Mikael Simberg

cscs-ci run

cscs-ci run

cscs-ci run

This could have been triggered by #643, right? I don't mind keeping it open until we're more sure, but it seems like that same failure could lead to similar failures...

Precision error on `test_eigensolver`: https://gitlab.com/cscs-ci/ci-testing/webhook-ci/mirrors/4700071344751697/7514005670787789/-/jobs/3465819735#L1355. (Ran on #668, but unlikely to be caused by it.)

A couple more (again `test_eigensolver` and `test_gen_eigensolver`): https://gitlab.com/cscs-ci/ci-testing/webhook-ci/mirrors/4700071344751697/7514005670787789/-/jobs/3477987284.

Possible hang in the tridiagonal eigensolver tests: https://gitlab.com/cscs-ci/ci-testing/webhook-ci/mirrors/4700071344751697/7514005670787789/-/jobs/5543375695#L2169. However, this may also be caused by external problems.

Another possible hang, this time in the hermitian multiplication test: https://gitlab.com/cscs-ci/ci-testing/webhook-ci/mirrors/4700071344751697/7514005670787789/-/jobs/5621888281#L2326. This one is on https://github.com/eth-cscs/DLA-Future/pull/993 and may be caused by the changes there, but it seems unlikely.

https://github.com/pika-org/pika/pull/976 seems to fix issues we've seen in the tridiagonal eigensolver test.

Segfault in `test_multiplication_hermitian` (after tests have run): https://gitlab.com/cscs-ci/ci-testing/webhook-ci/mirrors/4700071344751697/7514005670787789/-/jobs/7643366038#L2553. This one has the following backtrace: ``` 60: Backtrace: 60: /root/DLA-Future.bundle/usr/lib/libmimalloc.so.2(__libc_free+0x39)[0x7ffff6e0d2d9] 60: /root/DLA-Future.bundle/usr/lib/libgtest.so.1.13.0(_ZN7testing8internal11ThreadLocalISt6vectorINS0_9TraceInfoESaIS3_EEE11ValueHolderD0Ev+0x4e)[0x7ffff7120f6e] 60: /lib/x86_64-linux-gnu/libc.so.6(+0x91691)[0x7ffff6355691] 60: /lib/x86_64-linux-gnu/libc.so.6(+0x9494a)[0x7ffff635894a] 60: /lib/x86_64-linux-gnu/libc.so.6(+0x126850)[0x7ffff63ea850] ``` Maybe indicating...