summaryrefslogtreecommitdiffstats
path: root/drivers/infiniband
AgeCommit message (Collapse)Author
2010-09-27RDMA/cxgb3: Turn off RX coalescing for iWARP connectionsSteve Wise
The HW by default has RX coalescing on. For iWARP connections, this causes a 100ms delay in connection establishement due to the ingress MPA Start message being stalled in HW. So explicitly turn RX coalescing off when setting up iWARP connections. This was causing very bad performance for NP64 gather operations using Open MPI, due to the way it sets up connections on larger jobs. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Cc: <stable@kernel.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-08Merge branches 'cxgb3' and 'nes' into for-linusRoland Dreier
2010-09-08RDMA/nes: Fix hang with modified FIN handling on A0 cardsFaisal Latif
Changing state to CLOSING when FIN is received causes A0 cards to hang. Fix this by checking for A0 cards in FIN handling. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-08RDMA/nes: Change state to closing after FINFaisal Latif
When the driver receives an AE for FIN received, it closes the connection without changing the state of the connection in the hardware to closing. By changing the state to closing, hardware will do a normal close sequence. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-08RDMA/nes: Fix double CLOSE event indication crashFaisal Latif
During a stress testing in a large cluster, multiple close event are detected and BUG() is hit in the iWARP core. The cause is that the active node gave up while waiting for an MPA response from the peer and tried to close the connection by sending RST. The passive node driver receives the RST but is waiting for MPA response from the user. When the MPA accept is received, the driver offloads the connection and sends a CLOSE event. The driver gets an AE indicating RESET received and also sends a CLOSE event, hitting a BUG(). Fix this by correcting RESET handling and sending CLOSE events. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-08RDMA/nes: Write correct register write to set TX pause paramChien Tung
Setting TX pause param writes to the wrong register location causing the adapter to hang. Correct the define used to write the reigster. Addresses: https://bugs.openfabrics.org/show_bug.cgi?id=2116 Reported-by: Shiri Franchi <shirif@voltaire.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-02RDMA/cxgb3: Don't exceed the max HW CQ depthSteve Wise
The max depth supported by T3 is 64K entries. This fixes a bug introduced in commit 9918b28d ("RDMA/cxgb3: Increase the max CQ depth") that causes stalls and possibly crashes in large MPI clusters. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Cc: <stable@kernel.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-12Merge branch 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6Linus Torvalds
* 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6: mmc_spi: Fix unterminated of_match_table of/sparc: fix build regression from of_device changes of/device: Replace struct of_device with struct platform_device
2010-08-07RDMA/cxgb4: Obtain RDMA QID ranges from LLD/FWSteve Wise
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-07Merge branch 'for-linus' of ↵Linus Torvalds
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: (42 commits) IB/qib: Add missing <linux/slab.h> include IB/ehca: Drop unnecessary NULL test RDMA/nes: Fix confusing if statement indentation IB/ehca: Init irq tasklet before irq can happen RDMA/nes: Fix misindented code RDMA/nes: Fix showing wqm_quanta RDMA/nes: Get rid of "set but not used" variables RDMA/nes: Read firmware version from correct place IB/srp: Export req_lim via sysfs IB/srp: Make receive buffer handling more robust IB/srp: Use print_hex_dump() IB: Rename RAW_ETY to RAW_ETHERTYPE RDMA/nes: Fix two sparse warnings RDMA/cxgb3: Make needlessly global iwch_l2t_send() static IB/iser: Make needlessly global iser_alloc_rx_descriptors() static RDMA/cxgb4: Add timeouts when waiting for FW responses IB/qib: Fix race between qib_error_qp() and receive packet processing IB/qib: Limit the number of packets processed per interrupt IB/qib: Allow writes to the diag_counters to be able to clear them IB/qib: Set cfgctxts to number of CPUs by default ...
