Commit Graph

327 Commits

Author SHA1 Message Date
Jack Morgenstein
94361cf74a IB/mthca: check RDMA limits
Add limit checking on rd_atomic and dest_rd_atomic attributes:
especially for max_dest_rd_atomic, a value that is larger than HCA
capability can cause RDB overflow and corruption of another QP.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-12-09 16:32:21 -08:00
Jack Morgenstein
52d0df153c IB/mthca: fix memory user DB table leak
Free the memory allocated in mthca_init_user_db_tab() when releasing
the db_tab in mthca_cleanup_user_db_tab().

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-12-09 13:48:50 -08:00
Jack Morgenstein
0efc4883a6 IB/umad: fix memory leaks
Don't leak packet if it had a timeout, and don't leak timeout struct
if queue_packet() fails.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-12-09 13:46:32 -08:00
Sean Hefty
de1bb1a64c IB/cm: avoid reusing local ID
Use an increasing local ID to avoid re-using identifiers while
messages may still be outstanding on the old ID.  Without this, a
quick connect-disconnect-connect sequence can fail by matching
messages for the new connection with the old connection.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-30 10:01:13 -08:00
Sean Hefty
227eca8369 IB/cm: correct reported reject code
Change reject code from TIMEOUT to CONSUMER_REJECT when destroying a
cm_id in the process of connecting.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-30 10:00:25 -08:00
Jack Morgenstein
a3c8ab4fe8 IB/mthca: fix QP size limits for mem-free HCAs
Unlike tavor, the max work queue size is an exact power of 2 for arbel
mode, despite what the documentation (of the QUERY_DEV_LIM firmware
command) says.  Without this patch, on Arbel, we can start with a QP
of a valid size and get above the reported limit after rounding to the
next power of two.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-30 09:55:22 -08:00
Jack Morgenstein
f4e401562c IB/uverbs: track multicast group membership for userspace QPs
uverbs needs to track which multicast groups is each qp
attached to, in order to properly detach when cleanup
is performed on device file close.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 16:57:01 -08:00
Michael S. Tsirkin
e0ae9ecf46 IB/mthca: fix posting of send lists of length >= 255 on mem-free HCAs
On mem-free HCAs, when posting a long list of send requests, a
doorbell must be rung every 255 requests.  Add code to handle this.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 11:33:46 -08:00
Roland Dreier
267ee88ed3 IPoIB: fix error handling in ipoib_open
If ipoib_ib_dev_up() fails after ipoib_ib_dev_open() is called, then
ipoib_ib_dev_stop() needs to be called to clean up.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 10:55:58 -08:00
Michael S. Tsirkin
4f71055a45 IPoIB: protect child list in ipoib_ib_dev_flush
race condition: ipoib_ib_dev_flush is accessing child list without locks.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 10:53:30 -08:00
Roland Dreier
2e86541ec8 IPoIB: don't zero members after we allocate with kzalloc
ipoib_mcast_alloc() uses kzalloc(), so there's no need to zero out
members of the mcast struct after it's allocated.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 10:25:23 -08:00
Michael S. Tsirkin
de92248789 IPoIB: reinitialize mcast structs' completions for every query
Make sure mcast->done is initialized to uncompleted value before we
submit a new query, so that it's safe to wait on.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 10:18:45 -08:00
Roland Dreier
5872a9fc28 IPoIB: always set path->query to NULL when query finishes
Always set path->query to NULL when the SA path record query
completes, rather than only when we don't have an address handle.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-29 10:13:54 -08:00
Roland Dreier
65c7eddaba IPoIB: reinitialize path struct's completion for every query
It's possible that IPoIB will issue multiple SA queries for the same
path struct.  Therefore the struct's completion needs to be
initialized for each query rather than only once when the struct is
allocated, or else we might not wait long enough for later queries to
finish and free the path struct too soon.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-28 21:20:34 -08:00
Michael S. Tsirkin
bf6d9e23a3 IB/umad: fix RMPP handling
ib_umad_write in user_mad.c is looking at rmpp_hdr field in MAD before
checking that the MAD actually has the RMPP header.  So for a MAD
without RMPP header it looks like we are actually checking a bit
inside M_Key, or something.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-28 13:07:20 -08:00
Michael S. Tsirkin
187a25863f IB/mthca: reset QP's last pointers when transitioning to reset state
last pointer is not updated when QP is modified to reset state.  This
causes data corruption if WQEs are already posted on the queue.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-28 11:19:43 -08:00
Adrian Bunk
2012a116d9 [PATCH] drivers/infiniband/core/mad.c: fix use-after-release case
The Coverity checker spotted this obvious use-after-release bug caused
by a wrong order of the cleanups.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-27 20:23:13 -08:00
Linus Torvalds
9b152d53b7 Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband 2005-11-18 14:33:03 -08:00
Roland Dreier
eabc77935d IB/umad: make sure write()s have sufficient data
Make sure that userspace passes in enough data when sending a MAD.  We
always copy at least sizeof (struct ib_user_mad) + IB_MGMT_RMPP_HDR
bytes from userspace, so anything less is definitely invalid.  Also,
if the length is less than this limit, it's possible for the second
copy_from_user() to get a negative length and trigger a BUG().

