Simplify and optimize the logic in trace_buffer_iter() to use a conditional
operation instead of an if conditional.
Link: http://lkml.kernel.org/r/20180408113631.3947-1-cugyly@163.com
Signed-off-by: yuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The comment in create_filter() states that the passed in filter pointer
(filterp) will either be NULL or contain an error message stating why the
filter failed. But it also expects the filter pointer to point to NULL when
passed in. If it is not, the function create_filter_start() will warn and
return an error message without updating the filter pointer. This is not
what the comment states.
As we always expect the pointer to point to NULL, if it is not, trigger a
WARN_ON(), set it to NULL, and then continue the path as the rest will work
as the comment states. Also update the comment to state it must point to
NULL.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
'err' is used as a NUL-terminated string, but using strncpy() with the length
equal to the buffer size may result in lack of the termination:
kernel/trace/trace_events_hist.c: In function 'hist_err_event':
kernel/trace/trace_events_hist.c:396:3: error: 'strncpy' specified bound 256 equals destination size [-Werror=stringop-truncation]
strncpy(err, var, MAX_FILTER_STR_VAL);
This changes it to use the safer strscpy() instead.
Link: http://lkml.kernel.org/r/20180328140920.2842153-1-arnd@arndb.de
Cc: stable@vger.kernel.org
Fixes: f404da6e1d ("tracing: Add 'last error' error facility for hist triggers")
Acked-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
On repeated module load/unload cycles, its possible for the pvrmda driver
to encounter this crash:
...
[ 297.032448] RIP: 0010:[<ffffffff839e4620>] [<ffffffff839e4620>] netdev_walk_all_upper_dev_rcu+0x50/0xb0
[ 297.034078] RSP: 0018:ffff95087780bd08 EFLAGS: 00010286
[ 297.034986] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff95087a0c0000
[ 297.036196] RDX: ffff95087a0c0000 RSI: ffffffff839e44e0 RDI: ffff950835d0c000
[ 297.037421] RBP: ffff95087780bd40 R08: ffff95087a0e0ea0 R09: abddacd03f8e0ea0
[ 297.038636] R10: abddacd03f8e0ea0 R11: ffffef5901e9dbc0 R12: ffff95087a0c0000
[ 297.039854] R13: ffffffff839e44e0 R14: ffff95087a0c0000 R15: ffff950835d0c828
[ 297.041071] FS: 0000000000000000(0000) GS:ffff95087fc00000(0000) knlGS:0000000000000000
[ 297.042443] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 297.043429] CR2: ffffffffffffffe8 CR3: 000000007a652000 CR4: 00000000003607f0
[ 297.044674] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 297.045893] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 297.047109] Call Trace:
[ 297.047545] [<ffffffff839e4698>] netdev_has_upper_dev_all_rcu+0x18/0x20
[ 297.048691] [<ffffffffc05d31af>] is_eth_port_of_netdev+0x2f/0xa0 [ib_core]
[ 297.049886] [<ffffffffc05d3180>] ? is_eth_active_slave_of_bonding_rcu+0x70/0x70 [ib_core]
...
This occurs because vmw_pvrdma on probe stores a pointer to the netdev
that exists on function 0 of the same bus/device/slot (which represents
the vmxnet3 ethernet driver). However, it never removes this pointer if
the vmxnet3 module is removed, leading to crashes resulting from use after
free dereferencing incidents like the one above.
The fix is pretty straightforward. vmw_pvrdma should listen for
NETDEV_REGISTER and NETDEV_UNREGISTER events in its event listener code
block, and update the stored netdev pointer accordingly. This solution
has been tested by myself and the reporter with successful results. This
fix also allows the pvrdma driver to find its underlying ethernet device
in the event that vmxnet3 is loaded after pvrdma, which it was not able to
do before.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: ruquin@redhat.com
Tested-by: Adit Ranadive <aditr@vmware.com>
Acked-by: Adit Ranadive <aditr@vmware.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Currently the driver sets the mask of the gre_protocol to 0xffff
without consideration in the user request.
