Andrew Stubbs ams@codesourcery.com wrote:
I suggest teaching the kernel to rewrite that path when it finds a non-existent interpreter. Presumably the kernel can "know" what multiarch corresponds to the traditional ABI for any given ELF flags.
The problem with this is that is this brings namespace policy into the kernel, which kernel folks have been very opposed to in the past.
And I guess that's for good reason. You probably don't want to have to update your kernel in order to support a new ABI / different dynamic linker / or just some different filesystem layout that some userspace use case comes up with for whatever reason ...
If there's kernel support needed, the only way that would be acceptable is if all naming policy involved is actually configured from user space. That's another point that would be satisfied by a bind mount solution ...
Mit freundlichen Gruessen / Best Regards
Ulrich Weigand
-- Dr. Ulrich Weigand | Phone: +49-7031/16-3727 STSM, GNU compiler and toolchain for Linux on System z and Cell/B.E. IBM Deutschland Research & Development GmbH Vorsitzender des Aufsichtsrats: Martin Jetter | Geschäftsführung: Dirk Wittkopp Sitz der Gesellschaft: Böblingen | Registergericht: Amtsgericht Stuttgart, HRB 243294