<arigato>
failing pypyjit tests on default: yes, that's expected, that's because of the latest fix in the jit there are a few more guards produced. We need to think if we can do something, or else fix the tests
<arigato>
failing pypyjit tests on default: yes, that's expected, that's because of the latest fix in the jit there are a few more guards produced. We need to think if we can do something, or else fix the tests
<arigato>
failing pypyjit tests on default: yes, that's expected, that's because of the latest fix in the jit there are a few more guards produced. We need to think if we can do something, or else fix the tests
<arigato>
failing pypyjit tests on default: yes, that's expected, that's because of the latest fix in the jit there are a few more guards produced. We need to think if we can do something, or else fix the tests
<arigato>
failing pypyjit tests on default: yes, that's expected, that's because of the latest fix in the jit there are a few more guards produced. We need to think if we can do something, or else fix the tests
<mattip>
ok. How will we decide?
<mattip>
ok. How will we decide?
<mattip>
ok. How will we decide?
<mattip>
ok. How will we decide?
<mattip>
ok. How will we decide?
<arigato>
see if the extra guards are actually making sense
<arigato>
see if the extra guards are actually making sense
<arigato>
see if the extra guards are actually making sense
<arigato>
see if the extra guards are actually making sense
<arigato>
see if the extra guards are actually making sense
<arigato>
unicode-utf8: in this branch, there should be no more unicodegetitem() at all (or mostly), instead a more complicated sequence decoding utf-8. what you see is probably OK
<arigato>
unicode-utf8: in this branch, there should be no more unicodegetitem() at all (or mostly), instead a more complicated sequence decoding utf-8. what you see is probably OK
<arigato>
unicode-utf8: in this branch, there should be no more unicodegetitem() at all (or mostly), instead a more complicated sequence decoding utf-8. what you see is probably OK
<arigato>
unicode-utf8: in this branch, there should be no more unicodegetitem() at all (or mostly), instead a more complicated sequence decoding utf-8. what you see is probably OK
<arigato>
unicode-utf8: in this branch, there should be no more unicodegetitem() at all (or mostly), instead a more complicated sequence decoding utf-8. what you see is probably OK
<arigato>
the 64-bit buildslave on bencher4: it is run inside a chroot, it's documented in /root/README-CHROOT
<arigato>
the 64-bit buildslave on bencher4: it is run inside a chroot, it's documented in /root/README-CHROOT
<arigato>
the 64-bit buildslave on bencher4: it is run inside a chroot, it's documented in /root/README-CHROOT
<arigato>
the 64-bit buildslave on bencher4: it is run inside a chroot, it's documented in /root/README-CHROOT
<arigato>
the 64-bit buildslave on bencher4: it is run inside a chroot, it's documented in /root/README-CHROOT
<mattip>
thanks
<mattip>
thanks
<mattip>
thanks
<mattip>
thanks
<mattip>
thanks
<mattip>
I guess the problem is pypy itself needs hypothesis for -A tests, not the host python
<mattip>
I guess the problem is pypy itself needs hypothesis for -A tests, not the host python
<mattip>
I guess the problem is pypy itself needs hypothesis for -A tests, not the host python
<mattip>
I guess the problem is pypy itself needs hypothesis for -A tests, not the host python
<mattip>
I guess the problem is pypy itself needs hypothesis for -A tests, not the host python
<arigato>
so far, we didn't need that
<arigato>
so far, we didn't need that
<arigato>
so far, we didn't need that
<arigato>
so far, we didn't need that
<arigato>
so far, we didn't need that
<mattip>
I will fix the tests to deal with missing hypothesis
<mattip>
I will fix the tests to deal with missing hypothesis
<mattip>
I will fix the tests to deal with missing hypothesis
<mattip>
I will fix the tests to deal with missing hypothesis
<mattip>
I will fix the tests to deal with missing hypothesis