Explicit vs implicit barriers

Article: Fall Processor Forum 2006: IBM's POWER6
By: Rob Thorpe (rthorpe.delete@this.realworldtech.com), October 31, 2006 12:15 pm
Room: Moderated Discussions
Linus Torvalds (torvalds@osdl.org) on 10/27/06 wrote:
---------------------------
>Tzvetan Mikov (tzvetanmi@yahoo.com) on 10/27/06 wrote:
>>
>>I am blatantly attempting to hijack the subject:
>>
>>Are there any technical reasons, other than ifetch
>>bandwidth, which make explicit barriers more expensive than
>>implicit ones, if the goal is to make them fast ?
>
>Well, you could always make all barriers implicit, and then
>the explicit barriers are no-ops, so in that sense, they'd
>be free (apart from I$). But they'd be free only if they are
>pointless, so that's not a very interesting case, and not
>one that argues for explicit barriers..
>
>I'd argue that there are two costs to real barriers:
>
>- the software cost. The reason x86 doesn't have them
>is that a lot of software doesn't have them, and a
>lot of programmers don't understand them. I think the
>example of the alpha memory ordering issue should have
>shown that even clueful programmers simply don't
>know or even realize that they don't know the
>subtle pitfalls in this area.
>
>So not having to have barriers is actually a real
>advantage. Like it or not, it's simply easier to
>generate code that actually works on x86 than it is on
>alpha, and it has nothing to do with performance. Trust
>me, even among kernel programmers, most don't really
>realize how subtle these things can be. And we're the
>best of the best (*), and I'd claim that we work more
>with true SMP threads than anybody else does.
>
>(*) And so very modest, too.

I agree about that, SMP and threading are hard enough as it is. But this is only an argument for the availability of strong ordering, not for it's use everywhere.

>So don't ignore this cost. It's probably the biggest one,
>and it makes any performance issues secondary.

Not really. Threading itself is evil in my book, I avoid it like the plague.

But other folks than us have figured these things out and written software working with weak memory orderings, eg those writing NUMA code.

>- I don't think how many people realize this, but a
>"memory barrier" is not actually all that well-defined.
>You actually have a lot of cases, and it's not
>like there is only one type. A true memory barrier is
>actually more like a half-permeating barrier that can
>recognize the type of the memory operation: you can have
>barriers that do not allow later reads to pass
>the barrier one way, but allows earlier reads to
>pass it, or any writes to be buffered past it.

Yes. This is true in terms of GC memory barriers too AFAIK.

>In other words, pretty much every memory barrier
>model out there actually simplifies the barrier model
>to a few common cases, but in doing so, it may force the
>programmer to use a sledgehammer where a more subtle
>ordering would make more sense. And yes, that can mean
>that implicit barriers can actually do better.
>
>More importantly, a lot of barriers are totally and
>utterly unnecessary 99.999% of the time.

All of the above applies to the hardware barriers needed for strong ordering too. They also are sledgehammers that don't matter 99.999% of the time.

> A really
>highperformance implementation may re-order things
>very aggressively indeed, and then just check the end
>result - for example, if all the memory locations
>involved are in the L1 cache, and none of them take a
>cache miss, then by definition the ordering didn't
>matter, because nobody else could have seen it.
>This last point is also important: you don't actually have
>to care about ordering, unless there was traffic from
>another core that causes you to have to care. That's some-
>what simplified, of course: one of the reasons that the
>memory pipeline is the most complex part of a modern CPU
>is not even SMP effects, but making sure that the memory
>accesses appear "in program order" even for just the
>local thread.
>
>So notice how a CPU actually does a lot of this careful
>work to make execution look serialized already. In
>many ways, SMP memory ordering is not really any different.
>It just means that you may have to check that the cache
>lines you accessed are still under your own ownership
>until you can really commit the operation, but the point
>here is that you could actually do an out-of-order CPU that
>treated the (very rare) ordering mistake as a micro-fault
>and just replayed it.

