Resent-Date: Wed, 28 Apr 1999 10:16:36 +0200 (MET DST) To: Michael Schmitz Cc: Jes Sorensen , Linux/m68k Mailing List Subject: Re: 2.2.6 References: <199904231758.TAA62202@rsplus04.cern.ch> <19990424145207.C2337@quango.watervalley.net> <19990425183054.A5089@localhost.watervalley.net> <19990425174828.B20693@schmitzm> <37249EE1.45867BE2@lbl.gov> X-Yow: I haven't been married in over six years, but we had sexual counseling every day from Oral Roberts!! From: Andreas Schwab Date: 28 Apr 1999 10:15:00 +0200 In-Reply-To: Michael Schmitz's message of "Mon, 26 Apr 1999 10:14:09 -0700" Resent-From: linux-m68k@phil.uni-sb.de Michael Schmitz writes: |> Jes Sorensen wrote: |> > Michael> My guess is it was changed from m68k_register_serial to |> > Michael> register_serial for a reason? It's consistent in all serial |> > Michael> drivers, at least. |> > |> > Actually it is the other way round, this was part of Richard's Q40 |> > patch and I kinda liked the approach as I want the current m68k serial |> > drivers to go away. |> |> Sorry for crediting you with the idea :-) Any guess on why it fails with |> modules? And I've not tested if the Atari serials get rearranged that |> way either. (I think that's one of the things m68k_register_serial does |> different, right?) Use this patch: --- linux/arch/m68k/kernel/m68k_ksyms.c.~1~ Wed Feb 24 20:48:04 1999 +++ linux/arch/m68k/kernel/m68k_ksyms.c Mon Apr 26 23:50:26 1999 @@ -18,6 +18,7 @@ #include #include #include +#include asmlinkage long long __ashrdi3 (long long, int); extern char m68k_debug_device[]; @@ -54,6 +55,8 @@ EXPORT_SYMBOL(disable_irq); EXPORT_SYMBOL(kernel_set_cachemode); EXPORT_SYMBOL(kernel_thread); +EXPORT_SYMBOL(register_serial); +EXPORT_SYMBOL(unregister_serial); /* Networking helper routines. */ EXPORT_SYMBOL(csum_partial_copy); Andreas. -- Andreas Schwab "And now for something schwab@issan.cs.uni-dortmund.de completely different" schwab@gnu.org