From c7d8722922a2599dc4d77f8808d8e6c2fde5e7a2 Mon Sep 17 00:00:00 2001
From: liminfei-amd <91481003+liminfei-amd@users.noreply.github.com>
Date: Thu, 16 Jul 2026 17:10:08 +0800
Subject: [PATCH] ggml-cuda : restore prop.integrated on HIP builds (#24233)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PR #16308 set info.devices[id].integrated = false unconditionally for all
CUDA/HIP devices as a workaround for corrupted output on Jetson Orin
(#15034). On HIP/ROCm the device's real hipDeviceProp_t.integrated flag is
needed: with the cached field forced to false, supports_buft() refuses
CUDA host buffers on AMD APU/UMA parts, while get_type() already reads
prop.integrated (#23007) — an inconsistency that breaks integrated-GPU
host-buffer use on ROCm.

Guard the workaround so it only applies to non-HIP (CUDA) builds and
restore prop.integrated for HIP, keeping the Jetson workaround intact for
CUDA.

Fixes #23977

Signed-off-by: liminfei-amd <91481003+liminfei-amd@users.noreply.github.com>
---
 ggml/src/ggml-cuda/ggml-cuda.cu | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index e27ab1526..d644436e5 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -248,7 +248,11 @@ static ggml_cuda_device_info ggml_cuda_init() {
 
         info.default_tensor_split[id] = total_vram;
         total_vram += prop.totalGlobalMem;
+#if defined(GGML_USE_HIP)
+        info.devices[id].integrated = prop.integrated;
+#else
         info.devices[id].integrated = false; // Temporarily disabled due to issues with corrupted output (e.g. #15034)
+#endif
         info.devices[id].nsm        = prop.multiProcessorCount;
         info.devices[id].smpb       = prop.sharedMemPerBlock;
         info.devices[id].warp_size  = prop.warpSize;