Re: Automated report: NetBSD-current/macppc ATF test suite completion failure

[Andreas Gustafsson <gson%NetBSD.org@localhost>]

S.P.Zeidler wrote:
> I just updated babylon5 and all its packages (10.1 and 2024Q4-to-be)
> and if I understand
> https://releng.netbsd.org/b5reports/macppc/2024/2024.12.21.07.31.23/test.log
> correctly, /usr/pkg/bin/qemu-system-ppc doesn't work so well now.
> 
> Anybody know if that is a known issue in qemu-9.1.2nb1 or if it needs
> extra twiddles that a qemu-8 didn't need?

This does look like a qemu regression to me.  Rerunning qemu manually
shows a more detailed error message; for some reason everything after
"fatal:" was missing from the testbed log:

  qemu: fatal: Raised an exception without defined vector 94

  NIP fdb4d440   LR fdb4dc20 CTR fd61a340 XER 20000000 CPU#0
  MSR 0200d032 HID0 809400a4  HF 02004012 iidx 0 didx 0
  TB 00000000 889032049 DECR 160653
  GPR00 0000000000000125 00000000ffffe8b0 00000000fde90008 0000000000000000
  GPR04 0000000000000000 00000000fdcf3bac 00000000fdcf3b48 0000000000000001
  GPR08 0000000000000000 00000000fde90008 00000000ffffe8b0 00000000fdb4dc14
  GPR12 0000000044004482 0000000000000000 00000000fdee0f40 00000000fdef5830
  GPR16 00000000fdef5808 0000000000000000 00000000ffffea94 00000000ffffeac8
  GPR20 00000000fdee0f80 0000000001800114 0000000000000000 0000000000000001
  GPR24 0000000000000000 00000000fdef5820 0000000000000001 00000000fdb4da2c
  GPR28 00000000ffffe9c0 00000000ffffe8f8 00000000fdcf4b4c 00000000fdcf3b48
  CR 24004482  [ E  G  -  -  G  G  L  E  ]     RES 004@ffffffff
  FPR00 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR04 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR08 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR12 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR16 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR20 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR24 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPR28 0000000000000000 0000000000000000 0000000000000000 0000000000000000
  FPSCR 00000000
   SRR0 fd61a360  SRR1 0200d032    PVR 000c0209 VRSAVE 00000000
  SPRG0 00c03000 SPRG1 44004482  SPRG2 fde90008  SPRG3 00000000
  SPRG4 00000000 SPRG5 00000000  SPRG6 00000000  SPRG7 00000000
   SDR1 00e0001f   DAR cd528000  DSISR 42000000
  Abort trap

The same thing happens with a qemu hosted on Linux and built from
up-to-date sources (65cb7129f4160c7e07a0da107f888ec73ae96776),
so it's not specific to NetBSD hosts, nor fixed.
-- 
Andreas Gustafsson, gson%NetBSD.org@localhost