On Jun 13, 2014, at 10:49 AM, Glen Barber <g...@freebsd.org> wrote:

> On Fri, Jun 13, 2014 at 10:34:50AM -0600, Warner Losh wrote:
>> 
>> On Jun 10, 2014, at 11:22 AM, Nathan Whitehorn <nwhiteh...@freebsd.org> 
>> wrote:
>> 
>>> On 06/10/14 10:05, Julio Merino wrote:
>>>> Author: jmmv
>>>> Date: Tue Jun 10 17:05:41 2014
>>>> New Revision: 267332
>>>> URL: http://svnweb.freebsd.org/changeset/base/267332
>>>> 
>>>> Log:
>>>>  Strip out the test suite from the installation media.
>>>>     This affects the disc1.iso, dvd1.iso and bootonly.iso files.
>>>>     Obtained from: gjb
>>>> 
>>>> Modified:
>>>>  head/release/Makefile
>>>> 
>>>> Modified: head/release/Makefile
>>>> ==============================================================================
>>>> --- head/release/Makefile  Tue Jun 10 17:04:30 2014        (r267331)
>>>> +++ head/release/Makefile  Tue Jun 10 17:05:41 2014        (r267332)
>>>> @@ -155,7 +155,7 @@ system: packagesystem
>>>>    mkdir -p release
>>>>    cd ${WORLDDIR} && ${IMAKE} installkernel installworld distribution \
>>>>            DESTDIR=${.OBJDIR}/release WITHOUT_RESCUE=1 
>>>> WITHOUT_KERNEL_SYMBOLS=1 \
>>>> -          WITHOUT_PROFILE=1 WITHOUT_SENDMAIL=1 WITHOUT_ATF=1 
>>>> WITHOUT_LIB32=1
>>>> +          WITHOUT_PROFILE=1 WITHOUT_SENDMAIL=1 MK_TESTS=no WITHOUT_LIB32=1
>>>> # Copy distfiles
>>> 
>>> Shouldn't this be "WITHOUT_TESTS" or the like? I thought we weren't 
>>> supposed to set MK_* at the command line.
>> 
>> We recently started allowing MK_TESTS=no on the command line. In general, 
>> all of those WITHOUT_xxx should become MK_xxx=no in release/Makefile. Having 
>> the MK_xxxx version isolates us further from (a) the current defaults and 
>> (b) config files that turn it to the other direction.
>> 
> 
> The WITH*_FOO -> MK_FOO change is on my TODO this weekend.  I want to
> run at least one full build before committing it.

There’s no rush. What’s there now works, except in extreme circumstances. 
Testing of this stuff is good. It is way too easy to think a change is a NOP 
when in fact it breaks something subtle...

Warner

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

Reply via email to