Daniel Hiltgen 4fe5609563 metal: harden for ggml initialization failures (#15755) ...

* metal: harden for ggml initialization failures

ggml_metal_device_init performs a probe to verify the tensor API compiles.  On
some systems this passes, even though kernel coverage isn't complete, which
results in a later crash when compiling the real kernels.  This change adds a
single retry if any of the error strings match this failure mode to disable the
tensor API.  It also hardens an error case in the Go initDevices to detect
device initialization failures and panic instead of crashing later on a nil
array entry.

Fixes #15734

* review comments

* review comments

2026-04-30 16:28:03 -07:00

..

cpu_linux_test.go

DRY out the runner lifecycle code (#12540)

2025-10-23 11:20:02 -07:00

cpu_linux.go

discover: Support cgroups cores and memory limitations (#10292)

2025-11-17 16:13:03 -08:00

cpu_windows_test.go

…

cpu_windows.go

win: revert CPU discovery logic to 0.12.3 (#12969)

2025-11-05 10:32:38 -08:00

gpu_darwin.go

…

gpu_info_darwin.h

…

gpu_info_darwin.m

…

gpu.go

log: instrument CPU discovery timing (#12960)

2025-11-04 16:23:37 -08:00

runner_test.go

metal: harden for ggml initialization failures (#15755)

2026-04-30 16:28:03 -07:00

runner.go

metal: harden for ggml initialization failures (#15755)

2026-04-30 16:28:03 -07:00

types.go

discovery: only retry AMD GPUs (#12894)

2025-11-04 15:33:46 -08:00