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-18 14:18:26 -08:00
Michael S. Tsirkin
48fd0d1fdd IB/mthca: Safer max_send_sge/max_recv_sge calculation
Calculation of QP capabilities still isn't exactly right in mthca:
max_send_sge/max_recv_sge fields returned in create_qp can exceed the
handware supported limits.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-18 14:11:17 -08:00
Roland Dreier
cbc5b2bb9e [IB] mthca: don't disable RDMA writes if no responder resources
Responder resources are only required to handle RDMA reads and atomic
operations, not RDMA writes.  So the driver should allow RDMA writes
even if responder resources are set to 0.  This is especially
important for the UC transport -- with the old code, it was impossible
to enable RDMA writes for UC QPs.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-15 00:24:23 -08:00
Roland Dreier
47f2bce902 [IB] srp: don't post receive if no send buf available
Have __srp_get_tx_iu() fail if the target port's request limit will
not allow the initiator to post a send.  This avoids continuing on and
posting a receive, and then failing to post a corresponding send.  If
that happens, then the initiator will end up with an extra receive
posted, and if this happens to much, the receive queue will overflow.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-15 00:19:21 -08:00
Roland Dreier
5f068992a1 [IB] srp: increase max_luns
Increase SRP max_luns to 512 to match the kernel's default, since SRP
storage targets can have lots of LUNs and the SRP initiator itself
doesn't have any particular limit.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-11 14:06:01 -08:00
Greg Kroah-Hartman
249bb070f5 [PATCH] PCI: removed unneeded .owner field from struct pci_driver
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-11-10 16:09:17 -08:00
Linus Torvalds
78b9c0f91c Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband 2005-11-10 13:27:06 -08:00
Roland Dreier
94382f3562 [IB] umad: further ib_unregister_mad_agent() deadlock fixes
The previous umad deadlock fix left ib_umad_kill_port() still
vulnerable to deadlocking.  This patch fixes that by downgrading our
lock to a read lock when we might end up trying to reacquire the lock
for reading.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:51 -08:00
Michael S. Tsirkin
ae57e24a40 [IB] mthca: fix posting long lists of receive work requests
In Tavor mode, when posting a long list of receive work requests, a
doorbell must be rung every 256 requests.  Add code to do this when
required.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:51 -08:00
Roland Dreier
64044bcf75 [IB] mthca: fix wraparound handling in mthca_cq_clean()
Handle case where prod_index has wrapped around and become less than
cq->cons_index by checking that their difference as a signed int is
positive rather than comparing directly.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:51 -08:00
Michael S. Tsirkin
62abb8416f [IB] mthca: fix posting of atomic operations
The size of work requests for atomic operations was computed
incorrectly in mthca: all sizeofs need to be divided by 16.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Jack Morgenstein
77369ed31d [IB] uverbs: have kernel return QP capabilities
Move the computation of QP capabilities (max scatter/gather entries,
max inline data, etc) into the kernel, and have the uverbs module
return the values as part of the create QP response.  This keeps
precise knowledge of device limits in the low-level kernel driver.

This requires an ABI bump, so while we're making changes, get rid of
the max_sge parameter for the modify SRQ command -- it's not used and
shouldn't be there.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
ec914c52d6 [IB] umad: get rid of unused mr array
Now that ib_umad uses the new MAD sending interface, it no longer
needs its own L_Key.  So just delete the array of MRs that it keeps.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
40de2e548c [IB] Have cq_resize() method take an int, not int*
Change the struct ib_device.resize_cq() method to take a plain integer
that holds the new CQ size, rather than a pointer to an integer that
it uses to return the new size.  This makes the interface match the
exported ib_resize_cq() signature, and allows the low-level driver to
update the CQ size with proper locking if necessary.

No in-tree drivers are exporting this method yet.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
0b4ff2c0e6 [IB] mthca: fix typo in catastrophic error polling
Fix a typo in the rearming of the catastrophic error polling timer: we
should rearm the timer as long as the stop flag is _not_ set.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
8c608a32e3 [IPoIB] no need to set skb->dev right before freeing skb
For cut-and-paste reasons, the IPoIB driver was setting skb->dev right
before calling dev_kfree_skb_any().  Get rid of this.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
2f76e82947 [IB] umad: avoid potential deadlock when unregistering MAD agents
ib_unregister_mad_agent() completes all pending MAD sends and waits
for the agent's send_handler routine to return.  umad's send_handler()
calls queue_packet(), which does down_read() on the port mutex to look
up the agent ID.  This means that the port mutex cannot be held for
writing while calling ib_unregister_mad_agent(), or else it will
deadlock.  This patch fixes all the calls to ib_unregister_mad_agent()
in the umad module to avoid this deadlock.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:50 -08:00
Roland Dreier
1732b0ef3b [IPoIB] add path record information in debugfs
Add ibX_path files to debugfs that contain information about the IPoIB
path cache.  IPoIB ARP only gives GIDs, which the IPoIB driver must
resolve to real IB paths through the ib_sa module.  For debugging,
when the ARP table looks OK but traffic isn't flowing, it's useful to
be able to see if the resolution from GID to path worked.

Also clean up the formatting of the existing _mcg debugfs files.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-10 10:22:49 -08:00
Olaf Hering
733482e445 [PATCH] changing CONFIG_LOCALVERSION rebuilds too much, for no good reason
This patch removes almost all inclusions of linux/version.h.  The 3
#defines are unused in most of the touched files.

A few drivers use the simple KERNEL_VERSION(a,b,c) macro, which is
unfortunatly in linux/version.h.

There are also lots of #ifdef for long obsolete kernels, this was not
touched.  In a few places, the linux/version.h include was move to where
the LINUX_VERSION_CODE was used.

quilt vi `find * -type f -name "*.[ch]"|xargs grep -El '(UTS_RELEASE|LINUX_VERSION_CODE|KERNEL_VERSION|linux/version.h)'|grep -Ev '(/(boot|coda|drm)/|~$)'`

search pattern:
/UTS_RELEASE\|LINUX_VERSION_CODE\|KERNEL_VERSION\|linux\/\(utsname\|version\).h

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-09 07:55:57 -08:00
Jesper Juhl
6044ec8882 [PATCH] kfree cleanup: misc remaining drivers
This is the remaining misc drivers/ part of the big kfree cleanup patch.

Remove pointless checks for NULL prior to calling kfree() in misc files in
drivers/.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Acked-by: Aristeu Sergio Rozanski Filho <aris@cathedrallabs.org>
Acked-by: Roland Dreier <rolandd@cisco.com>
Acked-by: Pierre Ossman <drzeus@drzeus.cx>
Acked-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Len Brown <len.brown@intel.com>
Acked-by: "Antonino A. Daplas" <adaplas@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-07 07:54:05 -08:00
Tim Schmielau
8c65b4a604 [PATCH] fix remaining missing includes
Fix more include file problems that surfaced since I submitted the previous
fix-missing-includes.patch.  This should now allow not to include sched.h
from module.h, which is done by a followup patch.

Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-07 07:53:41 -08:00
Michael S. Tsirkin
8b37b94721 [IB] umad: two small fixes
Two small fixes for the umad module:
 - set kobject name for issm device properly
 - in ib_umad_add_one(), s is subtracted from the index i when
   initializing ports, so s should be subtracted from the index when
   freeing ports in the error path as well.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-06 15:47:02 -08:00
Jack Morgenstein
0f69ce1e44 [IB] mthca: report page size capability
Report the device's real page size capability in mthca_query_device().

