Commit 8cb0d98cc for llama.cpp

commit 8cb0d98ccd0a310879ae8eaa1ce2c41e4b40a052
Author: Eddie Zhou <eddiezhou1996@gmail.com>
Date:   Sun Nov 22 04:06:12 2026 -0700

    HIP: fix double halving of NVFP4 scales on CDNA3 (#32790)

    Fixes #32789

diff --git a/ggml/src/ggml-cuda/common.cuh b/ggml/src/ggml-cuda/common.cuh
index 8cd2366ed..5b3c32858 100644
--- a/ggml/src/ggml-cuda/common.cuh
+++ b/ggml/src/ggml-cuda/common.cuh
@@ -911,7 +911,7 @@ static __device__ __forceinline__ float ggml_cuda_ue4m3_to_fp32(uint8_t x) {
     // but CDNA3 supports only e4m3_fnuz (no inf).
     const uint32_t bits = x * (x != 0x7F && x != 0xFF); // Convert NaN to 0.0f to match CPU implementation.
     const __hip_fp8_e4m3_fnuz xf = *reinterpret_cast<const __hip_fp8_e4m3_fnuz *>(&bits);
-    return static_cast<float>(xf) / 2;
+    return static_cast<float>(xf);
 #else
 #if defined(FP8_AVAILABLE) && !defined(GGML_USE_HIP)
     const uint32_t bits = x * (x != 0x7F && x != 0xFF); // Convert NaN to 0.0f to match CPU implementation.