Files
tensorflow/third_party
Sohaib Iftikhar a1ae4db3a8 [XLA:GPU]: Fix caching issue with kernels when unrolling while loops
When while loops are unrolled the same thunk is called with `Record` multiple times for different loop iterations. Currently `CustomCallRecordState` stored a single sequence of `XLA_FFI_Command*`, so during update only the last recorded sequence of commands was passed to the FFI record API, overwriting the last iteration and leaving earlier iterations of the while loop with stale buffer pointers.

Keying the state by the sink command pointer returned on `RecordCreate` (and passed back via `RecordUpdate::command`) so each unrolled iteration updates its own recorded command sequence.

PiperOrigin-RevId: 987596254
2026-09-25 12:09:05 -07:00
..
…
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
…
2026-08-07 10:09:35 -07:00
2026-08-29 14:07:56 +02:00
2026-08-29 14:07:56 +02:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
…
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
…
2026-08-07 10:09:35 -07:00
2026-08-07 10:09:35 -07:00
…