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-04 21:28:21 -08:00
Linus Torvalds
127f2fa31a Merge branch 'srp' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband 2005-11-04 16:32:36 -08:00
Linus Torvalds
ba77df570c Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband 2005-11-04 16:31:54 -08:00
Jack Morgenstein
d09e327641 [IB] mthca: check P_Key index in modify QP
Make sure that the P_Key index passed into mthca_modify_qp() is
within the device's P_Key table.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-03 14:58:33 -08:00
Roland Dreier
0c99cb6d5f [IB] umad: fix hot remove of IB devices
Fix hotplug of devices for ib_umad module: when a device goes away,
kill off all MAD agents for open files associated with that device,
and make sure that the device is not touched again after ib_umad
returns from its remove_one function.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-03 12:01:18 -08:00
Roland Dreier
87cfe32375 [IB] mthca: fix format of FW version
Mellanox has decided that the components of the firmware version are
really meant to be displayed in decimal, e.g. 0x000400070190 is
version 4.7.400.  Change the format we use from "%x.%x.%x" to
"%d.%d.%d" to match this convention.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 22:59:37 -08:00
Roland Dreier
8ae5a8a24f [IPoIB] don't compile debug code if debugging isn't enabled
Don't build ipoib_mcast_iter_ functions if CONFIG_INFINIBAND_IPOIB_DEBUG
is not enabled -- their only callers will not be built either.

Also move the prototype for ipoib_open() to ipoib.h to fix a sparse warning.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 20:51:01 -08:00
Roland Dreier
aef9ec39c4 IB: Add SCSI RDMA Protocol (SRP) initiator
Add an InfiniBand SCSI RDMA Protocol (SRP) initiator.  This driver is
used to talk talk to InfiniBand SRP targets (storage devices).

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 14:07:13 -08:00
Roland Dreier
21a384897d [IPoIB] remove unneeded initializations to 0
Shrink our source and .text a little by removing a few assignments of
NULL and 0 to memory that is already cleared as part of the allocation.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 10:07:59 -08:00
Roland Dreier
de6eb66b56 [IB] kzalloc() conversions
Replace kmalloc()+memset(,0,) with kzalloc(), for a net savings of 35
source lines and about 500 bytes of text.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-11-02 07:23:14 -08:00
Roland Dreier
3bc12e75b2 [IPoIB] cleanups: fix comment, remove useless variables
Minor cleanups: fix a misleading comment, and get rid of attr_mask
variables that are only used to hold constants (just use the constants
directly).

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-31 07:10:33 -08:00
Roland Dreier
e5b251a24a [IB] mthca: Avoid SRQ free WQE list corruption
Fix wqe_to_link() to use a structure field that we know is definitely
always unused for receive work requests, so that it really avoids the
free list corruption bug that the comment claims it does.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-31 07:10:32 -08:00
Roland Dreier
7162a3e0db [IB] uverbs: Avoid NULL pointer deref on CQ async event
Userspace CQs that have no completion event channel attached end up
with their cq_context set to NULL.  However, asynchronous events like
"CQ overrun" can still occur on such CQs, so add a uverbs_file member
to struct ib_ucq_object that we can follow to deliver these events.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-31 07:10:32 -08:00
Tim Schmielau
4e57b68178 [PATCH] fix missing includes
I recently picked up my older work to remove unnecessary #includes of
sched.h, starting from a patch by Dave Jones to not include sched.h
from module.h. This reduces the number of indirect includes of sched.h
by ~300. Another ~400 pointless direct includes can be removed after
this disentangling (patch to follow later).
However, quite a few indirect includes need to be fixed up for this.

In order to feed the patches through -mm with as little disturbance as
possible, I've split out the fixes I accumulated up to now (complete for
i386 and x86_64, more archs to follow later) and post them before the real
patch.  This way this large part of the patch is kept simple with only
adding #includes, and all hunks are independent of each other.  So if any
hunk rejects or gets in the way of other patches, just drop it.  My scripts
will pick it up again in the next round.

Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 17:37:32 -08:00
Roland Dreier
a20583a7c2 [IPoIB] use spin_trylock_irqsave()
Use spin_trylock_irqsave() in ipoib_start_xmit() instead of
reinventing it out of local_irq_save(), spin_trylock() and
local_irq_restore().

