Messages in this thread | | | Date | Sat, 18 Dec 1999 00:55:20 -0500 (EST) | From | Stephen Frost <> | Subject | re: RasterMan on linux and threads |
| |
On Sat, 18 Dec 1999, Ingo Molnar wrote:
> On Fri, 17 Dec 1999, Stephen Frost wrote: > > > No, pthreads changed that, from my understanding. linuxthreads did > > it all in one thing w/o ever actually calling the kernel 'clone'. pthreads > > properly calles 'clone' and therefore each thread gets it's own PID and as > > such can be scheduled on any CPU. (Well, that's not the direct reason, but > > you know what I mean). > > LinuxThreads was using clone() ever since it became production quality. I > do not know what you mean by 'pthreads', if you mean glibc's pthread.h > API, that is the merged-in LinuxThreads code. (which merge was done some 2 > years ago)
I was thinking older than 2-3 years. By 'pthreads' I do mean the glibc2 pthread.h API.
> > [...] The first > > threading in Linux was done all in user-space, IIRC. Which meant that it > > couldn't because a single process (to the kernel) won't get multiple CPUs. > > a clone() capable LinuxThreads has been around for at least 3 years.
Hrm. For some reason I had thought that there was, for a time, a version of linuxthreads that did it ALL in user-space, and didn't use clone(). This would have been 3+ (or so) years ago, IIRC. I also seem to recall Linus talking about hoping someone would use clone (Or something similar if I'm not remembering exactly) so that he could see if it would work well for threading or not...
This is digging pretty far back for me, so it's not unlikely I'm wrong in this. :)
Stephen
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |