On Wed, 6 Sep 2000, Arnaldo Carvalho de Melo wrote:
> Em Wed, Sep 06, 2000 at 05:37:50PM +0100, Tigran Aivazian escreveu:
> > On Mon, 4 Sep 2000, Tigran Aivazian wrote:
> >
> > > On Fri, 25 Aug 2000, Arnaldo Carvalho de Melo wrote:
> > > > now the driver init sequence is not
> > > > serialized anymore, so races are possible
> > >
> > > since when? In 2.4.0-test8-pre2 mod->init and mod->cleanup are called
> > > under global kernel lock. As for static drivers they are initialised from
> > > either start_kernel() or init()->do_basic_setup() both of which take
> > > global kernel lock.
> > >
> >
> > Ok, I understand this now - just cheched the cardbus/pcihotplug code.
> >
> > Ok, now if you do this, please leave out ne.c and hp100.c - I have the
> > cards for these two so I can do it and test it myself.
>
> humm, In fact I've already sent a patch to Paul Gortmaker only, asking
> for feedback about the PCI case (that I think should be removed from it,
> as ne2k-pci already has this), here it is, maybe it'll help you in the
> process.
excellent idea (about removing PCI code from ne.c) - what did Paul say?
Also, omit pcnet32.c from your list as I have it too (well, vmware virtual
one) so I'll do it as well.
Regards,
Tigran
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Sep 07 2000 - 21:00:27 EST