Re: [patch 1/4] Linux Kernel Markers - Architecture Independent Code

[Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>]

* Frank Ch. Eigler (fche@redhat.com) wrote:
> Hi -
> 
> On Fri, Sep 21, 2007 at 08:58:19AM -0400, Mathieu Desnoyers wrote:
> > [...]
> > > > Current systemtap marker support code relies on the __markers_strings
> > > > section.
> > > Let users know that in comment above section definition in ld script.
> > [...]
> >         /* Markers: strings (used by SystemTAP) */                      \
> > [...]
> 
> I did not mean to imply that this was a necessary state of affairs.
> 
> The marker metadata must be stored in at least one place in the kernel
> image - this just happens to be a convenient one that David Smith's
> recent systemtap code used.  Without it, we'd probably have to do a
> more complicated search, following the pointers within the __markers
> structs.  That could work, but it hasn't been built/tested.
> 
> So, this proposed change (removal of this section) would break
> systemtap, and we have to jump through more hoops to make it work
> again.  Is the change worth it?
> 

I guess so. Getting the markers as clean as we can is very important for
kernel inclusion.

Mathieu

-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68