aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband
AgeCommit message (Collapse)Author
2008-04-16IB/ipath: Enable 4KB MTUDave Olson
Enable use of 4KB MTU. Since the driver uses more pinned memory for receive buffers when the 4KB MTU is enabled, whether or not the fabric supports that MTU, add a "mtu4096" module parameter that can be used to limit the MTU to 2KB when it is known that 4KB MTUs can't be used anyway. Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ipath: Shared context code needs to be sure device is usableDave Olson
The code was checking if units are present, but not that present units were usable (link up, etc.) Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ipath: Provide I/O bus speeds for diagnostic purposesArthur Jones
Modern I/O buses like PCIe and HT can be configured for multiple speeds and widths. When an ipath HCA seems to have lower than expected performance, it is very useful to be able to display what the driver thinks the bus speed is. Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ipath: Make some constants chip-specific, related cleanupDave Olson
This patch makes some constants chip-specific, and makes some related changes to prepare for supporting another HCA. Signed-off-by: Dave Olson <dave.olson@qlogic.com Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ipath: Misc sparse warning cleanupArthur Jones
Recent sparse versions and kernel cleanups knock down the false positive rate of the ipath driver code to a point where having it be sparse clean is worthwhile. Here we fixup the sparse warnings. Some of these warnings (and the impetus to run sparse again) are due to work by Roland Dreier. Signed-off-by: Arthur Jones <arthur.jones@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/mthca: Add IPoIB checksum offload supportEli Cohen
Arbel and Sinai devices support checksum generation and verification of TCP and UDP packets for UD IPoIB messages. This patch checks if the HCA supports this and sets the IB_DEVICE_UD_IP_CSUM capability flag if it does. It implements support for handling the IB_SEND_IP_CSUM send flag and setting the csum_ok field in receive work completions. Signed-off-by: Eli Cohen <eli@mellnaox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/mlx4: Add IPoIB checksum offload supportEli Cohen
ConnectX devices support checksum generation and verification of TCP and UDP packets for UD IPoIB messages. This patch checks if the HCA supports this and sets the IB_DEVICE_UD_IP_CSUM capability flag if it does. It implements support for handling the IB_SEND_IP_CSUM send flag and setting the csum_ok field in receive work completions. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Ali Ayub <ali@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IPoIB: Use checksum offload support if availableEli Cohen
For HCAs that support checksum offload (ie that set IB_DEVICE_UD_IP_CSUM in the device capabilities flags), have IPoIB set NETIF_F_IP_CSUM and use the HCA to generate and verify IP checksums. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB: Replace remaining __FUNCTION__ occurrences with __func__Harvey Harrison
__FUNCTION__ is gcc-specific, use __func__ instead. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ehca: Make symbols used only in a single source file staticRoland Dreier
Allow the compiler to optimize better and generate smaller code: add/remove: 0/6 grow/shrink: 2/0 up/down: 1528/-1864 (-336) function old new delta .ehca_set_pagebuf 1344 2172 +828 .ehca_probe 2312 3012 +700 ehca_set_pagebuf_phys 24 - -24 ehca_set_pagebuf_fmr 24 - -24 ehca_init_device 24 - -24 .ehca_set_pagebuf_fmr 480 - -480 .ehca_set_pagebuf_phys 512 - -512 .ehca_init_device 800 - -800 Also this fixes warnings like: drivers/infiniband/hw/ehca/ehca_mrmw.c:2015:5: warning: symbol 'ehca_set_pagebuf_fmr' was not declared. Should it be static? Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/nes: Make symbols used only in a single source file staticRoland Dreier
Avoid namespace pollution and allow the compiler to optimize better. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/nes: Use proper format and cast to print dma_addr_tRoland Dreier
On some platforms, eg sparc64, dma_addr_t is not the same size as a pointer, so printing dma_addr_t values by casting to void * and using a %p format generates warnings. Fix this by casting to unsigned long and using %lx instead. This fixes the warnings: drivers/infiniband/hw/nes/nes_verbs.c: In function 'nes_setup_virt_qp': drivers/infiniband/hw/nes/nes_verbs.c:1047: warning: cast to pointer from integer of different size drivers/infiniband/hw/nes/nes_verbs.c:1078: warning: cast to pointer from integer of different size drivers/infiniband/hw/nes/nes_verbs.c:1078: warning: cast to pointer from integer of different size drivers/infiniband/hw/nes/nes_verbs.c: In function 'nes_reg_user_mr': drivers/infiniband/hw/nes/nes_verbs.c:2657: warning: cast to pointer from integer of different size Reported by Andrew Morton <akpm@linux-foundation.org>. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/nes: Remove unused nes_netdev_exit() functionRoland Dreier
