mem: Clarify cache behaviour for pending dirty responses

This patch adds a bit of clarification around the assumptions made in
the cache when packets are sent out, and dirty responses are
pending. As part of the change, the marking of an MSHR as in service
is simplified slightly, and comments are added to explain what
assumptions are made.

1 files changed, 6 insertions, 5 deletions

diff --git a/src/mem/cache/cache.hh b/src/mem/cache/cache.hh
index e0bd29752..21a00dbbd 100644
--- a/src/mem/cache/cache.hh
+++ b/src/mem/cache/cache.hh
@@ -375,12 +375,13 @@ class Cache : public BaseCache
     PacketPtr getTimingPacket();
 
     /**
-     * Marks a request as in service (sent on the bus). This can have side
-     * effect since storage for no response commands is deallocated once they
-     * are successfully sent.
-     * @param pkt The request that was sent on the bus.
+     * Marks a request as in service (sent on the bus). This can have
+     * side effect since storage for no response commands is
+     * deallocated once they are successfully sent. Also remember if
+     * we are expecting a dirty response from another cache,
+     * effectively making this MSHR the ordering point.
      */
-    void markInService(MSHR *mshr, PacketPtr pkt = NULL);
+    void markInService(MSHR *mshr, bool pending_dirty_resp);
 
     /**
      * Return whether there are any outstanding misses.