Age | Commit message (Collapse) | Author | Files | Lines |
|
Fix build failure on various openSUSE distributions in OBS, where
rpmlint runs after the build and complains about these missing
post/postun scriptlets.
Fix for:
libosmo-hnbap0.x86_64: E: library-without-ldconfig-postin (Badness: 300) /usr/lib64/libosmo-hnbap.so.0.0.0
...
(none): E: badness 1200 exceeds threshold 1000, aborting.
Related: SYS#5516
Change-Id: I898aa67dbc7deaa9a8235f508a34dd29cc83b385
|
|
Change-Id: I6ecbce6aab11e1382ff9f71d901e1c6ac145f835
|
|
Change-Id: Iba106dcc18e3d429b4b9808610a44ac29b798172
|
|
Change-Id: I10ce4a9414768cdcf850feeca4088296e339919f
|
|
Change-Id: Id4e337cdc82c26922b7ca7f0fe8af523e9a0945c
|
|
Change-Id: I1bcbe047d233684647d6ffe9febb14a12c579696
|
|
This is an option enabled by default and not existent as cmdline param
since asn1c 0.9.26.
For some reason, when dropping it, asn1c generates different description
comment section on related files. In any case, this makes hnbap look
like other libraries which already got this param removed and hence also
generateds code looks the same.
Change-Id: Iccf81539a5045416185e619d24c3d542204437e6
|
|
Other libraries already use ASN1C_PREFIX except the HNBAP one. Let's use
it here to avoid collisions.
All new include/osmocom/hnbap/HNBAP_*.h and src/hnbap/HNBAP_*.c are
generated using the updated src/hnbap/Makefile.am.
Files such as src/*.c, test/* and include/osmocom/hnbap/hnbap_common.h
are modified manually to update structs/enums to new prefix.
Related: SYS#5516
Change-Id: I6c062f631527ae42811bfec134933671c05ecba6
|
|
Change-Id: Ib7a03973d43828f47ff59b561d9ddce43186f8af
|
|
Change-Id: I5666a6dc1ec9e56fe133642fb0054e913265c081
|
|
Change-Id: I203a25a8059218bb9fd6d10cbc028a296d347382
|
|
Warning observed compiling with gcc 11.1.0.
Change-Id: I6ca53dd57572338f15971633441da895843a164d
|
|
Change-Id: Icc8b90866a8edce9d10c87f4ced7137edf1c46f0
|
|
Change-Id: I9e21b504e4fac072d8481d0a699bc58d1b94a12d
|
|
Fix the conflict from having %{_includedir}/* in multiple packages by
replacing it with the correct subdir path.
Fix for:
found conflict of libosmo-hnbap-devel-0.7.0.11.b110.202110260026-1.1.armv7hl with libosmo-ranap-devel-0.7.0.11.b110.202110260026-1.1.armv7hl:
- /usr/include/osmocom/hnbap/Access-stratum-release-indicator.h
- /usr/include/osmocom/hnbap/AccessResult.h
- /usr/include/osmocom/hnbap/AdditionalNeighbourInfoList.h
...
I've verified that it builds again in my own OBS namespace.
Related: SYS#5516
Fixes: 35f2bd ("Provide libosmo-hnbap as a public shared library")
Fixes: dba8b0 ("Provide libosmo-rua as a public shared library")
Change-Id: Ia401caf01fa9ea351f811d7c22a04ccff556d6bc
|
|
Change-Id: Ie3e5b7fb88caf2a795fccdcf214f559f9683070c
|
|
header files
This makes builds reproducible and keep the existing relative path.
This should have been done in a previous commit changing the paths being
passed to asn1c to be absolute (top_srcdir).
Fixes: f67803684f47077de3c9e06eb7dc007e0bc86b28
Change-Id: I19537f136cbb654950172ee532027920d353a556
|
|
It is needed by the new osmo-hnodeb project.
Related: SYS#5516
Change-Id: I8b800baacbfe3a3c424c51b418b5ac2e2848b3ba
|
|
It is needed by the new osmo-hnodeb project.
Related: SYS#5516
Change-Id: I0d0862e9d59d3b206f3f7dfae4b39ae4032bb762
|
|
Change-Id: Ie22df146cfaf6aaf9c7cef884a964e6a3dfcc8ad
|
|
It was no longer needed since .la was used.
Change-Id: I3f196d5c2bb35826709bfc32d10e6b3683b552ee
|
|
Change-Id: I934b7be7ca1fac05410df6640428ee8e9c70c3e7
|
|
Change-Id: Id33e2e8caa37dc359d564188747cd4c1b8bc4064
|
|
This allows the SGSN always having feedback on the resolution of the
release, hence being able to stay in PMM CONNECTED state until the
resolution is received, then moving to PMM IDLE.
