From 562f3ef09834ea103efc41eda0d40e5890b09dac Mon Sep 17 00:00:00 2001
From: German Andryeyev <German.Andryeyev@amd.com>
Date: Tue, 16 Apr 2024 17:41:21 -0400
Subject: [PATCH] SWDEV-440746 - Fix the hostcall buffer creation

Avoid a deadlock on the host call buffer creation. Since the buffer will be
allocated in the queue thread, then use direct device memory allocation
skipping the global context lock.

Change-Id: I09b55ee03bb42ab5d320c152b52a8c842c5fdcc1


[ROCm/clr commit: 62559a6e5ab26880576cf1a80863ad8628a0deed]
---
 projects/clr/rocclr/device/devhostcall.cpp    | 5 ++++-
 projects/clr/rocclr/device/pal/palvirtual.cpp | 5 +++--
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/projects/clr/rocclr/device/devhostcall.cpp b/projects/clr/rocclr/device/devhostcall.cpp
index ca68f3ede5..9a0fb9629b 100644
--- a/projects/clr/rocclr/device/devhostcall.cpp
+++ b/projects/clr/rocclr/device/devhostcall.cpp
@@ -284,7 +284,10 @@ static struct Init {
   ~Init() {
     if (state == State::kInit) {
       state = State::kDestroy;
-      while (state == State::kDestroy) {}
+      // @note: Under Linux thread destruction can be delayed and
+      // ROCR may crash in a wait for event occasionally. Hence, runtime needs
+      // an early exit. The logic isn't required for Windows.
+      while (IS_LINUX && (state == State::kDestroy)) {}
     }
   }
 } kHostThreadActive;
diff --git a/projects/clr/rocclr/device/pal/palvirtual.cpp b/projects/clr/rocclr/device/pal/palvirtual.cpp
index 388b68a028..a20577b24c 100644
--- a/projects/clr/rocclr/device/pal/palvirtual.cpp
+++ b/projects/clr/rocclr/device/pal/palvirtual.cpp
@@ -1154,7 +1154,7 @@ VirtualGPU::~VirtualGPU() {
             "deleting hostcall buffer %p for virtual queue %p",
             hostcallBuffer_, this);
     disableHostcalls(hostcallBuffer_);
-    dev().context().svmFree(hostcallBuffer_);
+    dev().svmFree(hostcallBuffer_);
   }
 }
 
@@ -3766,7 +3766,8 @@ void* VirtualGPU::getOrCreateHostcallBuffer() {
   auto size = getHostcallBufferSize(numPackets);
   auto align = getHostcallBufferAlignment();
 
-  hostcallBuffer_ = dev().context().svmAlloc(size, align, CL_MEM_SVM_FINE_GRAIN_BUFFER | CL_MEM_SVM_ATOMICS);
+  hostcallBuffer_ = dev().svmAlloc(dev().context(), size, align,
+                                   CL_MEM_SVM_FINE_GRAIN_BUFFER | CL_MEM_SVM_ATOMICS, nullptr);
   if (!hostcallBuffer_) {
     ClPrint(amd::LOG_ERROR, amd::LOG_QUEUE,
             "Failed to create hostcall buffer");