Messages in this thread | | | Date | Wed, 24 Nov 1999 14:50:42 +0100 (CET) | From | Andrea Arcangeli <> | Subject | Re: [patch] alpha port 2.3.29pre3 |
| |
On Tue, 23 Nov 1999, Richard Henderson wrote:
>If you havn't already, you'll soon get angry mail from DaveM >about this, as this breaks Sparc. I of course tried for that >very change when the code was first written.
Hmm, maybe we can let Sparc to work fine with the change applyed? The thing I did looks like the natural thing to do. It's not a black magic hack. So IMHO if possible Sparc should be made to work with these changes applyed. Dave? Is it possible?
>> +/* The IO address space is larger than 0xffff */ >> +#define TSUNAMI_IO_SPACE (TSUNAMI_CONF(0) - TSUNAMI_IO(0)) >> +#define TSUNAMI_MEM_SPACE (_TSUNAMI_IACK_SC(0) - TSUNAMI_MEM(0)) > >Why this change? The IO address space may _be_ larger than 64k, but
hose->io_space->start = TSUNAMI_IO(index) - TSUNAMI_IO_BIAS; hose->io_space->end = hose->io_space->start + TSUNAMI_IO_SPACE;
Some device here wakeup (after SRM) with an _I/O_ base_address set at an offset larger than 64k and so without telling the kernel that the iospace of such bus is larger than 64k, the find_parent_resource breaks and we can't reinitialize the device (and reinitializing if possible we'll put the base address of the device in the first 64k). Please tell me if you don't agree on this change. If you don't agree the pdev_assign_unassigned_resources() stuff should be redesigned. Of course I tried to do the smaller changes possible in the common code (and I just broken Sparc...).
Andrea
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |