c5ff3", "parents": [ "a81c994d5eef87ed77cb30d8343d6be296528b3f" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 08 15:51:13 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:12 2007 -0700" }, "message": "RDMA/ucma: Allow user space to set service type\n\nExport the ability to set the type of service to user space. Model\nthe interface after setsockopt.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a81c994d5eef87ed77cb30d8343d6be296528b3f", "tree": "2d5be6dcb6283ab32cbc434b43f04df3034de5d8", "parents": [ "733d65fe33b3002a6f2694c0fd8bd760dc13141f" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 08 15:51:06 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:12 2007 -0700" }, "message": "RDMA/cma: Add ability to specify type of service\n\nProvide support to specify a type of service for a communication\nidentifier. A new function call is used when dealing with IPv4\naddresses. For IPv6 addresses, the ToS is specified through the\ntraffic class field in the sockaddr_in6 structure.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\n\n[ The comments Eitan Zahavi and myself have made over the v1 post at \n \u003chttp://lists.openfabrics.org/pipermail/general/2007-August/039247.html\u003e\n were fully addressed. ]\n \nReviewed-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e \nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "733d65fe33b3002a6f2694c0fd8bd760dc13141f", "tree": "cf9460740b86c49c704497d8e911488863e72aec", "parents": [ "81668838c4583b19276b16382e0c61e21ef5adf0" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Wed Aug 08 15:41:28 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:12 2007 -0700" }, "message": "IB/sa: Add new QoS fields to path record\n\nThe QoS annex defines new fields for path records. Add them to the\nib_sa for consumers that want to use them.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\nReviewed-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "81668838c4583b19276b16382e0c61e21ef5adf0", "tree": "eae4f3250ea914332ed4e46ddbd267b858bedd4c", "parents": [ "08c283ac262d7ab21c5733ff469ff88985381ca9" ], "author": { "name": "Sean Hefty", "email": "sean.hefty@intel.com", "time": "Thu Aug 02 12:21:31 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:11 2007 -0700" }, "message": "IPoIB: Specify Traffic Class with path record queries for QoS support\n\nTo support QoS within and between subnets, modify IPoIB to request\nspecific Traffic Class values with path record queries, using\nthe value associated with the IPoIB broadcast group.\n\nSigned-off-by: Sean Hefty \u003csean.hefty@intel.com\u003e\n\n[ See some comments I made on this at v1 and v2 of the posts\n \u003chttp://lists.openfabrics.org/pipermail/general/2007-August/039275.html\u003e\n \u003chttp://lists.openfabrics.org/pipermail/general/2007-September/040312.html\u003e ]\n\nReviewed-by: Or Gerlitz \u003cogerlitz@voltaire.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "08c283ac262d7ab21c5733ff469ff88985381ca9", "tree": "e2463e8ef65706bdb1f66d4029a8b8325ab27b40", "parents": [ "39089e77741a53874eb8a29e4516bbafcc29298a" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@de.ibm.com", "time": "Thu Sep 13 18:14:58 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:11 2007 -0700" }, "message": "IB/ehca: Fix large page HW cap defines\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "39089e77741a53874eb8a29e4516bbafcc29298a", "tree": "0dfe69303702c20eff8ada363a4c9da0313d561b", "parents": [ "5110e4de4995db1c28457b08ed1e291f9b38f2e7" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:35:32 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:11 2007 -0700" }, "message": "IB/ehca: Bump version number and change its format\n\nNobody needed the SVNEHCA_ prefix anyway.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5110e4de4995db1c28457b08ed1e291f9b38f2e7", "tree": "cbbf59d405c9383e44322bc6e231f60f44311e96", "parents": [ "0b5de96858e516311f2d3ca45073c2afd2eb5d94" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Wed Sep 12 16:44:11 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:11 2007 -0700" }, "message": "IB/ehca: Replace get_paca()-\u003epaca_index by the more portable raw_smp_processor_id()\n\nWe can use raw_smp_processor_id() here because the processor ID is\nonly used for debug output and therefore our use is preemption-unsafe.