st-api: Rework how drawables are invalidated v3.

The api and the state tracker manager code as well as the state tracker code
assumed that only a single context could be bound to a drawable. That is not
a valid assumption, since multiple contexts can bind to the same drawable.

Fix this by making it the state tracker's responsibility to update all
contexts binding to a drawable

Note that the state trackers themselves don't use atomic stamps on
frame-buffers. Multiple context rendering to the same drawable should
be protected by the application.

Signed-off-by: Thomas Hellstrom <[email protected]>

1 files changed, 11 insertions, 4 deletions

diff --git a/src/mesa/state_tracker/st_cb_viewport.c b/src/mesa/state_tracker/st_cb_viewport.c
index 049755e45c0..d4742eb897d 100644
--- a/src/mesa/state_tracker/st_cb_viewport.c
+++ b/src/mesa/state_tracker/st_cb_viewport.c
@@ -56,13 +56,20 @@ static void st_viewport(struct gl_context * ctx, GLint x, GLint y,
    if (!st->invalidate_on_gl_viewport)
       return;
 
+   /*
+    * Normally we'd want the state tracker manager to mark the drawables
+    * invalid only when needed. This will force the state tracker manager
+    * to revalidate the drawable, rather than just update the context with
+    * the latest cached drawable info.
+    */
+
    stdraw = st_ws_framebuffer(st->ctx->DrawBuffer);
    stread = st_ws_framebuffer(st->ctx->ReadBuffer);
 
-   if (stdraw)
-      p_atomic_set(&stdraw->revalidate, TRUE);
-   if (stread && stread != stdraw)
-      p_atomic_set(&stread->revalidate, TRUE);
+   if (stdraw && stdraw->iface)
+      stdraw->iface_stamp = p_atomic_read(&stdraw->iface->stamp) - 1;
+   if (stread && stread != stdraw && stread->iface)
+      stread->iface_stamp = p_atomic_read(&stread->iface->stamp) - 1;
 }
 
 void st_init_viewport_functions(struct dd_function_table *functions)