nes_netdev_exit() has no callers, so delete it. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/nes: Remove redundant NULL check in nes_unregister_ofa_device()Roland Dreier
nes_unregister_ofa_device() dereferences the nesibdev pointer before testing if it's NULL. Also, the test is doubly redundant because the only caller of nes_unregister_ofa_device() is nes_destroy_ofa_device(), which already tests if nesibdev is NULL. Remove the unnecessary test. This was spotted by the Coverity checker (CID 2190). Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/uverbs: Use alloc_file() instead of get_empty_filp()Roland Dreier
Christoph Hellwig wants to unexport get_empty_filp(), which is an ugly internal interface. Change the modular user in ib_uverbs_alloc_event_file() to use the better alloc_file() interface; this makes the code cleaner too. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/uverbs: Don't store struct file * for event filesRoland Dreier
The file member of struct ib_uverbs_event_file was only used to keep track of whether the file had been closed or not. The only thing we ever did with the value was check if it was NULL or not. Simplify the code and get rid of the need to keep track of the struct file * we allocate by replacing the file member with an is_closed member. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16mlx4_core: Fix confusion between mlx4_event and mlx4_dev_event enumsRoland Dreier
The struct mlx4_interface.event() method was supposed to get an enum mlx4_dev_event, but the driver code was actually passing in the hardware enum mlx4_event values. Fix up the callers of mlx4_dispatch_event() so that they pass in the right type of value, and fix up the event method in mlx4_ib so that it can handle the enum mlx4_dev_event values. This eliminates the need for the subtype parameter to the event method, so remove it. This also fixes the sparse warning drivers/net/mlx4/intf.c:127:48: warning: mixing different enum types drivers/net/mlx4/intf.c:127:48: int enum mlx4_event versus drivers/net/mlx4/intf.c:127:48: int enum mlx4_dev_event Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/amso1100: Endian annotate mqsq allocatorRoland Dreier
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/amso1100: Start of endianness annotationRoland Dreier
Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Steve Wise <swise@opengridcomputing.com>
2008-04-16RDMA/nes: Delete unused variablesRoland Dreier
None of the cqp_reqs_XXX counters were ever used anywhere, and neither was the nics_per_function variable. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/nes: Trivial endianness annotationsRoland Dreier
Fix a couple of htonl() that should really be ntohl(). Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/ucma: Endian annotationRoland Dreier
Add __force cast of node_guid to __u64, since we are sticking it into a structure whose definition is shared with userspace. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/cm: Endianness annotationsRoland Dreier
Mostly update the RB tree comparisons to force __be types to normal integers, but the change to cm_format_sidr_req() is a real fix: param->path->pkey is already __be16. Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Sean Hefty <sean.hefty@intel.com>
2008-04-16IB/mlx4: Endianness annotationsRoland Dreier
Trivial fixes to stamp_send_wqe(). Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/ipath: Fix sparse warning about shadowed symbolRoland Dreier
Fix drivers/infiniband/hw/ipath/ipath_init_chip.c:526:10: warning: symbol 'val' shadows an earlier one drivers/infiniband/hw/ipath/ipath_init_chip.c:473:6: originally declared here by giving the second val a different name. Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Arthur Jones <arthur.jones@qlogic.com>
2008-04-16IB/ipath: Fix sparse warning about pointer signednessArthur Jones
There's no reason for the third parameter of ipath_count_units() to be a u32 *, so change it to be an int * instead. This fixes the sparse warning: drivers/infiniband/hw/ipath/ipath_file_ops.c:1654:47: warning: incorrect type in argument 3 (different signedness) drivers/infiniband/hw/ipath/ipath_file_ops.c:1654:47: expected unsigned int [usertype] *maxportsp drivers/infiniband/hw/ipath/ipath_file_ops.c:1654:47: got int *<noident> Signed-off-by: Arthur Jones <arthur.jones@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16RDMA/cxgb3: IDR IDs are signedRoland Dreier
Fix sparse warnings about pointer signedness by using a signed int when calling idr_get_new_above(). Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Steve Wise <swise@opengridcomputing.com>
2008-04-16RDMA/amso1100: Don't use 0UL as a NULL pointerRoland Dreier
Write tests for NULL pointers as if (!ptr) instead of if (ptr == 0UL) to fix sparse warnings. Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Steve Wise <swise@opengridcomputing.com>
2008-04-16IB/mlx4: Convert "if(foo)" to "if (foo)"Roland Dreier