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "0b5de96858e516311f2d3ca45073c2afd2eb5d94", "tree": "83fbe9083a68e5a1456a44111dd823b49c71e69a", "parents": [ "e90d0b3daede2bae2e78f8bf88c19182961cd19d" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:34:35 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:11 2007 -0700" }, "message": "IB/ehca: Serialize MR alloc and MR free hvCalls\n\nSome firmware levels exhibit a race condition between H_ALLOC_RESOURCE(MR)\nand H_FREE_RESOURCE(MR). Work around this problem by locking these hvCalls\nagainst each other.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e90d0b3daede2bae2e78f8bf88c19182961cd19d", "tree": "a04090695e126a4f7b12743084679eebb7abf83e", "parents": [ "b708fba3c2942a175c3cb04a7bb4c89f907b497b" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:34:04 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:10 2007 -0700" }, "message": "IB/ehca: Path migration support\n\nFix some modify_qp() issues related to path migration.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b708fba3c2942a175c3cb04a7bb4c89f907b497b", "tree": "8e87c303b56f691199debfa690de0c3870251047", "parents": [ "86dce445e01a50339f8f86c466c64a863e5fd18a" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:33:40 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:10 2007 -0700" }, "message": "IB/ehca: Add check for max #SGE to create_qp()\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "86dce445e01a50339f8f86c466c64a863e5fd18a", "tree": "41ceaab2cd5b6a0addf9af41d2c3115aede5696a", "parents": [ "e37221928bf685d63ba5319746eafe463d61e330" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:32:50 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:10 2007 -0700" }, "message": "IB/ehca: ehca_gen_warn() should always print\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e37221928bf685d63ba5319746eafe463d61e330", "tree": "ee74c9b24f41ab930579d51cd80b2ff4d173da09", "parents": [ "2863ad4bddf366790a733cfd71f2f480afdf36fc" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:32:22 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:10 2007 -0700" }, "message": "IB/ehca: Print return codes as signed decimal integers\n\n...because -12 is easier to read than FFFFFFF4.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2863ad4bddf366790a733cfd71f2f480afdf36fc", "tree": "48b0d1b705d43a22069554d1a520599c9db7a1fb", "parents": [ "e390d3b52f791fcea26312ba4982cda82052727b" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Tue Sep 11 15:31:49 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:10 2007 -0700" }, "message": "IB/ehca: Refactor hvcall tracing\n\nChange hvcall trace output towards better readability: reg numbers\ninstead of argument numbers, return code as signed decimal instead of\nunsigned hex.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e390d3b52f791fcea26312ba4982cda82052727b", "tree": "399af315b321e57c88cca886cd3a69c05360891b", "parents": [ "5281a4b8a0c6bac0c070913ec25868faa06a3115" ], "author": { "name": "Hoang-Nam Nguyen", "email": "hnguyen@de.ibm.com", "time": "Tue Sep 11 15:31:06 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:09 2007 -0700" }, "message": "IB/ehca: Use remap_4k_pfn() to map firmware contexts to user space\n\nUse Paul\u0027s new remap_4k_pfn() function to map our 4K firmware contexts\ninto user space on 64K-page machines without exposing neighboring\nfirmware contexts. Return the context\u0027s offset within a 64K page to\nuser space so it can determine the proper virtual address.