summaryrefslogtreecommitdiff
path: root/drivers/infiniband
AgeCommit message (Collapse)AuthorFilesLines
2010-05-27ehca: convert cpu notifier to return encapsulate errno valueAkinobu Mita1-1/+1
By the previous modification, the cpu notifier can return encapsulate errno value. This converts the cpu notifiers for ehca. Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Hoang-Nam Nguyen <hnguyen@de.ibm.com> Cc: Christoph Raisch <raisch@de.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-05-25Merge branch 'for-linus' of ↵Linus Torvalds77-5126/+51061
git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: RDMA/nes: Fix incorrect unlock in nes_process_mac_intr() RDMA/nes: Async event for closed QP causes crash RDMA/nes: Have ethtool read hardware registers for rx/tx stats RDMA/cxgb4: Only insert sq qid in lookup table RDMA/cxgb4: Support IB_WR_READ_WITH_INV opcode RDMA/cxgb4: Set fence flag for inv-local-stag work requests RDMA/cxgb4: Update some HW limits RDMA/cxgb4: Don't limit fastreg page list depth RDMA/cxgb4: Return proper errors in fastreg mr/pbl allocation RDMA/cxgb4: Fix overflow bug in CQ arm RDMA/cxgb4: Optimize CQ overflow detection RDMA/cxgb4: CQ size must be IQ size - 2 RDMA/cxgb4: Register RDMA provider based on LLD state_change events RDMA/cxgb4: Detach from the LLD after unregistering RDMA device IB/ipath: Remove support for QLogic PCIe QLE devices IB/qib: Add new qib driver for QLogic PCIe InfiniBand adapters IB/mad: Make needlessly global mad_sendq_size/mad_recvq_size static IB/core: Allow device-specific per-port sysfs files mlx4_core: Clean up mlx4_alloc_icm() a bit mlx4_core: Fix possible chunk sg list overflow in mlx4_alloc_icm()
2010-05-25Merge branches 'cxgb4', 'misc', 'mlx4', 'nes' and 'qib' into for-nextRoland Dreier71-5051/+50957
2010-05-25RDMA/nes: Fix incorrect unlock in nes_process_mac_intr()Chien Tung1-1/+2
Commit ce6e74f2 ("RDMA/nes: Make nesadapter->phy_lock usage consistent") introduced a problem where phy_lock was only unlocked within an if statement and so nes_process_mac_intr() could return with phy_lock still held. Fix this. This was discovered because of the sparse warning: drivers/infiniband/hw/nes/nes_hw.c:2643:9: warning: context imbalance in 'nes_process_mac_intr' - different lock contexts for basic block Reported-by: Roland Dreier <rdreier@cisco.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/nes: Async event for closed QP causes crashFaisal Latif1-0/+9
Under abnormal termination, modify_qp() closes the QP, and async event (AE) handling also attempts to close the same QP, causing a crash. Fix this by checking the state of the QP before processing the AE. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/nes: Have ethtool read hardware registers for rx/tx statsFaisal Latif1-17/+55
Enhance ethtool to read hardware registers for rcv/tx error stats. Also add support for free pbl resources. Remove cq depth stats, which are not used. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Only insert sq qid in lookup tableSteve Wise1-16/+9
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Support IB_WR_READ_WITH_INV opcodeSteve Wise1-1/+5
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Set fence flag for inv-local-stag work requestsSteve Wise1-0/+2
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Update some HW limitsSteve Wise1-8/+9
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Don't limit fastreg page list depthSteve Wise1-3/+0
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Return proper errors in fastreg mr/pbl allocationSteve Wise1-2/+6
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Fix overflow bug in CQ armSteve Wise1-19/+12
- wrap cq->cqidx_inc based on cq size. - optimize t4_arm_cq logic. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Optimize CQ overflow detectionSteve Wise2-12/+17
1) save the timestamp flit in the cq when we consume a CQE. 2) always compare the saved flit with the previous entry flit when reading the next CQE entry. If the flits don't compare, then we have overflowed. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: CQ size must be IQ size - 2Steve Wise1-1/+4
We need 1 extra entry for the status page and 1 to always have 1 free entry to detect when the queue is full. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Register RDMA provider based on LLD state_change eventsSteve Wise3-10/+39
The LLD now supports proper UP state change events, so move the RDMA provider registration to UP path. This fixes a crash when loading iw_cxgb4 _after_ the NFS/RDMA transport is up and running. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-25RDMA/cxgb4: Detach from the LLD after unregistering RDMA deviceSteve Wise1-3/+1
In the RDMA core unregister path, kernel users will be calling down into the T4 provider to release resources. So we cannot detach from the LLD until this process completes. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-24IB/ipath: Remove support for QLogic PCIe QLE devicesRalph Campbell9-7131/+7
