re: RasterMan on linux and threads
raster@rasterman.com
Wed, 15 Dec 1999 19:57:11 -0500 (EST)
On 17 Dec, Richard Gooch scribbled:
-> raster@rasterman.com writes:
-> > On 17 Dec, Dan Kegel scribbled:
-> > -> re http://kernelnotes.org/lnxlists/linux-kernel/lk_9912_03/msg00480.html
-> > ->
-> > -> Rasterman is wrong in saying that all threads run on the same
-> >
-> > hmm - when did that change ? i thought that was the case and was
-> > baked up on hat asumtion by someone else a few weeks ago (primarily
-> > the reason being to make sure the threads share caches for speed
-> > reasons and to make sure cache concurrency issues are moe easiyl
-> > dealt with... well thats what i unerstood... i may be wrong (2.2 or
-> > 2.3 may have changed that)
->
-> It hasn't changed. Linux always scheduled tasks on available CPUs
-> (sans cache affinity heuristics).
->
-> IIRC, 1.3.38 was the first kernel with stable SMP support that I
-> played with, and it definately scheduled threads on separate CPUs
-> (otherwise my threaded compute applications wouldn't have sped up by a
-> factor of 2).
hmm - so threads form one process do get scheduled across multiple
cpus? in that case i migt yet get a decent speedup (altho i'm giving
the memory bus a good kicking whilst i'm at it so i probably wont get a
2x speedup since i'll get memopry contention - or likely will - ut
thats to be seen form tests if i can get any decent speedup via threads)
--
--------------- Codito, ergo sum - "I code, therefore I am" --------------------
The Rasterman (Carsten Haitzler) raster@rasterman.com raster@valinux.com
raster@enlightenment.org raster@linux.com
raster@zip.com.au
-
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/