The current INIT_HCA firmware command timeout is sufficient for the
default number of resources (QPs, CQs, etc) being allocated, but if
the HCA profile is modified to increase the amount of resources, then
a spurious timeout is detected and HCA initialization fails.
Increase the timeout for the INIT_HCA command to 10 seconds, which
also brings it into line with all the other command timeouts.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use the same CQ for CM send completions as for all other IPoIB
completions. This means all completions are processed via the same
NAPI polling routine. This should help reduce the number of
interrupts for bi-directional traffic (such as TCP) and fixes "driver
is hogging interrupts" errors reported for IPoIB send side, e.g.
<https://bugs.openfabrics.org/show_bug.cgi?id=508>
To do this, keep a per-interface counter of outstanding send WRs, and
stop the interface when this counter reaches the send queue size to
avoid CQ overruns.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit 9ead190b ("IB/uverbs: Don't serialize with ib_uverbs_idr_mutex")
rewrote how userspace objects are looked up in the uverbs module's
idrs, and introduced a severe bug in the process: there is no checking
that an operation is being performed by the right process any more.
Fix this by adding the missing check of uobj->context in __idr_get_uobj().
Apparently everyone is being very careful to only touch their own
objects, because this bug was introduced in June 2006 in 2.6.18, and
has gone undetected until now.
Cc: stable <stable@kernel.org>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add sanity checks to send queue sizes passed in from userspace. The
minimum sq stride value below is taken from the MT25408 PRM (section
11.10, Table 306, log_sq_stride definition).
Without this check, userspace can submit arbitrarily large/small
values for the number of WQEs and the stride, which can crash the
kernel.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
It's too hard to figure out what "!likely(...)" really means, and who
knows how compilers interpret the hint.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ehca_shca.hca_cap_mr_pgsize now contains all supported page sizes ORed
together. This makes some checks easier to code and understand, plus
we can return this value verbatim in query_hca(), fixing a problem
with SRP (reported by Anton Blanchard -- thanks!).
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Simplify ehca_encode_hwpage_size(), fixing an infinite loop for pgsize == 0
in the process. Fix the bug in alloc_fmr() that triggered the loop.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Because hardware reports the SRQ token in RWQEs of SRQ base QPs, supply the
base QP token as SRQ token, so we can properly find the SRQ base QP.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use round_jiffies() to align the 1 second ah_reap_task with other work
and potentially save power by sleeping cores for longer.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Deadlock condition reported by Kanoj Sarcar <kanoj@netxen.com>.
The deadlock occurs when a connection request arrives at the same
time that a wildcard listen is being destroyed.
A wildcard listen maintains per device listen requests for each
RDMA device in the system. The per device listens are automatically
added and removed when RDMA devices are inserted or removed from
the system.
When a wildcard listen is destroyed, rdma_destroy_id() acquires
the rdma_cm's device mutex ('lock') to protect against hot-plug
events adding or removing per device listens. It then tries to
destroy the per device listens by calling ib_destroy_cm_id() or
iw_destroy_cm_id(). It does this while holding the device mutex.
However, if the underlying iw/ib CM reports a connection request
while this is occurring, the rdma_cm callback function will try
to acquire the same device mutex. Since we're in a callback,
the ib_destroy_cm_id() or iw_destroy_cm_id() calls will block until
their callback thread returns, but the callback is blocked waiting for
the device mutex.
Fix this by re-working how per device listens are destroyed. Use
rdma_destroy_id(), which avoids the deadlock, in place of
cma_destroy_listen(). Additional synchronization is added to handle
device hot-plug events and ensure that the id is not destroyed twice.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
If a user allocates a QP on an rdma_cm_id, the rdma_cm will automatically
transition the QP through its states (RTR, RTS, error, etc.) While the
QP state transitions are occurring, the QP itself must remain valid.
Provide locking around the QP pointer to prevent its destruction while
accessing the pointer.
This fixes an issue reported by Olaf Kirch from Oracle that resulted in
a system crash:
"An incoming connection arrives and we decide to tear down the nascent
connection. The remote ends decides to do the same. We start to shut
down the connection, and call rdma_destroy_qp on our cm_id. ... Now
apparently a 'connect reject' message comes in from the other host,
and cma_ib_handler() is called with an event of IB_CM_REJ_RECEIVED.
It calls cma_modify_qp_err, which for some odd reason tries to modify
the exact same QP we just destroyed."