The ib_qib driver is taking over support for QLogic PCIe QLE devices, so remove support for them from ib_ipath. The ib_ipath driver now supports only the obsolete QLogic Hyper-Transport IB host channel adapter (model QHT7140). Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-24IB/qib: Add new qib driver for QLogic PCIe InfiniBand adaptersRalph Campbell51-0/+52965
Add a low-level IB driver for QLogic PCIe adapters. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-24IB/mad: Make needlessly global mad_sendq_size/mad_recvq_size staticRoland Dreier1-2/+2
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-22Merge remote branch 'origin' into secretlab/next-devicetreeGrant Likely51-267/+9820
Merging in current state of Linus' tree to deal with merge conflicts and build failures in vio.c after merge. Conflicts: drivers/i2c/busses/i2c-cpm.c drivers/i2c/busses/i2c-mpc.c drivers/net/gianfar.c Also fixed up one line in arch/powerpc/kernel/vio.c to use the correct node pointer. Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2010-05-22of: Remove duplicate fields from of_platform_driverGrant Likely1-3/+4
.name, .match_table and .owner are duplicated in both of_platform_driver and device_driver. This patch is a removes the extra copies from struct of_platform_driver and converts all users to the device_driver members. This patch is a pretty mechanical change. The usage model doesn't change and if any drivers have been missed, or if anything has been fixed up incorrectly, then it will fail with a compile time error, and the fixup will be trivial. This patch looks big and scary because it touches so many files, but it should be pretty safe. Signed-off-by: Grant Likely <grant.likely@secretlab.ca> Acked-by: Sean MacLennan <smaclennan@pikatech.com>
2010-05-21IB/core: Allow device-specific per-port sysfs filesRalph Campbell11-15/+32
Add a new parameter to ib_register_device() so that low-level device drivers can pass in a pointer to a callback function that will be called for each port that is registered in sysfs. This allows low-level device drivers to create files in /sys/class/infiniband/<hca>/ports/<N>/ without having to poke through the internals of the RDMA sysfs handling. There is no need for an unregister function since the kobject reference will go to zero when ib_unregister_device() is called. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-21Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6Linus Torvalds2-14/+8
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1674 commits) qlcnic: adding co maintainer ixgbe: add support for active DA cables ixgbe: dcb, do not tag tc_prio_control frames ixgbe: fix ixgbe_tx_is_paused logic ixgbe: always enable vlan strip/insert when DCB is enabled ixgbe: remove some redundant code in setting FCoE FIP filter ixgbe: fix wrong offset to fc_frame_header in ixgbe_fcoe_ddp ixgbe: fix header len when unsplit packet overflows to data buffer ipv6: Never schedule DAD timer on dead address ipv6: Use POSTDAD state ipv6: Use state_lock to protect ifa state ipv6: Replace inet6_ifaddr->dead with state cxgb4: notify upper drivers if the device is already up when they load cxgb4: keep interrupts available when the ports are brought down cxgb4: fix initial addition of MAC address cnic: Return SPQ credit to bnx2x after ring setup and shutdown. cnic: Convert cnic_local_flags to atomic ops. can: Fix SJA1000 command register writes on SMP systems bridge: fix build for CONFIG_SYSFS disabled ARCNET: Limit com20020 PCI ID matches for SOHARD cards ... Fix up various conflicts with pcmcia tree drivers/net/ {pcmcia/3c589_cs.c, wireless/orinoco/orinoco_cs.c and wireless/orinoco/spectrum_cs.c} and feature removal (Documentation/feature-removal-schedule.txt). Also fix a non-content conflict due to pm_qos_requirement getting renamed in the PM tree (now pm_qos_request) in net/mac80211/scan.c
2010-05-20Merge branch 'for-linus' of ↵Linus Torvalds3-4/+4
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (44 commits) vlynq: make whole Kconfig-menu dependant on architecture add descriptive comment for TIF_MEMDIE task flag declaration. EEPROM: max6875: Header file cleanup EEPROM: 93cx6: Header file cleanup EEPROM: Header file cleanup agp: use NULL instead of 0 when pointer is needed rtc-v3020: make bitfield unsigned PCI: make bitfield unsigned jbd2: use NULL instead of 0 when pointer is needed cciss: fix shadows sparse warning doc: inode uses a mutex instead of a semaphore. uml: i386: Avoid redefinition of NR_syscalls fix "seperate" typos in comments cocbalt_lcdfb: correct sections doc: Change urls for sparse Powerpc: wii: Fix typo in comment i2o: cleanup some exit paths Documentation/: it's -> its where appropriate UML: Fix compiler warning due to missing task_struct declaration UML: add kernel.h include to signal.c ...