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-29 13:54:40 -07:00
Al Viro
2d3c0b7bed [PATCH] missing include in infiniband
use of IS_ERR/PTR_ERR in infiniband/core/agent.c, without a portable
chain of includes pulling err.h (breaks on a bunch of platforms).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 10:35:07 -07:00
Michael S. Tsirkin
affcd50546 [IB] mthca: report asynchronous CQ events
Implement reporting asynchronous CQ events in Mellanox HCA driver.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-29 07:39:42 -07:00
Roland Dreier
4cce3390c9 [IB] fix up class_device_create() calls
Fix class_device_create() calls to match the new prototype which
takes a parent device pointer.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-28 16:38:15 -07:00
Roland Dreier
a4da0628ef Merge master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6 2005-10-28 15:50:51 -07:00
Roland Dreier
70a30e16a8 [IB] uverbs: Fix device lifetime problems
Move ib_uverbs module to using cdev_alloc() and class_device_create()
so that we can handle device lifetime properly.  Now we can make sure
we keep all of our data structures around until the last way to reach
them is gone.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-28 15:38:26 -07:00
Roland Dreier
a74968f8c3 [IB] umad: Fix device lifetime problems
Move ib_umad module to using cdev_alloc() and class_device_create() so
that we can handle device lifetime properly.  Now we can make sure we
keep all of our data structures around until the last way to reach
them is gone.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-28 15:37:23 -07:00
Roland Dreier
1993d683f3 [IPoIB] Drop RX packets when out of memory
Change the way IPoIB handles RX packets when it can't allocate a new
receive skbuff.  If the allocation of a new receive skb fails, we now
drop the packet we just received and repost the original receive skb.
This means that the receive ring always stays full and we don't have
to monkey around with trying to schedule a refill task for later.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-28 15:30:34 -07:00
Greg KH
6fbfddcb52 Merge ../bleed-2.6 2005-10-28 10:13:16 -07:00
Greg Kroah-Hartman
53f4654272 [PATCH] Driver Core: fix up all callers of class_device_create()
The previous patch adding the ability to nest struct class_device
changed the paramaters to the call class_device_create().  This patch
fixes up all in-kernel users of the function.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-10-28 09:52:52 -07:00
Al Viro
87b750dc4b [PATCH] gfp_t: drivers/infiniband
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-28 08:16:50 -07:00
Sean Hefty
cb0f0910f4 [IB] ib_umad: various cleanups
Simplify user_mad.c code in a few places, and convert from kmalloc() +
memset() to kzalloc().  This also fixes a theoretical race window by
not accessing packet->length after posting the send buffer (the send
could complete and packet could be freed before we get to the return
statement at the end of ib_umad_write()).

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-27 20:48:11 -07:00
Roland Dreier
089a1bedd8 [IB] ib_umad: fix crash when freeing send buffers
The conversion of user_mad.c to the new MAD send API was slightly off:
in a few places, we used packet->msg instead of packet->msg->mad when
referring to the actual data buffer, which ended up corrupting the
underlying data structure and crashing when we free an invalid pointer.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-27 20:33:43 -07:00
Roland Dreier
3d155f8cd0 [IB] mthca: first pass at catastrophic error reporting
Add some initial support for detecting and reporting catastrophic
errors reported by Mellanox HCAs.  We start a periodic timer which
polls the catastrophic error reporting buffer in device memory.  If an
error is detected, we dump the contents of the buffer for port-mortem
debugging, and report a fatal asynchronous error to higher levels.

In the future we can try to recover from these errors by resetting the
device, but this will require some work in higher-level code as well.
Let's get this in now, so that we at least get catastrophic errors
reported in logs.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-27 11:03:38 -07:00
Roland Dreier
7cc656efb5 [IB] simplify mad_rmpp.c:alloc_response_msg()
Change alloc_response_msg() in mad_rmpp.c to return the struct
it allocates directly (or an error code a la ERR_PTR), rather than
returning a status and passing the struct back in a pointer param.
This simplifies the code and gets rid of warnings like

    drivers/infiniband/core/mad_rmpp.c: In function nack_recv:
    drivers/infiniband/core/mad_rmpp.c:192: warning: msg may be used uninitialized in this function

with newer versions of gcc.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-25 15:13:54 -07:00
Roland Dreier
547e309073 [IB] mthca: correct modify QP attribute masks for UC
The UC transport does not support RDMA reads or atomic operations, so
we shouldn't require or even allow the consumer to set attributes
relating to these operations for UC QPs.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-25 10:57:32 -07:00
Sean Hefty
34816ad98e [IB] Fix MAD layer DMA mappings to avoid touching data buffer once mapped
The MAD layer was violating the DMA API by touching data buffers used
for sends after the DMA mapping was done.  This causes problems on
non-cache-coherent architectures, because the device doing DMA won't
see updates to the payload buffers that exist only in the CPU cache.

