On 21/03/14 17:49, Justin Cormack wrote:
>> Is there any reason why we need a .ro for the build phase?  Could we
>> just put all the magic in a rumprun libc.so and be able to run
>> off-the-shelf NetBSD dynamic binaries?  We may have discussed this
>> earlier, but I've forgotten again.  I'm guessing your cache is hot on
>> the details of rumprun, so now's a good time to (re-)discuss.
>
> The issue is that it is linked against host and NetBSD libc so there would
> be symbol collisions.

Right, but can't all the symbols in the binary be renamed, and then we 
hook up the necessary calls to host in libc.so?  I'm sure it's easier 
said than done, though ...

> The way it works now is to resolve all NetBSD symbols except the syscalls
> but this is not so easy in the dynamic case.
>
> I think it is possible though. That was why I was working on static linking
> before. It might require a custom host toolchain.

Ok, please keep us posted on that research!

p.s. updated the rump-pktgenif configure script to rumprun executive. 
The script, IMHO, looks very approachable now:
https://github.com/rumpkernel/rump-pktgenif/blob/master/tool/config.sh.example

------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
rumpkernel-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rumpkernel-users

Reply via email to