[XPU] Add forward_xpu to Mixer2RMSNormGated and FusedRMSNormGated (#54968)

Signed-off-by: Marceli Fylcek <marceli.fylcek@intel.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Marceli Fylcek
2026-09-11 07:58:28 +08:00
committed by GitHub
co-authored by Claude Opus 5
parent a89de950a4
commit 5b6cf93e8e
2 changed files with 23 additions and 0 deletions
@@ -180,6 +180,13 @@ class Mixer2RMSNormGated(CustomOp):
norm_before_gate=False,
)
def forward_xpu(
self,
x: torch.Tensor,
gate: torch.Tensor,
) -> torch.Tensor | tuple[torch.Tensor, torch.Tensor]:
return self.forward_cuda(x, gate)
def mamba_v2_sharded_weight_loader(
shard_spec: list[tuple[int, int, float]],
+16
View File
@@ -529,6 +529,22 @@ class FusedRMSNormGated(CustomOp):
residual_in_fp32=residual_in_fp32,
)
def forward_xpu(
self,
x: torch.Tensor,
g: torch.Tensor,
residual: torch.Tensor | None = None,
prenorm: bool = False,
residual_in_fp32: bool = False,
) -> torch.Tensor:
return self.forward_cuda(
x,
g,
residual=residual,
prenorm=prenorm,
residual_in_fp32=residual_in_fp32,
)
@triton.heuristics({"IS_VARLEN": lambda args: args["cu_seqlens"] is not None})
@triton.autotune(