Fix this by having all MAD consumers use ib_create_send_mad() to
allocate their send buffers, and moving the DMA mapping into the MAD
layer so it can be done just before calling send (and after any
modifications of the send buffer by the MAD layer).

Tested on a non-cache-coherent PowerPC 440SPe system.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-25 10:51:39 -07:00
Sean Hefty
ae7971a770 [IB] CM: Fix initialization of QP attributes for UC QPs.
Fix cm_init_qp_init_attr(), cm_init_qp_rtr_attr() and cm_init_qp_rts_attr()
so that they correctly handle the differences between UC and RC QPs.  This
fixes problems with setting up UC QPs through the CM.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-24 12:33:56 -07:00
Roland Dreier
ec329a1359 Manual merge of for-linus to upstream (fix conflicts in drivers/infiniband/core/ucm.c) 2005-10-24 10:55:29 -07:00
Roland Dreier
5d7edb3c1a [IB] Add idr_destroy() calls on module unload
Add idr_destroy() calls to the module_exit() functions of the four IB
driver modules that use idrs, so we don't leak idr_layer_cache objects
when these modules are unloaded.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-24 10:53:25 -07:00
Roland Dreier
75eeec2f3f [PATCH] ib: mthca: Always re-arm EQs in mthca_tavor_interrupt()
We should always re-arm an event queue's interrupt in
mthca_tavor_interrupt() if the corresponding bit is set in the event cause
register (ECR), even if we didn't find any entries in the EQ.  If we don't,
then there's a window where we miss an EQ entry and then get stuck because
we don't get another EQ event.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-23 16:38:39 -07:00
Roland Dreier
c8e0ca683d [IB] mthca: Always re-arm EQs in mthca_tavor_interrupt()
We should always re-arm an event queue's interrupt in
mthca_tavor_interrupt() if the corresponding bit is set in the event
cause register (ECR), even if we didn't find any entries in the EQ.
If we don't, then there's a window where we miss an EQ entry and then
get stuck because we don't get another EQ event.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-22 09:43:29 -07:00
Roland Dreier
bbf2078609 [IB] user_mad: Use class_device.devt
Use devt member of struct class_device so that we don't have to create
our own "dev" file in sysfs.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-20 12:54:01 -07:00
Roland Dreier
2e0c512aff [IB] user_mad: trivial coding style fixes
Add spaces after "sizeof" operator to match the rest of file.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-20 12:30:16 -07:00
Roland Dreier
3910f44d79 [IB] cm: Add missing break in switch
Add missing "break" in switch statement.  Without the break, the
CM ended up always falling through and setting every connection
request to use RC transport, which meant that UC connections
didn't work.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-20 12:29:36 -07:00
Jack Morgenstein
7150bf8a98 [IB] mthca: Don't enter QP into MCG more than once.
Avoid entering a QP as member of a multicast group multiple times.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-18 14:46:38 -07:00
Roland Dreier
ba8e931024 [IB] Fail sysfs queries after device is unregistered
We keep IB device structures around until the last sysfs reference is
gone, but we shouldn't ask the low-level driver to do anything after
the LLD unregisters the device.  To handle this, check the reg_state
field and just fail sysfs show() requests if the device has already
been unregistered.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-18 14:14:56 -07:00
Roland Dreier
d476306f1c [IB] mthca: Add struct pci_driver.owner field
Set mthca_driver.owner to THIS_MODULE.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-18 14:07:07 -07:00
Roland Dreier
c6f5cb7be0 [IB] mthca: Use enum in mthca_alloc_db() prototype
Make the type parameter of mthca_alloc_db() be an enum mthca_db_type
instead of an int.  This doesn't have any practical effect but
documents the functions a little better.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-18 13:22:16 -07:00
Roland Dreier
4b2d319b53 [IPoIB] Improve ipoib_timeout() output
Use jiffies_to_msecs() so we print a human-readable time so
we don't have to worry about what HZ is configured to, and
print out a few values to make post-mortem analysis easier.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-18 12:20:06 -07:00
Roland Dreier
762a03e21e [IB] ucm: quiet sparse warnings
Make ctx_id_mutex and ctx_id_table static to quiet sparse warnings.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:38:50 -07:00
Sean Hefty
07d357d0cb [IB] CM: bind IDs to a specific device
Bind communication identifiers to a device to support device removal.
Export per HCA CM devices to userspace.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
2005-10-17 15:37:43 -07:00
Sean Hefty
595e726a1f [IB] merge ucm.h into ucm.c
Eliminate ucm.h.  Replace ucm_dbg with direct call to printk KERN_ERR.