Yes. It could work like that, its certainly the kind of way I'd make it work :). But by your previous view of "robust" would this be robust? The problem is that if you write something that relies extensively on the strong-ordering then it may replay very often.

>And what is the result of that kind of advanced CPU core?
>It just means that all explicit ordering is pointless.
>The core already tracks the ordering requirements for
>purely local reasons (ie it may have executed a store
>earlier, but it can't really retire it and "commit" it
>until all previous work has finished).

The core tracks it's action on the parts that it is modifying, certainly. The problem is those parts that other cores are modifying. What strong ordering implies is that for some level of access, say X the load/store logic of one core must contact the load/store logic of another and say "I'm about to write into memory your messing with, so appear serial". Then the two must arrange things then so they do appear serial. This is the problem, since it's difficult for the cores involved to continue with whatever they were doing while it's going on.

This problem is inevitable anyway if ordering actually does matter. But if it doesn't then it's just a waste of time.

>In many ways, memory ordering instructions can just be
>seen as a sign that the CPU isn't tracking everything it
>could track, so the software needs to tell it to flush
>some queue explicitly. That's not really software's job,
>and especially since a lot of this is dynamic, the hw could
>do better if it just tracked itself instead of flushing
>things unnecessarily!

No. For it's own purposes a system with weak-ordering between cores and one with strong-ordering between cores must store the same info. Because each core needs this info to remain consistent itself - cores don't see their own stores out of order.

The difference is how the info is used between cores. In an implicit system if the info indicates that something could go awry then all outstanding stores must complete. In an explicit system the programmer has the choice over whether to treat the problem in this way.

>But see above on the reason why you shouldn't expect sw
>to do a great job. And besides - judging by past behaviour,
>I suspect we'll continue to see smarter and smarter CPU's,
>with less and less upside from explitic barriers. And at
>some point, the upside just doesn't really exist at all,
>and explicit memory barriers have only problems.
>
>Will we get there? Maybe not. But I seriously believe that
>the x86 is a better model for a reasonable future than the
>alpha was (or powerpc, or mips, or ..). It has some
>barriers, but they are pretty minimal (and I'm actually
>told that even the read barrier may not actually be
>needed by any ever shipping x86 core).

Yes, I think your right about that. This is not a good thing though, it just means everyone is paying the memory barrier tax whether they need it or not.

< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
POWER6 article onlineDavid Kanter2006/10/16 10:31 PM
  POWER6 article onlineNiels Jørgen Kruse2006/10/16 11:59 PM
    POWER6 article onlineDavid Kanter2006/10/17 12:03 AM
      Interesting cachesPaul Aaron Clayton2006/10/17 06:35 AM
  POWER6 article onlinerwessel2006/10/17 12:17 AM
    POWER6 article onlineGabriele Svelto2006/10/17 12:49 AM
      POWER6 article onlinerwessel2006/10/17 12:55 AM
      POWER6 article onlineDavid Kanter2006/10/17 01:01 AM
        POWER6 article onlineGabriele Svelto2006/10/17 01:06 AM
  BCD arithmetic remarkRob Thorpe2006/10/17 06:57 AM
    BCD arithmetic remarkDavid Kanter2006/10/17 08:04 AM
      BCD arithmetic remarkRob Thorpe2006/10/17 08:33 AM
        Java perfSerge Monkewitz2006/10/17 12:25 PM
          Java perfRob Thorpe2006/10/18 04:27 AM
            Java perfsp2006/10/18 04:50 AM
              Java perfRob Thorpe2006/10/18 08:36 AM
                Data parallelism in .NET using GPUsStephen Goss2006/10/18 11:48 PM
                  Data parallelism in .NET using GPUssp2006/10/19 01:08 AM
                    Data parallelism in .NET using GPUsStephen Goss2006/10/21 02:22 AM
                      Data parallelism in .NET using GPUsfastpathguru2006/11/01 02:30 PM
                        Data parallelism in .NET using GPUsIan Ameline2006/11/01 05:31 PM
                Java perfGabriele Svelto2006/10/19 12:23 AM
                  Java perfRob Thorpe2006/10/19 02:58 AM
              Numerical Java perfWes Felter2006/10/18 10:25 AM
                Numerical Java perfsp2006/10/18 10:46 AM
                  Numerical Java perfRob Thorpe2006/10/18 11:23 AM
                    Numerical Java perfTzvetan Mikov2006/10/18 11:44 AM
                      Numerical Java perfRob Thorpe2006/10/19 04:30 AM
                        Numerical Java perfsp2006/10/19 05:22 AM
                          Numerical Java perfGabriele Svelto2006/10/19 05:33 AM
                          Numerical Java perfRob Thorpe2006/10/19 07:08 AM
                            Numerical Java perfJuho Snellman2006/10/20 08:38 AM
                              Numerical Java perfRob Thorpe2006/10/23 04:54 AM
                                SBCL garbage collectorJuho Snellman2006/10/25 03:29 AM
                          Numerical Java perfTzvetan Mikov2006/10/19 09:04 AM
                        Language safety in a multithread environmentTzvetan Mikov2006/10/19 09:34 AM
                          Language safety in a multithread environmentRicardo B2006/10/19 11:43 AM
                            Language safety in a multithread environmentTzvetan Mikov2006/10/19 12:11 PM
                              Language safety in a multithread environmentRicardo B2006/10/19 02:23 PM
                                Language safety in a multithread environmentTzvetan Mikov2006/10/19 04:08 PM
                                  Language safety in a multithread environmentRicardo B2006/10/20 05:04 AM
                                    Language safety in a multithread environmentTzvetan Mikov2006/10/20 01:53 PM
                                      Language safety in a multithread environmentRicardo B2006/10/20 02:06 PM
                                      PSRicardo B2006/10/20 02:12 PM
                                        PSTzvetan Mikov2006/10/20 03:30 PM
                          Language safety in a multithread environmentGabriele Svelto2006/10/20 01:26 AM
                            Language safety in a multithread environmentTzvetan Mikov2006/10/20 01:34 PM
                              Language safety in a multithread environmentS. Rao2006/10/20 08:39 PM
                                Language safety in a multithread environmentTzvetan Mikov2006/10/21 09:39 AM
                                  Language safety in a multithread environmentS. Rao2006/10/21 03:54 PM
                                    Language safety in a multithread environmentRicardo B2006/10/21 04:45 PM
                                    Language safety in a multithread environmentTzvetan Mikov2006/10/22 09:42 AM
                                      Language safety in a multithread environmentRob Thorpe2006/10/23 09:35 AM
                                        Language safety in a multithread environmentLinus Torvalds2006/10/23 04:06 PM
                                          Language safety in a multithread environmentTzvetan Mikov2006/10/23 11:55 PM
                                            Language safety in a multithread environmentLinus Torvalds2006/10/24 07:35 AM
                                              Language safety in a multithread environmentTzvetan Mikov2006/10/24 11:26 AM
                                                ClarificationTzvetan Mikov2006/10/24 12:15 PM
                                                Language safety in a multithread environmentLinus Torvalds2006/10/25 07:37 AM
                                                  Language safety in a multithread environmentRob Thorpe2006/10/25 07:52 AM
                                                  Language safety in a multithread environmentTzvetan Mikov2006/10/25 09:01 AM
                                                    Language safety in a multithread environmentLinus Torvalds2006/10/25 05:24 PM
                                                      Language safety in a multithread environmentS. Rao2006/10/25 06:08 PM
                                                        Other approachesRob Thorpe2006/10/26 10:09 AM
                                                          Hang on I forgot the other approachesRob Thorpe2006/10/26 10:27 AM
                                                            Hang on I forgot the other approachesS. Rao2006/10/26 02:16 PM
                                                      Language safety in a multithread environmentGabriele Svelto2006/10/26 03:56 AM
                                                        Language safety in a multithread environmentLinus Torvalds2006/10/26 07:52 AM
                                                          Language safety in a multithread environmentGabriele Svelto2006/10/26 08:11 AM
                                                            "Safety" and different types of safetyRob Thorpe2006/10/26 08:58 AM
