The SOC code can set whether it's known that the CPU
is immune, don't default to turning this off.
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Retpolines were never completely implemented, even on x86.
Move this particular Kconfig to only concern itself with
the assembly code, and don't default it on ever since we
prefer SSBD instead.
We can restore the common kernel-wide CONFIG_RETPOLINE once
we have an end-to-end implementation.
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
- Remove redundant dependencies on the PWM symbol (which show up as
PWM && PWM in the documentation). The 'source's in
drivers/pwm/Kconfig are already within an 'if PWM' block.
- Turn some repeated 'depends on FOO' into 'if FOO' blocks.
- Turn some 'if FOO's that surround a single symbol into
'depends on FOO'.
'if FOO' is equivalent to adding a 'depends on FOO' to each symbol
within the 'if'.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
The feature that used this symbol was removed in 18b07e09e0.
This is just a cleanup commit that removes the unused symbol.
Signed-off-by: Tomasz Gorochowik <tgorochowik@antmicro.com>
This test was written to wait on a fifo with a timeout, return, and
check the timing between the start and end using k_cycle_get_32() to
see that it didn't run long. But timeouts expire on tick boundaries,
and so if tick expires between the start of the test and the entry to
k_fifo_get(), the timeout will take one full tick longer than expected
due to aliasing.
As it happened this passed everywhere except nRF (whose cycle timer is
32 kHz and thus more susceptible to coarser aliasing like this), and
even there it passed for a while until the spinlock validation layer
went in and added just enough time to the userspace code paths
(i.e. the code between the start time fetch and the point where the
fifo blocks takes longer) to open the window and push us over the
limit.
The workaround here is just to add a k_sleep(1) call, which is
guaranteed to block and wake up synchronously at the next tick.
Fixes#13289
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
nRF52810 Flash and SRAM sizes do not have power-of-two values.
This commit reduces the amount of MPU regions reserved at boot
time to enforce basic permissions in Flash and SRAM memory. It
does this by selecting the next power-of-two size for the Flash
and SRAM regions, respectively. In this way we can cover the
entire Flash and SRAM space each with a single MPU region, thus,
save two MPU regions to be used by the kernel/application.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
These options were removed from kconfig in previous patch. So rename
the leftovers as shared_irq driver options are defined in DTS now.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
These options were removed from kconfig in previous patch. So rename
the leftovers as shared_irq driver options are defined in DTS now.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
These options were removed from kconfig in previous patch. So rename
the leftovers as shared_irq driver options are defined in DTS now.
Fix#13746
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Some old peer central controllers respond with Unknown
Response PDU to our local Slave Initiated Feature Request
after the peer initiated the Encryption Setup Procedure.
The peer has voilated the Bluetooth Specification by sending
the Unknown Response during the Encryption Setup Procedure,
but as a workaround to IOP with such controller our local
implementation is updated to not disconnect the connection
but accept the voilating Unknown Response PDU.
Relates to #12362.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Following changes done:
* While looking through generated net_buf HTML documentation
I noticed that some of the macros were not documented.
* Removed extern variable declarations from generated
documentation (because those variables were not documented).
* Replaced "/** @brief xxx" by "/**\n * @brief xxx" as
checkpatch complained about them (@brief being in the same
line as the start of the block comment).
* Went through all the block comments and made them look similar
and removed extra space character.
* Removed duplicate lines from function documentations. So
if the @brief text is the same as the detailed one, then the
generated output was looking funny.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
CCC cfg value was not being cleared on disconnection of a non bonded
peer if another peer was connected, had non zero value and occurred
earlier in the ccc cfg array.
Signed-off-by: Nick Ward <nick.ward@setec.com.au>
These options are oftentimes used when implementing servers, and
thus required to port existing socket apps. These options are also
safe to just ignore, e.g. SO_REUSEADDR has effect only for repeated
recreation of server socket (not an expected usecase for a Zephyr
app), while TCP_NODELAY is effectively the default for Zephyr, as we
don't implement TCP buffering (aka Nagle algorithm).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
These can be used as getsockopt/getsockopt params and required for
compatibility with existing socket applications.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
We expect that identifier in yaml will match board name for sanitycheck.
Change the identifier to match so sanitycheck runs propertly.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Checking the return of fs_stat to ensure that there is not hidden error.
Problem spotted by coverity.
CID 190949
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Multiple flag bits were set so the ACK flag set was not checked
properly which meant that connection establishment was not
successfull.
Fixes#13943
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Repalces send fifo_fill implementation with poll_out,
this makes the receiver usefull on most SoC.
Signed-off-by: Georgij Cernysiov <g.cernysiov@elco-automation.de>
Rearranges functions to public and private groups,
and adds missing function comments.
Signed-off-by: Georgij Cernysiov <g.cernysiov@elco-automation.de>
Log possible uart device binding error and change
the error type to a more correct one.
Signed-off-by: Georgij Cernysiov <g.cernysiov@elco-automation.de>
No need to hide the symbols in the header file if CONFIG_NET_LLDP
is not enabled. This also allows the documentation to be generated
properly.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The Link Layer Discovery Protocol chapter was missing Overview
and also the API reference was not populated properly.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Port all users of 'set_conf_file' to use the built-in rules
instead. This follows the convention-over-configuration principle to
make the system as a whole simpler and more consistent.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
In-tree, all usage of 'set_conf_file' is implementing the same
pattern (with some inconsequential differences).
To reduce copy-paste and get closer to dropping support for
'set_conf_file', we include this pattern into the built-in set of
patterns.
'set_conf_file' is causing issues for future extensions,
e.g. multi-image support, and is also the source of a lot of
duplicated code.
Note support for 'set_conf_file' is not dropped, it is only marked as
deprecated.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
The return of hwinfo_get_device_id is a signed size and it returns a
negative number in case of error. This test was using an unsigned
variable invalidating the errror check.
CID 190929
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Zephyr Continuous Integration made the choice of rebasing pull requests
on its fast moving master branch before testing them. Unfortunately the
Shippable interface assumes the other choice[*] and highlights the SHA1
of the pull request _before_ rebasing which is very misleading. For now
the only way to realize this disconnect is to scrutinize the build logs
and/or this source code (or tribal knowledge).
To mitigate and reduce a bit this confusion, run a short:
git log -n 5 --oneline --decorate
before and after rebase.
Making it worse, Shippable sub-runs are not pegged to a common base
version. For instance the reason why shippable sub-run 35578.1 passed
while all other 35578.x sub runs failed is very likely because PR #13803
was merged to master in the mean time. This commit would have been able
to prove that.
[*] In an ideal world with infinite time and resources, Continuous
Integration would test pull requests BOTH alone and combined with the
moving tip.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
On (at least) Fedora "pytest" is python2 and python2 only.
Also update remote name in sample invocation from "upstream" to "origin"
because that's how west sets things up by default (and this script
requires west)
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
MISRA-C says that char type should not be used in arithmetically as the
data doesn't represent numbers.
MISRA-C rules 10.1 and 10.2
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Variable evt_prop could be used without being initialized. This problem
was spotted by coverity.
CID 190970
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
- PR #13722 moved 3 gen_*.py to /arch/x86/ without updating CODEOWNERS
- PR #13745 forgot a trailing slash in /dts/bindings/can
- PR #13675 forgot a trailing slash in /samples/boards/intel_s1000_crb
No idea why github doesn't keep reporting these every time and
impossible to tell as CI hides most of its logs.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Console subsystem doesn't depend on older consoles-in-drivers, the
only common thing between them is CONFIG_UART_CONSOLE_ON_DEV_NAME
setting, so make it so (by depending on either UART_CONSOLE or
CONSOLE_SUBSYS.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Potentially overflowing expression 1U << register_demand
I changed to (u64_t)1U << register_demand
To avoid overflow, I casted 1U to type u64_t.
Coverity-CID: 190996
Fixes: #13829
Signed-off-by: Maksim Masalski <maxxliferobot@gmail.com>
The newer series of timer drivers will compare counters vs. the last
tick boundary to compute a number of ticks to announce to the kernel.
In the case of CONFIG_TICKLESS=n, this actually represents a change of
behavior from our old scheme where "ticks" always reflected the number
of interrupts received.
The distinction only matters when an interrupt is delayed more than a
full tick, of course. But that actually makes a difference to some
timekeeping code. Restore the old behavior.
This also has the benefit of further reducing code size when !TICKLESS
and improving performance of the ISR by removing the division
(remember Cortex M0 has no hardware divide!).
Fixes#12409
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>