This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: testroot.pristine creation falls over copying dynamic linker


On Wed, 7 Aug 2019, DJ Delorie wrote:

> Joseph Myers <joseph@codesourcery.com> writes:
> > I have now done a MIPS test of this patch in the 
> > --enable-hardcoded-path-in-tests case and confirmed it also works in that 
> > case as well - of course in that case it copies libraries to paths under 
> > <builddir>/testroot.pristine/<builddir> so that execution within the 
> > testroot can find the libraries at the hardcoded paths.
> 
> Does that work?  IIRC I mount builddir into the testroot, so if you
> create files under testroot/builddir from *outside* the testroot,
> they'll get hidden when the container runs.

If they're hidden by identical files, that's fine; the copies may have 
been unnecessary in that case, but are harmless.  It's simply a 
consequence of the LD_TRACE_LOADED_OBJECTS=1 calls, both before and after 
the proposed patch, that in the --enable-hardcoded-path-in-tests case the 
dynamic linker gets copied to testroot.pristine/builddir because that's 
the hardcoded path.  (In fact some files get copied to 
testroot.pristine/builddir even in the absence of 
--enable-hardcoded-path-in-tests and without the patch.)

-- 
Joseph S. Myers
joseph@codesourcery.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]