                                                              "Safety" and different types of safetyLinus Torvalds2006/10/26 09:26 AM
                                                                "Safety" and different types of safetyRob Thorpe2006/10/26 09:50 AM
                                                                  "Safety" and different types of safetyTzvetan Mikov2006/10/26 10:56 AM
                                                                    Exactly!Gabriele Svelto2006/10/27 12:29 AM
                                                                  message passing languagesFoo_2006/10/28 01:19 AM
                                                        Language safety in a multithread environmentRob Thorpe2006/10/26 10:45 AM
                                                          RecapTzvetan Mikov2006/10/26 11:19 AM
                                                            Information and ickiness on JavaRob Thorpe2006/10/27 04:38 AM
                                                              Excellent threadJasonB2006/10/27 02:57 PM
                                                                Excellent threadsp2006/10/28 12:38 AM
                                                              Information and ickiness on JavaTzvetan Mikov2006/10/27 05:12 PM
                                                      Language safety in a multithread environmentTzvetan Mikov2006/10/26 08:17 AM
                                                        Language safety in a multithread environmentLinus Torvalds2006/10/26 08:57 AM
                                                          Language safety in a multithread environmentRob Thorpe2006/10/26 09:38 AM
                                                            Language safety in a multithread environmentS. Rao2006/10/26 02:51 PM
                                                              Language safety in a multithread environmentRob Thorpe2006/10/27 01:26 AM
                                                                Language safety in a multithread environmentLinus Torvalds2006/10/27 08:40 AM
                                                                  Language safety in a multithread environmentRob Thorpe2006/10/27 09:06 AM
                                                                    Language safety in a multithread environmentLinus Torvalds2006/10/27 10:53 AM
                                                                      Language safety in a multithread environmentrwessel2006/10/27 03:21 PM
                                                                        Explicit vs implicit barriersTzvetan Mikov2006/10/27 05:39 PM
                                                                          Explicit vs implicit barriersLinus Torvalds2006/10/27 06:28 PM
                                                                            Explicit vs implicit barriersTzvetan Mikov2006/10/28 06:02 PM
                                                                              Explicit vs implicit barriersLinus Torvalds2006/10/28 07:55 PM
                                                                                Explicit vs implicit barriersRob Thorpe2006/10/31 12:37 PM
                                                                            Explicit vs implicit barriersRob Thorpe2006/10/31 12:15 PM
                                                                          Explicit vs implicit barriersrwessel2006/10/27 06:39 PM
                                                                      Why OoO stores are evil but OoO loads are o.k.?Michael S2006/10/28 02:10 PM
                                                                        Why OoO stores are evil but OoO loads are o.k.?Linus Torvalds2006/10/28 04:29 PM
                                                                          Why OoO stores are evil but OoO loads are o.k.?Ricardo B2006/10/30 06:36 AM
                                                                            Why OoO stores are evil but OoO loads are o.k.?Linus Torvalds2006/10/30 09:57 AM
                                                                              Why OoO stores are evil but OoO loads are o.k.?Tzvetan Mikov2006/10/30 11:55 AM
                                                                                Why OoO stores are evil but OoO loads are o.k.?Michael S2006/10/30 12:14 PM
                                                                                  Why OoO stores are evil but OoO loads are o.k.?Tzvetan Mikov2006/10/30 12:20 PM
                                                                                    Why OoO stores are evil but OoO loads are o.k.?Michael S2006/10/30 01:11 PM
                                                                                      A reference would be nice :-) (NT)Tzvetan Mikov2006/10/30 01:32 PM
                                                                                        A reference Michael S2006/10/30 03:37 PM
                                                                                          A reference Tzvetan Mikov2006/11/03 08:57 PM
                                                                                            check your e-mbox (NT)Michael S2006/11/04 12:47 PM
                                                                                      Why OoO stores are evil but OoO loads are o.k.?David Kanter2006/10/30 01:40 PM
                                                                                        Why OoO stores are evil but OoO loads are o.k.?Michael S2006/10/30 02:28 PM
                                                                                          Why OoO stores are evil but OoO loads are o.k.?David Kanter2006/10/30 02:43 PM
                                                                                            Why OoO stores are evil but OoO loads are o.k.?Michael S2006/10/30 04:36 PM
                                                                                              Why OoO stores are evil but OoO loads are o.k.?David Kanter2006/10/30 10:14 PM
