This magically didn't really affect normal operation, because in
normal operation the demodulator can consume data faster than it
is produced, so the fifo never grows beyond 0-1 items.
However in cases where the demodulator is slow and the producer
is fast (e.g. reading from a file and running with lots of debug on)
the fifo would lose buffers as soon as it had more than one pending.
Previously, json stats updates were fixed to 60 seconds, and using --stats--every
with an interval less than 60 seconds produced confusing results.
This commit fixes --stats-every with short intervals, and adds --json-stats-every
to control the json stats update interval. Note that if --json-stats-every is not
a multiple of 60 seconds, then the 1-/5-/15-minute data may not include all data
up to the current time (in that case, the most recent data is reflected in
the "latest" stats)
This implements the alternative approach discussed in PR #89
The numbers displayed on the stats output were also very hard to
keep track of because the display widths kept changing depending
on the value. They are now displayed with fixed widths.
(courtesy @gtjoseph)
(cherry picked from commit 19decfee0cef5f9a93a533f45f072cb1b0a6da2a)
* Switch all conversion routines to use starch.
main user-visible changes:
* ensure you check out submodules ('git clone --recurse-submodules")
* --version shows the CPU features and DSP implementations in use
* --wisdom allows overriding of the built-in architecture wisdom
* --dcfilter no longer supported
* "starch-benchmark" binary will benchmark all options on the
current machine and can produce a wisdom file to feed to
the --wisdom option
If you have a usecase for --dcfilter, please get in touch and
let me know - it's an edge case and for now there's no starch/DSP
support for it, but support can be written if needed.
In almost all cases the new conversion routines are slightly or
substantially faster than the old conversion routines. The only case
that is slower is SC16/SC16Q11 on a Pi 0, which is around 10% slower
due to changing from heavily approximated lookup tables to higher
quality results (but SC16 is probably already out of reach of a Pi 0)
* No need to build with SC16Q11_TABLE_BITS any more
* Add oneoff/uc8_capture_stats
(reads a UC8 capture; measures min/max/mean I and Q)
* Switch UC8 conversion to 127.4 center, 128 range.
Looking at actual UC8 captures from a RTL2832, the mean I and Q
are actually at 127.4, so use that as the zero point.
This means that the resulting I/Q maximum values could be as large as
127.6. Switch to 128 for simplicity.
* Switch to the new UC8 zero offset in benchmarks, fix some bugs
* Fix some bugs in SC16/SC16Q11 validation, tighten the max error requirements
* Ditch UC8 approximation path, add a NEON VRQSQRTE path.
* Tweak the SC16 exact path, add a new impl that uses a mix of
u32 & floats.
* SC16Q11 impl tweaks:
* add a u32->float exact path
* ditch the approximation path
* add a NEON VRSQRTE path
* add a 12-bit table path (using the full signed I/Q value, not absolute value)
* Ditch SC16 approximation path, add NEON vrsqrte path
* Add oneoff/dsp_error_measurement
This runs sample input through the DSP functions that are
allowed to be inexact and dumps the results as a TSV suitable for
feeding to gnuplot to look at the actual errors.
* Update make clean, make wisdom targets
* Update wisdom based on benchmarking
* Preserve the raw wisdom benchmark data
* Update to latest starch
* Update .gitignore for new wisdom files
* Update starch generated code
* Build starch-benchmark as part of the 'all' target
* Use wisdom from /etc/dump1090-fa/wisdom.local if present
* Package starch-benchmark and a helper script to generate local wisdom data
* Remove submodules in preparation for importing them directly
* Import cpu_features v0.6.0 from https://github.com/google/cpu_features/releases/tag/v0.6.0
* Import starch at commit a725c8491dc33a321565d451b385131e589d8490
from https://github.com/flightaware/starch
It was possible, due to an off-by-one error in normalize_timespec, that
the computed deadline passed to pthread_cond_timedwait had tv_nsec == 1000000000.
glibc would reject this with EINVAL. While I never caught this in the act, there
seem to be two follow-on problems from this:
1) we were looking for a negative return from pthread_cond_timedwait and testing errno.
This is not how errors are reported, instead pthread functions return the error value
directly. So the enclosing loop could spin forever, repeatedly passing the bad deadline
value.
2) even if the error handling caught the EINVAL return, it assumed that the mutex was
no longer held and didn't release it; but on error returns, the mutex state is actually
untouched.
Fix both cases (probably), and log about unusual returns from pthread_cond_timedwait
George Joseph
Oliver Jowett
eric1tran
Edwin Groothuis
Eric Tran