I was compiling a new 1.5 kernel, and ran into this:

jon@bullbox$ cc  -O2 -msb -Wall -Wstrict-prototypes -Wmissing-prototypes  -Wpointer-arith -Wno-format -Wno-uninitialized -Wno-main -I. -I../../../../arch -I../../../.. -nostdinc -DACCOUNTING -DLKM -DSHMMAXPGS="0x400" -DSETUIDSCRIPTS -DNS381 -DMAXUSERS=8 -D_KERNEL  -c ../../../../arch/pc532/pc532/db_trace.c
../../../../arch/pc532/pc532/db_trace.c:80: warning: initialization from incompatible pointer type
../../../../arch/pc532/pc532/db_trace.c:81: warning: initialization from incompatible pointer type

The diff below fixes it, but I'm not sure what to do with that
(*pr)("error") thing.  As a quick hack, I changed a parameter which
will have no effect, but get's past the error I get when I used
the statement 0; instead.

diff -r1.9 db_trace.c
56,57c56
< static int    db_spec_regs __P((struct db_variable *, db_expr_t *, int,
<     void (*)(const char *, ...)));
---
> static int    db_spec_regs __P((struct db_variable *, db_expr_t *, int));
387c386
< db_spec_regs(vp, valp, what, pr)
---
> db_spec_regs(vp, valp, what)
391d389
<       void (*pr) __P((const char *, ...));
451,452c449,451
<               (*pr)("Internal error, unknown register in db_spec_regs");
<       return(0);
---
>               what = 0;
> /*            (*pr)("Internal error, unknown register in db_spec_regs");
> */    return(0);

Is there some other way to get that ddb error printing function,
or a better way to handle the possible error than to ignore it?
I checked all the other ports, and the ones that treat registers
this way only have zero or one special registers, so they don't
bother with checking for a possible error like this.

I don't have the ability to put this in -current right now, but
think it should be pulled up on the 1.5 branch, especially if the
port gets orphaned.  Anyone have any suggestions on that?

Jon