                                                                                Why OoO stores are evil but OoO loads are o.k.?Linus Torvalds2006/10/30 12:53 PM
                                                                      Language safety in a multithread environmentRob Thorpe2006/10/31 11:25 AM
                                                                        Language safety in a multithread environmentLinus Torvalds2006/10/31 12:59 PM
                                                                          Language safety in a multithread environmentRob Thorpe2006/10/31 01:38 PM
                                                                            Language safety in a multithread environmentLinus Torvalds2006/10/31 02:02 PM
                                                                              Language safety in a multithread environmentLinus Torvalds2006/10/31 03:54 PM
                                                                                Will reply tommorrow, very busy today (NT)Rob Thorpe2006/11/01 11:14 AM
                                                                              Language safety in a multithread environmentRob Thorpe2006/11/02 01:30 PM
                                                                                Language safety in a multithread environmentrwessel2006/11/02 02:06 PM
                                                                                  Language safety in a multithread environmentGabriele Svelto2006/11/03 02:39 AM
                                                                                    Language safety in a multithread environmentLinus Torvalds2006/11/03 09:40 AM
                                                                                      Language safety in a multithread environmentGabriele Svelto2006/11/03 10:12 AM
                                                                                        Language safety in a multithread environmentLinus Torvalds2006/11/03 01:20 PM
                                                                                          Language safety in a multithread environmentGabriele Svelto2006/11/06 03:42 AM
                                                                                      Language safety in a multithread environmentEric Bron2006/11/04 04:47 AM
                                                                                        Language safety in a multithread environmentGabriele Svelto2006/11/06 04:13 AM
                                                                                      Language safety in a multithread environmentWilco2006/11/04 12:29 PM
                                                                                        Language safety in a multithread environmentRicardo B2006/11/04 02:54 PM
                                                                                        Language safety in a multithread environmentGabriele Svelto2006/11/06 03:54 AM
                                                                                          Language safety in a multithread environmentWilco2006/11/06 12:26 PM
                                                                                            Language safety in a multithread environmentGabriele Svelto2006/11/07 01:54 AM
                                                                                              Language safety in a multithread environmentWilco2006/11/07 04:34 PM
                                                                                                Language safety in a multithread environmentGabriele Svelto2006/11/08 02:29 AM
                                                                                      Language safety in a multithread environmentrwessel2006/11/04 10:13 PM
                                                                                        Language safety in a multithread environmentrwessel2006/11/04 10:23 PM
                                                                                      Language safety in a multithread environmentMichael S2006/11/05 04:05 AM
                                                                                        Language safety in a multithread environmentLinus Torvalds2006/11/05 10:14 AM
                                                                                          unaligned SSE testsEric Bron2006/11/05 02:26 PM
                                                                                            unaligned SSE testsMichael S2006/11/05 03:47 PM
                                                                                              unaligned SSE testsEric Bron2006/11/05 04:23 PM
                                                                                                unaligned SSE testsMichael S2006/11/05 05:03 PM
                                                                                                  unaligned SSE testsEric Bron2006/11/05 05:28 PM
                                                                                                    Now results make sense. How about Psc? (NT)Michael S2006/11/05 05:37 PM
                                                                                                      tomorrow, it's 01:43 PM where I'm (NT)Eric Bron2006/11/05 05:44 PM
                                                                                                        I mean 01:43 AM (NT)Eric Bron2006/11/05 05:45 PM
                                                                                                      PSC scoresEric Bron2006/11/06 02:25 AM
                                                                                                        PSC scoresMichael S2006/11/06 04:28 AM
                                                                                                          PSC scoresIan Ameline2006/11/06 09:00 AM
                                                                                                            meaningful performance gainMichael S2006/11/06 09:33 AM
                                                                                                              meaningful performance gainLinus Torvalds2006/11/06 11:15 AM