2010-05-19of: Always use 'struct device.of_node' to get device node pointer.Grant Likely1-5/+6
The following structure elements duplicate the information in 'struct device.of_node' and so are being eliminated. This patch makes all readers of these elements use device.of_node instead. (struct of_device *)->node (struct dev_archdata *)->prom_node (sparc) (struct dev_archdata *)->of_node (powerpc & microblaze) Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2010-05-16Merge branches 'amso1100', 'bkl', 'cma', 'cxgb3', 'cxgb4', 'ipoib', 'iser', ↵Roland Dreier41-236/+9795
'masked-atomics', 'misc', 'mthca' and 'nes' into for-next
2010-05-16IB/core: Use kmemdup() instead of kmalloc()+memcpy()Julia Lawall2-5/+3
Use kmemdup when some other buffer is immediately copied into the allocated region. A simplified version of the semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression from,to,size,flag; statement S; @@ - to = \(kmalloc\|kzalloc\)(size,flag); + to = kmemdup(from,size,flag); if (to==NULL || ...) S - memcpy(to, from, size); // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-12IB/iser: Fix error flow in iser_create_ib_conn_res()Dan Carpenter1-16/+9
We shouldn't free things here because we free them later. The call tree looks like this: iser_connect() ==> initiating the connection establishment and later iser_cma_handler() => iser_route_handler() => iser_create_ib_conn_res() if we fail here, eventually iser_conn_release() is called, resulting in a double free. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-12IB/iser: Enhance disconnection logic for multi-pathingOr Gerlitz3-38/+46
The iser connection teardown flow isn't over until the underlying Connection Manager (e.g the IB CM) delivers a disconnected or timeout event through the RDMA-CM. When the remote (target) side isn't reachable, e.g when some HW e.g port/hca/switch isn't functioning or taken down administratively, the CM timeout flow is used and the event may be generated only after relatively long time -- on the order of tens of seconds. The current iser code exposes this possibly long delay to higher layers, specifically to the iscsid daemon and iscsi kernel stack. As a result, the iscsi stack doesn't respond well: this low-level CM delay is added to the fail-over time under HA schemes such as the one provided by DM multipath through the multipathd(8) service. This patch enhances the reference counting scheme on iser's IB connections so that the disconnect flow initiated by iscsid from user space (ep_disconnect) doesn't wait for the CM to deliver the disconnect/timeout event. (The connection teardown isn't done from iser's view point until the event is delivered) The iser ib (rdma) connection object is destroyed when its reference count reaches zero. When this happens on the RDMA-CM callback context, extra care is taken so that the RDMA-CM does the actual destroying of the associated ID, since doing it in the callback is prohibited. The reference count of iser ib connection normally reaches three, where the <ref, deref> relations are 1. conn <init, terminate> 2. conn <bind, stop/destroy> 3. cma id <create, disconnect/error/timeout callbacks> With this patch, multipath fail-over time is about 30 seconds, while without this patch, multipath fail-over time is about 130 seconds. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-12IB/iser: Remove buggy back-pointer settingOr Gerlitz1-2/+0
The iscsi connection object life cycle includes binding and unbinding (conn_stop) to/from the iscsi transport connection object. Since iscsi connection objects are recycled, at the time the transport connection (e.g iser's IB connection) is released, it is not valid to touch the iscsi connection tied to the transport back-pointer since it may already point to a different transport connection. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-12IB/iser: Add asynchronous event handlerOr Gerlitz2-1/+16
Add handler to handle events such as port up and down. This is useful when testing high-availability schemes such as multi-pathing. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-05-06MAINTAINERS: Add cxgb4 and iw_cxgb4 entriesRoland Dreier6-164/+224
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-29RDMA/cxgb3: Shrink .text with compile-time init of handlers arraysRoland Dreier2-66/+65
Using compile-time designated initializers for the handler arrays instead of open-coding the initialization in iwch_cm_init() is (IMHO) cleaner, and leads to substantially smaller code: on my x86-64 build, bloat-o-meter shows: add/remove: 0/1 grow/shrink: 4/3 up/down: 4/-1682 (-1678) function old new delta tx_ack 167 168 +1 state_set 55 56 +1 start_ep_timer 99 100 +1 pass_establish 177 178 +1 act_open_req_arp_failure 39 38 -1 sched 84 82 -2 iwch_cm_init 442 91 -351 work_handlers 1328 - -1328 Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-23Merge branch 'master' into for-nextJiri Kosina83-21/+158
