A non straw man view of hardware transactional memory

By: anon2 (anon.delete@this.anon.com), April 2, 2021 5:28 am
Room: Moderated Discussions
Carson (carson.delete@this.example.edu) on April 2, 2021 1:11 am wrote:
> anon2 (anon.delete@this.anon.com) on March 31, 2021 11:03 pm wrote:
> > But that's not what TSX is up against. This has the same smell as those early stupid GPGPU benchmarks
> > that tried to claim 10x, 100x, 1000x performance improvement over CPU despite not even a single component
> > of the GPU having better than 10x advantage over the CPU (FLOPS, memory bandwidth, etc), and many things
> > being worse (BW and latency to system memory, instruction latencies, branches, performance per thread
> > etc). Most of their speedups came because they completely rewrote the algorithms for vectorization
> > and parallelism for the GPU and did not re-test the modern baseline on the CPU.
>
> This is what Linus recently called "optimizing bubblesort", and people have been using it for
> decades to sell flocks of their particular chickens, from Transputers to SPARC T-series.
>
> It's easy to get great parallel speedup if you hobble the uniprocessor with a shitty algorithm.
>
> Likewise, as you say, it's easy to make your hardware lock elision shine
> if you encumber the implementation with lots of elidable locks.
>
> > The modern baseline for a central highly contended b-tree
> > is using RCU (or similar) for persistence guarantees
> > and fine grained (per node) sequence counters for validity/atomicity of lookup vs updates, and fine grained
> > locking pushed down to the lowest common nodes to protect
> > updates from one another (or possibly for simplicity
> > a global lock for rebalancing and leaf-node locks for non-rebalancing
> > updates, but b-trees are quite amenable
> > to have fine graned rebalance operations because splits and joins frequency drops by an order of magnitude
> > or two as you move up levels, if parameters are chosen correctly.
> >
> > So now what the naive TSX implementation is up against is something
> > with zero atomic operations on the lookup-side,
> > compared with what is quite a heavy transaction begin/end
> > in Intel implementations. So TSX is slower right out
> > of the gate, which is a big problem (in theory you would
> > think an implementation might be able to make it almost
> > zero cost, although complexity and cost is obviously then hidden in the microarchitecture).
>
> You're not wrong, but...
>
> RCUing a non-trivial data structure is decidedly non-trivial.

The data structure in the test referenced is relatively trivial though. B-trees are fairly simple to make lock free lookups after you have a persistence guarantee.

> It took years to get
> lock-free dcache access in the Linux kernel and that complexity imposes an ongoing
> maintenance cost because it forces complex interfaces on all the surrounding code.

It took years to get any kind of SMP support at all in Linux.

But if you mean directory entry cache, then the lookup data structure was made lock free a long time ago. Later this was significantly expanded to be able to do several dependent lookups without locks. Which I'm sure is very complicated but TM is unlikely to be able to help you in such situations let alone make them simpler.

>
> Remember, transactional memory started life as a software technique for writing
> complex locking code without multiplicative (exponential) complexity blowup.

I'm not sure what there is to remember. I know what people hoped it would do, but it doesn't (really).

>
> A hardware technique that makes such code easier to write and maintain, even at the cost
> of ultimate performance, is valuable. It can even give you a net performance win by making
> it practical to parallelize software that otherwise wouldn't be worth the bother.
>
> (This is a variant of the "why use compilers when hand-tuned assembler is faster?" argument.)
>
> Second, not all data structures can be RCUed or otherwise given a lock-free fast path. This is the limit
> case of the preceding problem. Sometimes the semantics you need just don't lend themselves to RCU.

No, and the simple matter is not all approaches to a problem scale to multiple CPUs in general. Any requirement you put on a data structure will pare down the viable solution space, which is just a reality. I don't claim what I said were "modern parallel software techniques" solve all such problems so you can just make any data structure parallel, I said that TM does not seem to significantly expand your solution space.

>
> It's not at all clear that the SAP programmers are stupid or lazy; it's quite plausible
> that they tries quite hard and couldn't find any better alternatives. It's certainly
> clear that they did not set out to create a star vehicle for TSX

I didn't say they were stupid or lazy or it's a simple problem. I'm sure it's not, which is also why the simple problem in the referenced paper that was microbenchmarked is not relevant.

