HAL2 problem solved

Hi,

look what my Indy tells me now:

resetting global isr:0018
reset done isr:0000
reactivation done isr:0018
SGI HAL2 Processor, Revision 4.1.0
hal2: checking registers
hal2: waiting isr:0019 idr0:0123 idr1:ffff idr2:0000 idr3:0000
hal2: finished waiting at cnt:1000 isr:0018 idr0:0123 idr1:ffff idr2:0000 idr3:0000
hal2: wrote #1
hal2: waiting isr:0019 idr0:0132 idr1:0231 idr2:0000 idr3:0000
hal2: finished waiting at cnt:1000 isr:0018 idr0:0132 idr1:0231 idr2:0000 idr3:0000
hal2: wrote #2
hal2: waiting isr:0019 idr0:0123 idr1:0231 idr2:0000 idr3:0000
hal2: finished waiting at cnt:1000 isr:0018 idr0:0123 idr1:0231 idr2:0000 idr3:0000
hal2: read #1
hal2: waiting isr:0019 idr0:0132 idr1:0231 idr2:0000 idr3:0000
hal2: finished waiting at cnt:1000 isr:0018 idr0:0132 idr1:0231 idr2:0000 idr3:0000
hal2: read #2
hal2: waiting isr:0019 idr0:0231 idr1:0231 idr2:0000 idr3:0000
hal2: finished waiting at cnt:1000 isr:0018 idr0:0231 idr1:0231 idr2:0000 idr3:0000
hal2: read #3
hal2: card found
sgiaudio: initializing

The problem was, that the access to HAL2 register must be 32bit wide, but they were only 16bit (I'll check in the fixed hal2.h in a couple of minutes). Silly me did the change yesterday, but the .o didn't get recompiled due to messed up dependencies:-( So I did some changes today and it worked, removed the changes and it still worked.

Thomas.

   This device has completely bogus header. Compaq scores again :-|
It's a host bridge, but it should be called ghost bridge instead ;^)
                                        [Martin `MJ' Mares on linux-kernel]