Re: [PATCH] Link dynamic tests with newly built glibc

[Roland McGrath <roland at hack dot frob dot com>]

> > The status quo has the nice property that you can just take a test binary
> > and run it against an installed libc manually, which people actually do
> > fairly often (including manual versions of cross-testing).  I certainly
> > want us to support cross-testing in a first-class way as soon as we can
> > manage to do it cleanly.
> 
> My proposal doesn't change it.

Huh?  The patch unconditionally changes the links of test binaries to embed
a build directory name for ld.so rather than the installed name.  You can't
use such a binary elsewhere.

> > There is no way to get the behavior HJ wants without regressing on some of
> > the other cases in which test binaries as we build them today can be used
> 
> The only thing my proposal tests less is "ld.so --library-path" since it
> is used less.

I never said you'd regress what 'make check' tests.  I said you'd regress
other scenarios in which the test binaries can be used.

> My proposal makes testing closer to what is actually used than the
> current scheme and it exposed a real bug:

I clearly didn't say that it's not desireable to be able to test this way.
I said it's not desireable to preclude (or make more difficult) other
things we can do today, as a side-effect of enabling this kind of testing.