This is the mail archive of the
guile@sourceware.cygnus.com
mailing list for the Guile project.
more gc stuff
- To: Michael Livshin <mlivshin at bigfoot dot com>
- Subject: more gc stuff
- From: Dirk Herrmann <dirk at ida dot ing dot tu-bs dot de>
- Date: Wed, 28 Jun 2000 14:56:20 +0200 (MEST)
- cc: Guile Mailing List <guile at sourceware dot cygnus dot com>
Hello,
it looks as scm_gc_for_newcell is broken: When gc is disabled via
scm_block_gc, scm_igc won't bring any new cells. However, in this case
master->grow_heap_p won't be set also, because this flag is only set
within gc_sweep_freelist_finish, which is only called if gc is not
disabled.
The error does not become obvious as long as no gc is required while
scm_block_gc is 1. Since gs is only blocked during initialization,
everything appears to be fine.
Am I missing something here, or should scm_gc_for_newcell be fixed to cope
with the fact that the call to scm_igc does not make any new cells
available?
Best regards
Dirk