On 23/11/2020 13:46, Josh Rovero wrote:
Having some issues compiling RC2 from the source tarball on Fedora
Core 32, 64it. All FC32 rpms are up to date. Two Readline references
in rigctl.c:
rl_readline_name on line 566
using_history() on line 569
Is there a specific readline version that mus
Hi Bill,
> Are we sure that a sysctl.conf with the necessary systctl parameter changes
> is not being read on reboot?
The only evidence for that is the fact that WSJT-X on an M1 Mac fails with a
shared memory error and has no problems on an Intel Mac. I now have a report
from a VK about a me
On 19/11/2020 14:54, John Nelson via wsjt-devel wrote:
Hi Bill,
I have just had a reply from W4 about info sysctl. This is the same as that in
use now - dated 2007. However, the line marked BUG might explain what has
been discovered in the present case.
— John G4KLA
BUGS
If loadable
Hi Bill,
I have just had a reply from W4 about info sysctl. This is the same as that in
use now - dated 2007. However, the line marked BUG might explain what has
been discovered in the present case.
— John G4KLA
BUGS
If loadable kernel modules are used to introduce additional kernel fu
On 18/11/2020 21:54, John Nelson via wsjt-devel wrote:
Hi Bill,
Further to the WSJT-X + Big Sur + M1 problem, three conclusions have been
reached following tests with my correspondent:
1. The M1 page size is 4k and not 16k
2. It seems that the M1 version of Big Sur does not get the shared
On 18/11/2020 21:54, John Nelson via wsjt-devel wrote:
Hi Bill,
Further to the WSJT-X + Big Sur + M1 problem, three conclusions have been
reached following tests with my correspondent:
1. The M1 page size is 4k and not 16k
2. It seems that the M1 version of Big Sur does not get the shared
Hi Bill,
Further to the WSJT-X + Big Sur + M1 problem, three conclusions have been
reached following tests with my correspondent:
1. The M1 page size is 4k and not 16k
2. It seems that the M1 version of Big Sur does not get the shared memory
information from /etc/sysctl.conf
3. If kern.s
On 18/11/2020 20:50, John Nelson via wsjt-devel wrote:
Hi Bill,
I have had a report from someone failing to run 2.3.0-rc2 on a new Mac with
Apple M1 chip and Big Sur. The shared memory error is the only message. I
have established that the sysctl.conf file is in place and has not been
corru
Hi Bill,
I have had a report from someone failing to run 2.3.0-rc2 on a new Mac with
Apple M1 chip and Big Sur. The shared memory error is the only message. I
have established that the sysctl.conf file is in place and has not been
corrupted.This OM confirms that WSJT-X + Big Sur + Intel w
Hi Bill,
> the symbol namespace would include 'v2s' for a static link rather than ‘v2’
Agreed. Which is why I am puzzled that, given all the libboost*.dylib
libraries, that the linker cannot find a dynamic linked symbol in a dylib.
> Once I have a few more pressing issues resolved I will retur
On 16/11/2020 18:57, John Nelson via wsjt-devel wrote:
Hi Bill,
Understood. But then I will await your further investigations on the linking
problem which I initially sent to you.
I would prefer not to have to modify files.
---
It has found what it seems to need. The problem about linking r
Hi Bill,
Understood. But then I will await your further investigations on the linking
problem which I initially sent to you.
I would prefer not to have to modify files.
---
It has found what it seems to need. The problem about linking remains:
Undefined symbols for architecture x86_64:
On 16/11/2020 18:32, John Nelson via wsjt-devel wrote:
Hi Bill,
I have found a solution.
1. remade the boost libraries with:b2 link=static runtime-link=static
install
2. made three changes to CMakeLists.txt
79,80d78
< # JMN Added next line
< set (Boost_USE_STATIC_RUNTIME ON)
1067
Hi Bill,
I have found a solution.