2010-08-06of/device: Replace struct of_device with struct platform_deviceGrant Likely
of_device is just an alias for platform_device, so remove it entirely. Also replace to_of_device() with to_platform_device() and update comment blocks. This patch was initially generated from the following semantic patch, and then edited by hand to pick up the bits that coccinelle didn't catch. @@ @@ -struct of_device +struct platform_device Signed-off-by: Grant Likely <grant.likely@secretlab.ca> Reviewed-by: David S. Miller <davem@davemloft.net>
2010-08-05Merge branches 'cxgb3', 'cxgb4', 'ehca', 'ipath', 'misc', 'nes', 'qib' and ↵Roland Dreier
'srp' into for-next
2010-08-05IB/qib: Add missing <linux/slab.h> includeDavid Miller
Fix build failure on sparc64 which is missing the include of <linux/slab.h> via <asm/pci.h> that x86, powerpc, ia64, etc. have. Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-05IB/ehca: Drop unnecessary NULL testJulia Lawall
list_for_each_entry binds its first argument to a non-null value, and thus any null test on the value of that argument is superfluous. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ iterator I; expression x; statement S,S1,S2; @@ I(x,...) { <... - if (x == NULL && ...) S ...> } // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Acked-by: Alexander Schmidt <alexs@linux.vnet.ibm.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-05RDMA/nes: Fix confusing if statement indentationRoland Dreier
Fix confusing indentation that makes a statement look as if it's part of an if statement when in fact it isn't. Reported-by: Julia Lawall <julia@diku.dk> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04IB/ehca: Init irq tasklet before irq can happenAlexander Schmidt
Initialize tasklet before interrupts are requested to prevent scheduling of an uninitialized tasklet. Signed-off-by: Alexander Schmidt <alexs@linux.vnet.ibm.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04Merge branch 'for-next' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial * 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (48 commits) Documentation: update broken web addresses. fix comment typo "choosed" -> "chosen" hostap:hostap_hw.c Fix typo in comment Fix spelling contorller -> controller in comments Kconfig.debug: FAIL_IO_TIMEOUT: typo Faul -> Fault fs/Kconfig: Fix typo Userpace -> Userspace Removing dead MACH_U300_BS26 drivers/infiniband: Remove unnecessary casts of private_data fs/ocfs2: Remove unnecessary casts of private_data libfc: use ARRAY_SIZE scsi: bfa: use ARRAY_SIZE drm: i915: use ARRAY_SIZE drm: drm_edid: use ARRAY_SIZE synclink: use ARRAY_SIZE block: cciss: use ARRAY_SIZE comment typo fixes: charater => character fix comment typos concerning "challenge" arm: plat-spear: fix typo in kerneldoc reiserfs: typo comment fix update email address ...
2010-08-04RDMA/nes: Fix misindented codeRoland Dreier
In nes_probe(), a bit of code is indented one tab stop too far. Fix this. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/nes: Fix showing wqm_quantaRoland Dreier
In nes_show_wqm_quanta(), the wrong value is printed. Fix this. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/nes: Get rid of "set but not used" variablesRoland Dreier
Delete dead code in various places that is shown by gcc 4.6's new -Wunused-but-set-variable warnings. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/nes: Read firmware version from correct placeMiroslaw Walukiewicz
Signed-off-by: Mirek Walukiewicz <miroslaw.walukiewicz@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04IB/srp: Export req_lim via sysfsBart Van Assche
Export req_lim via sysfs for debugging. Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com> Acked-by: David Dillow <dave@thedillows.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6Linus Torvalds
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1443 commits) phy/marvell: add 88ec048 support igb: Program MDICNFG register prior to PHY init e1000e: correct MAC-PHY interconnect register offset for 82579 hso: Add new product ID can: Add driver for esd CAN-USB/2 device l2tp: fix export of header file for userspace can-raw: Fix skb_orphan_try handling Revert "net: remove zap_completion_queue" net: cleanup inclusion phy/marvell: add 88e1121 interface mode support u32: negative offset fix net: Fix a typo from "dev" to "ndev" igb: Use irq_synchronize per vector when using MSI-X ixgbevf: fix null pointer dereference due to filter being set for VLAN 0 e1000e: Fix irq_synchronize in MSI-X case e1000e: register pm_qos request on hardware activation ip_fragment: fix subtracting PPPOE_SES_HLEN from mtu twice net: Add getsockopt support for TCP thin-streams cxgb4: update driver version cxgb4: add new PCI IDs ... Manually fix up conflicts in: - drivers/net/e1000e/netdev.c: due to pm_qos registration infrastructure changes - drivers/net/phy/marvell.c: conflict between adding 88ec048 support and cleaning up the IDs - drivers/net/wireless/ipw2x00/ipw2100.c: trivial ipw2100_pm_qos_req conflict (registration change vs marking it static)