\n\nFor details about remap_4k_pfn(), see commit 721151d0 or\nhttp://patchwork.ozlabs.org/linuxppc/patch?id\u003d10281\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5281a4b8a0c6bac0c070913ec25868faa06a3115", "tree": "49e379688fca00a0a3eb4f031aab90929f784f34", "parents": [ "441633b968a5be0ef9be7c37ae24c35eda5b730d" ], "author": { "name": "Stefan Roscher", "email": "stefan.roscher@de.ibm.com", "time": "Tue Sep 11 15:29:39 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:08 2007 -0700" }, "message": "IB/ehca: Support more than 4k QPs for userspace and kernelspace\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "441633b968a5be0ef9be7c37ae24c35eda5b730d", "tree": "292b4799419ae19eda2b3d0c4c81af05cfba9342", "parents": [ "a855b1a7423ac83c76638f156d79c854b0feb94d" ], "author": { "name": "Stefan Roscher", "email": "stefan.roscher@de.ibm.com", "time": "Tue Sep 11 15:26:33 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:07 2007 -0700" }, "message": "IB/ehca: Small QP userspace support\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "a855b1a7423ac83c76638f156d79c854b0feb94d", "tree": "548ea2d78e8850dadffd7cc664045ac5adb7b8d5", "parents": [ "3c10c7c929b30e8813d15960cb97f95a0b2ba615" ], "author": { "name": "Peter Oruba", "email": "peter.oruba@amd.com", "time": "Fri Aug 10 13:54:33 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:07 2007 -0700" }, "message": "IB/mthca: Use PCI-X/PCI-Express read control interfaces\n\nThese driver changes incorporate the proposed PCI-X / PCI-Express read\nbyte count interface. Reading and setting those values doesn\u0027t take\nplace \"manually\", instead wrapping functions are called to allow\nquirks for some PCI bridges.\n\nSigned-off by: Peter Oruba \u003cpeter.oruba@amd.com\u003e\nBased on work by Stephen Hemminger \u003cshemminger@linux-foundation.org\u003e\nCc: Roland Dreier \u003crolandd@cisco.com\u003e\nCc: Michael S. Tsirkin \u003cmst@mellanox.co.il\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "3c10c7c929b30e8813d15960cb97f95a0b2ba615", "tree": "afa80536856b47a173c0e850c6acb02baf4e75e7", "parents": [ "339e2640a9f403f7b7acb2ea67f3568b8ac3eebf" ], "author": { "name": "Ali Ayoub", "email": "ali@mellanox.co.il", "time": "Sun Sep 09 14:55:11 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:07 2007 -0700" }, "message": "IB/sa: Error handling thinko fix\n\nib_create_send_mad() returns an error code pointer on error, not NULL.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "339e2640a9f403f7b7acb2ea67f3568b8ac3eebf", "tree": "84ddd9c4c6964a730ce115c885fe6b7f49b9d8ed", "parents": [ "1f794483025bc1e9e8e3321089a7f29ad3cc448f" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Aug 29 12:43:01 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IB/ehca: Export module parameters in sysfs\n\nAt the moment the ehca module parameters are not exported in sysfs.\nExport them with 0444 permissions.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nAcked-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1f794483025bc1e9e8e3321089a7f29ad3cc448f", "tree": "17653e1be8354f435275d5504e9f066fd42f1f4b", "parents": [ "d7dc3ccbe4dd1d37950da0138079e61d5be06ca9" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Aug 29 11:05:35 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IB/ehca: Make output clearer by removing some debug messages\n\nehca spits out a lot of debugging information. I had to look closely to\nsee the \"Port 1 is not active\" message within all the debug:\n\neHCA Infiniband Device Driver (Rel.: SVNEHCA_0022)\neHCA scaling code enabled\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_define_sqp Port 1 is not active.\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_create_qp ehca_define_sqp() failed rc\u003dffffffffffffffff\nib_mad: Couldn\u0027t create ib_mad QP1\nib_mad: Couldn\u0027t open ehca0 port 1\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_alloc_fmr unsupported fmr_attr-\u003epage_shift\u003d9\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_alloc_fmr rc\u003dffffffffffffffea pd\u003dc000000b4b5b2420 mr_access_flags\u003d7 fmr_attr\u003dc0000005afd37394\nfmr_create failed for FMR 0\n\nRemove a few debug statements so that things are clearer:\n\neHCA Infiniband Device Driver (Rel.: SVNEHCA_0022)\neHCA scaling code enabled\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_define_sqp Port 1 is not active.