                                                                                                                Dynamically resolved alignmentDavid Kanter2006/11/06 12:49 PM
                                                                                                                  Dynamically resolved alignmentRicardo B2006/11/06 01:46 PM
                                                                                                                  Dynamically resolved alignmentLinus Torvalds2006/11/06 01:48 PM
                                                                                                                meaningful performance gainEric Bron2006/11/07 02:23 AM
                                                                                                        RE: "I suppose we will have to continue to align" ...Not really :)redpriest2006/11/06 05:11 PM
                                                                                                          Your point?Michael S2006/11/06 05:47 PM
                                                                                                            Actually Michael, it's quite a bit more than that.redpriest2006/11/06 06:30 PM
                                                                                                              Actually Michael, it's quite a bit more than that.anonymous2006/11/06 08:36 PM
                                                                                                                Actually Michael, it's quite a bit more than that.Linus Torvalds2006/11/06 09:01 PM
                                                                                                                  Actually Michael, it's quite a bit more than that.Rob Thorpe2006/11/07 03:31 AM
                                                                                                                  Actually Michael, it's quite a bit more than that.anonymous2006/11/07 07:27 AM
                                                                                                              link vs linqMichael S2006/11/07 02:04 AM
                                                                                                                link vs linqanonymous2006/11/07 07:13 AM
                                                                                                                  Thank you (NT)Michael S2006/11/07 08:04 AM
                                                                                                          Barcelona dreamsEric Bron2006/11/07 01:08 AM
                                                                                          Language safety in a multithread environmentWilco2006/11/05 03:14 PM
                                                                                            Language safety in a multithread environmentRicardo B2006/11/05 03:56 PM
                                                                                              Language safety in a multithread environmentWilco2006/11/05 06:52 PM
                                                                                                Language safety in a multithread environmentRicardo B2006/11/05 09:09 PM
                                                                                                  Language safety in a multithread environmentWilco2006/11/06 02:25 PM
                                                                                                    Language safety in a multithread environmentRicardo B2006/11/06 03:48 PM
                                                                                                      Language safety in a multithread environmentWilco2006/11/06 04:44 PM
                                                                                                        Language safety in a multithread environmentRicardo B2006/11/06 04:53 PM
                                                                                                        A common problemRob Thorpe2006/11/07 09:45 AM
                                                                                                          A common problemWilco2006/11/09 04:46 PM
                                                                                                            A common problemRob Thorpe2006/11/10 08:29 AM
                                                                                                              A common problemrwessel2006/11/13 04:47 PM
                                                                                                                A common problemWilco2006/11/13 05:51 PM
                                                                                                                  A common problemRob Thorpe2006/11/14 03:38 AM
                                                                                                                    A common problemLinus Torvalds2006/11/14 09:37 AM
                                                                                                                      A common problemRob Thorpe2006/11/14 11:18 AM
                                                                                                                        A common problemagony2006/11/14 02:57 PM
                                                                                                                          A common problemrwessel2006/11/14 04:13 PM
                                                                                                                            A common problemAndi Kleen2006/11/15 04:24 AM
                                                                                                                              A common problemrwessel2006/11/15 04:42 PM
                                                                                                                        A common problemDoug Siebert2006/11/14 09:50 PM
                                                                                                                          x86 versus other platformsDavid Kanter2006/11/15 12:58 AM
                                                                                                                            x86 versus other platformsj2006/11/15 01:57 AM