>
> I don't think either of us really understands the size of the class of problems for which HTM is beneficial.
>
> (Perhaps an intermediate case is a multiprocessor scheduler. The basic design of one
> priority queue which everyone fights over the head of runs out of steam pretty fast,
> so a lot of designs are approximations which sacrifice strict prioritization in order
> to limit contention. If HTM allows more accurate prioritization, that's a win.)
>
> Jumping back to the anti-HTM side, it helps handle more contention, but if your data structure suffers a
> lot of contention, there are basic speed-of-light limits on the performance that can be achieved. I don't
> see a way to not need to think very carefully about locking and contention in any parallel application.

Well I'm not so sure of that. Linus said it better than I could in another sub thread here. TLE in particular (the one which claims to offer a simple scalable solution) suffers badly from the everyone-falls-off-the-cliff problem, and doing explicit TM programming in a scalable way does not seem to be at all simple or yielding of good results (with the exception perhaps of Linus' constrained TM usage but that also doesn't have good results on Intel CPUs at least because of the straight line performance overhead).

I mean so far it's all still just handwaving to me. Not that I've worked extensively with TM myself, but at this point of the game I haven't seen really anything at all use it, and certainly not to great advantage. You assert above that it _helps_ handle more contention. Well it is _intended_ to help to handle more contention. In some ideal world or specific microbenchmark maybe. In real world processors that have implemented working on real world code, we still don't have much evidence for the assertion.

>
> > I would say though that I have less confidence in Intel's implementations of such things
> > particularly looking at what Apple have done with M1 atomic operations, so it's entirely
> > possible a really good design team could come up with an implementation that's significantly
> > better and opens up a more significant niche of legacy code that could benefit from it.
>
> Agreed.

< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Armv9 officially announcedJon Masters2021/03/30 11:41 AM
  Armv9 officially announcedGabriele Svelto2021/03/30 01:27 PM
    HTM and TLEFoo_2021/03/30 01:31 PM
      HTM and TLEdmcq2021/03/30 03:03 PM
      Intel RTM and HLE is a successGanon2021/03/30 05:22 PM
        Intel RTM and HLE is a success (is it?)Foo_2021/03/31 01:16 AM
          Intel RTM and HLE was an abject failureanonymou52021/03/31 03:04 AM
          Intel RTM and HLE is a success (is it?)Andrey2021/03/31 05:27 AM
            Intel RTM and HLE is a success (is it?)Foo_2021/03/31 05:58 AM
              Intel RTM and HLE is a success (is it?)Andrey2021/03/31 07:45 AM
                Intel RTM and HLE is a success (is it?)Foo_2021/03/31 09:32 AM
                  Intel RTM and HLE is a success (is it?)Andrey2021/03/31 09:57 AM
                    Intel RTM and HLE is a success (is it?)anonymou52021/03/31 10:39 AM
                      ^ feel free to delete this one -- broken HTML there (NT)anonymou52021/03/31 10:40 AM
                      Intel RTM and HLE is a success (is it?)Andrey2021/03/31 10:47 AM
                    Intel RTM and HLE is a success (is it?)anonymou52021/03/31 10:40 AM
                Intel RTM and HLE is a success (is it?)Ganon2021/03/31 09:58 AM
                  Intel RTM and HLE is a success (is it?)anonymou52021/03/31 10:42 AM
                  Intel RTM and HLE is a success (is it?)Linus Torvalds2021/03/31 11:54 AM
                    Intel RTM and HLE is a success (is it?)Linus Torvalds2021/03/31 12:00 PM
                      Any idea about IBM?Mark Roulo2021/03/31 12:15 PM
                        Any idea about IBM?Linus Torvalds2021/03/31 12:37 PM
                          Any idea about IBM?dmcq2021/03/31 03:04 PM
                            Any idea about IBM?Linus Torvalds2021/03/31 04:44 PM
                              A non straw man view of hardware transactional memoryGanon2021/03/31 07:52 PM
                                A non straw man view of hardware transactional memoryanon22021/03/31 11:03 PM
                                  A non straw man view of hardware transactional memoryCarson2021/04/02 01:11 AM
                                    A non straw man view of hardware transactional memoryanon22021/04/02 05:28 AM
                              IBM zArch TM - guaranteed progressDavid Kanter2021/03/31 08:37 PM
                              Any idea about IBM?Andrey2021/03/31 11:31 PM
                                Any idea about IBM?Linus Torvalds2021/04/01 10:54 AM
                                  Any idea about IBM?Andrey2021/04/02 12:50 PM
                        Any idea about IBM?someone2021/04/01 12:02 AM
                      Intel RTM and HLE is a success (is it?)anon22021/03/31 03:46 PM
                        Intel RTM and HLE is a success (is it?)Linus Torvalds2021/03/31 05:08 PM
                          Leaving it to software is tricky!David Kanter2021/03/31 08:41 PM
                            And a hardware predicttor needs access to fallback timingCarson2021/04/01 11:13 PM
                              Hardware fallback pathAnon2021/04/02 10:51 AM
                              And a hardware predicttor needs access to fallback timingLinus Torvalds2021/04/03 10:41 AM
                                And a hardware predicttor needs access to fallback timingLinus Torvalds2021/04/03 11:11 AM
                                  And a hardware predicttor needs access to fallback timingsr2021/04/03 11:30 AM
                                    And a hardware predicttor needs access to fallback timingLinus Torvalds2021/04/03 12:14 PM
                                      And a hardware predicttor needs access to fallback timingsr2021/04/03 12:39 PM
                                        And a hardware predicttor needs access to fallback timingAnon2021/04/03 02:08 PM
                                          And a hardware predicttor needs access to fallback timingsr2021/04/03 02:33 PM
                                            And a hardware predicttor needs access to fallback timingdmcq2021/04/04 05:35 AM
                                        And a hardware predicttor needs access to fallback timingLinus Torvalds2021/04/03 02:22 PM
                                          Transactional memory isn't exclusive to lockingsr2021/04/04 12:17 AM
                                            Transactional memory isn't exclusive to lockingAspect of Anonimity2021/04/04 03:49 AM
                                            Transactional memory isn't exclusive to lockingAndrey2021/04/04 04:58 AM
                                              Transactional memory isn't exclusive to lockingsr2021/04/04 10:10 AM
                                                Transactional memory isn't exclusive to lockingAndrey2021/04/04 10:33 AM
                                                  Transactional memory isn't exclusive to lockingsr2021/04/05 02:41 AM
                                      And a hardware predicttor needs access to fallback timingGeertB2021/04/04 07:08 PM
                            Leaving it to software is tricky!Andrey2021/04/02 03:00 PM
                          Intel RTM and HLE is a success (is it?)@never_released2021/04/01 08:21 AM
                            Intel RTM and HLE is a success (is it?)@never_released2021/04/01 08:30 AM
                            Intel RTM and HLE is a success (is it?)Linus Torvalds2021/04/01 10:00 AM
                              Intel RTM and HLE is a success (is it?)dmcq2021/04/01 10:35 AM
                                Intel RTM and HLE is a success (is it?)Linus Torvalds2021/04/01 10:59 AM
                                  Is HTM actually in ARMv9?dncq2021/04/01 11:26 AM
                                    Is HTM actually in ARMv9?Linus Torvalds2021/04/01 12:13 PM
                              Intel RTM and HLE is a success (is it?)Jörn Engel2021/04/01 09:15 PM