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-16IB/mthca: Formatting cleanupsRoland Dreier
Fix a few whitespace and other coding style problems. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-30trivial endianness annotations: infiniband coreAl Viro
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-03-28RDMA/cxgb3: Program hardware IRD with correct valueRoland Dreier
Because of a typo in iwch_accept_cr(), the cxgb3 connection handling code programs the hardware IRD (incoming RDMA read queue depth) with the value that is passed in for the ORD (outgoing RDMA read queue depth). In particular this means that if an application passes in IRD > 0 and ORD = 0 (which is a completely sane and valid thing to do for an app that expects only incoming RDMA read requests), then the hardware will end up programmed with IRD = 0 and the app will fail in a mysterious way. Fix this by using "ep->ird" instead of "ep->ord" in the intended place. Signed-off-by: Roland Dreier <rolandd@cisco.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-03-21RDMA/nes: Fix MSS calculation on RDMA pathChien Tung
Fix the calculation of the MSS for RDMA connections: we need to allow space in frames for a VLAN tag too. Signed-off-by: Chien Tung <ctung@neteffect.com> Signed-off-by: Glenn Streiff <gstreiff@neteffect.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-12IPoIB: Allocate priv->tx_ring with vmalloc()Roland Dreier
Commit 7143740d ("IPoIB: Add send gather support") made struct ipoib_tx_buf significantly larger, since the mapping member changed from a single u64 to an array with MAX_SKB_FRAGS + 1 entries. This means that allocating tx_rings with kzalloc() may fail because there is not enough contiguous memory for the new, much bigger size. Fix this regression by allocating the rings with vmalloc() instead. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IPoIB/cm: Set tx_wr.num_sge in connected mode post_send()Roland Dreier
Commit 7143740d ("IPoIB: Add send gather support") made it possible for tx_wr.num_sge to be != 1 -- this happens if send gather support is enabled. However, the code in the connected mode post_send() function assumes the old invariant, namely that tx_wr.num_sge is always 1. Fix this by explicitly setting tx_wr.num_sge to 1 in the CM post_send(). Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IPoIB: Don't drop multicast sends when they can be queuedOr Gerlitz
When set_multicast_list() is called the multicast task is restarted and the IPOIB_MCAST_STARTED bit is cleared. As a result for some window of time, multicast packets are not transmitted nor queued but rather dropped by ipoib_mcast_send(). These dropped packets are painful in two cases: - bonding fail-over which both calls set_multicast_list() on the new active slave and sends Gratuitous ARP through that slave. - IP_DROP_MEMBERSHIP code which both calls set_multicast_list() on the device and issues IGMP leave. In both these cases, depending on the scheduling of the IPoIB multicast task, the packets would be dropped. As a result, in the bonding case, the failover would not be detected by the peers until their neighbour is renewed the neighbour (which takes a few tens of seconds). In the IGMP case, the IP router doesn't get an IGMP leave and would only learn on that from further probes on the group (also a delay of at least a few tens of seconds). Fix this by allowing transmission (or queuing) depending on the IPOIB_FLAG_OPER_UP flag instead of the IPOIB_MCAST_STARTED flag. Signed-off-by: Olga Shern <olgas@voltaire.com> Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IB/ipath: Reset the retry counter for RDMA_READ_RESPONSE_MIDDLE packetsPatrick Marchand Latifi
Reset the retry counter when we get a good RDMA_READ_RESPONSE_MIDDLE packet. This fix will prevent the requester from reporting a retry exceeded error too early. Signed-off-by: Patrick Marchand Latifi <patrick.latifi@qlogic.com>
2008-03-11IB/ipath: Fix error completion put on send CQ instead of recv CQPatrick Marchand Latifi
A work completion entry could be placed on the wrong completion queue when an RC QP is placed in the error state. Signed-off-by: Patrick Marchand Latifi <patrick.latifi@qlogic.com> Acked-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IB/ipath: Fix RC QP initializationPatrick Marchand Latifi
This patch fixes the initialization of RC QPs, since we would rely on the queue pair type (ibqp->qp_type) being set, but this field is only initialized when we return from ipath_create_qp (it is initialized by the user-level verbs library). The fix is to not depend on this field to initialize the send and the receive state of the RC QP. Signed-off-by: Patrick Marchand Latifi <patrick.latifi@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IB/ipath: Fix potentially wrong RNR retry counter returned in ipath_query_qp()Patrick Marchand Latifi