                                                                                                                          A common problemRob Thorpe2006/11/15 11:56 AM
                                                                                                                            A common problemWilco2006/11/15 01:54 PM
                                                                                                                              A common problemCarlie Coats2006/11/16 06:44 AM
                                                                                                                              A common problemLinus Torvalds2006/11/16 10:03 AM
                                                                                                                      A common problemCarlie Coats2006/11/16 06:38 AM
                                                                                                                        A common problemRob Thorpe2006/11/16 07:29 AM
                                                                                                                    A common problemWilco2006/11/15 02:13 PM
                                                                                                                      A common problemRob Thorpe2006/11/16 09:20 AM
                                                                                                                        A common problemrwessel2006/11/16 04:27 PM
                                                                                                                        A common problemWilco2006/11/17 02:20 PM
                                                                                                                          A common problemRob Thorpe2006/11/20 04:14 AM
                                                                                          You are aiming too lowMichael S2006/11/05 04:19 PM
                                                                                            And you're aiming quite highRob Thorpe2006/11/06 08:00 AM
                                                                                    Language safety in a multithread environmentWilco2006/11/04 11:20 AM
                                                                                      Language safety in a multithread environmentGabriele Svelto2006/11/06 04:10 AM
                                                                                        Language safety in a multithread environmentWilco2006/11/06 10:59 AM
                                                                                          Language safety in a multithread environmentGabriele Svelto2006/11/08 02:47 AM
                                                                                            Language safety in a multithread environmentWilco2006/11/13 05:13 PM
                                                                                              Language safety in a multithread environmentGabriele Svelto2006/11/14 02:15 AM
                                                                                                Language safety in a multithread environmentWilco2006/11/17 02:51 PM
                                                                                    Language safety in a multithread environmentrwessel2006/11/04 10:11 PM
                                                                                  Language safety in a multithread environmentRob Thorpe2006/11/03 09:58 AM
                                                                  Language safety in a multithread environmentbandwidth2006/10/27 06:47 PM
                                                                    Language safety in a multithread environmentLinus Torvalds2006/10/27 07:52 PM
                                                                      Language safety in a multithread environmentLinus Torvalds2006/10/27 08:10 PM
                                                                        Language safety in a multithread environmentbandwidth2006/10/28 03:17 AM
                                                                      Language safety in a multithread environmentbandwidth2006/10/28 03:09 AM
                                                                        Language safety in a multithread environmentLinus Torvalds2006/10/28 09:23 AM
                                                                      Language safety in a multithread environmentGabriele Svelto2006/10/30 03:26 AM
                                                                        Language safety in a multithread environmentRicardo B2006/10/30 06:52 AM
                                                                        Language safety in a multithread environmentJasonB2006/10/30 07:12 AM
                                                                          Language safety in a multithread environmentGabriele Svelto2006/10/30 07:28 AM
                                                                            Language safety in a multithread environmentJasonB2006/10/30 08:32 AM
                                                                              Language safety in a multithread environmentGabriele Svelto2006/10/30 08:48 AM
                                                                                Language safety in a multithread environmentJasonB2006/10/30 09:21 AM
                                                                          Language safety in a multithread environmentRicardo B2006/10/30 07:31 AM
                                                                            Language safety in a multithread environmentGabriele Svelto2006/10/30 07:43 AM
                                                                              Language safety in a multithread environmentRicardo B2006/10/30 07:54 AM
                                                                                Language safety in a multithread environmentGabriele Svelto2006/10/30 08:02 AM
                                                                                  Language safety in a multithread environmentRicardo B2006/10/30 08:12 AM