Signed-off-by: Sean Hefty <sean.hefty@intel.com>
2005-10-17 15:33:47 -07:00
Roland Dreier
67cdb40ca4 [IB] uverbs: Implement more commands
Add kernel support for userspace calling poll CQ, request CQ
notification, post send, post receive, post SRQ receive, create AH and
destroy AH commands.  These commands allow us to support userspace
verbs for devices that can't perform these operations directly from
userspace (eg the PathScale HCA).

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:31 -07:00
Roland Dreier
883a99c702 [IB] uverbs: Add a mask of device methods allowed for userspace
Give each device a uverbs_cmd_mask, so that a low-level driver can
control which methods may be called on behalf of userspace.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:30 -07:00
Roland Dreier
56c202d6e4 [IB] fail SA queries if device initialization failed
If the SA query module's initialization fails for a device, then that
device won't have a struct ib_sa_device associated.  We should fail SA
queries in that case, rather than blindly dereferencing the NULL
pointer we get back from ib_get_client_data().

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:29 -07:00
Roland Dreier
305a7e8705 [IB] uverbs: unlock correctly in error paths
A couple of functions were missing spin_unlock calls in error paths.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:29 -07:00
Roland Dreier
5b6810e048 [IPoIB] Rename ipoib_create_qp() -> ipoib_init_qp() and fix error cleanup
ipoib_create_qp() no longer creates IPoIB's QP, so it shouldn't
destroy the QP on failure -- that unwinding happens elsewhere, so the
current code can cause a double free.  While we're at it, the
function's name should match what it actually does, so rename it to
ipoib_init_qp().

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:29 -07:00
Jack Morgenstein
efaae8f71f [IB] mthca: Better limit checking and reporting
Check the sizes of CQs, QPs and SRQs when creating objects, and fail
instead of creating too-big queues.  Also return real limits instead
of just plausible-sounding values from mthca_query_device().

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:29 -07:00
Roland Dreier
4ab6fb7e5b [IB] Fix leak on MAD initialization failure
There is a bug in ib_mad_init_device(): if ib_agent_port_open() fails
for a given port, then the current code doesn't call ib_mad_port_close()
for that port.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:28 -07:00
Roland Dreier
e23d6d2b09 [IB] mthca: detect SRQ overflow
The hardware relies on us keeping one extra work request that never
gets used in SRQs.  Add checks to the SRQ work request posting
functions so that they fail when someone is about to use up that extra
work request, rather than when someone uses the very last work request.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:28 -07:00
Roland Dreier
90f104da22 [IB] mthca: SRQ limit reached events
Our hardware supports generating an event when the number of receives
posted to a shared receive queue (SRQ) falls below a user-specified
limit.  Implement mthca_modify_srq() to arm the limit, and add code to
handle dispatching SRQ events when they occur.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:28 -07:00
Roland Dreier
116c0074ec [IB] Check port number in ib_query_port()/ib_modify_port()
Check port number before passing query_port or modify_port operations
on to device driver.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:27 -07:00
Roland Dreier
f575394f1d [IB] uverbs: reject invalid memory registration permission flags
Reject userspace memory registrations with invalid permission flags:
"local write" is required if "remote write" or "remote atomic" is also
requested.

Pointed out by Jack Morgenstein <jackm@mellanox.co.il>

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:27 -07:00
Jack Morgenstein
9825051e8c [IB] mthca: Fill in more fields in query_port method
Add code to fill in the bad_pkey_cntr, max_mtu, active_mtu and
subnet_timeout fields in mthca_query_port().

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:26 -07:00
Roland Dreier
274c089163 [IB] uverbs: Add device-specific ABI version attribute
Add abi_version attribute to uverbs class devices to allow for
ABI versioning of device-specific interfaces.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
2005-10-17 15:20:26 -07:00