Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

test solaris #265

Merged
merged 1 commit into from Oct 2, 2017
Merged

test solaris #265

merged 1 commit into from Oct 2, 2017

Conversation

rgerhards
Copy link
Member

No description provided.

@rgerhards rgerhards force-pushed the solaris branch 8 times, most recently from a2f6e14 to 7ac516d Compare October 2, 2017 09:51
@rgerhards rgerhards added this to the v2.0.4 milestone Oct 2, 2017
@rgerhards rgerhards merged commit bcbec4e into rsyslog:master Oct 2, 2017
@rgerhards rgerhards deleted the solaris branch October 2, 2017 10:49
@andreasstieger
Copy link

This change is weird. It introduces an int function without return?

I: Program returns random data in a function
E: liblognorm no-return-in-nonvoid-function strndup.c:42

@rgerhards
Copy link
Member Author

Right, should be a void. The sole point is that there is any symbol at all in the lib, else Solaris build fails. It is never used.

@andreasstieger
Copy link

I see. Could this be wrapped pre-compiler conditionals for Solaris only?

rgerhards added a commit to rgerhards/liblognorm that referenced this pull request Nov 25, 2017
@rgerhards
compatibility layer had an unclean function definition
0a044b1 
which triggerred build errors/warnings on some platforms
We could correct this with detection for Solaris, but for that we
would need to change our autoconf configuration, for what I
currently have no time. So I just correct the invalid type.
This causes no harm except for maybe some bytes of unnecessary
object code. Comment now describes the use case.

Thanks to Andreas Stieger for bringing this up.

see also rsyslog#265 (comment)
@rgerhards rgerhards mentioned this pull request Nov 25, 2017
Closed
@rgerhards
Copy link
Member Author

I have now just corrected the invalid type -- digging out the right OS detection is time-consuming for me. It's better to have this solution now. The commit comment says it all.

@rgerhards
Copy link
Member Author

@andreasstieger the change is merged, please let me know if this works for you (at least as a temporary solution). Do you need a public release (or would prefer it)?

@andreasstieger
Copy link

#273 0a044b1 works, although we went for a different solution. There is no need for a new release just for build fixes, proper distributions are happy to patch the source with upstream fixes or their own.

@rgerhards
Copy link
Member Author

Thanks for the reply. I hope you go without the patch in the future. As time permits, I'll also see that we do a check for Solaris, but I am wading through a very long backlog.

rgerhards added a commit to rgerhards/liblognorm that referenced this pull request Apr 25, 2018
@rgerhards
compatibility layer had an unclean function definition
0cccee2 
which triggerred build errors/warnings on some platforms
We could correct this with detection for Solaris, but for that we
would need to change our autoconf configuration, for what I
currently have no time. So I just correct the invalid type.
This causes no harm except for maybe some bytes of unnecessary
object code. Comment now describes the use case.

Thanks to Andreas Stieger for bringing this up.

see also rsyslog#265 (comment)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
v2.0.4
Development

Successfully merging this pull request may close these issues.

None yet
2 participants
@rgerhards @andreasstieger