There can be a case where the requester's rnr retry counter (s_rnr_retry) is less than the number of rnr retries allowed per QP (s_rnr_retry_cnt). This can happen if the s_rnr_retry counter is being decremented and an ipath_query_qp call is issued during that time frame. The fix is to always return the number of rnr retries allowed per QP instead of the requester's rnr counter. Found by code review. Signed-off-by: Patrick Marchand Latifi <patrick.latifi@qlogic.com> Acked-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-11IB/ipath: Fix IB compliance problems with link state vs physical stateRalph Campbell
Subnet manager SetPortinfo messages distingush between changing the link state (DOWN, ARM, ACTIVE) and the link physical state (POLL, SLEEP, DISABLED). These are somewhat independent commands and affect when link width and speed changes take effect. Without this patch, a link DOWN physical state NOP command was causing the link width and speed settings to take effect which should only happen when the link physical state is goes down (either by a SMP or some link physical error like link errors exceeding the threshold). Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-10RDMA/iwcm: Don't access a cm_id after dropping referenceSteve Wise
cm_work_handler() can access cm_id_priv after it drops its reference by calling iwch_deref_id(), which might cause it to be freed. The fix is to look at whether IWCM_F_CALLBACK_DESTROY is set _before_ dropping the reference. Then if it was set, free the cm_id on this thread. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-10IB/iser: Handle iser_device allocation error gracefullyArne Redlich
"iser_device" allocation failure is "handled" with a BUG_ON() right before dereferencing the NULL-pointer - fix this! Signed-off-by: Arne Redlich <arne.redlich@xiranet.com> Signed-off-by: Erez Zilber <erezz@voltaire.com>
2008-03-10IB/iser: Fix list iteration bugArne Redlich
The iteration through the list of "iser_device"s during device lookup/creation is broken -- it might result in an infinite loop if more than one HCA is used with iSER. Fix this by using list_for_each_entry() instead of the open-coded flawed list iteration code. Signed-off-by: Arne Redlich <arne.redlich@xiranet.com> Signed-off-by: Erez Zilber <erezz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-03-09RDMA/cxgb3: Fix iwch_create_cq() off-by-one errorJon Mason
The cxbg3 driver is unnecessarily decreasing the number of CQ entries by one when creating a CQ. This will cause the CQ not to have as many entries as requested by the user if the user requests a power of 2 size. Signed-off-by: Jon Mason <jon@opengridcomputing.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-29RDMA/cxgb3: Return correct max_inline_data when creating a QPJon Mason
Set cap.max_inline_data to the actual max inline data that the adapter support, so that userspace apps see the right value returned. Signed-off-by: Jon Mason <jon@opengridcomputing.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-29IB/fmr_pool: Flush all dirty FMRs from ib_fmr_pool_flush()Pete Wyckoff
Commit a3cd7d90 ("IB/fmr_pool: ib_fmr_pool_flush() should flush all dirty FMRs") caused a regression for iSER and was reverted in e5507736. This change attempts to redo the original patch so that all used FMR entries are flushed when ib_flush_fmr_pool() is called without affecting the normal FMR pool cleaning thread. Simply move used entries from the clean list onto the dirty list in ib_flush_fmr_pool() before letting the cleanup thread do its job. Signed-off-by: Pete Wyckoff <pw@osc.edu> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-29Revert "IB/fmr_pool: ib_fmr_pool_flush() should flush all dirty FMRs"Pete Wyckoff
This reverts commit a3cd7d9070be417a21905c997ee32d756d999b38. The original commit breaks iSER reliably, making it complain: iser: iser_reg_page_vec:ib_fmr_pool_map_phys failed: -11 The FMR cleanup thread runs ib_fmr_batch_release() as dirty entries build up. This commit causes clean but used FMR entries also to be purged. During that process, another thread can see that there are no free FMRs and fail, even though there should always have been enough available. Signed-off-by: Pete Wyckoff <pw@osc.edu> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-29IB/cm: Flush workqueue when removing deviceSean Hefty
When a CM MAD is received, it is queued to a CM workqueue for processing. The queued work item references the port and device on which the MAD was received. If that device is removed from the system before the work item can execute, the work item will reference freed memory. To fix this, flush the workqueue after unregistering to receive MAD, and before the device is be freed. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-26RDMA/nes: Fix interrupt moderation low thresholdJohn Lacombe
Interrupt moderation low threshold value was incorrectly triggering, indicating that the threshold should be lowered. The impact was the timer was likely to become 40usecs and get stuck there. The biggest side effect was too many interrupts and nonoptimal performance. Signed-off-by: John Lacombe <jlacombe@neteffect.com> Signed-off-by: Glenn Streiff <gstreiff@neteffect.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>