5139d052d8
The evolution of the original WAIT_FOR trick in this test, first to move it to a generic spot and then to make it take a delay paramatrization, have had the unfortunate side effect of breaking the original usage. And... it's really not following the design intent anyway. The idea here was to spin hard waiting on external hardware (a python script running on the host) to deliver an interrupt. Both of the new tricks (suspending the thread with a sleep and hammering the HDA wall clock for time) can affect interrupt delivery. Just restore the original implementation (albeit with a new name, "AWAIT", since the original was stolen) until someone has a chance to make this work properly with the new API. There's no reason it can't, it just doesn't yet. Fixes #43828 Signed-off-by: Andy Ross <andrew.j.ross@intel.com> |
||
|---|---|---|
| .. | ||
| altera_max10 | ||
| espressif_esp32/cache_coex | ||
| intel_adsp | ||
| intel_s1000_crb | ||
| mec15xxevb_assy6853 | ||
| native_posix | ||