Greg Gallagher authored Feb 02, 2018 and Philippe Gerum committed Feb 05, 2018

Some SOCs will have two GPIO chips that require two independent drivers.  In my
case I was working with the Zynq SOC and needed a driver for the gpios on the
FPGA and a driver for the gpios available on the ARM core. This change allows
for two rt gpio drivers to be built and loaded at the same time.
The rt gpio driver will be renamed from xeno-gpio to the following:

xeno-gpio-bcm2835.ko
xeno-gpio-mxc.ko
xeno-gpio-sun8i-h3.ko
xeno-gpio-zynq7000.ko

This also moves the gpio-core logic into the kernel image.

Greg Gallagher authored Feb 02, 2018 and Philippe Gerum committed Feb 05, 2018

If the device name from the device tree starts with a forward slash (/) the
rtdm device stores it in the registry including the forward slash.  When we
go to use that device and do the registry lookup we use a relative path from
/dev/rtdm which doesn't contain the forward slash and fails to find a match.
Open won't return an error but IO calls will fail.  To fix this when we
register an RTDM device skip the first character if it's a forward slash.

In my case the path from the device tree is
        /amba_pl/gpio@41200000/gpio905
which gets stored in the registry. When we want to use the device and look up
the device path in the registry we use
        amba_pl/gpio@41200000/gpio905
which won't find a match in the registry.

Tested on Zynq7000 gpio drivers

Jan Kiszka authored Jan 26, 2018 and Philippe Gerum committed Jan 27, 2018

It can be confusing for people reading a trace when a context switch due
to a wake up in the IRQ handler comes before the handler was left. Move
the exit instrumentation before the schedule, just like we already do in
the clock handler.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>

This thread must wait for work in interruptible mode, does not wait
for a device event. This fixes the bad process time accounting which
otherwise would converge to 1.00 for an idle system when the Cobalt
core is enabled.

rtdm_gpiochip_alloc() is a way to instantiate a single GPIO descriptor
instead of mapping all GPIO controllers defined by the firmware which
belong to a particular type. This routine is matched by
rtdm_gpiochip_remove() for deallocation.

We need the former to return a pointer to the allocated descriptor we
may then pass to the latter.

Some client code may want to deal with GPIO interrupt handling
directly, without support for an irq_chip controller paired with the
GPIO chip.

Assume that a failure to retrieve the IRQ number associated to a pin
means exactly that, offloading interrupt handling to an open coded
implementation the client code should provide for.

In parallel, introduce rtdm_gpiochip_post_event() for notifying the
GPIO core from an incoming interrupt event on a given pin.

For creating a new RTDM-based GPIO chip descriptor (struct
rtdm_gpio_chip), mapped to its regular counterpart (struct gpio_chip).

Marek Vasut authored Dec 19, 2017 and Philippe Gerum committed Dec 19, 2017

Prefix CPPFLAGS and LDFLAGS with AM_ to make automake actually
use those and propagate external CPPFLAGS and LDFLAGS settings
into the build.

Signed-off-by: Marek Vasut <marex@denx.de>

lib/boilerplate/bootstrap.c is meant to be glued to the application
executable for bootstrapping the Cobalt services automatically at load
up.

From the standpoint of its sole author (i.e. me), doing so adds no
requirement with respect to the licensing terms of the containing
application.

Allow lib/boilerplate/bootstrap.c to be distributed under the terms of
the MIT license to reflect this fact.

Leopold Palomo-Avellaneda authored Nov 27, 2017 and Philippe Gerum committed Dec 10, 2017

the local debug variable has a name-clash with a identically named
function in arch/x86/include/asm/traps.h
also a macro guard apparently used an outdated name.

This patch is equivalent than the proposed one by Norbert Lange
for the intel cards.

Greg Gallagher authored Dec 05, 2017 and Philippe Gerum committed Dec 05, 2017

For the zynq platform (and possibly others in the future) we need to modify
gpio-core to request gpio pins before using them. The open() function will
now request a gpio and fail if it's already reserved. This should make
the pin request transparent to the user. The ability to request and release
pins is also available in an ioctrl message.  Tested on Microzed Zynq-7010
platform and the raspberry pi 2 board.

Greg Gallagher authored Dec 04, 2017 and Philippe Gerum committed Dec 05, 2017

split rtdm_fd_enter, move the functionality where we store the fd
until after the open() call succeeds.  Calls where open() fail a fd is
left in the tree even after the cleanup code is executed.  If this fd
number is used again we will fail the call to open until a different
fd is used.  This patch addresses this situation by not adding the fd
into the tree until open has succeeded and the fd is valid.

git repository hosting authored Dec 01, 2017 and Philippe Gerum committed Dec 01, 2017

Module: xenomai-jki
Branch: for-forge
Commit: dfec6f1f9448b4e207e36280a0e48923687a1cee
URL:    http://git.xenomai.org/?p=xenomai-jki.git;a=commit;h=dfec6f1f9448b4e207e36280a0e48923687a1cee



Author: Jan Kiszka <jan.kiszka@siemens.com>
Date:   Thu Nov 30 21:39:44 2017 +0100

cobalt/rtdm: Avoid dereferencing NULL device in __rtdm_dev_ioctl_core

File descriptors created by message queues or timerfds do not have
devices associated with them. Dereferencing that pointer prior to
checking if we have device fd in the first place is a recipe for
crashes.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>

TLSF's init_memory_pool() wants heaps to be larger than 4k on 64bit
architectures, due to the increased size of the meta-data compared to
their 32bit counterpart (struct tlsf).

Use 8k as the minimum heap size instead of PAGE_SIZE in the pshared
case, and make sure the main pool size amounts to at least 8k in the
process-private case, so that init_memory_pool() never fails in
heapobj_pkg_init_private().

Norbert Lange authored Nov 21, 2017 and Philippe Gerum committed Nov 22, 2017

the local debug variable has a name-clash with a identically named
function in arch/x86/include/asm/traps.h
also a macro guard apparently used an outdated name

Signed-off-by: Norbert Lange <norbert.lange@andritz.com>

This fixes a regression introduced by #4b4001c7.

Often reported as a failure to build Xenomai over Debian Stretch,
which ships with gcc6 built with --enable-default-pie, ld's -Ur option
as given by wrap-link.sh does not mix well with -pie which is
implicitly enabled, causing the linker to bail out on error.

Since this script eventually completes the link stage to produce a
fully resolved executable, collecting the ctors/dtors and resolving
all references from the partially linked object file is implied
anyway, making -Ur useless in stage2 at best.

For details, see:
http://xenomai.org/pipermail/xenomai/2017-October/037799.html
https://xenomai.org/pipermail/xenomai/2017-August/037648.html
http://www.xenomai.org/pipermail/xenomai/2017-May/037312.html

We need unwind tables to be precise at the instruction level, so they
can be used for accurate stack unwinding (e.g. backtrace()) from
asynchronous events like signals. Slackspot depends on this ability.

Failure to do so may cause applications to report incomplete
backtraces with glibc's backtrace(), or even crash over libunwind, see
http://xenomai.org/pipermail/xenomai/2017-October/037763.html

Enable -fasynchronous-unwind-tables for building the Xenomai libraries
and the user applications (via xeno-config --cflags).

Report and suggestion for a fix by Andreas Glatz
<andi.glatz@gmail.com>.

We do not want a syscall which is neither a Cobalt nor a Linux syscall
to trigger a switch to secondary mode when issued from primary
mode. Catch the situation early, returning ENOSYS.