From 354c91dcfb44110652dc2122ddb0e0a4e211eade Mon Sep 17 00:00:00 2001
From: ROCm CI Service Account <66695075+rocm-ci@users.noreply.github.com>
Date: Wed, 19 Oct 2022 16:50:51 +0530
Subject: [PATCH] SWDEV-356559 - update the note on hipIpcOpenMemHandle showing
 the different from CUDA (#2988)

Change-Id: I4116eaa3a8f31283dda7d7c0c680acdb954036b9

[ROCm/hip commit: acf96881652ae83dae6d436f7c53c30dd046715b]
---
 projects/hip/include/hip/hip_runtime_api.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/projects/hip/include/hip/hip_runtime_api.h b/projects/hip/include/hip/hip_runtime_api.h
index aa36787636..0685876aef 100644
--- a/projects/hip/include/hip/hip_runtime_api.h
+++ b/projects/hip/include/hip/hip_runtime_api.h
@@ -1707,8 +1707,9 @@ hipError_t hipIpcGetMemHandle(hipIpcMemHandle_t* handle, void* devPtr);
  * hipErrorInvalidHandle,
  * hipErrorTooManyPeers
  *
- * @note No guarantees are made about the address returned in @p *devPtr.
- * In particular, multiple processes may not receive the same address for the same @p handle.
+ * @note During multiple processes, using the same memory handle opened by the current context,
+ * there is no guarantee that the same device poiter will be returned in @p *devPtr.
+ * This is diffrent from CUDA.
  *
  */
 hipError_t hipIpcOpenMemHandle(void** devPtr, hipIpcMemHandle_t handle, unsigned int flags);