This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Thread-, Signal- and Cancellation-safety documentation
- From: Torvald Riegel <triegel at redhat dot com>
- To: Alexandre Oliva <aoliva at redhat dot com>
- Cc: Rich Felker <dalias at aerifal dot cx>, libc-alpha at sourceware dot org
- Date: Wed, 05 Jun 2013 15:15:51 +0200
- Subject: Re: Thread-, Signal- and Cancellation-safety documentation
- References: <1368788184 dot 3054 dot 3161 dot camel at triegel dot csb> <orehd3h2j4 dot fsf at livre dot home> <1369588322 dot 16968 dot 2933 dot camel at triegel dot csb> <orip20qzua dot fsf at livre dot home> <1369936586 dot 16968 dot 9405 dot camel at triegel dot csb> <orehcmq122 dot fsf at livre dot home> <orehclih8s dot fsf at livre dot home> <ory5atgkyr dot fsf at livre dot home> <20130602143136 dot GM20323 at brightrain dot aerifal dot cx> <orppw4h1v1 dot fsf at livre dot home> <20130602215932 dot GC29800 at brightrain dot aerifal dot cx> <or7gicdtac dot fsf at livre dot home>
On Sun, 2013-06-02 at 19:47 -0300, Alexandre Oliva wrote:
> On Jun 2, 2013, Rich Felker <dalias@aerifal.cx> wrote:
> > POSIX requires actual shared memory. An implementation that fakes it
> > is not conforming.
>
> We're all screwed, then, because NUMA is hardly actual shared memory :-)
If you can implement the language's/POSIX' memory model on top of it,
and this implementation provides the same guarantees as process-local
memory, it is shared memory. For example, cache-coherent NUMA isn't an
issue at all.