| Age | Commit message (Collapse) | Author |
|---|
|
Create a utility function for loading a Python module by its name
in UTF-8.
|
|
|
|
Limit usage of the Python C API to the stable ABI subset as defined
by PEP 384. This removes some type definitions and functions which
libsigrokdecode made use of. Convert all affected code to suitable
API alternatives. Also fix a few leaks that became apparent while
working on the code.
The most visible change is that PyTypeObject is now an opaque type.
Thus, the custom Decoder and srd_logic types are now created on the
heap via an alternative API. Unfortunately, since tp_name is now
inaccessible, type names had to be removed from the log output.
Stack traces after Python exceptions are now formatted by calling
into Python, since the trace object C API is no longer available.
|
|
The new decoder stacks on top of the usb_packet PD. It adds one new
annotation row, and is able to save the decoded data as PCAP trace.
It has been successfully tested against all traces in sigrok-dumps and
some more traces.
|
|
|
|
|
|
USB low/full speed allows for frequency tolerance of 1.5%/0.25%. At
maximum packet size (sync + PID + data + CRC16) of 12 bytes/1027 bytes
this amounts to 1.4 bits/20 bits, so the decoder has to lock to the
actual symbol frequency to avoid any symbol misdetections.
The signal is sampled twice, once at the symbol center and once at
the expected edge position. Comparing the symbol at both positions gives
an indication if the current bit width is too low or too high. Adjust
accordingly.
|
|
In case CRC validation fails, an crc5-err/crc16-err annotation is issued,
instead of crc5-ok/crc16-ok, and the error is noted in the annotation
text.
|
|
There is a small typo in the rfm12 protocol decoder that I just noticed:
|
|
Rely on the compiler include path instead.
|
|
Avoid confusion resulting from using the same macro name for
different purposes.
|
|
|
|
|
|
Don't mess with colon-separated search path strings -- instead,
simply prepend items to the sys.path list object. Also keep the
internal decoder path list in the same order as the items added
to sys.path.
|
|
It seems the Python we ship along our Windows binaries does not
support u"..." strings, even though it's been a language feature
since Python 2. Remove the "u" prefix to avoid a syntax error.
Also, consistently use format "%.1f" at all scales.
This fixes bug #569.
|
|
|
|
|
|
|
|
Just as in libsigrok, use g_vfprintf() in the log callback to make
sure we are linking to a standard-conforming printf implementation.
|
|
Do not redefine it though when already set, so that it can be
overridden by the user, or indirectly by the compiler settings.
|
|
Since Autoconf places some important feature flags only into the
configuration header, it is necessary to include it globally to
guarantee a consistent build.
|
|
|
|
Decorate srd_log() with the GCC format printf attribute so that the
compiler can check the format string and arguments for consistency.
Also remove the srd_err(), sr_warn(), etc. functions and replace
them with macro wrappers around srd_log().
|
|
The confusingly named srd_log_logdomain_set() simply set a global
string prefixed to the log message by the default log callback.
This is pretty much useless, misleadingly named, and not used by
either sigrok-cli or PulseView.
|
|
|
|
Also output the compiler version in the configuration summary.
|
|
|
|
This is so that downstream modules can find the install location
of the decoders.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The STM32F10xxx has two serially connected JTAG TAPs, the boundary scan tap
(5 bits) and the Cortex-M3 TAP (4 bits). See UM 31.5 "STM32F10xxx JTAG TAP
connection" for details.
Due to this, we need to ignore the last bit of each data shift (and we
currently ignore the 5 bits of the boundary scan tap).
|
|
The code was previously not decoding a bit for the first state change
to SHIFT-IR/-DR, which was incorrect.
|
|
|
|
|
|
|
|
|
|
This basically makes glibc expose the same set of features as
if gcc was invoked without any restricting -std=c* option. Unlike
_GNU_SOURCE however, it does not enable GNU-specific extensions.
So, with this macro defined the behavior of Linux with glibc
should match that of other platforms.
|
|
In order to avoid confusion of the flags-gathering pkg-config
result with the actual test for the availability of "check",
change the pkg-config output variable prefix from CHECK to TESTS.
|
|
|
|
|
|
Use autoreconf instead of invoking the various Auto tools
separately. Get rid of the Darwin-specific guesswork -- it does
not make sense to handle this at the level of libsigrokdecode.
People should set up their ACLOCAL_PATH themselves as appropriate
for their own system; just as they already need to set up various
other paths.
|
|
|
|
The buttons are pressed if the respective bit is 0.
|
|
Don't pass unitsize to srd_inst_channel_set_all(), have that only
set the channel map. Instead, srd_session_send() now has a parameter
for the unitsize which is passed with every new chunk to be decoded.
This is in preparation to fix issues with devices or files which
have a unitsize != 1 and where the "guessed" unitsize based on the
number of channels is not correct.
This also allows for (potential) future changes where every chunk can
indeed have a different unitsize.
This fixes (parts of) bug #352.
|
|
|
|
Instead of showing raw values in the annotations, calculate the
actual voltage (based on raw value, gain, and Vref) and use that.
|