                          Intel RTM and HLE is a success (is it?)---2021/04/02 10:00 AM
                      Intel RTM and HLE is a success (is it?)Jon Masters2021/04/01 10:56 AM
                        RockMichael S2021/04/01 12:29 PM
            Intel RTM and HLE is a success (is it?)Linus Torvalds2021/03/31 11:50 AM
              Intel RTM and HLE is a success (is it?)anon22021/03/31 03:57 PM
                Intel RTM and HLE is a success (is it?)anon32021/03/31 04:09 PM
          Intel RTM and HLE is a success (is it?)someone2021/03/31 11:56 PM
            Intel RTM and HLE is a success (is it?)someone2021/04/01 12:21 AM
              Intel RTM and HLE is a success (is it?)none2021/04/01 04:31 AM
                Intel RTM and HLE is a success (is it?)anonymou52021/04/01 09:24 AM
                  Intel RTM and HLE is a success (is it?)anony2021/04/01 10:26 AM
                  Intel RTM and HLE is a success (is it?)none2021/04/01 11:20 AM
                Intel RTM and HLE is a success (is it?)Brendan2021/04/01 05:23 PM
                  Intel RTM and HLE is a success (is it?)Adrian2021/04/02 01:03 AM
      Transactional memory similarity to garbage collectionPaul A. Clayton2021/04/05 01:53 PM
        Maybe notMark Roulo2021/04/05 03:07 PM
        Transactional memory similarity to garbage collectionAnon2021/04/05 03:14 PM
        No conflict between theory and practiceAspect of Anonimity2021/04/05 07:57 PM
          No conflict between theory and practiceNoSpammer2021/04/05 09:38 PM
            No conflict between theory and practicedmcq2021/04/06 07:05 AM
              No conflict between theory and practicesr2021/04/06 09:22 AM
                No conflict between theory and practiceBen LaHaise2021/04/06 03:43 PM
                  No conflict between theory and practicesr2021/04/07 09:42 AM
                    No conflict between theory and practiceAnon2021/04/07 10:06 AM
                    No conflict between theory and practiceLinus Torvalds2021/04/07 10:35 AM
                      No conflict between theory and practicesr2021/04/07 12:34 PM
                        If HTM were well implemented, nobody would complain (NT)Anon2021/04/07 12:54 PM
                          If HTM were well implemented, nobody would complaindmcq2021/04/07 03:36 PM
                            If HTM were well implemented, nobody would complainAnon2021/04/07 04:04 PM
                          If HTM were well implemented, nobody would complain---2021/04/08 09:37 AM
                        No conflict between theory and practiceLinus Torvalds2021/04/07 01:20 PM
                          No conflict between theory and practiceAndrey2021/04/07 02:32 PM
                            No conflict between theory and practicedmcq2021/04/07 03:32 PM
                              No conflict between theory and practiceanonymou52021/04/07 04:26 PM
                              No conflict between theory and practiceAndrey2021/04/07 05:54 PM
                                No conflict between theory and practiceLinus Torvalds2021/04/08 08:41 AM
                                  No conflict between theory and practiceAndrey2021/04/08 09:12 AM
                                  No conflict between theory and practiceRobert Williams2021/04/08 09:15 AM
                                    No conflict between theory and practiceLinus Torvalds2021/04/08 09:56 AM
                                      No conflict between theory and practiceRobert Williams2021/04/08 07:50 PM
                                        No conflict between theory and practiceLinus Torvalds2021/04/09 09:25 AM
                                          TSX for all?Robert Williams2021/04/09 12:46 PM
                                            It helps adoption when developers can run the code an their machines.Mark Roulo2021/04/09 12:54 PM
                                              It helps adoption when developers can run the code an their machines.me2021/04/09 02:21 PM
                                                It helps adoption when developers can run the code an their machines.Andrey2021/04/10 07:08 AM
                                                  It helps adoption when developers can run the code an their machines.me2021/04/10 12:43 PM
                                                    It helps adoption when developers can run the code an their machines.Robert Williams2021/04/10 07:05 PM
                                                      It helps adoption when developers can run the code an their machines.Andrey2021/04/11 01:42 AM
                                                        It helps adoption when developers can run the code an their machines.Michael S2021/04/11 04:23 AM
                                              It helps adoption when developers can run the code an their machines.Robert Williams2021/04/10 08:24 AM
                                                It helps adoption when developers can run the code an their machines.Andrey2021/04/10 10:36 AM
                                                  It helps adoption when developers can run the code an their machines.Michael S2021/04/10 10:58 AM
                                                    It helps adoption when developers can run the code an their machines.Robert Williams2021/04/10 11:42 AM
                                                    It helps adoption when developers can run the code an their machines.Brendan2021/04/10 11:27 PM
                                                      It helps adoption when developers can run the code an their machines.Michael S2021/04/11 03:34 AM
                                                        It helps adoption when developers can run the code an their machines.Brendan2021/04/11 04:18 PM
                                            TSX for all?wumpus2021/04/09 01:10 PM
                                            TSX for all?Linus Torvalds2021/04/09 03:03 PM
                                              [CLICK BAIT?] Torvalds to recommend arm64!anonymous22021/04/09 03:17 PM
