Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Fri, 8 Dec 2017 13:54:46 -0600

On 12/08/2017 01:45 PM, Mark Rivers wrote:
> I think you are missing quotes in the path name.  My batch file for 2017 has this:
> 
> "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" amd64
> 

Thanks, I just fixed that but it then died in the batch file with no
message or error. Should I be CALLing the .bat instead?

- Andrew

>> -----Original Message-----
>> From: [email protected] [mailto:[email protected]] On Behalf
>> Of Andrew Johnson
>> Sent: Friday, December 08, 2017 1:19 PM
>> To: Michael Davidsaver; Ralph Lange; EPICS core-talk
>> Subject: Re: moving initHooks into libCom
>>
>> On 12/08/2017 12:20 PM, Michael Davidsaver wrote:
>>> On 12/08/2017 12:24 PM, Andrew Johnson wrote:
>>>> Hi Michael,
>>>>
>>>> Assuming you're intending it to be merged *after* the 7.0.1 release then
>>>> yes, please go ahead and work on it.
>>>
>>> "Before or after" is my question.  Before would let me eliminate libpvAccessIOC
>>> in this release.  This has never been a concern for me, but was for Ralph.
>>
>> I would have liked to eliminate that library, but I think it's too late
>> now. The EPICS_BASE_PVA_CORE_LIBS variable should probably exclude the
>> pvAccessIOC library, and pva2pva could export EPICS_BASE_PVA_IOC_LIBS
>> say, which should include both pvAccessIOC and qsrv. That change I would
>> still accept. There might need to be an adjustment to the pvDatabase
>> module which is now using that variable and was previously linking
>> against pvAccessIOC, but I don't know if it actually needs it or not.
>>
>>>> However I would prefer that you fix the MSVC compiling issues that have
>>>> been reported in pva2pva first, since we should really fix them *before*
>>>> the final release which is due out on Tuesday.
>>>
>>> I'm fighting limited time, and internet access, this week.
>>> Also the usual problem that I don't have access to MSVC!
>>>
>>> How would you like me to approach this?
>>> I can't say that I relish spending time on appveyer config,
>>> but if APS jenkins doesn't show the failure, then I guess
>>> this is my only option.
>>
>> Mark Rivers also responded
>>> How hard would it be to upgrade the APS Jenkins server from 2010 to 2017
>>> Community?  I installed 2017 on my machine this week and it was pretty
>>> quick and painless.
>>
>> I have less than 12 working hours left until our telecon on Tuesday
>> morning when we'll start the final release, and I promised our IT group
>> I would continue working on moving the APS EPICS website to a new
>> web-server, so I'm running out of time myself.
>>
>> We do have MS Visual Studio 14.0 already installed on our Windows build
>> agent so I just adjusted the APS Jenkins epics-master-windows job to use
>> that instead of 2010 and triggered a rebuild, which seems to be running.
>> This job does update the submodules to the git tip, but it will only
>> check hourly and trigger automatically when the core/master branch is
>> updated so it will be faster to email me to run another build (use my
>> gmail if you're still working on it over the weekend).
>>
>>> Michael, if you have a pretty good idea how to fix it I could test
>>> compiling on 2015 and 2017, either via a git branch or just e-mail.  So
>>> far the problem just seems to be 1 file.
>>
>> It's probably worth testing those too if you can spare the time, maybe
>> after Michael has fixed the build with 14.0 (if it fails!).
>>
>> Thanks guys,
>>
>> - Andrew
>>
>> --
>> Arguing for surveillance because you have nothing to hide is no
>> different than making the claim, "I don't care about freedom of
>> speech because I have nothing to say." -- Edward Snowdon

-- 
Arguing for surveillance because you have nothing to hide is no
different than making the claim, "I don't care about freedom of
speech because I have nothing to say." -- Edward Snowdon

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System