1. remade the boost libraries with:b2 link=static runtime-link=static
install
2. made three changes to CMakeLists.txt
79,80d78
< # JMN Added next line
< set (Boost_USE_STATIC_RUNTIME ON)
1067c1065
< #target_compile_definitions (wsjt_cxx PUBLIC BOOS
On 16/11/2020 14:25, John Nelson via wsjt-devel wrote:
Hi Bill,
How did you build and install he Boost C++ libraries?
bootstrap.sh —prefix=/usr/local/ —libdir=/usr/local/lib
—includedir=/usr/local/include
b2 install
If my configure is finding the static libraries — what change in configu
Hi Bill,
> How did you build and install he Boost C++ libraries?
bootstrap.sh —prefix=/usr/local/ —libdir=/usr/local/lib
—includedir=/usr/local/include
b2 install
If my configure is finding the static libraries — what change in configuration
must be made to force it to find the dynamic libr
On 16/11/2020 13:16, John Nelson via wsjt-devel wrote:
Hi Bill,
all that should be needed after building and installing the Boost libraries is
to include the path where they are installed in the CMAKE_PREFIX_PATH
CMake knows there the boost libraries are. For instance:
-- Building wsjtx v2.
Hi Bill,
> all that should be needed after building and installing the Boost libraries
> is to include the path where they are installed in the CMAKE_PREFIX_PATH
CMake knows there the boost libraries are. For instance:
-- Building wsjtx v2.3.0.0-rc2
-- Found Boost 1.70.0 at /usr/local/lib/cma
On 16/11/2020 09:02, John Nelson via wsjt-devel wrote:
Hi Bill,
I have installed Boost 1.70 and updated CMake. Could you send me the cmake
build command that you use for Mac build? Are there particular link options
for boost?
Thanks…
— John G4KLA
Hi John,
all that should be needed aft
HI Bill,
Further to my previous email when I requested:
> Could you send me the cmake build command that you use for Mac build? Are
> there particular link options for boost?
The problem at link time is:
Undefined symbols for architecture x86_64:
"boost::log::v2_mt_posix::attributes::timer
Hi Bill,
I have installed Boost 1.70 and updated CMake. Could you send me the cmake
build command that you use for Mac build? Are there particular link options
for boost?
Thanks…
— John G4KLA
smime.p7s
Description: S/MIME cryptographic signature
On 15/11/2020 20:07, Stefan Wernli wrote:
Hi Bill and David,
libboost-dev didn't help, but libboost-all-dev solved the problem. It
compiled and runs fine so far.
Thanks and best 73
Stefan
Hi Stefan,
thanks for the update, glad you are up and running with WSJT-X v2.3.0 RC2.
73
Bill
G4WJS.
Hi Bill and David,
libboost-dev didn't help, but libboost-all-dev solved the problem. It
compiled and runs fine so far.
Thanks and best 73
Stefan
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinf
Versions.cmake is missing from the tar file.
Mike W9MDB
On Sunday, November 15, 2020, 11:12:10 AM CST, Stefan Wernli
wrote:
Hi,
I get an error when compiling the new rc2 on debian testing. 2.3.0-rc1
compiles successfully on the same system.
-- Building wsjtx v2.3.0.0-rc2
-- Perfor
I couldn’t compile it either. Just installed the binary.
Sent from my iPhone
> On Nov 15, 2020, at 9:12 AM, Stefan Wernli wrote:
>
> Hi,
>
> I get an error when compiling the new rc2 on debian testing. 2.3.0-rc1
> compiles successfully on the same system.
>
>
> -- Building wsjtx v2.3.0.0-r
You need the boost libraries:
sudo apt install libboost-dev
On Sun, Nov 15, 2020 at 12:09 PM Stefan Wernli wrote:
> Hi,
>
> I get an error when compiling the new rc2 on debian testing. 2.3.0-rc1
> compiles successfully on the same system.
>
>
> -- Building wsjtx v2.3.0.0-rc2
> -- Performing Tes
On 15/11/2020 15:51, Stefan Wernli wrote:
Hi,
I get an error when compiling the new rc2 on debian testing. 2.3.0-rc1
compiles successfully on the same system.
-- Building wsjtx v2.3.0.0-rc2
-- Performing Test HAVE_MATH
-- Performing Test HAVE_MATH - Failed
-- Performing Test HAVE_LIBM_MATH
--
27 matches
Mail list logo
