Re: build failures on Solaris

[Denis Ovsienko via tcpdump-workers <tcpdump-workers () lists tcpdump org>]

> --- Begin Message ---
>
>
> From: Denis Ovsienko <denis () ovsienko info>
>
> Date: Mon, 2 Aug 2021 02:08:05 +0100
>
> On Sun, 1 Aug 2021 15:45:39 -0700
> Guy Harris <gharris () sonic net> wrote:
>
> [...]
> > So where do the Solaris 11 hosts show up on the buildbot site?
>
> They don't. Solaris 9 and 11 are not in any CI at all. OpenCSW Buildbot
> runs on Solaris 10, but uses a single build script. I used OpenCSW
> shell access to see how realistic it is to do a manual full matrix CI
> on each Solaris version, and it had chased some bugs out. The gcc211
> Solaris host had added to the list a bit (not yet described). That was
> on SPARC.
>
> If you would like to host a Solaris 11 Buildbot worker, it would need
> Python 3.6 or newer plus the usual tools (CMake, git...). Also it would
> need to be firewalled from private networks and to have means to
> restore to a known-good snapshot conveniently if the VM becomes
> compromised through a pull request. Buildbot workers run fine behind
> NAT.
>
> [...]
> > > As I have discovered just now, it does not reproduce on OpenCSW host
> > > gcc211:  
> >
> > Probably some annoying combination of one or more of "different
> > compilers", "later version of CMake", "at least some versions of cc
> > and gcc build 32-bit binaries by default even on Solaris 11 on a
> > 64-bit machine(!)", and so on.
> >
> > This is going to take a fair bit of cleanup, not the least of which
> > includes forcing build with both autotools *and* CMake to default to
> > 64-bit builds on 64-bit Solaris.
>
> For clarity, there is no rush to fix every obscure issue in this
> problem space, but it is useful to have the problem space mapped.
>
> One purpose for that would be stating something like "Only these
> particular kinds of Solaris are expected/known to work if they meet
> this and this requirement." in the documentation. I do not have a use
> case for Solaris, so I cannot define such a baseline, so I had tried
> everything that was available, for worse or better.
>
> Another purpose would be to cover as much as possible at least with
> easy manual pre-release matrix builds, and ideally with CI.
>
> -- 
>     Denis Ovsienko
>
> --- End Message ---
_______________________________________________
tcpdump-workers mailing list
tcpdump-workers () lists tcpdump org
https://lists.sandelman.ca/mailman/listinfo/tcpdump-workers