remove comment about probable leak. I think its fixed

since we removed the odd commSetSelect() call for new
store interface.

diff --git a/src/store_client.cc b/src/store_client.cc
index 5f2a1a40ee73b4b31e9ed2f1fd4dc5d32f4e877e..e7568ecfa336782e09fb30a10b97711b6546d81b 100644 (file)
--- a/src/store_client.cc
+++ b/src/store_client.cc
@@ -1,6 +1,6 @@
 
 /*
- * $Id: store_client.cc,v 1.71 1999/05/22 07:42:10 wessels Exp $
+ * $Id: store_client.cc,v 1.72 1999/06/19 16:36:05 wessels Exp $
  *
  * DEBUG: section 20    Storage Manager Client-Side Interface
  * AUTHOR: Duane Wessels
@@ -419,10 +419,8 @@ storeUnregister(StoreEntry * e, void *data)
     sc->flags.disk_io_pending = 0;
     if (e->store_status == STORE_OK && e->swap_status != SWAPOUT_DONE)
        storeSwapOut(e);
-    if (sc->swapin_sio) {
+    if (sc->swapin_sio)
        storeClose(sc->swapin_sio);
-       /* XXX this probably leaks file_read handler structures */
-    }
     if ((callback = sc->callback) != NULL) {
        /* callback with ssize = -1 to indicate unexpected termination */
        debug(20, 3) ("storeUnregister: store_client for %s has a callback\n",