On Thu, Mar 24, 2011 at 9:22 AM, Kim Culhan <w8hdkim@gmail.com> wrote:
On Wed, March 23, 2011 9:15 am, Michal Bielicki wrote:
The problem is a bug in solaris perl development package that I haven't been doing anything about since I do not use perl for roughly a decade. Its not the hash module, its the perl module.
Am 23.03.2011 um 11:44 schrieb A E [Gmail]:
Hello FreeSwitchers!
Am attempting to compile the latest (off of Git) code for FS on Solaris 11 Express on an 64-Bit x86 machine. However, I'm running into the error that Bruce McAllister has reported on 23rd Nov 2010. The bug is still sitting unresolved. So I'm assuming there's a backdoor trick to fixing this OR that some other version of FS code doesn't have this issue. Could someone please let me on the secret on which build compiles on x86 Solaris v10+ and/or what the backdoor trick to bypass this is? Should I not compile the hash module where this bug seems to be originating from?
I'm seeing the same error here with the perl module commented-out in modules.conf
Please describe the workaround you are using to avoid the problem caused by the perl module.
thanks -kim
Hello Kim, for now, I've just commented out mod_hash in modules.conf to see if I can compile the rest of it successfully. But there's a LOT of problems along the way. So I will come back to this later. I don't want to open a Jira on it as it's already there. However, if you're not getting this with mod_hash but something else, then you will have to look carefully what is causing it. I have noticed other places where the make process is crashing due to unrecognised options being passed to the Solaris Studio express CC compiler. So I will probably open a Jira on other issues I have encountered. I'm still hoping Michal would reply and let us know how he is getting it to compile every time. For now, I'm just editing the Makefile for each of such application/module which is encountering this problem so I can keep moving forward HTH