On Fri, Aug 11, 2006 at 11:09:34PM +0200, Hans de Goede wrote: > Andrew Haley wrote: > > Jesse Keating writes: > > > On Friday 11 August 2006 16:29, Andrew Haley wrote: > > > > No. It requires execmem because it really needs it. > > > > > > Then it really needs to be fixed. > > > > It doesn't need to be fixed becasue it's not a bug. We generate code > > on the fly because it's the right thing to do. There's nothing > > special about gcj in this regard: generating code on the fly is not > > unusual, and it's a valid techinique. > > > > Yes and there are ways to make it work without tripping over selinux's > execmem check, see: > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=185214#c4 Well, in libgcj case it is slightly complicated by the fact that the executable memory needs to be garbage collected and I'm not sure how complicated would it be to 1) teach boehm-gc about 2 pools of memory, executable and non-executable and especially 2) that the executable memory has the executable and writable window into the same file FYI, ATM libgcj uses mmap with PROT_READ|PROT_WRITE|PROT_EXEC, so it at least doesn't expect heap to be executable. Jakub -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
