On Jun 7, 2014, at 3:48 PM, Marius Strobl <[email protected]> wrote:
> On Sat, Jun 07, 2014 at 09:23:47AM -0600, Warner Losh wrote:
>> Why? There should be no -gdwarf-2 in the tree, that was a workaround that
>> has since been obsoleted by (a) -g automatically changes to this in the
>> makefile system and (b) the default was reverted back to dwarf 2. This
>> wasn?t a mismerge, but an intentional fix.
>
> Because despite what you say, -gdwarf-2 actually still is in stable/9 and
> stable/10 as of now:
> marius@alchemy:/home/marius/co > grep -l -- -gdwarf-2 {9,10}/src/sys/*/conf/*
> 9/src/sys/amd64/conf/GENERIC
> 9/src/sys/i386/conf/GENERIC
> 9/src/sys/i386/conf/XEN
> 9/src/sys/sparc64/conf/GENERIC
> 10/src/sys/amd64/conf/GENERIC
> 10/src/sys/i386/conf/GENERIC
> 10/src/sys/i386/conf/XEN
> 10/src/sys/sparc64/conf/GENERIC
>
> And also because replacing -gdwarf-2 with -g just wasn't part of r266820,
> doing the former when MFCing that revision as r267049 simply was a mismerge
> on my part.
> Apart from these I'm fully aware that -gdwarf-2 was/is only transient, but
> we're just not fully back to -g, yet.
Oh, yea, the MFC issue…. Yea, you’re right. That mess hasn’t been sorted out
yet :( I’ll sort 10 out this weekend and 9 out after 9.3, per the RE’s request.
Warner
signature.asc
Description: Message signed with OpenPGP using GPGMail
