Re: anita tests timing out

[Martin Husemann <martin%homeworld.netbsd.org@localhost>]

On Wed, Nov 23, 2011 at 07:27:10PM +0200, Andreas Gustafsson wrote:
> Christos Zoulas wrote:
> > Curiously:
> > 
> > atf-run
> [...]
> > info: uname.version, NetBSD 5.99.56 (QUASAR) #72: Sat Nov 19 21:23:32 EST 
> > 2011 
> > christos%quasar.astron.com@localhost:/usr/src/sys/arch/amd64/compile/QUASAR
> 
> Of course a build from Nov 19 works.  The bug wasn't even introduced
> until the 20th, and then not fixed by your commit of the 22nd.

On sparc64 a 

  cd /usr/tests/modules
  atf-run | atf-report

results in an unkillable process t_modctl in the first test with this backtrace:

vsnprintf(e2cd800, 400, 176d278, f05da18, 39a7d30, 0) at netbsd:vsnprintf+0x1c
snprintf(e2cd800, 400, 176d278, 18c4688, e2cdc00, e2cdc00) at netbsd:snprintf+0x
24                                                                             
module_load_vfs(2, 0, 0, 39a7d30, f05db30, 0) at netbsd:module_load_vfs+0x224
module_do_load(e2cdc00, 0, 0, 39deb70, 0, 0) at netbsd:module_do_load+0x394  
module_load(e2cdc00, 0, 39deb70, 0, f05a000, e493bd0) at netbsd:module_load+0x50
                                                                               
sys_modctl(16, f05ddd0, f05de10, 0, 2, 0) at netbsd:sys_modctl+0x384
syscall_plain(f05ded0, f05ddd0, 41144fd8, ffffffffffffa531, 41144fd8, 2) at netb
sd:syscall_plain+0x134                                                         


Martin