Commit Graph

15 Commits

Author SHA1 Message Date
Steve Wise
9efe10a1e1 RDMA/cxgb4: Fail RDMA initialization for unsupported cards
The iw_cxgb4 module crashes at init time if the T4 card does not
support RDMA.  So clean up the init logic to correctly deal with
non-RDMA cards.

 - If any RDMA resources are not available, then fail the initialization
   logging an info message.
 - Clean up properly on initialization failures.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-10-06 09:32:44 -07:00
Steve Wise
2f25e9a540 RDMA/cxgb4: EEH errors can hang the driver
A few more EEH fixes:

c4iw_wait_for_reply(): detect fatal EEH condition on timeout and
return an error.

The iw_cxgb4 driver was only calling ib_deregister_device() on an EEH
event followed by a ib_register_device() when the device was
reinitialized.  However, the RDMA core doesn't allow multiple
iterations of register/deregister by the provider. See
drivers/infiniband/core/sysfs.c: ib_device_unregister_sysfs() where
the kobject ref is held until the device is deallocated in
ib_deallocate_device().  Calling deregister adds this kobj reference,
and then a subsequent register call will generate a WARN_ON() from the
kobject subsystem because the kobject is being initialized but is
already initialized with the ref held.

So the provider must deregister and dealloc when resetting for an EEH
event, then alloc/register to re-initialize.  To do this, we cannot
use the device ptr as our ULD handle since it will change with each
reallocation.  This commit adds a ULD context struct which is used as
the ULD handle, and then contains the device pointer and other state
needed.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-05-09 22:06:23 -07:00
Steve Wise
bbe9a0a2bc RDMA/cxgb4: Initialization errors can cause crash
c4iw_uld_add() must return ERR_PTR() values instead of NULL on failure.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-05-09 22:06:22 -07:00
Steve Wise
db5d040d7b RDMA/cxgb4: Debugfs dump_qp() updates
- Show whether the SQ is in onchip memory or not.
- Dump both SQ and RQ QIDs.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-03-14 12:09:14 -07:00
Steve Wise
767fbe8151 RDMA/cxgb4: Dispatch FATAL event on EEH errors
This at least kicks the user mode applications that are watching for
device events.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-03-14 12:09:13 -07:00
Steve Wise
2942813739 RDMA/cxgb4: Remove db_drop_task
Unloading iw_cxgb4 can crash due to the unload code trying to use
db_drop_task, which is uninitialized.  So remove this dead code.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-03-14 12:09:10 -07:00
Steve Wise
da411ba1da RDMA/cxgb4: Use cxgb4 service for packet gl to skb
Remove the local service t4_pktgl_to_skb() and use cxgb4_pktgl_to_skb()
exported by cxgb4.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-10-22 21:58:50 -07:00
Steve Wise
3160977a6e RDMA/cxgb4: Use simple_read_from_buffer() for debugfs handlers
We can replace our equivalent open-coded version.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-10-11 20:15:14 -07:00
Steve Wise
8bbac892fb RDMA/cxgb4: Add default_llseek to debugfs files
Incorporate BKL removal changes.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-10-11 20:14:00 -07:00
Steve Wise
c6d7b26791 RDMA/cxgb4: Support on-chip SQs
T4 support on-chip SQs to reduce latency.  This patch adds support for
this in iw_cxgb4:

 - Manage ocqp memory like other adapter mem resources.
 - Allocate user mode SQs from ocqp mem if available.
 - Map ocqp mem to user process using write combining.
 - Map PCIE_MA_SYNC reg to user process.

Bump uverbs ABI.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-28 10:46:35 -07:00
Steve Wise
9e8d1fa342 RDMA/cxgb4: debugfs files for dumping active stags
Add "stags" debugfs file.  This is useful for examining the TPTE and
PBL entries in adapter memory.  It allows scripts to dump just the
active entries.

Also clean up the "qps" file handlers and shared common code.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-28 10:46:33 -07:00
Steve Wise
93fb72e443 RDMA/cxgb4: Obtain RDMA QID ranges from LLD/FW
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-08-07 23:08:47 -07:00
Steve Wise
1c01c53883 RDMA/cxgb4: Register RDMA provider based on LLD state_change events
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>
2010-05-24 21:07:59 -07:00
Steve Wise
fd388ce677 RDMA/cxgb4: Detach from the LLD after unregistering RDMA device
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>
2010-05-24 21:07:59 -07:00
Steve Wise
cfdda9d764 RDMA/cxgb4: Add driver for Chelsio T4 RNIC
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>
2010-04-21 15:30:06 -07:00