[pypy-commit] stmgc use-gcc: add another question about light finalizers

[Raemi]

Author: Remi Meier <remi.meier at inf.ethz.ch>
Branch: use-gcc
Changeset: r1946:9076b244ab08
Date: 2015-08-26 17:22 +0200
http://bitbucket.org/pypy/stmgc/changeset/9076b244ab08/

Log:	add another question about light finalizers

diff --git a/c8/stm/finalizer.c b/c8/stm/finalizer.c
--- a/c8/stm/finalizer.c
+++ b/c8/stm/finalizer.c
@@ -486,6 +486,10 @@
    XXX: what happens in _execute_finalizer if the transaction
         conflicts (or fails to become inevitable) in a finalizer?
         (the run_finalizers list is half-way cleared?)
+   XXX: according to translator.backendopt.finalizer, getfield_gc
+        for primitive types is a safe op in light finalizers.
+        I don't think that's correct in general (maybe if
+        getfield on *dying obj*).
 */
 
 static void _invoke_general_finalizers(stm_thread_local_t *tl)