Bug#257001: more on m68k kernel build bugs...
On Fri, Jul 16, 2004 at 01:26:17PM +0200, Sven Luther wrote:
> On Fri, Jul 16, 2004 at 09:51:22AM +0200, Christian T. Steigies wrote:
> > On Fri, Jul 16, 2004 at 02:30:33PM +0900, Carsten Haitzler wrote:
> > > After some more digging I have found even more issues. as of kernel 2.3.17 the
> > > mac 68k specific code changed a fair bit - they introduced calls that do not
> > > exist anywhere in the kernel. specifically i have found:
> > >
> > > extern __u32 adb_read_time(void);
> > >
> > > it is defined as an extern symbol - but NOWHERE in the code for the kernel can i
> > > find that function being provided or any #define changing its name etc. this
> > > code was introduced as of 2.3.17 and would break the kernel for 68k mac builds
> > > as of this version.
> >
> > Please install and apply kernel-patch-2.4.26-m68k.
> >
> > arch/m68k/mac/misc.c
> >
> > /*
> > * Return the current time as the number of seconds since January 1, 1904.
> > */
> >
> > static long adb_read_time(void)
> > {
> > volatile struct adb_request req;
> > long time;
> >
> > adb_request((struct adb_request *) &req, NULL,
> > ADBREQ_RAW|ADBREQ_SYNC,
> > 2, CUDA_PACKET, CUDA_GET_TIME);
> >
> > time = (req.reply[3] << 24) | (req.reply[4] << 16)
> > | (req.reply[5] << 8) | req.reply[6];
> > return time - RTC_OFFSET;
> > }
> >
> > Now I am not saying you will get a working kernel with this, but you can
> > compile one, as you can see with kernel-image-2.4.26-mac. Upstream, ie the
> > mac kernel hackers, are not working on 2.4 anymore, instead they are working
> > on 2.6. Please see the linux-m68k mailing list for more.
>
> On more arch which should go with 2.6 for sarge ?
>From Raster's information it seems that 2.4 for mac is not in a good state.
I have asked him to test kernel-patch-2.4.26-m68k (he is one of my
colleagues) and let me know how that goes. However, even if the kernel
builds neither he nor I have a way to test it. Also, Crest is currently
down, so he can't do anything right now.
I am wondering if we can really support the kernel-source-2.4.X (X>=26)
package for m68k. If not, what is the best thing to do? Do any other
members of the kernel team have any thoughts? I have CCed debian-68k to
see if they have any thoughts.
> Christian, what is the current status of 2.6 m68k kernels ? Do they run
> on at least some hardware (amiga, or so the rumor goes) ?
--
Horms
Reply to: