From ae9a7f34c05e8b1548501771fc054428bc0146b7 Mon Sep 17 00:00:00 2001
From: Julia Jiang <56359287+jujiang-del@users.noreply.github.com>
Date: Wed, 10 Sep 2025 12:22:24 -0400
Subject: [PATCH] SWDEV-504376 - Update definition for
 hipHostAllocWriteCombined (#887)

---
 projects/hip/include/hip/hip_runtime_api.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/projects/hip/include/hip/hip_runtime_api.h b/projects/hip/include/hip/hip_runtime_api.h
index d0f2b26092..8001bc341d 100644
--- a/projects/hip/include/hip/hip_runtime_api.h
+++ b/projects/hip/include/hip/hip_runtime_api.h
@@ -830,14 +830,17 @@ enum hipLimit_t {
 
 /** Allocates the memory as write-combined. On some system configurations, write-combined allocation
  * may be transferred faster across the PCI Express bus, however, could have low read efficiency by
- * most CPUs. It's a good option for data transfer from host to device via mapped pinned memory.*/
+ * most CPUs. It's a good option for data transfer from host to device via mapped pinned memory.
+ * @note  This flag is only for CUDA source compatibility but not functional within HIP runtime,
+ * because the allocation path is currently not supported on the AMD platform.*/
 #define hipHostAllocWriteCombined 0x4
 
 /** Allocates the memory as write-combined. On some system configurations, write-combined allocation
  * may be transferred faster across the PCI Express bus, however, could have low read efficiency by
  * most CPUs. It's a good option for data transfer from host to device via mapped pinned memory.
  * @note  This flag is the same definition as #hipHostAllocWriteCombined which is equivalent to
- * cudaHostAllocWriteCombined.*/
+ * cudaHostAllocWriteCombined. It is only for CUDA source compatibility but not functional within
+ * HIP runtime, because the allocation path is currently not supported on the AMD platform.*/
 #define hipHostMallocWriteCombined 0x4
 
 /**