msvcp90: Use atomic write for releasing threadsafe_queue spin lock.

This fixes data race in ARM/ARM64 platforms, and prevents potential memory access reordering by the compiler.

msvcp90: Use atomic write for releasing threadsafe_queue spin lock.
dce7a61f · Jinoh Kang · Alexandre Julliard · 94512ecc · dce7a61f
Commit dce7a61f authored Sep 19, 2022 by Jinoh Kang Committed by Alexandre Julliard Sep 20, 2022
Show whitespace changes
Inline Side-by-side

Showing with 2 additions and 2 deletions

details.c dlls/msvcp90/details.c +2 -2

No files found.
--- a/dlls/msvcp90/details.c
+++ b/dlls/msvcp90/details.c
@@ -239,7 +239,7 @@ static void threadsafe_queue_push(threadsafe_queue *queue, size_t id,
        queue->tail = p;
        if(!queue->head)
            queue->head = p;
-        queue->lock = 0;
+        WriteRelease(&queue->lock, 0);
    }
    else
    {
@@ -289,7 +289,7 @@ static BOOL threadsafe_queue_pop(threadsafe_queue *queue, size_t id,
        queue->head = p->_Next;
        if(!queue->head)
            queue->tail = NULL;
-        queue->lock = 0;
+        WriteRelease(&queue->lock, 0);

        /* TODO: Add exception handling */
        call__Concurrent_queue_base_v4__Deallocate_page(parent, p);