Related: SYS#5389
Change-Id: Iac822c74e56750dc40e94573eae0e20853ff68c0
|
|
Change-Id: I2233df2b09ae46f28dd99d7540d71fee8e47b6ff
|
|
We don't want to print usage information after the VTY reference XML.
Change-Id: Ia729134a8f18d53274bbc063aa39508c7dcc1f72
|
|
Change-Id: I3b592c39646a77681efde5debb5ac2a844208949
|
|
Remove hnbgw_vty_reference from the source tree.
In manuals/Makefile.am use the new BUILT_REFERENCE_XML feature recently added
to osmo-gsm-manuals, and add a build target to generate the XML using the new
osmo-hnbgw --vty-ref-xml cmdline switch.
Change-Id: Ia1e978ad9139c41368cd2720c339e1738d345c4a
Depends: I613d692328050a036d05b49a436ab495fc2087ba
Related: OS#5041
|
|
For some unknown reason, using size=0 instead of 1 makes LeakSanitizer
report a leak when calling osmo-hnbgw --vty-ref-xml:
"""
==19857==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 96 byte(s) in 1 object(s) allocated from:
#0 0x7f083b4d0d28 in malloc (/usr/lib/x86_64-linux-gnu/libasan.so.3+0xc1d28)
#1 0x7f083a8e5661 in talloc_named_const (/usr/lib/x86_64-linux-gnu/libtalloc.so.2+0x6661)
"""
It seems iu_client.c is already using size=1.
Change-Id: I383653c981e68a053babd33dee1f0df8731752b0
|
|
This global variable is already defined in iu_client.c provided by
libosmo-ranap (and used by libasn1c).
Fixes ASan warning at starutp.
Change-Id: I7dbae83ce0dc4ac25dc10879e223280414f7b561
|
|
Change-Id: If07b00e928fbecc596a7f75fc55e811bd9199b72
Depends: Ie2022a7f9e167e5ceacf15350c037dd43768ff40
Relates: OS#5041
|
|
Change-Id: Ic53a0a9afb8038759bb3ab1d4c5970f23632e84e
|
|
Let's disable category here since we don't care about its formatting here.
In any case, every test relying on logging output validation should
always explicitly state the config to avoid issues in the future if
default values change.
Change-Id: Ia633cae0a26f294e276b431a82256abb1b57ebba
Related: OS#5034
|
|
The related compiler warnings were overlooked as due to the
asn1c-generated code there always are tons of warnings printed compiling
this project :(
Change-Id: I40b1265ba696501cc72e674f3ef4146c47aacf1b
|
|
Change-Id: I6d2d033b0427bdc84fee61e0f3cb7b29935214bf
Closes: OS#4143
|
|
So far we don't really have any way of matching a given log message
to a specific hNB. Let's use the newly-introduced LOGHNB() macro.
Change-Id: Ia16eed4ca5e9ef231259e94c65105f7586f366bc
|
|
Change-Id: I4bf2f1cb724bc3e3c56d5a5c1270bb152b0e6e0a
|
|
Change-Id: I17f20cea8aa63ed79168653292b99c90b8c03caf
|
|
Change-Id: Ic92e60dda41a9e4d139294b1e851fff5c4f26143
|
|
Change-Id: I320ad86b83b0308b43489e8d95f63d6ed55821b5
|
|
Change-Id: I168e4f7cd48cbaa387e6b05868107250795a0f0c
|
|
Change-Id: If8b799777598377048512c338f6384560b445ba3
|
|
Change-Id: I04b1c0446124d66dae1e8fb99d9b82782d12bde7
|
|
gcc warns about set but not used variable otherwise.
Change-Id: Ia49b7e88a66a4a744c2b14db6331cc0de431f60b
|
|
Change-Id: I3b0a6218ac6242d03642972d5b244815ce9d97a0
|
|
Change-Id: I12870a1cf2462b7cc2868a54f58416e03a856084
|
|
osmo_hexdump expect it to be a const unsigned char*, and gcc warns about
incorrect signedness.
Change-Id: I7e0622f502a4442d700340b0f1c2f8cd01e90ecd
|
|
Related: OS#4912
Change-Id: I339cc37a259e30b131fadbdaebd413c54b04f0a0
|
|
we unconditionally depend on this header file, yet nothing
in configure.ac ensures its presence. Building on a clean
Debian 10 with all existing dependencies installed passes
configure but fails at compilation time.
Change-Id: I0a4407ef6ee512ad5ea8775d90acfe1c4fa67cd0
|
|
for some strange reason we didn't yet build the manual in the debian
package build.
Change-Id: Id53956494144b46e78fc25a176a9b8917788d86e
|