A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 13:30 UTC
[MISC][AMD] Add unused annotation to rocm kernel file (#17097)
Signed-off-by: Lu Fang <lufang@fb.com>
Lu Fang committed
Apr 25, 2025 at 20:33 UTC
c8e5be35f74b2d52fb9f9546d0e4151879164034
1 file changed
+2
-2
csrc/rocm/skinny_gemms.cu
+2
-2
@@ -150,7 +150,7 @@ __global__ void LLGemm1_kernel(const scalar_t* in_a, const scalar_t* in_b,
150
colB_elem4w = bf4[threadid * 4 + 3];
151
152
scalar2_t Af2;
153
- scalar2_t Bf2;
153
+ [[maybe_unused]] scalar2_t Bf2;
154
float2 S;
155
156
auto Ah2ptr = reinterpret_cast<scalar2_t*>(&rowA_elem4);
@@ -1597,4 +1597,4 @@ void wvSplitKQ(at::Tensor& in_a, at::Tensor& in_b, at::Tensor& out_c,
1597
}
1598
});
1599
});
1600
-}
\ No newline at end of file
1600
+}