|
From: Mark W. <ma...@kl...> - 2023-04-03 22:26:53
|
On Tue, Apr 04, 2023 at 07:42:15AM +1000, Nicholas Nethercote wrote: > On Tue, 4 Apr 2023 at 07:08, Mark Wielaard <ma...@kl...> wrote: > > > > > > I looked at some of the failing builders, they all fail like this: > > > > > > > [...] > > > > integration > > > > -FAIL: qawo(f456) elist (7.25063790881233303e-15 observed vs > > 7.25922435194575979e-15 expected) > > > > interpolation > > > > [...] > > > > rng > > > > +FAIL: random32-bsd, 10000 steps (852261210 observed vs 1663114331 > > expected) > > > > +FAIL: random64-bsd, 10000 steps (210970120 observed vs 864469165 > > expected) > > > > +FAIL: random32-libc5, 10000 steps (367802360 observed vs 1967452027 > > expected) > > > > +FAIL: random64-libc5, 10000 steps (221021662 observed vs 2106639801 > > expected) > > > > roots > > > > So the (now new) failures are the random tests. > > > > And it looks like `qawo` has changed to passing, too? qawo was only failing on i386. so the logic in the makefile assumes everything is passing and if not, that it must be a run on i386 where there is only one difference, the FAIL: qawo. See the (still passing) i386 run: https://builder.sourceware.org/buildbot/#/builders/40/builds/310/steps/6/logs/stdio Cheers, Mark |