Cleanup '...' strings to reduce binary size. Change version
information.
With this change it's much more obvious what is and what is not included
in a profile.
Examples:
dump
dump profile
dump master
This makes the binary considerably smaller (73852 text with gcc
4.7.3, 71620 text with gcc 4.9.0 from TNT).
Build-tested only.
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
This saves a bit more space and a reasonable thing to do since a
custom startup is used anyway.
Build-tested only.
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
This removes a function that's no longer used, and inline specifiers
from include file (where they're useless anyway). This makes the whole
default project warnings-free with -Wall -pedantic.
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
This rearranges Makefile a bit and enables Link Time Optimisation
which leads to a smaller code (77724 text with gcc 4.7.3).
Unconditionally adding -ggdb3 doesn't affect code size anyhow as all
the debugging info is stored in separate sections.
Build-tested only.
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Conflicts:
Makefile
Output files were being left in the working directory.
Build needs to be tested on cygwin, if it fails find a different
solution or revert this commit.
Added documentation.
Added LED_STRIP feature, can only be enabled under certain circumstances
depending on target due to pin/timer mappings - see documentation.
This is an alternative to #ifdef's everywhere which is getting out of
hand.
Confirmed that the VCP is working on the STM32F3Discovery board and
ChebuzzF3 variant.
Latest naze binary.
Removing other binaries, git repo isn't really the place to store
binaries and they keep getting in the way. This is will be kept here
until the configurator is updated.
The primary reason is to support the D4R-II with it's much faster PWM
frequency. The PWM RSSI code could not keep up, and since there are no
timers free for using capture compare of PWM signals in hardware one
solution is to use the ADC at a slow sample rate.
RC2 is used as before and it expects a signal between 0 and 3.3v. An
inline smoothing capacitor may help.
This commit also removes the cli command adc_power_channel since the
reading was never actually exposed anywhere.
Also support FrSky 1khz RSSI. See documentation also added in this
commit.
This commit also cleans up the PWM mapping code. 'mask' didn't need to
be a mask and it wasn't possible to add another 'type' since there were
only 4 possible values when it was a mask and they were already defined.
Combined with switching to using 16 bits instead of 8 for the mapping
configurations, it's now possible to have 256 types instead of 4 at the
expense of a few bytes of flash.
Moved the RSSI calculation into rx_common.c, previously it was in the
main loop.
Dominic Clifton
Paul Fertser
dongie
treymarc