2010-08-04IB/srp: Make receive buffer handling more robustBart Van Assche
The current strategy in ib_srp for posting receive buffers is: * Post one buffer after channel establishment. * Post one buffer before sending an SRP_CMD or SRP_TSK_MGMT to the target. As a result, only the first non-SRP_RSP information unit from the target will be processed. If that first information unit is an SRP_T_LOGOUT, it will be processed. On the other hand, if the initiator receives an SRP_CRED_REQ or SRP_AER_REQ before it receives a SRP_T_LOGOUT, the SRP_T_LOGOUT won't be processed. We can fix this inconsistency by changing the strategy for posting receive buffers to: * Post all receive buffers after channel establishment. * After a receive buffer has been consumed and processed, post it again. A side effect is that the ib_post_recv() call is moved out of the SCSI command processing path. Since __srp_post_recv() is not called directly any more, get rid of it and move the code directly into srp_post_recv(). Also, move srp_post_recv() up in the file to avoid a forward declaration. Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com> Acked-by: David Dillow <dave@thedillows.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04IB/srp: Use print_hex_dump()Bart Van Assche
Replace an open-coded dump of the receive buffer with a call to print_hex_dump(). Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04IB: Rename RAW_ETY to RAW_ETHERTYPEAleksey Senin
Change abbreviated IB_QPT_RAW_ETY to IB_QPT_RAW_ETHERTYPE to make the special QP type easier to understand. cf http://www.mail-archive.com/linux-rdma@vger.kernel.org/msg04530.html Signed-off-by: Aleksey Senin <alekseys@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/nes: Fix two sparse warningsOr Gerlitz
Simple changes to fix warnings: CHECK drivers/infiniband/hw/nes/nes_verbs.c nes_verbs.c:1944:45: warning: Using plain integer as NULL pointer nes_verbs.c:1944:48: warning: Using plain integer as NULL pointer CHECK drivers/infiniband/hw/nes/nes_cm.c nes_cm.c:2645:43: warning: mixing different enum types nes_cm.c:2645:43: int enum iw_cm_event_type versus nes_cm.c:2645:43: int enum iw_cm_event_status Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Acked-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/cxgb3: Make needlessly global iwch_l2t_send() staticOr Gerlitz
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04IB/iser: Make needlessly global iser_alloc_rx_descriptors() staticOr Gerlitz
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04RDMA/cxgb4: Add timeouts when waiting for FW responsesSteve Wise
Don't hang a host thread if the FW stops responding. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-04Merge branch 'master' into for-nextJiri Kosina
2010-08-03IB/qib: Fix race between qib_error_qp() and receive packet processingRalph Campbell
When transitioning a QP to the error state, in progress RWQEs need to be marked complete. This also involves releasing the reference count to the memory regions referenced in the SGEs. The locking in the receive packet processing wasn't sufficient to prevent qib_error_qp() from modifying the r_sge state at the same time, thus leading to kernel panics. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-03IB/qib: Limit the number of packets processed per interruptRalph Campbell
Don't processes too many packets without allowing other IRQ functions a chance to run. Otherwise, there is a chance of getting a "soft lockup" messages and poor application response times. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-03IB/qib: Allow writes to the diag_counters to be able to clear themIra Weiny
Signed-off-by: Ira Weiny <weiny2@llnl.gov> Acked-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-03IB/qib: Set cfgctxts to number of CPUs by defaultRalph Campbell
Up to now, we have set the number of available user contexts based on the number of hardware contexts which is set according to the number of available CPUs. This was fine since most CPUs had a power of two number of cores and the chip supported 4, 8, or 16 user contexts. Now that some systems have 12 cores, the default isn't optimal and should be set to 12 even though 16 hardware contexts need to be enabled. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-02RDMA/cxgb4: Set/reset the EP timer inside EP lockSteve Wise
Endpoint timer manipulation needs to be done inside the lock. Otherwise we can get into a situation where a timer is stopped before it is started, which hits the WARN_ON() in stop_ep_timer(). Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-02RDMA/cxgb4: Use correct control txqSteve Wise
There is only one control txq per tx channel. So use the port number as the queue index when sending. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-02RDMA/cxgb4: Fix race in fini pathSteve Wise