                                                                                    Language safety in a multithread environmentGabriele Svelto2006/10/30 08:29 AM
                                                                                      Language safety in a multithread environmentRicardo B2006/10/30 08:40 AM
                                                                                Language safety in a multithread environmentdraugaz2006/10/30 12:31 PM
                                                                              Language safety in a multithread environmentLinus Torvalds2006/10/30 10:09 AM
                                                                                Language safety in a multithread environmentGabriele Svelto2006/10/31 02:01 AM
                                                                                  Language safety in a multithread environmentLinus Torvalds2006/10/31 08:59 AM
                                                                    Language safety in a multithread environmentpgerassi2006/10/27 08:00 PM
                                                                      Language safety in a multithread environmentbandwidth2006/10/28 03:25 AM
                                                                        Language safety in a multithread environmentpgerassi2006/10/31 01:32 AM
                                                                      Language safety in a multithread environmentRob Thorpe2006/10/31 06:52 AM
                                                        pythonFoo_2006/10/28 01:51 AM
                                                          pythonTzvetan Mikov2006/10/28 04:55 PM
                                                            pythonFoo_2006/10/29 03:51 AM
                                                          pythonRob Thorpe2006/10/31 05:01 AM
                                        Language safety in a multithread environmentTzvetan Mikov2006/10/24 10:53 AM
                                          Language safety in a multithread environmentRob Thorpe2006/10/25 01:00 AM
                                          Language safety in a multithread environment (Lock-free structures)Ray2006/10/28 11:25 AM
                                            Language safety in a multithread environment (Lock-free structures)Linus Torvalds2006/10/28 11:58 AM
                                              Language safety in a multithread environment (Lock-free structures)Ray2006/10/28 04:47 PM
                                                Language safety in a multithread environment (Lock-free structures)Linus Torvalds2006/10/28 08:02 PM
                                        Language safety in a multithread environmentJuho Snellman2006/10/25 03:59 AM
                                          Language safety in a multithread environmentRob Thorpe2006/10/25 04:38 AM
                                            Language safety in a multithread environmentJuho Snellman2006/10/25 06:38 AM
                                              Language safety in a multithread environmentRob Thorpe2006/10/25 07:06 AM
                                                More on this topicRob Thorpe2006/10/25 08:41 AM
                                                  More on this topicTzvetan Mikov2006/10/25 10:17 AM
                                                    More on this topicRob Thorpe2006/10/26 02:02 AM
                                                Language safety in a multithread environmentJuho Snellman2006/10/25 10:33 AM
                                                  Language safety in a multithread environmentRob Thorpe2006/10/26 01:48 AM
                                      Language safety in a multithread environmentS. Rao2006/10/23 03:40 PM
                                        Language safety in a multithread environmentTzvetan Mikov2006/10/23 11:38 PM
                                        Language safety in a multithread environmentGabriele Svelto2006/10/24 01:02 AM
                                          Language safety in a multithread environmentS. Rao2006/10/24 12:40 PM
                                        Comparison to stack allocationRob Thorpe2006/10/24 03:09 AM
                                          Comparison to stack allocationS. Rao2006/10/24 12:32 PM
                                            Exactly, see my other post (NT)Rob Thorpe2006/10/25 12:53 AM
                          Language safety in a multithread environmentDavid Wragg2006/10/20 02:46 AM
                            Language safety in a multithread environmentTzvetan Mikov2006/10/21 09:24 AM
                          Ugh, evil problemRob Thorpe2006/10/20 03:56 AM
                            Ugh, evil problemGabriele Svelto2006/10/20 04:38 AM
                              Ugh, evil problemRob Thorpe2006/10/20 05:47 AM
                                Ugh, evil problemGabriele Svelto2006/10/20 05:54 AM
                                Ugh, evil problemRicardo B2006/10/20 06:43 AM
                            Ugh, evil problemJasonB2006/10/20 06:17 AM
                              Ugh, evil problemS. Rao2006/10/20 08:09 PM
                                Ugh, evil problemJasonB2006/10/21 07:58 AM
                                SecondedRob Thorpe2006/10/23 01:53 AM
                          Copying GC Rob Thorpe2006/10/23 02:51 AM
                            Copying GC Tzvetan Mikov2006/10/23 11:13 PM
                              Copying GC Rob Thorpe2006/10/24 01:22 AM
              Python + native extensionsFoo_2006/10/18 02:29 PM
                Python + native extensionssp2006/10/19 01:05 AM
  z9 - Power5 connection?foobar2006/10/17 01:15 PM
    z9 - Power5 connection?David Kanter2006/10/17 02:03 PM
      z9 - Power5 connection?foobar2006/10/17 02:33 PM
        z9 - Power5 connection?rwessel2006/10/17 07:39 PM
          z9 - Power5 connection?foobar2006/10/17 08:38 PM
            Sounds like I should edit the article...David Kanter2006/10/17 10:48 PM
            z9 - Power5 connection?Jesper Frimann2006/10/17 11:59 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell avocado?