Date: Fri, 2 Oct 1998 11:17:16 +0100 (BST) From: Doug Rabson <dfr@nlsystems.com> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Cc: Alex Nash <nash@mcs.net>, Poul-Henning Kamp <phk@critter.freebsd.dk>, John Birrell <jb@freebsd.org>, cvs-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/lib/libc_r/uthread uthread_gc.c Makefile.inc uthread_init.c uthread_find_thread.c uthread_kern.c uthread_create.c uthread_exit.c Message-ID: <Pine.BSF.4.01.9810021116300.414-100000@herring.nlsystems.com> In-Reply-To: <199810011440.KAA20723@khavrinen.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 1 Oct 1998, Garrett Wollman wrote: > <<On Thu, 1 Oct 1998 09:28:57 -0500, Alex Nash <nash@mcs.net> said: > > > On Thu, Oct 01, 1998 at 10:02:06AM +0200, Poul-Henning Kamp wrote: > >> Alex, can we constrain ourselves to >actual< problems rather than > >> theoretically-possible-on-really-stupid-machines-problems ? > > > The atomicity of malloc_active++ is compiler dependent, not > > really-stupid-machine dependent. > > And, in particular, it may not be atomic on a three-operand RISC > machine. Imagine the following code sequence on an imaginary > processor: > > addil $malloc_active, %r0, %r1 > addl (%r1), %r0, %r2 > addil $1, %r2, %r3 > addl %r3, %r0, (%r1) I had to fix a whole stack of this kind of thing in the kernel before it would work reliably on the alpha. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 Fax: +44 181 381 1039
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.01.9810021116300.414-100000>