Fix it by copy the mask from the verbs spec.
Fixes: da2f22ae77 ("IB/mlx5: Add support for GRE flow specification")
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Reviewed-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Intel Ice Lake has the same SMBus host controller than Intel Cannon
Lake. Add the PCI ID to the drivers list of supported devices.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
[jarkko.nikula@linux.intel.com: Add entries to Documentation and Kconfig]
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Decrement the number of elements in the map in case the allocation
of a new node fails.
Fixes: 6c90598174 ("bpf: pre-allocate hash map elements")
Signed-off-by: Mauricio Vasquez B <mauricio.vasquez@polito.it>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Make list of supported chipsets a little bit shorter by consolidating
names that belong to the same family.
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
In order to support advanced features, the I2C FIFO interface was
changed in the version of the Tegra I2C controller found in Tegra194.
The changes are backwards incompatible, so the driver needs to be
programmed in a slightly different way on new chips.
Add support for MST FIFO programming and add an OF match entry for
Tegra194. At the same time, mark all prior generations of this
controller as not having the MST FIFO interface.
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Add another injector for an incomplete transfer. As mentioned in the
docs, this one is important to check bus recovery algorithms with it.
Otherwise random data may be sent to devices!
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
For a ppgtt that we are constructing, there is no struct_mutex
dependence so skip it. In the process, also ping the scheduler
frequently to try and avoid the NMI watchdog.
v2: gen6 requires struct_mutex to clean up (currently)
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
References: https://bugs.freedesktop.org/show_bug.cgi?id=107094
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180703135331.12265-1-chris@chris-wilson.co.uk
Make the incomplete_transfer routine reusable, so we can add other test
cases with different patterns later. Prepare the docs for that, too.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Trivial added debug print for dev->clk_freq doesn't necessarily tell the
actual bus speed or mode the controller is operating. For instance it
may indicate 1 MHz Fast Mode Plus or 3.4 MHz High Speed but driver ends up
using 400 kHz Fast Mode due missing timing parameters or missing support
from HW.
Add a debug print that prints the bus speed based on the validated speed
that gets programmed into a HW.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
SDA hold time is an important timing parameter and often reason for
arbitration lost errors if not set to a correct value. Add a debug print
for it in order to see what value gets programmed to a HW.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Mixed timing parameter validation, calculation and their debug prints
with HW initialization in i2c_dw_init_master() and i2c_dw_init_slave()
as been bothering me some time.
It makes function a little bit unclear to follow, doesn't show what steps
are needed to do only once during probe and what are needed whenever HW
needs to be reinitialized. Also those debug prints show information that
doesn't change runtime and thus are also needlessly printed multiple times
whenever HW is reinitialized.
Thus let the i2c_dw_init_master() and i2c_dw_init_slave() to do only HW
initialization and move out one time parameter setting and debug prints
to separate functions which are called only during probe.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
SDA hold time configuration is common to both master and slave code. It
is also something that can be done once during probe and do only
register write when HW needs to be reinitialized.
Remove duplication and move SDA hold time configuration to common code.
It will be called from slave probe and for master code from a new
i2c_dw_set_timings_master() to where we will populate more probe time
timing parameter setting.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
This is rather readability update than micro-optimization, or if not
optimization at all. We take the input clock rate to a variable and pass
that to SCL timing parameter calculation functions.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
It looks ___constant_swab32 is an internal byte swap detail. Switch to
swap32() and include header file for it explicitly. Now it comes
implicitly via linux/i2c.h include.
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Move register access detection out from master and slave HW
initialization code to common code. Motivation for this is to have
register access configured before HW initialization and remove
duplicated code.
This allows to do further separation between probe time initialization
and runtime reinitialization code.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Commit 43df43e6ba ("i2c: designware: Don't set SCL timings and speed
mode when in slave mode") made variable comp_param1 needless in
i2c_dw_init_slave(). Remove it now.
