svn commit: r265581 - in head: . share/mk

[Warner Losh]

On May 8, 2014, at 3:20 PM, Glen Barber <gjb at FreeBSD.org> wrote:

> On Wed, May 07, 2014 at 07:09:05PM -0700, Warner Losh wrote:
>> 
>> On May 8, 2014, at 3:04 PM, Glen Barber <gjb at freebsd.org> wrote:
>> 
>>> On Thu, May 08, 2014 at 05:57:20PM -0400, Glen Barber wrote:
>>>> On Wed, May 07, 2014 at 06:15:03PM +0000, Warner Losh wrote:
>>>>> Author: imp
>>>>> Date: Wed May  7 18:15:02 2014
>>>>> New Revision: 265581
>>>>> URL: http://svnweb.freebsd.org/changeset/base/265581
>>>>> 
>>>>> Log:
>>>>> bsd.compiler.mk was implicitly included by bsd.own.mk in historical
>>>>> versions. With its movement to src.opts.mk, bsd.prog.mk was testing
>>>>> COMPILER_TYPE without including the bsd.compiler.mk anymore. In the
>>>>> source tree, this caused no problems, for reasons that aren't clear,
>>>>> but does cause problems outside of the source tree. Allow
>>>>> bsd.compiler.mk to be included multiple times safely, and always
>>>>> include bsd.compiler.mk at the top of bsd.prog.mk. Resist the urge to
>>>>> put it in bsd.init.mk, since that would reintroduce the implicit
>>>>> include.
>>>>> 
>>>>> Modified:
>>>>> head/UPDATING
>>>>> head/share/mk/bsd.compiler.mk
>>>>> head/share/mk/bsd.prog.mk
>>>>> 
>>>> 
>>>> Something here is breaking head/ release builds.  I don't know if it is
>>>> this exact change set or not.
>>>> 
>>>> --------------------------------------------------------------
>>>>>>> Kernel build for GENERIC completed on Thu May  8 01:47:57 UTC 2014
>>>> --------------------------------------------------------------
>>>> make: "/usr/share/mk/bsd.obj.mk" line 43: Could not find bsd.own.mk
>>>> make: "/usr/share/mk/bsd.init.mk" line 15: Could not find bsd.own.mk
>>>> make: Fatal errors encountered -- cannot continue
>>>> make: stopped in /usr/src/release
>>>> 
>>> 
>>> The revision I'm building against is r265621, for what it is worth.
>> 
>> OK. That’s a weird error.  I haven’t deleted bsd.own.mk...
>> 
>> Any chance you can do some bisection to see if there’s a specific
>> change you can narrow this down to? But /usr/share/mk suggests
>> there’s some host contamination going on, which implies needing to
>> have a synchronized host environment (I can’t recall if make release
>> is fully virtualized or not).
>> 
>> But before all that, can you confirm you have a /usr/share/mk/bsd.own.mk?
>> 
> 
> Ugh...  This is the problem...
> 
>  root at grind:/releng/11-amd64-GENERIC-snap # ll usr/share/mk/bsd.own.mk
>  ls: usr/share/mk/bsd.own.mk: No such file or directory
> 
> This doesn't make any sense to me though, unless I misunderstand a prior
> change here.  The host does the buildworld in
> /releng/11-amd64-GENERIC-snap/usr/src with MAKEOBJDIRPREFIX set to
> a non-default location, and installworld from that.
> 
> In this specific case, the build was done as an "upgrade" build, not
> "clean" build (meaning, to seed the build chroot for a clean release
> build).  I tend to alternate if MAKEOBJDIR is pristine every other week
> to try to capture cases where we expect "just 'rm -rf /usr/obj'" as
> a "fix" for problems.

That’s really odd…

Doh, one too many s/bsd.own.mk/src.opts.mk/g is the cause. fix coming.

Warner
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.freebsd.org/pipermail/svn-src-head/attachments/20140507/6cb90713/attachment.sig>