clang-native-arm-cortex-a9

Renato Golin renato.golin at linaro.org
Thu Aug 7 13:20:43 PDT 2014


On 7 August 2014 19:33, Rafael EspĂ­ndola <rafael.espindola at gmail.com> wrote:
> XFALI: arm
>
> in the test. Any idea why it is not marking the test as XFAIL?

Ha, that's a funny typo. JIT tests tend to fail in random
configurations, so that's probably the reason why it hasn't failed on
the bots. I occasionally see them fail during releases, but I ignore
them every time as we don't really care for the old JIT.

I'm not sure how those tests are setup, but I thought they were tied
with the old JIT, so should have gone with the bath water...

Marking XFAIL, though, would make the tests "fail" when they pass, so
adding noise back to the bots. If no one is interested in checking
those problems on ARM/Linux, that we might restrict to x86 or Darwin
configurations only.

cheers,
--renato




More information about the llvm-commits mailing list