[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-libs/glibc/files/eblits: pkg_preinst.eblit

[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-libs/glibc/files/eblits: pkg_preinst.eblit

[Mike Frysinger]

On Fri, Oct 8, 2010 at 11:33 AM, Diego Petteno (flameeyes)
<flameeyes@gentoo.org> wrote:
> flameeyes    10/10/08 15:33:13
>
>  Modified:             pkg_preinst.eblit
>  Log:
>  QA: avoid using $D when testing the just-built loader, as it might cause trouble when it contains colon characters; the fix is trivial.

you'd think trivial fixes wouldnt contain bugs ...

> +               pushd "${D}"/$(get_libdir) 2>/dev/null
> +               ./ld-*.so --library-path . ${x} > /dev/null \
>                        || die "simple run test (${x}) failed"
> +               popd 2>/dev/null

yet another reason (strace also readily comes to mind) i keep asking
you to post fixes and/or notify maintainers (base-system/toolchain)
instead of just committing things.  pushd/popd do not write to stderr.
 so this change now echoes the dirs when people emerge it.  plus, this
really should be outside of the loop.
-mike

[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-libs/glibc/files/eblits: pkg_preinst.eblit

[Diego Elio Pettenò]

[-- Attachment #1: Type: text/plain, Size: 1352 bytes --]

Il giorno mer, 13/10/2010 alle 20.12 -0400, Mike Frysinger ha scritto:
> 
> yet another reason (strace also readily comes to mind) i keep asking
> you to post fixes and/or notify maintainers (base-system/toolchain)

For your information I _am_ part of base-system, so at least for strace
you really can't say anything. I also told you before that if you want
to be notified personally you should add _yourself_ to metadata beside
the herd.

> instead of just committing things.  pushd/popd do not write to stderr.
>  so this change now echoes the dirs when people emerge it.  plus, this
> really should be outside of the loop. 

I admit the mistake, fwiw it was part of debugging:

pushd &> /dev/null

became

pushd 2> /dev/null

and then forgot about changing it back. So sorry to have added ... four
lines of noise on a merge log?

On the other hand if I left it to the same process I left my personal
problem with binary merging of glibc (i.e. no news about it in months),
the user who reported to me that glibc segfaulted during binary merges
would probably still be hitting the same problem. Hmm I wonder what's
worse.

-- 
Diego Elio Pettenò — “Flameeyes”
http://blog.flameeyes.eu/

If you found a .asc file in this mail and know not what it is,
it's a GnuPG digital signature: http://www.gnupg.org/


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 490 bytes --]