I just tried a 2.0 beta MP kernel (from sources checked out on July 26) on a
couple of Sparc 20's.  On the first machine:

  mainbus0 (root): SUNW,SPARCstation-20: hostid 7274ea58
  cpu0 at mainbus0: mid 8: TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
  cpu0: physical 20K instruction (64 b/l), 16K data (32 b/l), 1024K external (32 b/l): cache enabled
  cpu1 at mainbus0: mid 10: TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
  cpu1: physical 20K instruction (64 b/l), 16K data (32 b/l), 1024K external (32 b/l): cache enabled

it crashed once before getting to multi-user with:

  panic: dma0: cannot allocate DVMA address
  Begin traceback...
  0x0(0xf0c4c800, 0xf0b6e19c, 0xf0b6e1a0, 0x0, 0xf3176244, 0x1) at netbsd:esp_dma_setup+0x18
  esp_dma_setup(0xf0b6e000, 0xf0b6e19c, 0xf0b6e1a0, 0x0, 0xf3176244, 0x30300) at netbsd:ncr53c9x_select+0x24c
  ncr53c9x_select(0xf0b6e000, 0xf0c9e000, 0x0, 0xf0190c44, 0x0, 0x8) at netbsd:ncr53c9x_scsipi_request+0x23c

and next when compiling a package with:

  panic: kernel fault
  Begin traceback...
  0x0(0x9, 0x5336, 0xf0d37df0, 0xf37dfba0, 0x0, 0xf02e0340) at 0xf0006254
  0xf0006254(0xf0d86cf8, 0x0, 0x1000000, 0x0, 0x700, 0xde000) at netbsd:pv_link4m+0x6c
  pv_link4m(0xf04bd820, 0xf37e1100, 0x10145000, 0xf37dfd18, 0xf0d91700, 0xf0d91000) at netbsd:pmap_enu4m+0x228

On the second machine (3 x HS100), it never makes it multi-user.  As it
crashes again when it tries to dump, I don't have a record of the original
crash.  This second machine is the one that I tested the SMP fixes pullup
(ticket 179) on back in April.  It ran perfectly then (I tested it using
build.sh -j 4).

Is anyone sucessfully running MP on a recent 2.0 beta?

Thanks,

J

-- 
  My other computer also runs NetBSD    /        Sailing at Newbiggin
        http://www.netbsd.org/        /   http://www.newbigginsailingclub.org/