We don't have C++ unsafe headers (anymore).
I added a test to fix_headers.sh that checks if certain "unsafe"
headers are ONLY included inside a __BEGIN_DECLS ... __END_DECLS
(because after all, they are unsafe), as well as checking that
no other header files are included inside such a block. The rationale
of the latter is that if a file is a C header and it declares
function prototypes (otherwise it doesn't matter) and is sometimes
included outside a __BEGIN_DECLS ... __END_DECLS block (from a C++
source file) then it has to be C++ safe and doesn't ever to be
included from inside such a block; while if a file is a C++ header
then obviously it should never be included from such a block.
fix_headers.sh subsequently found several safe headers to be
included from inside such a block, and those that were (apparently
in the past) unsafe were included only sometimes inside such a
block and often outside it. I had a look at those files and saw
that at least an attempt has been made to make them C++ safe,
but especially because they already are included OFTEN outside
a __BEGIN_DECLS ... __END_DECLS (from C++ source files) the
best decision seems to treat them as safe.
This is not risky: .c files that define such functions still
generate C-linkage for their functions. If a C++ unsafe C header
is included outside a __BEGIN_DECLS ... __END_DECLS block then
the only possible result would be an undefined reference to
a function with C++-linkage that will not exist. Aka, when
something links after this commit, then the commit was correct.
I did build all targets and they all linked.
This patch reorders px4_defines.h to make it more readable (I think)
but more importantly, cleans up the #include <math.h>/<cmath>
and [std::]isfinite stuff.
My main goal was to completely get rid of including math.h/cmath,
because that doesn't really belong in a header that is supposed to
define macro's and is included in almost every source file (if not
all).
I'm not sure what it did before ;) (pun intended), but now it does
the following:
PX4_ISFINITE is only used in C++ code (that was already the case,
but hereby is official; for C code just use 'isfinite()') and is
defined to be std::isfinite, except on __PX4_QURT because that uses
the HEXAGON toolset which (erroneously) defines isfinite as macro.
I would have liked to remove PX4_ISFINITE completely from the code
and just use std::isfinite whereever that is needed, but that would
have required changing the libecl submodule, and at the moment I'm
getting tired of changing submodules... so maybe something for the
future.
Also, all includes of <math.h> or <cmath> have been removed except
for __PX4_NUTTX. Like the HEXAGON toolset NuttX currently defines
isfinite as macro for C++. So, we could have solved this in the
same was as __P4_QURT; but since we can fix NuttX ourselves I chose
to add a kludge to px4_defines.h instead that fixes this problem,
until the time that NuttX can be fixed (again postponing changing
a submodule). The kludge still demands including <cmath>, thus.
After removal of the math header file, it needed to be included
in source files that actually need it, of course.
Finally, I had a look at the math macro's (like M_PI, M_PI_F,
M_DEG_TO_RAD etc). These are sometimes (erroneously) defined in
certain math.h header files (like both, hexagon and nuttx).
This is incorrect: neither the C nor the C++ standard defines
math constants (neither as macro nor otherwise). The "problem"
here was that px4_defines.h defined some of the M_*_F float
constants in terms of the M_* double constant, which are
sometimes not defined either thus. So, I cleaned this up by
defining the M_*_F math constants as float literals in px4_defines.h,
except when they are defined in math.h for that platform.
This means that math.h has to be always included when using those
constants, but well; not much difference there as those files
usually also need/use the macro NAN (which *is* a standard macro
defined by math.h).
Finally finally, DEFAULT_PARAM_FILE was removed as it isn't
used anymore.
All in all I think the resulting px4_defines.h is nice, giving me
much less the feeling of a nearly unmaintainable and over time
slowly growing collection of kludges and hacks.
These are some changes that I needed to compile
most of the unsupported targets.
After this all (make list_config_targets) compile
for me except: posix_eagle_muorb, posix_sdflight_default
and qurt_eagle_legacy_driver_default.
This is the equivalent to Greg's upstream solution for ctype.h &
cctype with the necessary shadow waning fix.
It also deletes the Hack in pa4_defines.h
This change fixes a wrong behaviour when a takeoff command is sent.
An example:
- MIS_TAKEOFF_ALT set to 10 meters
- Takeoff command with alt setpoint of 2 meters
Old behaviour:
1. Climb to 10 meters -> takeoff WP reached
2. Go to setpoint at 2 meters
New behaviour:
1. Climb to 10 meters and stay there but notify that altitude was
overridden.
In the normal sitl `commander takeoff` case, the takeoff jump was never
actually carried out because the default altitude radius is set to 3m
and the takeoff altitude to ~2m which means that the takeoff waypoint is
"reached" immediately.
This works around this edge case by using the altitude between the home
altitude and takeoff altitude divided by 2 as a acceptance radius.
Since orb_copy is guarded with a orb_check, we ended up not having the
corrent landed state in the beginning which could lead to some
unexpected behaviour on takeoff.
In most cases, _extern_address was unused, thus wasting cycles & RAM. This
adds a separate class BlockParamExt with the field and uses it in ekf2_main
Frees roughly 0.5KB of RAM on Pixracer
The lifetime of the string is guaranteed because we never delete
DeviceNode objects and the strings in question are already on the heap.
This frees roughly 2.2KB of RAM on Pixracer
reorder & reduce instance to uint8_t (which is more than enough).
Reduces sizeof(MavlinkOrbSubscription) from 40 to 24.
On Pixracer this frees almost 2KB of RAM
The limits the maximum queue size to 255, which I think is ok for the
forseable future.
sizeof(uORB::DeviceNode) is reduces from 128 to 112 on NuttX, and with
~80 instances, this saves over 1KB of RAM.
Instead use a single timestamp for subscription checks. This frees up
~800B of RAM.
Also make sure the subscription checks are distributed over time. On each
update, at most 1 topic subscription is checked. Reduces the load of the
logger from 7.3% to 5.8% (Pixracer)
ulog_message_info_header_s *msg = reinterpret_cast<ulog_message_info_header_s *>(buffer);
members of msg could end up unaligned, because of the uint8_t buffer.
Carlo Wood