RFR: JDK-8141543 Propagate --disable-warnings-as-errors to hotspot
Magnus Ihse Bursie
magnus.ihse.bursie at oracle.com
Fri Nov 6 12:04:36 UTC 2015
On 2015-11-06 10:52, Volker Simonis wrote:
> On Fri, Nov 6, 2015 at 10:35 AM, Magnus Ihse Bursie
> <magnus.ihse.bursie at oracle.com> wrote:
>> On 2015-11-05 18:19, Volker Simonis wrote:
>>> Hi Magnus,
>>>
>>> great that finally somebody is addressing this issue!
>>>
>>> But shouldn't we also change the lines:
>>>
>>> WARNINGS_ARE_ERRORS = -Werror
>>>
>>> to:
>>>
>>> WARNINGS_ARE_ERRORS ?= -Werror
>>>
>>> in the various hotspot compiler files (e.g.
>>> make/linux/makefiles/gcc.make).
>> Hm..? I've just been following an old "recipe" to set WARNINGS_ARE_ERRORS to
>> empty for disabling Hotspot builds. I tested on my local machine that it did
>> indeed allow warnings to pass, and then I was happy, and didn't look into
>> the Hotspot makefiles.
>>
>> I agree that this does look a bit weird, but I think it works since make let
>> command line variables override even what's explicitely set in the
>> makefiles. I don't think it's worth fixing in the old hotspot build.
>>
> If it's set on the command line, you're right. But I couldn't find the
> place where it is set on the command line. As far as I saw, it is only
> passed through hotspot-spec.gmk. Could you please point me to the
> place, where WARNINGS_ARE_ERRORS is passed to make as a command line
> parameter? Maybe I'm a little confused, but I don't see it ...
Sorry, I mixed up with MAKE_ARGS from JDK-8141548.
Well, then, I have now answer. I may know more about make and the
OpenJDK build system than what's really healthy for you, but here I give
up. It works, though. And soon the old hotspot build system is history.
'nough for me. :)
/Magnus
More information about the build-dev
mailing list