Christoph Hellwig wrote: > On Sun, Nov 11, 2007 at 06:01:37PM +0100, Jochen Friedrich wrote: >> --- a/arch/powerpc/sysdev/commproc.c >> +++ b/arch/powerpc/sysdev/commproc.c >> @@ -51,6 +51,8 @@ static void m8xx_cpm_dpinit(void); >> static uint host_buffer; /* One page of host buffer */ >> static uint host_end; /* end + 1 */ >> cpm8xx_t __iomem *cpmp; /* Pointer to comm processor space */ >> +EXPORT_SYMBOL(cpmp); > > Sorry, but this is a nightmare waiting to happen. Please define > proper accessors instead.
We have proper accessors: in_be32, out_be16, etc. Now, is the huge struct encompassing all of immr space a bad idea? Sure. Are we working towards getting rid of it? Yes. Is that a reason to keep modules from working in the meantime? No. > (And get rid of the typedef while you're at it) Again, changing all the users is a bit beyond the scope of this patch. -Scott _______________________________________________ Linuxppc-embedded mailing list Linuxppc-embedded@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-embedded
