Summary: | net-analyzer/ntopng-5.6-r2 fails to compile: ntop_defines.h:494:99: error: invalid conversion from [-fpermissive] | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Agostino Sarubbo <ago> |
Component: | Current packages | Assignee: | Gentoo Netmon project <netmon> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | chain, diazona, eschwartz, esigra, ev, gentoo, gentoo, hyedad, luke, ostroffjh, scott, tanekliang |
Priority: | Normal | Keywords: | PATCH |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
See Also: |
https://github.com/ntop/ntopng/issues/8229 https://bugs.gentoo.org/show_bug.cgi?id=932616 |
||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
build.log
Patch fixing this build error. |
Description
Agostino Sarubbo
![]() Created attachment 794564 [details]
build.log
build log and emerge --info
gcc14_tinderbox has reproduced this issue with version 5.6-r2 - Updating summary. So... what can be done here? I am on GCC 13.2, stable. src/LuaEngineNtop.cpp: In function 'int ntop_rrd_tune(lua_State*)': src/LuaEngineNtop.cpp:5709:27: error: invalid conversion from 'char**' to 'const char**' [-fpermissive] 5709 | status = rrd_tune(argc, (char**)argv); | ^~~~~~~~~~~~ | | | char** In file included from src/LuaEngineNtop.cpp:25: Created attachment 885477 [details, diff]
Patch fixing this build error.
I've also encountered this issue. I've created a quick and dirty patch removing const from enough variables to silence this error.
After applying this patch I was able to build net-analyzer/ntopng-5.6-r2 on my system.
The error is pretty obviously not an LTO-specific error (and does indeed reproduce without -flto). Upstream says:
> I'm sorry, we won't be able to accept your fix because you are not
> using the same RRD library we are using. If you look at
> https://github.com/ntop/ntopng/blob/dev/third-party/rrdtool-1.4.8/src/rrd.h#L204
> if we accept your patch we'll break our code.
Then they closed the issue as "works for me"...
It's trivial to fix though, as I commented on the upstream bug report. It just needs to conditionalize the logic depending on whether the vendored copy is used.
Patch also works for me, but I'm curious, as the same version emerged without problem last October. Could the difference be due to a change in gcc version? commit 5729e747ac8bc212d7ccda0acc5a7eeffdcc6aa3 Author: Denis Pronin <dannftk@yandex.ru> Date: Wed Apr 24 09:24:26 2024 +0300 net-analyzer/ntopng: add 6.0 Closes: https://bugs.gentoo.org/933819 Signed-off-by: Denis Pronin <dannftk@yandex.ru> Closes: https://github.com/gentoo/gentoo/pull/37091 Signed-off-by: Sam James <sam@gentoo.org> (In reply to Jack from comment #7) > Patch also works for me, but I'm curious, as the same version emerged > without problem last October. Could the difference be due to a change in > gcc version? rrdtool version maybe? |