Bug#1101686: mpich: triggers test errors: MPII_init_gpu(51)....: gpu_init failed

2025-05-02 Thread Drew Parsons
Package: mpich Followup-For: Bug #1101686 Control: forwarded -1 https://github.com/pmodels/mpich/issues/7399 Post scriptum for this mpich gpu bug, upstream received notice about the problem with gpu support https://github.com/pmodels/mpich/issues/7399 and has now applied a patch https://github.c

Bug#1101686: mpich: triggers test errors: MPII_init_gpu(51)....: gpu_init failed

2025-04-08 Thread Drew Parsons
Package: mpich Version: 4.3.0-4 Followup-For: Bug #1101686 Control: reopen 1101686 Alistair, mpich 4.3.0-4 is still triggering gpu errors in armci-mpi, bagel and eztrace. -- System Information: Debian Release: trixie/sid APT prefers unstable-debug APT policy: (500, 'unstable-debug'), (500,

Bug#1101686: mpich: triggers test errors: MPII_init_gpu(51)....: gpu_init failed

2025-04-01 Thread Drew Parsons
Package: mpich Version: 4.3.0-4 Followup-For: Bug #1101686 Control: affects -1 src:armci-mpi src:bagel src:eztrace It looks like there might have been a problem with the 4.3.0-4 upload. Changelog says it disabled GPU (HIP) support, but looks like the wrong bug numbers might have been given (this

Bug#1101686: mpich: triggers test errors: MPII_init_gpu(51)....: gpu_init failed

2025-03-30 Thread Drew Parsons
Package: mpich Version: 4.3.0-3 Severity: serious Justification: FTBFS The new mpich 4.3.0 is doing something different with GPUs that is causing test failures. mpich itself was affected, Bug#1100880, but 4.3.0-3 hid the problem by disabling GPU support in autopkgtests. That doesn't help other pa