room: KLUDGE-reduction.

  * Now that the thread structure doesn't have a lowtag, we can
remove the KLUDGE that prevented even FIXNUMs from being treated
as though they were thread-object-sized instead of a single word.

diff --git a/src/code/room.lisp b/src/code/room.lisp
index 3ddbfc6..4f81df1 100644 (file)
--- a/src/code/room.lisp
+++ b/src/code/room.lisp
@@ -39,28 +39,6 @@
         (size (primitive-object-size obj)))
     (cond
      ((not lowtag))
-     (;; KLUDGE described in dan_b message "Another one for the
-      ;; collection [bug 108]" (sbcl-devel 2004-01-22)
-      ;;
-      ;; In a freshly started SBCL 0.8.7.20ish, (TIME (ROOM T))  causes
-      ;;   debugger invoked on a SB-INT:BUG in thread 5911:
-      ;;     failed AVER: "(SAP= CURRENT END)"
-      ;; [WHN: Similar things happened on one but not the other of my
-      ;; machines when I just run ROOM a lot in a loop.]
-      ;;
-      ;; This appears to be due to my [DB] abuse of the primitive
-      ;; object macros to define a thread object that shares a lowtag
-      ;; with fixnums and has no widetag: it looks like the code that
-      ;; generates *META-ROOM-INFO* infers from this that even fixnums
-      ;; are thread-sized - probably undesirable.
-      ;;
-      ;; This [the fix; the EQL NAME 'THREAD clause here] is more in the
-      ;; nature of a workaround than a really good fix. I'm not sure
-      ;; what a really good fix is: I /think/ it's probably to remove
-      ;; the :LOWTAG option in DEFINE-PRIMITIVE-OBJECT THREAD, then teach
-      ;; genesis to generate the necessary OBJECT_SLOT_OFFSET macros
-      ;; for assembly source in the runtime/genesis/*.h files.
-      (eql name 'thread))
      ((not widetag)
       (let ((info (make-room-info :name name
                                   :kind :lowtag))