There exists a race condition where the app disconnects, which initiates an orderly close (via rdma_fini()), concurrently with an ingress abort condition, which initiates an abortive close operation. Since rdma_fini() must be called without IRQs disabled, the fini can be called after the QP has been transitioned to ERROR. This is ok, but we need to protect against qp->ep getting NULLed. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-28IB/cm: Check LAP state before sending an MRASean Hefty
NULL pointer dereferences in ib_cm_init_qp_attr() were seen by some users. From a crash dump, I determined that we died in cm_init_qp_rts_attr() (it's inlined, so it doesn't show up in the traceback) on the line labeled below: static int cm_init_qp_rts_attr(struct cm_id_private *cm_id_priv, struct ib_qp_attr *qp_attr, int *qp_attr_mask) { ........ if (cm_id_priv->id.lap_state == IB_CM_LAP_UNINIT) { ..... } else { *qp_attr_mask = IB_QP_ALT_PATH | IB_QP_PATH_MIG_STATE; qp_attr->alt_port_num = cm_id_priv->alt_av.port->port_num; <-die The problem is that the rdma_cm can call ib_send_cm_mra() after a connection has been established. The ib_cm incorrectly assumes that the MRA is in response to a LAP (load alternate path) message, even though no LAP message has been received. The ib_cm needs to check the lap_state before sending an MRA if the cm_id state is established. Reported-by: Arthur Kepner <akepner@sgi.com> Reported-by: Josh England <jjengla@gmail.com> Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-28RDMA/nes: Fix hangs on ifdownFaisal Latif
When ib_unregister_device() is called from netdev stop during ifdown, it sometimes hangs. Changes made to indicate port_err to ib_dispatch_event() during netdev stop and port_active during netdev open. The ib_unregister_device() is only called during remove of the module. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-28RDMA/nes: Store and print eeprom versionChien Tung
Read and print eeprom version and save it off for later use. Also delete a tab. Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-28RDMA/nes: Convert pci_table entries to PCI_VDEVICEPeter Huewe
This patch converts pci_table entries, where .subvendor=PCI_ANY_ID and .subdevice=PCI_ANY_ID, .class=0 and .class_mask=0, to use the PCI_VDEVICE macro, and thus improves readability. Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21IB/ehca: Catch failing ioremap()Alexander Schmidt
When ioremap() fails with a NULL pointer, catch the error and pass it to the caller of create_qp() or create_cq() instead of trying to dereference the NULL pointer later on. Signed-off-by: Alexander Schmidt <alexs@linux.vnet.ibm.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21IB/qib: Allow PSM to select from multiple port assignment algorithmsDave Olson
We used to allow only full specification, or using all contexts within an HCA before moving to the next HCA. We now allow an additional method -- round-robining through HCAs -- and make that the default. Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21IB/qib: Turn off IB latency modeRalph Campbell
Turn off IB latency mode. This improves link quality for slower process chips. Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21IB/qib: Use generic_file_llseekArnd Bergmann
When the default llseek action gets changed to no_llseek, all file systems relying on the current behaviour need to set explicit .llseek operations. In case of qib_fs, we want the files to be seekable, so generic_file_llseek fits best. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21RDMA/cxgb4: Support variable sized work requestsSteve Wise
T4 EQ entries are in multiples of 64 bytes. Currently the RDMA SQ and RQ use fixed sized entries composed of 4 EQ entries for the SQ and 2 EQ entries for the RQ. For optimial latency with small IO, we need to change this so the HW only needs to DMA the EQ entries actually used by a given work request. Implementation: - add wq_pidx counter to track where we are in the EQ. cidx/pidx are used for the sw sq/rq tracking and flow control. - the variable part of work requests is the SGL. Add new functions to build the SGL and/or immediate data directly in the EQ memory wrapping when needed. - adjust the min burst size for the EQ contexts to 64B. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21RDMA/cxgb3: Clean up signed check of unsigned variableDan Carpenter
Q_FREECNT() returns the number of spaces free. This should never be a negative amount. Also the num_wrs is an unsigned int so it can never be less than zero. Signed-off-by: Dan Carpenter <error27@gmail.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21RDMA/cxgb4: Remove dependency on __GFP_NOFAILDavid Rientjes
The alloc_skb() in various allocations are failable, so remove __GFP_NOFAIL from their masks. Signed-off-by: David Rientjes <rientjes@google.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-07-21RDMA/cxgb4: Add module option to tweak delayed ackSteve Wise
Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>