* move class definition to .h file, like we do for other devices.
* move smpl_buf class to a different file inside uhd/.
* Preparation work to have smpl_buf being used in a generic way for
devices other than UHD (LMS).
Change-Id: Ib4594320da9bb7f6e9f52e7d70d11ecd11106aae
ALERT log level is not Osmocom standard level, it's just a define in
osmo-trx to keep compatibility with old code.
Same goes for one reference to "ERR" intead of "ERROR".
Change-Id: I0e171a8ac8a8bfa804ac97fba3d73efcfa6424b4
No need to use the heap here since buffer is only used as a temporary
trash. Using the stack is quicker.
Change-Id: Iede8dc0903ee3865a52c8e2fd811bcde444fee33
add device dependant max gain setup - limesdr mini and limenet micro need slightly reduced maximum gains to get a PASS on phase error measurements
rework clock reference setup - external clock needs to be selected before calling LMS_Init(), internal can only be set after.
remove now unused compat_LMS_VCTCXO* functions - we do not set the VCTXCO directly anymore
Change-Id: I3cf905b0a84bc1ec200891762a6646141ee37181
move enable: it is important that channels are actually enabled before applying any configuration (besides external clock)
move disable: move to close, so channels are not disabled and not enabled again while osmotrx is active.
Change-Id: I82878913254ce15a85db8d006e13d5eb639793e9
We haven't even released any tagged version of libosmocore yet which
includes support for the renamed OSMO_FD_READ constant. Let's avoid
any breakage and use the new constants only with considerable delay,
at the very least only when released libosmocore versions provide it.
Change-Id: Idb57077b2a4b2a71dd5d75a24ded8bb5887da188
Maximum number of carriers is fixed to 3 channels on a single
physical RF channel in multi-ARFCN mode. For some reason, it
was limited to 5.
Let's fix this, and also follow this limitation in the
following VTY command handlers:
- cfg_multi_arfcn_cmd,
- cfg_chan_cmd.
Change-Id: I66a1462f368458afd313ee6f0bc0abc496dde817
This should avoid prolematic scenarios where different signal handlers
are running on different thread in parallel. Furthermore, we make sure
those signals are always run by main loop thread.
Change-Id: I9b9d9793be9af11dbe433e0ce09b7ac57a3bdfb5
Recently a blocked osmo-trx process was found after ending SIGTERM to
it.
Apparently one thread was handling SIGTERM and calling fprintf()
(grabbing libc lock) while another thread was handling another signal
and also grabbing similar lock. Both thread looked deadlocked there.
Probably this change doesn't fix the block on its own, but at least
simplifies scenarios inside signal ctx which can go wrong.
Change-Id: If91621913b8b03d8a0f4c863be0b0d479f97e8a1
Besides just general cleanup, the major changes are :
- Fully internal generation of reference data that doesn't
depend on glibc or even on any floating point math
- Golden results are included in a .h
Due to varying precision of different implementation or
architecture, any kind of textual compare is impossible, so
we include golden values and compare results of both the
'base' implementation the potentially 'optimized' one again
this set of values with a small error tolerance
Change-Id: I4e203d2c4b778af77d630ed15d4cef6b0c0eb76d
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
Since I838c21db29c54f1924dd478c2b34b46b70aab2cd we have both TS1
and TS2 synch. sequences, in addition to "default" TS0. Let's
finally introduce the VTY configuration parameter, that can
be used to toggle optional detection of both TS1 and TS2.
Note: we keep this optional because of potentially bad impact on
performance. There's no point in paying the performance penalty
unless upper levels (BTS, PCU) actually make use of it.
Change-Id: I1aee998d83b06692d76a83f79748f9129a2547e8
Related: OS#3054
- Those are not used any where
- Those are not supported by the sse/neon accelerated versions
- And I see very little use cases for those.
Change-Id: Ic850269a0ed5d98c0ea68980afd31016ed555b48
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
There appears to have been some logic to operate a USRP1 in
transmit-only GSM mode. This is achieved using the skipRx member
of the transceiver object. However, there's nobody that ever
initializes it properly, and hence the feature is not possible to
use anyway.
I don't think this has any valid use case, so let's remove it.
Change-Id: I616193f1e9aaefbf4ceb26761657811093f28b6f
Link LMSDeviceTest against LMS_LIBS, so it does not only compile on
Debian, but also on Ubuntu and openSUSE. Thanks to roox for figuring
this out.
Related: OS#3654
Change-Id: I6980d4290f623485a77db10fea6d17de0321c092
boost is only used in USRPDevice class.
It looks like it can be removed entirely quite easily, since only
boost::shared_ptr is used for 2 variables. That's left for somebody with
the device and willingness to test the changes.
Change-Id: I4c3fa3ff58fd552d0cb4c4cf2033615d84c07c96
According to gettimeofday manual:
"Applications should use the clock_gettime() function instead of the
obsolescent gettimeofday() function."
Furthermore, it may be desirable in the future to use other clocks such
as monotonic.
Change-Id: I2286998c5eefbf3c3dfb105c223daec7a1083803
Allow selecting a specific LimeSDR device by setting dev-args in the
config file. Split up the given dev-args address by comma and select
the device where all substrings can be found.
I could not test this with real hardware, but I have added a test case
to make sure this works as expected.
Related: OS#3654
Change-Id: Ib9aaa066a01bf9de3f78234d7ada884d6f28c852
Make the "opt" argument const. This function will also be used by
LMSDevice.cpp in a follow-up commit.
Related: OS#3654
Change-Id: If3f0f682ca453c2b0a06175ec9626567932cfce6
Under failure, it could still be that stream status is updated, so let's
father that in all cases.
Change-Id: I4e2b8be06d2993db1bab233948a8ee774b8ac4ee
This log category is applied to messages related to TRX CTRL socket
interface, and it's printed in yellow, same color used in osmo-bts-trx
for TRX category (so same messages are printed with same color in both
sides).
Change-Id: I98ec5e416272783ad3fbadf70478a4e48ae64983
Original issue: In order to use SSE instructions, 16-byte aligned memory
chunks are needed, and C++ version < C++11 doesn't provide for a native
new/delete store. For that reason, memalign() must be used in the
implementation of convolve_h_alloc() for some buffers.
On the other side, The C++ code relies on C++ "new T[]" operator to
allocate a chunk of memory containing an array of class instances. As
classes are complex types, they cannot be allocated through C structures
(calling malloc). Experimentally can be seen too that it's unreliable
and the process will crash during startup if malloc() is used and then a
Complex<> deferred from it.
Previous implementation allowed for use of convolve_h_alloc or new[]
based on how the (signal)Vector is called, because then the buffer is
not going to be managed internally. But that's unreliable since resize()
calling resize() on it could use "delete" operator on a malloc'ed
buffer, and end up having a new new[] allocated buffer. It was also
found that some of the callers were actually leaking memory through ASan (because the
buffer is not managed by the Vector instance).
IMHO best option would be to rewrite all this code using C structures
and malloc/free exclusively, since it would make all this cod eeasier to
maintain.
But for now, let's extend the Vector class to allow specifying an
external alloc/free function and let the Vector instance take care of
the ownership of the buffer in all scenarios.
Change-Id: Ie484a4762a7f77fe1b105188ea03a6f025730b82
Pau Espin
Joachim Steiger
Harald Welte
Vadim Yanitskiy
Sylvain Munaut
Oliver Smith