* bug fix: INTERRUPT-THREAD on a dying thread could produce memory
fault.
* bug fix: finalizers, gc hooks never run in a WITHOUT-INTERRUPTS
+ * bug fix: fix random memory faults related to interrupts on alpha
* bug fix: fix deadlocks related to starting threads
changes in sbcl-1.0.25 relative to 1.0.24:
(declare (type (unsigned-byte 32) n)
(optimize (speed 3) (safety 0)))
(sb!alien:sap-alien (sb!vm::current-thread-offset-sap
- (+ sb!vm::thread-interrupt-contexts-offset n))
+ (+ sb!vm::thread-interrupt-contexts-offset
+ #!-alpha n
+ #!+alpha (* 2 n)))
(* os-context-t)))
#!+(or x86 x86-64)
(interrupt-data :c-type "struct interrupt_data *"
:length #!+alpha 2 #!-alpha 1)
(stepping)
+ ;; KLUDGE: On alpha, until STEPPING we have been lucky and the 32
+ ;; bit slots came in pairs. However the C compiler will align
+ ;; interrupt_contexts on a double word boundary. This logic should
+ ;; be handled by DEFINE-PRIMITIVE-OBJECT.
+ #!+alpha
+ (padding)
(interrupt-contexts :c-type "os_context_t *" :rest-p t))
;;; checkins which aren't released. (And occasionally for internal
;;; versions, especially for internal versions off the main CVS
;;; branch, it gets hairier, e.g. "0.pre7.14.flaky4.13".)
-"1.0.25.42"
+"1.0.25.43"