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Architectures such as ia64 see alignment traps when doing a 64-bit
read from __be32 doorbell[2] arrays to do doorbell writes in
mthca_write64(). Fix this by just passing the two halves of the
doorbell value into mthca_write64(). This actually improves the
generated code by allowing the compiler to see what's going on better.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
* 'release' of git://lm-sensors.org/kernel/mhoffman/hwmon-2.6: (53 commits)
hwmon: (vt8231) fix sparse warning
hwmon: (sis5595) fix sparse warning
hwmon: (w83627hf) don't assume bank 0
hwmon: (w83627hf) Fix setting fan min right after driver load
hwmon: (w83627hf) De-macro sysfs callback functions
hwmon: Add new combined driver for FSC chips
hwmon: (ibmpex) Release IPMI user if hwmon registration fails
hwmon: (dme1737) Add sch311x support
hwmon: (dme1737) group functions logically
hwmon: (dme1737) cleanups
hwmon: IBM power meter driver
hwmon: (coretemp) Add support for Celeron 4xx
hwmon: (lm87) Disable VID when it should be
hwmon: (w83781d) Add individual alarm and beep files
hwmon: VRM is not read from registers
MAINTAINERS: update hwmon subsystem git trees
hwmon: Fix the code examples in documentation
hwmon: update sysfs interface document - error handling
hwmon: (thmc50) Fix a debug message
hwmon: (thmc50) Don't create temp3 if not enabled
...
- killed paranoid NULL Pointer check
- human readable LED states
- support for "Eicon DIVA USB 4.0" (0x071d/0x1005)
Signed-off-by: Martin Bachem <info@colognechip.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
if your mask is host-endian, you should apply it after le64_to_cpu();
if it's little-endian - before. Doing both (for the same mask and
little-endian value) is broken.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
deal with signedness of the stuff passed to set_bit() et.al.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
... since that sucker is not 32bit-only and on 64bit skb->tail is an
offset, not a pointer.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: (21 commits)
HID: hidraw_connect() memleak fix
HID: add hidraw interface
USB HID: provide hook for hidraw write()
HID: hiddev: Add 32bit ioctl compatibilty
HID: Add GeneralTouch touchscreen to the blacklist
HID: add support for Microsoft Wireless Laser Keyboard 6000
Input: add KEY_LOGOFF
USBHID: report descriptor fix for MacBook JIS keyboard
HID: trivial fixes in hid-debug
HID: fix input mapping for Microsoft Ergonomic Keyboard
HID: use hid-plff driver for GreenAsia 0e8f:0003 devices
USBHID: Add HID_QUIRK_NOGET for ELO Touch Screen 2700 display
HID: enable hiddev for the SantaRosa MacBookPro IR receiver
USBHID: add CM109 device to blacklist
HID: Report usage codes of keys as EV_MSC scancode events
HID: ignore all non-LED usages in output fields in hid-input
HID: fix whitespace damage
HID: add support for Thrustmaster FGT Force Feedback wheel
HID: minimal autosuspend support for USB HID devices
HID: add support for Microsoft Natural Ergonomic Keyboard 4000
...
* git://git.infradead.org/mtd-2.6:
[MTD] fix mtdconcat for subpage-write NAND
[MTD] [OneNAND] Avoid deadlock in erase callback; release chip lock first.
[MTD] [OneNAND] Return only negative error codes
[MTD] [OneNAND] Synchronize block locking operations
UBI: return correct error code
UBI: remove useless inlines
UBI: fix atomic LEB change problems
UBI: use byte hexdump
UBI: do not use vmalloc on I/O path
UBI: allocate memory with GFP_NOFS
UBI: use linux print_hex_dump(), not home-grown one
UBI: don't use array index before testing if it is negative
UBI: add more prints
UBI: fix sparse warnings
UBI: fix leak in ubi_scan_erase_peb
Using readw() and friends => needs to pull io.h and not all targets are
doing that via indirect chains.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It looks like hidraw_connect() is leaking memory in case of failure.
Also it should return -ENOMEM when kzalloc fails.
Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
hidraw is an interface that is going to obsolete hiddev one
day.
Many userland applications are using libusb instead of using
kernel-provided hiddev interface. This is caused by various
reasons - the HID parser in kernel doesn't handle all the
HID hardware on the planet properly, some devices might require
its own specific quirks/drivers, etc.
hiddev interface tries to do its best to parse all the received
reports properly, and presents only parsed usages into userspace.
This is however often not enough, and that's the reason why
many userland applications just don't use hiddev at all, and
rather use libusb to read raw USB events and process them on
their own.
Another drawback of hiddev is that it is USB-specific.
hidraw interface provides userspace readers with really raw HID
reports, no matter what the low-level transport layer is (USB/BT),
and gives the userland applications all the freedom to process
the HID reports in a way they wish to.
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
This allows the mtdconcat driver to work with NAND flash devices that
support sub-page writes.
Signed-off-by: Chris Paulson-Ellis <chris@edesix.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
When the erase callback performs some other action on the flash, it's
highly likely to deadlock unless we actually release the chip lock
before calling it.
This patch mirrors that same change already done for NAND.
Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
The OneNAND driver was confusing JFFS2 by returning positive error
codes.
Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Ensure OneNAND's block locking operations are synchronized
like all other operations.
Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