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
live_gtt is a very slow test to run, simply because it tries to allocate
and use as much as the 48b address space as possibly can and in the
process will try to own all of the system memory. This leads to resource
exhaustion and CPU starvation; the latter impacts us when the NMI
watchdog declares a task hung due to a mutex contention with ourselves.
This we can prevent by releasing the struct_mutex and forcing our
i915/rcu workers to run, and in particular flushing the freed object
worker that is the cause for concern.
References: https://bugs.freedesktop.org/show_bug.cgi?id=107094
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180703101829.7360-1-chris@chris-wilson.co.uk
Calling the __i2c_smbus_xfer wrapper in __i2c_mux_smbus_xfer provides
retries and thus makes the parent-locked case consistent with the both
mux-locked (i2c_mux_smbus_xfer) and the I2C transfer cases.
Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
This simplifies the code, and you get retries for free if the adapter
does not support ->master_xfer.
Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
This simplifies the code, and you get retries for free if the adapter
does not support ->master_xfer.
Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
This simplifies the code, and you get retries for free if the adapter
does not support ->master_xfer.
Signed-off-by: Peter Rosin <peda@axentia.se>
Acked-by: Michael Shych <michaelsh@mellanox.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Removes all locking from i2c_smbus_xfer and renames it to __i2c_smbus_xfer,
then adds a new i2c_smbus_xfer function that simply grabs the lock while
calling the unlocked variant.
This is not perfectly equivalent, since i2c_smbus_xfer was callable from
atomic/irq context if you happened to end up emulating SMBus with an I2C
transfer, and that is no longer the case with this patch. It is unknown
(to me) if anything depends on that quirk, but it seems fragile enough to
simply break those cases and require them to call i2c_transfer directly
instead.
While at it, for consistency rename the 2nd to last argument (size) of
the i2c_smbus_xfer declaration to protocol and remove the surplus extern
marker.
Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
People have gone to all the effort of writing kernel-doc for these
functions; the least we can do is put them in the "Other functions"
part of the VFS documentation.
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The general interrupt handler is_rcv_avail_int() has two paths,
do_interrupt() (callback) and handle_user_interrupt(). The
do_interrupt() callback is for the threaded receive handling.
is_rcv_avail_int() cannot handle threaded IRQs.
If the do_interrupt() path is taken, and the IRQ returns
IRQ_WAKE_THREAD, the IRQ behavior will be indeterminate.
Remove incorrect call to do_interrupt() from is_rcv_avail_int(),
leaving the un-threaded (handle_user_interrupt()) path.
Fixes: f4f30031c3 ("staging/rdma/hfi1: Thread the receive interrupt.")
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Kamenee Arumugam <kamenee.arumugam@intel.com>
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The in_use_ctxts bitmask is for user receive contexts only. Setting it for
any other type of receive context is incorrect.
Move initial set of in_use_ctxts bits from the general context init to the
user context specific init. Having this bit set can allow contexts to be
incorrectly identified by some IRQ handlers. This will allow
handle_user_interrupt() will now filter user contexts correctly.
Clean up redundant is_rcv_urgent_int() user context check.
A follow on patch will clean up an incorrect code path in the
is_rcv_avail_int().
Fixes: 8737ce95c4 ("IB/hfi1: Fix an assign/ordering issue with shared context IDs")
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Kamenee Arumugam <kamenee.arumugam@intel.com>
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The LM36922 has one output but can sync current to 2 LED
strings. The user may only use one sync so the other
syncs need to be disabled.
The LM36923 has 3 LED syncs.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Update the binding to be able to configure the LED
sync to turn on.
The LM36922 has one output but can sync current to 2 LED
strings. The user may only use one sync so the other
syncs need to be disabled.
The LM36923 has 3 LED syncs.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Update the code to use the fwnode calls as opposed
to ARM DT node api's.
