On Mon, Jul 05, 2021 at 12:34:04PM +0200, Christoph Berg wrote: > Re: tony mancill > > However, the patched build alone does not resolve the crash of > > sdrangelove on my system until I also install the updated hamlib4 from > > experimental. That is, sdrangelove appears to be affected by > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=980472. > > > > I'm not sure what the plans are for the hamlib4 with respect to > > Bullseye. > > 4.0 is probably not the ideal version to have in bullseye, but it's a > great improvement over the version we had before (which was missing > support for many transceivers on the market today). > > Hamlib 4.1 is unfortunately incompatible with 4.0 in some internal > data structures (which are supposed to be opaque, but aren't always in > practise), so updating to that would need fixes in other programs (I > know about wsjtx, see libhamlib4's Breaks:). I don't think we can or > should try to push that into bullseye. Yes, wstjx is twice as popular as sdrangelove according to popcon, and I am surprised the delta isn't even greater. > > Any concerns with an upload of soapyosmo to get us one step > > closer? > > Seems sensible to me. Thank you for the response. I will to upload with this (minimal) change and request an unblock. Given that the upload addresses the retitled bug in soapysdr, I will include "Closes: #990240" but want to point out that the "affects" for sdrangelove isn't really applicable. As far as I can tell, the "ERROR" logged during startup isn't the actual cause of the crash, it simply happens to be the last thing logged before the crash. > [ERROR] SoapySDR::loadModule(/usr/lib/x86_64-linux-gnu/SoapySDR/modules0.7/librfspaceSupport.so) > dlopen() failed: /usr/lib/x86_64-linux-gnu/SoapySDR/modules0.7/librfspaceSupport.so: undefined symbol: _ZN5boost6chrono12steady_clock3nowEv > Hash collision!!! Fatal error!! > pure virtual method called Cheers, tony
Attachment:
signature.asc
Description: PGP signature