Merge branch '367-driver-source-formatting-to-silence-most-of-the-5-4-linux-kernel-checkpatch-reported-problems' into 'master'

Resolve "driver: source formatting to silence most of the 5.4 Linux kernel checkpatch reported problems"

Closes #367

See merge request !350

checkpatch does not like use of C++ style comments in header
files for some reason.

Identifier GPL-2.0-or-later seems to be recognized, GPL-2.0+
is not recognized correctly.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Some lines are longer than 80 characters to conform alignment rules
of follow-up lines in function definitions/declarations requested
by checkpatch. 80 characters limit has been relaxed to 100 in 5.7
kernel so this should be without problems for mainline inclusion.

Suggestion

  CHECK: Prefer kernel type 'u32' over 'uint32_t'

ignored in auto-generated header files. The files are used in QEMU
and user space tools and correct setup of u32 definition to not
clash in all cases could be problematic.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Added kernel-paths.diff with changes in paths required for build
when included in Linux kernel mainline tree.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Merge branch '366-doc-mention-ctu-can-fd-qemu-emulation-support-and-more-people-support' into 'master'

Resolve "doc: mention CTU CAN FD QEMU emulation support and more people support"

Closes #366

See merge request !349

Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Merge branch '365-driver-fix-long-ago-omitted-byte-swap-in-bing-endian-register-write' into 'master'

Resolve "driver: fix long ago omitted byte swap in bing-endian register write."

Closes #365

See merge request !348

Even on big-endian systems ioread32/iowrite32 accesses memory-mapped
register in little-endian manner to be compatible with PCI devices
so the problem of missing be in iowrite32be for ctucan_hw_write32_be
has not been found during MIPS Malta QEMU test, but it should
be corrected for case of native CTU CAN FD core mapping on big-endian
systems.

Closes #365Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Resolve "driver: build native driver for current running kernel by default"

Closes #364

See merge request !347

When no parameter or environment variable is set when make is
invoked then build driver for current running Linux kernel.

Closes #364Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Merge branch '363-ci-driver-build-fails-after-makefile-relaxed-to-easier-use-on-multiple-architectures' into 'master'

Resolve "ci: driver build fails after makefile relaxed to easier use on multiple architectures."

Closes #363

See merge request !346

When KDIR is specified then new Makefile version does not enforce
ARM target build because it complicates MIPS and native builds.
But when the Linux kernel build directory does not contain
GNUmakefile trick to setup ARCH and CROSS_COMPILE, then
plain make with KDIR fails after relaxation.
Specification of the ARM target explicitly in CI command
is required now.

Fixes #363Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Resolve "Byte order of CAN farme data is wappped on big-endian system."

Closes #362

See merge request !345

Tested on big-endian MIPS Malta system emulated by QEMU
with CTU CAN FD device model.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

This simplifies build for different kernels and architectures,
i.e. test build for big-endian MIPS Malta architecture.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>

Merge branch '361-vhdl-93-with-vivado-compatibility-troubles-rx_triggers-cannot-read-from-output' into 'master'

Resolve "VHDL 93 with Vivado compatibility troubles: rx_triggers cannot read from output"

Closes #361

See merge request !344

This reverts commit 26939ddd.

This reverts commit 5b97442e.

src: Fix trigger shifting. Block original trigger.

Closes #360

See merge request !342

This guarantees proper sending of next bit upon resynchronisation where
PH2 is shortened to 0!

test: Remove prescaler unit test from nightly

Closes #359

See merge request !341

Resolve "Prescaler bug-fixes"

Closes #359

See merge request !340

# Conflicts:
#	src/prescaler/bit_segment_meter.vhd

src,test: Fix No positive resynchronisation.

Closes #357

See merge request !339