This allows the device to be used with either DT
configurations or ACPI definitions.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Update the license header to the current
SPDX licensing format.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Avoid that the compiler complains about set-but-not-used variables when
building with W=1. This patch does not change any functionality.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Remove the driver version and release date information because such
information is not relevant for an upstream driver. See also commit
e1267b0124 ("RDMA: Remove useless MODULE_VERSION").
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Variable mode is assigned to pdata->led_pdata->mode[led->id] and yet
is not being used when calling function max8997_led_set_mode. Fix
this by using mode when calling max8997_led_set_mode.
Cleans up clang warning:
warning: variable 'mode' set but not used [-Wunused-but-set-variable]
Fixes: 8584cb82f1 ("leds: Add suuport for MAX8997-LED driver")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Update the email address in the module information and in the comments
and in the module information, and drop the existing boilerplate in
favor of a SPDX header.
Also sync the module license information with the header.
Signed-off-by: Daniel Mack <daniel@zonque.org>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
The binding details are described in an earlier commit that adds the
documentation.
Signed-off-by: Daniel Mack <daniel@zonque.org>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Clean up the code a bit and transition over to the gpiod based interface.
Signed-off-by: Daniel Mack <daniel@zonque.org>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
In preparation to DT probe functionality, merge create_lt3593_led() into
its only call-site. The DT based setup code will be quite different, so this
internal helper function is of no help.
This also changes the way the driver works by only handling one entry inside
'struct gpio_led_platform_data'. If multiple devices of the same type are
found in a design, there should be a platform device for each of them. The
only mainline user of this driver is not affected by this change.
Last, use devm_led_classdev_register() instead of
led_classdev_register(), so the driver's remove callback can go away.
Signed-off-by: Daniel Mack <daniel@zonque.org>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
This patch adds the bindings document for LT3593 LED drivers.
The binding is kept consistent with other LED driver bindings in that it
stores all the LED-specific properties in its own subnode. As the hardware
only supports one channel, there can consequently only be one sub-node.
Signed-off-by: Daniel Mack <daniel@zonque.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
IB_MULTICAST_LID_BASE is defined as follows:
#define IB_MULTICAST_LID_BASE cpu_to_be16(0xC000)
Hence use be16_to_cpu() to convert it to CPU endianness. Compile-tested
only.
Fixes: af808ece5c ("IB/SA: Check dlid before SA agent queries for ClassPortInfo")
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com>
Cc: Mike Marciniszyn <mike.marciniszyn@intel.com>
Cc: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The conversion from offsetof() calculations to sizeof()
wrongly behaved for missed exact size and in scenario with
more than one flow.
In such scenario we got "create flow failed, flow 10: 8 bytes
left from uverb cmd" error, which is wrong because the size of
kern_spec is exactly 8 bytes, and we were not supposed to fail.
Cc: <stable@vger.kernel.org> # 3.12
Fixes: 4fae7f1704 ("RDMA/uverbs: Fix slab-out-of-bounds in ib_uverbs_ex_create_flow")
Reported-by: Ran Rozenstein <ranro@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The cros-ec I2C and SPI transport drivers have been moved from MFD
subsystem to platform/chrome, at the same time, the config symbol
has been renamed and lost the MFD_ prefix So, update the help message
accordingly.
Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Benson Leung <bleung@chromium.org>
The cros-ec I2C and SPI transport drivers have been moved from MFD
subsystem to platform/chrome, at the same time, the config symbol has
been renamed and lost the MFD_ prefix, so update all configs to the new
config symbol name.
Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Acked-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Benson Leung <bleung@chromium.org>
There are some cros-ec transport drivers (I2C, SPI) living in MFD, while
others (LPC) living in drivers/platform. The transport drivers are more
platform specific. So, move the I2C and SPI transport drivers to the
platform/chrome directory. The patch also removes the MFD_ prefix of
their Kconfig symbols.
Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Benson Leung <bleung@chromium.org>