                                              TSX for all?Linus Torvalds2021/04/09 03:22 PM
                                                amen! (NT)anonymou52021/04/09 05:42 PM
                                            TSX for all?Emil Briggs2021/04/10 05:52 AM
                                              TSX for all?Michael S2021/04/10 11:13 AM
                                                TSX for all?Brendan2021/04/10 11:05 PM
                                          No conflict between theory and practiceanonymouse2021/04/09 03:02 PM
                            No conflict between theory and practiceLinus Torvalds2021/04/07 05:12 PM
                              No conflict between theory and practiceAndrey2021/04/07 05:33 PM
                                No conflict between theory and practiceanon22021/04/07 08:12 PM
                                  No conflict between theory and practiceAndrey2021/04/08 12:29 AM
                                    No conflict between theory and practiceAnon2021/04/08 12:50 AM
                                    No conflict between theory and practiceanon22021/04/08 12:52 AM
                                      No conflict between theory and practiceAndrey2021/04/08 02:41 AM
                                        No conflict between theory and practiceAdrian2021/04/08 03:09 AM
                                          No conflict between theory and practiceAndrey2021/04/08 03:50 AM
                                            No conflict between theory and practiceAdrian2021/04/08 05:07 AM
                                            No conflict between theory and practiceanon22021/04/08 05:18 AM
                                              No conflict between theory and practiceAndrey2021/04/08 08:18 AM
                                                No conflict between theory and practiceanon22021/04/08 09:11 AM
                                                  No conflict between theory and practiceAndrey2021/04/08 09:48 AM
                                                    No conflict between theory and practiceanon22021/04/08 04:10 PM
                                            No conflict between theory and practice---2021/04/08 08:30 PM
                                              No conflict between theory and practicedmcq2021/04/09 02:25 AM
                                                No conflict between theory and practiceLinus Torvalds2021/04/09 09:44 AM
                                        No conflict between theory and practiceanon22021/04/08 04:09 AM
                                          No conflict between theory and practicesr2021/04/10 01:22 AM
                                            No conflict between theory and practiceAnon2021/04/10 04:00 AM
                                              No conflict between theory and practiceEtienne Lorrain2021/04/12 12:56 AM
                                                No conflict between theory and practiceAnon2021/04/12 01:54 AM
                                                  No conflict between theory and practicedmcq2021/04/12 01:44 PM
                              No conflict between theory and practiceGabriele Svelto2021/04/08 01:03 AM
                                No conflict between theory and practicedmcq2021/04/12 01:53 PM
                                  No conflict between theory and practicedmcq2021/04/14 04:50 AM
                      transactional memory = memory lock/unlocksr2021/04/10 12:56 AM
                        transactional memory = memory lock/unlockAnon2021/04/10 04:05 AM
                    No conflict between theory and practiceanon22021/04/07 05:19 PM
                No conflict between theory and practicedmcq2021/04/07 01:16 PM
            No conflict between theory and practiceAnon2021/04/06 10:46 AM
            No conflict between theory and practiceAspect of Anonimity2021/04/06 02:16 PM
              No conflict between theory and practiceNoSpammer2021/04/10 12:02 AM
                No conflict between theory and practicesr2021/04/10 01:47 AM
                No conflict between theory and practiceAspect of Anonimity2021/04/10 11:30 PM
                  No conflict between theory and practiceAndrey2021/04/11 05:05 AM
                    HmAspect of Anonimity2021/04/12 07:29 PM
                      Special internal SRAM to store Mutex?Etienne Lorrain2021/04/13 01:01 AM
                        Special internal SRAM to store Mutex?Anon2021/04/13 01:26 AM
                        Special internal SRAM to store Mutex?Linus Torvalds2021/04/13 09:53 AM
                        Special internal SRAM to store Mutex?Dan Fay2021/04/13 12:03 PM
                          Special internal SRAM to store Mutex?Ben LaHaise2021/04/13 03:32 PM
                          Special internal SRAM to store Mutex?Gabriele Svelto2021/04/13 11:43 PM
                        Special internal SRAM to store Mutex?Carson2021/04/13 10:19 PM
          No conflict between theory and practicesr2021/04/06 07:12 AM
            No conflict between theory and practiceAnon2021/04/06 10:43 AM
            No conflict between theory and practiceAspect of Anonimity2021/04/06 03:20 PM
              No conflict between theory and practicesr2021/04/07 10:09 AM
                No conflict between theory and practiceanon22021/04/07 08:53 PM
  Armv9 officially announceddmcq2021/03/30 03:28 PM
    Totally backwads logic on compatibilityHeikki Kultala2021/03/30 03:44 PM
      Totally backwads logic on compatibilityDoug S2021/03/31 12:09 PM
        Totally backwads logic on compatibilitydmcq2021/03/31 03:15 PM
  Armv9 officially announcedv92021/04/12 03:57 PM
    Armv9 officially announcedj2021/04/13 05:07 AM
      Armv9 officially announcedDoug S2021/04/13 10:21 AM
        Armv9 officially announcedanonymou52021/04/13 11:06 AM
          Armv9 officially announcedDoug S2021/04/13 01:01 PM
        Armv9 officially announceddmcq2021/04/13 03:20 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell avocado?