2010-04-22IPoIB: Allow disabling/enabling TSO on the fly through ethtoolOr Gerlitz1-0/+20
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22IB/mlx4: Add support for masked atomic operationsVladimir Sokolovsky3-11/+48
Add support for masked atomic operations (masked compare and swap, masked fetch and add). Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/cma: Randomize local port allocationTetsuo Handa1-45/+25
Randomize local port allocation in the way sctp_get_port_local() does. Update rover at the end of loop since we're likely to pick a valid port on the first try. Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reviewed-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/nes: Make unnecessarily global functions staticRoland Dreier2-3/+3
This allows the compiler to do a bit better; on my x86-64 build: add/remove: 0/2 grow/shrink: 1/0 up/down: 2288/-2365 (-77) function old new delta nes_init_phy 273 2561 +2288 nes_init_1g_phy 469 - -469 nes_init_2025_phy 1896 - -1896 Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/nes: Make nesadapter->phy_lock usage consistentChien Tung3-20/+14
nes_{read,write}_1G_phy_reg() are using phy_lock while nes_{read,write}_10G_phy_reg() leave that to the caller. Remove phy_lock from 1G routines and leave the locking to the caller. Add additional phy_lock calls around 1G read/write. Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/cxgb4: Add driver for Chelsio T4 RNICSteve Wise16-0/+9447
Add an RDMA/iWARP driver for Chelsio T4 Ethernet adapters. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22IB/mthca: Use the dma state API instead of pci equivalentsFUJITA Tomonori3-8/+8
The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/amso1100: Use the dma state API instead of pci equivalentsFUJITA Tomonori6-13/+13
The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/cxgb3: Don't free skbs on NET_XMIT_* indications from LLDSteve Wise1-2/+2
The low level cxgb3 driver can return NET_XMIT_CN and friends. The iw_cxgb3 driver should _not_ treat these as errors. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-22RDMA/cxgb3: Use the dma state API instead of pci equivalentsFUJITA Tomonori3-9/+9
The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-21IB: Explicitly rule out llseek to avoid BKL in default_llseek()Roland Dreier4-10/+20
Several RDMA user-access drivers have file_operations structures with no .llseek method set. None of the drivers actually do anything with f_pos, so this means llseek is essentially a NOP, instead of returning an error as leaving other file_operations methods unimplemented would do. This is mostly harmless, except that a NULL .llseek means that default_llseek() is used, and this function grabs the BKL, which we would like to avoid. Since llseek does nothing useful on these files, we would like it to return an error to userspace instead of silently grabbing the BKL and succeeding. For nearly all of the file types, we take the belt-and-suspenders approach of setting the .llseek method to no_llseek and also calling nonseekable_open(); the exception is the uverbs_event files, which are created with anon_inode_getfile(), which already sets f_mode the same way as nonseekable_open() would. This work is motivated by Arnd Bergmann's bkl-removal tree. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-04-12Merge branch 'master' of ↵David S. Miller80-11/+101
master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 Conflicts: drivers/net/stmmac/stmmac_main.c drivers/net/wireless/wl12xx/wl1271_cmd.c drivers/net/wireless/wl12xx/wl1271_main.c drivers/net/wireless/wl12xx/wl1271_spi.c net/core/ethtool.c net/mac80211/scan.c
2010-04-09Merge branch 'for-linus' of ↵Linus Torvalds4-7/+7
git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: IB/mlx4: Check correct variable for allocation failure RDMA/nes: Correct cap.max_inline_data assignment in nes_query_qp() RDMA/cm: Set num_paths when manually assigning path records IB/cm: Fix device_create() return value check
2010-04-09Merge branches 'cma', 'misc', 'mlx4' and 'nes' into for-linusRoland Dreier3-7/+6
2010-04-08IB/mlx4: Check correct variable for allocation failureDan Carpenter1-1/+1
The intent here is to check the "mfrpl->mapped_page_list" allocation. We checked "mfrpl->ibfrpl.page_list" earlier. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
generated by cgit v1.2.3 (git 2.39.0) at 2024-11-20 19:14:41 +0300