Commit 231a3e8
authored
Remove unneeded
```
- max(
- next_power_of_2(
- V.graph.sizevars.size_hint(
- seq_len_q,
- fallback=torch._inductor.config.unbacked_symint_fallback, # type: ignore[arg-type]
- )
- * gqa_shared_heads
- ),
- 1 if torch.xpu.is_available() else 16,
+ next_power_of_2(
+ V.graph.sizevars.size_hint(
+ seq_len_q,
+ fallback=torch._inductor.config.unbacked_symint_fallback, # type: ignore[arg-type]
+ )
+ * gqa_shared_heads
```
On xpu, the results are equivalent, `max(next_power_of_2, 1) ==
next_power_of_2`.
CI:
https://github.com/intel/intel-xpu-backend-for-triton/actions/runs/17266774537/job/49000662056
(no regression)
Signed-off-by: Whitney Tsang <[email protected]>flex_decoding.patch (#4962)1 parent bfbdc55 commit 231a3e8
2 files changed
+0
-27
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
36 | 36 | | |
37 | 37 | | |
38 | 38 | | |
39 | | - | |
This file was deleted.
0 commit comments