Re: Back from the dead, with updates!
Tobias Nygren <[email protected]> Thu, 24 Sep 2015 22:08:56 +0200
| Newsgroups | gmane.os.netbsd.ports.sgimips |
|---|---|
| Message-ID | <[email protected]> |
On Thu, 24 Sep 2015 14:26:41 -0400 Michael <[email protected]> wrote: > Some things that happened to the sgimips port recently: > - it's usable again, thanks to skrll@ fixing a bunch of cache handling bugs months ago Hi! Exciting stuff! But sadly still not working for me. All three GENERIC kernels built from HEAD crash early. I will try to dust off my other O2 in case this one has gone bad. Could you please upload a kernel binary that works for you? Here are logs. Crash pc seems to be below the start of kernel text. Obtaining /netbsd.GENERIC32_IP3x from server 6377168+135024 entry: 0x80069000 phys segment: 0x5e000 @ 0x2000 adding 0x5e000 @ 0x2000 to freelist 0 phys segment: 0x636000 @ 0x60000 adding 0x8000 @ 0x60000 to freelist 0 skipping 0x62e000 @ 0x68000 (kernel) phys segment: 0x6ba000 @ 0x696000 adding 0x6ba000 @ 0x696000 to freelist 0 phys segment: 0xec00000 @ 0x1400000 adding 0xec00000 @ 0x1400000 to freelist 0 sgimips_bus_dma_init pmap_steal_memory: seg 0: 0x2 0x2 0x60 0x60 pmap_steal_memory: seg 0: 0x38 0x38 0x60 0x60 pid 0(system): trap: cpu0, address error (load or I-fetch) in kernel mode status=0x20000003, cause=0x8010, epc=0x803ca760, vaddr=0xdeadbeef tf=0x80039ca0 ksp=0x80039d40 ra=0x8008dfdc ppl=0 kernel: address error (load or I-fetch) trap nullvoid Stopped in pid 0.1 (system) at 803ca760: lw v0,88(a0) db> bt 0x80039d40: 0+803ca760 (0,1000,1000,400004) ra 0 sz 0 User-level: pid 0.1 db> Obtaining /netbsd.GENERIC64_32_IP3x from server 6106032+135424 entry: 0x80069000 phys segment: 0x5e000 @ 0x2000 adding 0x5e000 @ 0x2000 to freelist 0 phys segment: 0x5f4000 @ 0x60000 adding 0x8000 @ 0x60000 to freelist 0 skipping 0x5ec000 @ 0x68000 (kernel) phys segment: 0x6fc000 @ 0x654000 adding 0x6fc000 @ 0x654000 to freelist 0 phys segment: 0xec00000 @ 0x1400000 adding 0xec00000 @ 0x1400000 to freelist 0 sgimips_bus_dma_init pmap_steal_memory: seg 0: 0x2 0x2 0x60 0x60 pmap_steal_memory: seg 0: 0x38 0x38 0x60 0x60 pid 0(system): trap: cpu0, address error (load or I-fetch) in kernel mode status=0x20000003, cause=0x8010, epc=0x80393314, vaddr=0xdeadbeef tf=0x80039b80 ksp=0x80039cc0 ra=0xffffffff8008b71c ppl=0 kernel: address error (load or I-fetch) trap nullvoid Stopped in pid 0.1 (system) at 80393314: lw v0,88(a0) db> Obtaining /netbsd.GENERIC64_IP3x from server 6688752+192272 entry: 0xffffffffa0069000 phys segment: 0x5e000 @ 0x2000 adding 0x5e000 @ 0x2000 to freelist 0 phys segment: 0x690000 @ 0x60000 adding 0x8000 @ 0x60000 to freelist 0 skipping 0x688000 @ 0x68000 (kernel) phys segment: 0x660000 @ 0x6f0000 adding 0x660000 @ 0x6f0000 to freelist 0 phys segment: 0xec00000 @ 0x1400000 adding 0xec00000 @ 0x1400000 to freelist 0 sgimips_bus_dma_init pmap_steal_memory: seg 0: 0x2 0x2 0x60 0x60 pmap_steal_memory: seg 0: too small for 98 pages pmap_steal_memory: seg 1: 0x60 0x60 0x68 0x68 pmap_steal_memory: seg 1: too small for 98 pages pmap_steal_memory: seg 2: 0x6f0 0x6f0 0xd50 0xd50 pmap_steal_memory: seg 0: 0x2 0x2 0x60 0x60 (did not make it to db prompt) -Tobias