\nib_mad: Couldn\u0027t create ib_mad QP1\nib_mad: Couldn\u0027t open ehca0 port 1\nehca D.001.DQDXYCB-P1-C9: PU0006 EHCA_ERR:ehca_alloc_fmr unsupported fmr_attr-\u003epage_shift\u003d9\nfmr_create failed for FMR 0\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nAcked-by: Hoang-Nam Nguyen \u003chnguyen@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "d7dc3ccbe4dd1d37950da0138079e61d5be06ca9", "tree": "6391db1b61ae23c34c351b617803d31dc9255804", "parents": [ "ca6de177acef8f2c7c3901ea583a263364ca7bbb" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IB/mlx4: Fix up SRQ limit_watermark endianness\n\nmlx4_srq_query() returns a big-endian 16-bit value through an int *,\nwhich screws up sparse checking. Fix this so that a CPU-endian value\nis returned.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ca6de177acef8f2c7c3901ea583a263364ca7bbb", "tree": "04910ca60735e6ef99916f7b91d83fc468852af6", "parents": [ "b3ac60fc243f2312d27ecded058ef96f52f25fe0" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Tue Aug 21 18:46:10 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IPoIB: Fix error path memory leak\n\nClean up properly if ib_query_pkey() or ib_query_gid() fail.\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "b3ac60fc243f2312d27ecded058ef96f52f25fe0", "tree": "6e4f2e3f543fc4ed7356b8e8c7d1540597a45cf1", "parents": [ "017aadc4b505ad3ec2acc4e6ba96d63ae1c997a5" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.co.il", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IPoIB: Fix typo to end statement with \u0027;\u0027 instead of \u0027,\u0027\n\nSigned-off-by: Eli Cohen \u003celi@mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "017aadc4b505ad3ec2acc4e6ba96d63ae1c997a5", "tree": "40ba779d2a173bcbc9287622d214afd3f7c984d5", "parents": [ "08fb105540f4649eaa25270ec3ada1d35406afcc" ], "author": { "name": "Michael S. Tsirkin", "email": "mst@dev.mellanox.co.il", "time": "Tue Aug 07 16:10:34 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:06 2007 -0700" }, "message": "IB/mthca: Enable MSI-X by default\n\nRecover from MSI-X errors by automatically falling back on regular\ninterrupt, instead of asking the user to do this manually. This makes\nit possible to enable MSI-X by default, and will make it possible to\nget rid of the msi_x module option in the future.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "8a68bbe31d51cc75d2acf1c9f5ceff91f7662ea4", "tree": "2238a4f1cc3dba0b61e5bec2c4fca20ba27317c6", "parents": [ "1fea391039d1c4e876a164099bff475a02a29d96" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Aug 29 08:36:22 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:05 2007 -0700" }, "message": "IB/fmr_pool: Clean up some error messages in fmr_pool.c\n\nA number of printks in fmr_pool.c dont have newlines, eg:\n\n fmr_create failed for FMR 0\u003c5\u003eFS-Cache: Loaded\n\nFix them up.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "1fea391039d1c4e876a164099bff475a02a29d96", "tree": "5f5201806ecb8c2919b5c3c26fe22bd42f4cb22a", "parents": [ "2242fa4f04d670f02efb43ec537d677edc220880" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:05 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:05 2007 -0700" }, "message": "IB/ehca: Include \u003clinux/mutex.h\u003e from ehca_classes.h\n\nehca_classes.h uses struct mutex, so while \u003clinux/mutex.h\u003e seems to be\npulled in indirectly by one of the headers it includes, the right\nthing is to include \u003clinux/mutex.h\u003e directly.\n\nSigned-off-by: Michael S. Tsirkin \u003cmst@dev.mellanox.co.il\u003e\nAcked-by: Stefan Roscher \u003cstefan.roscher@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "2242fa4f04d670f02efb43ec537d677edc220880", "tree": "a3fc501a86dbbb752201b97cf963703bd01d0aa4", "parents": [ "eaf559bf566f76887533c077d425adce847f06c8" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:05 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:05 2007 -0700" }, "message": "IB/mlx4: Use __set_data_seg() in mlx4_ib_post_recv()\n\nUse a __set_data_seg() helper in mlx4_ib_post_recv() too; in addition\nto making the code easier to read, this also allows gcc to generate\nbetter code -- on x86_64:\n\nadd/remove: 0/0 grow/shrink: 0/1 up/down: 0/-8 (-8)\nfunction old new delta\nmlx4_ib_post_recv 359 351 -8\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "65d470b3ea52ee1402499d6fcb4632296452e5b1", "tree": "9c1e0a5d9c959b2b92d59c9e309e2ae68cb5c191", "parents": [ "ce423ef50ee1b6b7db63c748034423aa0afce224" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:04 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:04 2007 -0700" }, "message": "IB: find_first_zero_bit() takes unsigned pointer\n\nFix sparse warning\n\n drivers/infiniband/core/device.c:142:6: warning: incorrect type in argument 1 (different signedness)\n drivers/infiniband/core/device.c:142:6: expected unsigned long const *addr\n drivers/infiniband/core/device.c:142:6: got long *[assigned] inuse\n\nby making the local variable inuse unsigned. Does not affect generated\ncode at all.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "ce423ef50ee1b6b7db63c748034423aa0afce224", "tree": "9504d23398a8c03ca99e4a2fecdaee45b540a077", "parents": [ "e54664c0958acf14ef3a65d1b78f4a54b437cdf7" ], "author": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:04 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:04 2007 -0700" }, "message": "IPoIB: Make sure no receives are handled when stopping device\n\nThe current IPoIB code might process receive completions from\nipoib_drain_cq() when bringing down the interface. This could cause\npackets to be passed up the stack without the device\u0027s poll method\nbeing called. Avoid this by setting the status of any successful\ncompletions to IB_WC_WR_FLUSH_ERR.\n\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "e54664c0958acf14ef3a65d1b78f4a54b437cdf7", "tree": "6f99569d872294ac208dff7da83e097b05ecc25f", "parents": [ "bbf25010f1a6b761914430f5fca081ec8c7accd1" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Sun Jul 29 15:12:26 2007 -0500" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Oct 09 19:59:04 2007 -0700" }, "message": "RDMA/cxgb3: Make the iw_cxgb3 module parameters writable\n\nAllow changing parameter values without having to reload the module.\nThis is safe because these parameters are only looked at when a new\nconnection is established.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6e694ea33e7a7fad908d188c46f441f04fb633d4", "tree": "82325240b222299766b5cfa53e59306cb607878a", "parents": [ "40ffbfad6bb79a99cc7627bdaca0ee22dec526f6" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Wed Sep 19 09:52:25 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Sun Sep 23 13:03:22 2007 -0700" }, "message": "IB/mlx4: Fix data corruption triggered by wrong headroom marking order\n\nThis is an addendum to commit 0e6e7416 (\"IB/mlx4: Handle new FW\nrequirement for send request prefetching\"). We also need to handle\nprefetch marking properly for S/G segments, or else the HCA may end up\nprocessing S/G segments that are not fully written and end up sending\nthe wrong data. This can actually cause data corruption in practice,\nespecially on systems with relatively slow CPUs (where the HCA is more\nlikely to prefetch while the CPU is in the middle of writing a work\nrequest into memory).\n\nWe write S/G segments in reverse order into the WQE, in order to\nguarantee that the first dword of all cachelines containing S/G\nsegments is written last (overwriting the headroom invalidation\npattern). The entire cacheline will thus contain valid data when the\ninvalidation pattern is overwritten.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6db602d447fb6c3aeb020c5dff5219de317f8bb4", "tree": "379cfdff0c1f6eb66edfbc6a50dfde281602ca13", "parents": [ "e89a5a43b95cdc4305b7c8e8121a380f02476636", "5ff70cac3e98af64f9a1eaec9e762ff4927c26d1" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Aug 31 20:40:37 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Fri Aug 31 20:40:37 2007 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:\n IB/ehca: SRQ fixes to enable IPoIB CM\n IB/ehca: Fix Small QP regressions\n" }, { "commit": "5ff70cac3e98af64f9a1eaec9e762ff4927c26d1", "tree": "4a0b840f6cef113b175bfe63c75c0420611fca82", "parents": [ "fecea0ab3415bfab9a1964690e53b10c5d8f2e46" ], "author": { "name": "Joachim Fenkes", "email": "fenkes@de.ibm.com", "time": "Fri Aug 31 16:03:37 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Aug 31 13:58:04 2007 -0700" }, "message": "IB/ehca: SRQ fixes to enable IPoIB CM\n\nFix ehca SRQ support so that IPoIB connected mode works:\n\n - Report max_srq \u003e 0 if SRQ is supported\n - Report \"last wqe reached\" asynchronous event when base QP dies;\n this is required by the IB spec and IPoIB CM relies on receiving it\n when cleaning up.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "fecea0ab3415bfab9a1964690e53b10c5d8f2e46", "tree": "45c8a33154fad54c00dd86d9b5b989d6b3bd04d1", "parents": [ "18115f45374d19ada218fc013aa5308baf5d283e" ], "author": { "name": "Stefan Roscher", "email": "stefan.roscher@de.ibm.com", "time": "Fri Aug 31 16:02:59 2007 +0200" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Aug 31 13:56:42 2007 -0700" }, "message": "IB/ehca: Fix Small QP regressions\n\nThe new Small QP code had a few bugs that would also make it trigger\nfor non-Small QPs. Fix them.\n\nSigned-off-by: Joachim Fenkes \u003cfenkes@de.ibm.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "5fbf816fe7d72bfdbf22bfec05b4ec3aa6849f72", "tree": "62f35df3e9078e3f981b9c04176d9a9cb557a53a", "parents": [ "54d0f56e63edb1aa3901ea6389bf7ea5d43d0c22" ], "author": { "name": "Divy Le Ray", "email": "divy@chelsio.com", "time": "Wed Aug 29 19:15:47 2007 -0700" }, "committer": { "name": "Jeff Garzik", "email": "jeff@garzik.org", "time": "Fri Aug 31 07:29:08 2007 -0400" }, "message": "cxgb3 - Fix dev-\u003epriv usage\n\ncxgb3 used netdev_priv() and dev-\u003epriv for different purposes.\nIn 2.6.23, netdev_priv() \u003d\u003d dev-\u003epriv, cxgb3 needs a fix.\nThis patch is a partial backport of Dave Miller\u0027s changes in the\nnet-2.6.24 git branch.\n\nWithout this fix, cxgb3 crashes on 2.6.23.\n\nSigned-off-by: Divy Le Ray \u003cdivy@chelsio.com\u003e\nSigned-off-by: Jeff Garzik \u003cjeff@garzik.org\u003e\n" }, { "commit": "fe11cb6ba40afff15efb053fd0bcba45274636e0", "tree": "0de475e13fde7954d2136ad117a13ec4bb9450cd", "parents": [ "947b2a8083a03e6fff448ce8928956015614855e" ], "author": { "name": "Ilpo Järvinen", "email": "ilpo.jarvinen@helsinki.fi", "time": "Thu Aug 16 01:02:07 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Wed Aug 15 20:24:06 2007 -0700" }, "message": "IB/mlx4: Incorrect semicolon after if statement\n\nA stray semicolon makes us inadvertently ignore the value of err.\n\nSigned-off-by: Ilpo Järvinen \u003cilpo.jarvinen@helsinki.fi\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "6958e827f187c9c5cd39af075567f74f02bf3dd1", "tree": "705a1c3c2f1d3719e2b8bfbed80a0d4d8f7c11b4", "parents": [ "198919151dea65d83dd0fb66979b1df28402f2b0" ], "author": { "name": "Jack Morgenstein", "email": "jackm@dev.mellanox.co.il", "time": "Mon Aug 06 17:09:09 2007 +0300" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Tue Aug 07 12:40:56 2007 -0700" }, "message": "IPoIB: Fix leak in ipoib_transport_dev_init() error path\n\nipoib_transport_dev_init() calls ipoib_cm_dev_init(), so it needs to\ncall ipoib_cm_dev_cleanup() to unwind that on the error path.\n\nFound by Dotan Barak of Mellanox.\n\nSigned-off-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" }, { "commit": "198919151dea65d83dd0fb66979b1df28402f2b0", "tree": "e12875f99e99ddb41352332c09ae4e58a7187287", "parents": [ "3d1ff48da760968793f3c36672961ffd23088787" ], "author": { "name": "Vu Pham", "email": "vu@mellanox.com", "time": "Fri Aug 03 14:25:48 2007 -0700" }, "committer": { "name": "Roland Dreier", "email": "rolandd@cisco.com", "time": "Fri Aug 03 14:29:06 2007 -0700" }, "message": "IB/mlx4: Fix opcode returned in RDMA read completion\n\nCurrent code has a cut-and-paste error and returns IB_WC_SEND when it\nshould return IB_WC_RDMA_READ.\n\nSigned-off-by: Vu Pham \u003cvu@mellanox.com\u003e\nSigned-off-by: Roland Dreier \u003crolandd@cisco.com\u003e\n" } ], "next": "3d1ff48da760968793f3c36672961ffd23088787" }