7 ms·

I'm in school, so this may be oversimplified, but if the processor/assembly code is predicting the next result, it gets the result faster. The processor only do

by trolan 3y ago

I'm in school, so this may be oversimplified, but if the processor/assembly code is predicting the next result, it gets the result faster. The processor only does this prediction with conditional branches. The extra if for printing or finding the min invoke the prediction with the accuracies stated.

dataflow 3y ago
> The processor only does this prediction with conditional branches This sounds... wrong? Unless ARM64 is designed in an absurd way? I'd love to see the full disassembly; something seems funny here. If it was x86 I would say it's a conditional move causing this, but I don't know what's going on on ARM.
- fmstephe 3y ago
  It's interesting that you say conditional move here. I am confused by this behaviour, and although I definitely don't know what the answer is here; the non-lol version does have a CSEL (https://developer.arm.com/documentation/dui0802/b/CSEL https://developer.arm.com/documentation/dui0802/b/CSEL) which is totally missing from the lol version. Non-lol https://godbolt.org/z/ds1raTYc9 https://godbolt.org/z/ds1raTYc9 lol https://godbolt.org/z/c3afrb6bG https://godbolt.org/z/c3afrb6bG
  - dataflow 3y ago
    Ah there you go, that's the conditional move on ARM. Yeah, those are slow.
    - fmstephe 3y ago
      Yeah, after reading the blog post I felt reasonably confident that this is a compiler bug (in terms of perf). Hopefully someone with a deeper understanding can verify or discredit this here. I'm curious to see the fix for this (I assume it will generate a fix).
      - dataflow 3y ago
        Confused, why do you think this is a compiler bug? You should definitely expect a predictable branch to be faster than a conditional move, and the insertion of a conditional move in the original is totally sensible (albeit not intuitive if you haven't seen it).
        littlestymaar 3y ago
        > and the insertion of a conditional move in the original is totally sensible (albeit not intuitive if you haven't seen it). Would you mind expanding? If the conditional move isn't needed, and given that it's costly in terms of perfs, how is that “totally sensible” to have one here?
        dataflow 3y ago
        The performance of branches is data-dependent. The performance of conditional moves is data-independent. In this case the branches are predictable, so they perform better here. In general, though, the compiler has no idea whether that's the case, so it makes sense to insert a conditional move to avoid branch misprediction penalties.
        fmstephe 3y ago
        This is a very good answer. Thanks.
        mcv 3y ago
        But still, why the difference? With or without the "lol", the data dependence is identical, the predictability is identical. Yet with the "lol" it optimises differently. My best guess is that it's because of the added cost to the other branch: printing is expensive, so the compiler really doesn't want to do that, and would prefer to incorrectly predict not-printing than to incorrectly predict printing. If that tradeoff is what causes the difference in behaviour here, then I understand. But I don't like it, because it's ridiculous and we shouldn't have to con the compiler into doing what's right. If we're going to have to do this, I'd rather have the language provide a way for us to make explicit what we expect here.
    - tylerhou 3y ago
      "Conditional move is slow" is not the right takeaway. (In fact, conditional move can be much /faster/ than a branch in certain circumstances (e.g. hard-to-predict branches, like binary search).) The reason why this code is slow is because the conditional move is on the critical path for resolving a loop-carried data dependency. In other words, to execute the assignment `maxV = (v if (v > maxV) else maxV)`, the CPU has to test v > maxV. But in order to make this test, the CPU has to wait until the conditional move in the previous loop iteration finishes (to know what the new value of maxV is). So the overall execution time of the loop is worse, because each iteration depends on the previous iteration. (This is analogous to why traversing a linked list is slower than an array.) However, for the branch version, there is no data dependency on the previous loop, so the CPU really can (speculatively) execute multiple loop iterations in parallel. The the next loop iteration can start before the previous finishes. And with the large reorder buffers on modern CPUs, maybe up to 4-8 loop iterations can execute in parallel! On the other hand, conditional moves would be much faster for a loop like: bigs := make([]int, len(numsA)) for idx, v : range numsA { if numsA[idx] > numsB[idx] { bigs[idx] = numsA[idx] } else { bigs[idx] = numsB[idx] } } There is no loop-carried data dependency, so the conditional moves can resolve in parallel. The above code with a conditional move is likely to be much faster than code with an equivalent branch. (And even if branches are perfectly predicted, the conditional move version will only be slightly slower, not 2X slower.)
      - fmstephe 3y ago
        So would you say in the general case, i.e. the compiler obviously doesn't know how predictable a branch may be, using conditional move is a poor choice when writing to an address which is repeatedly written to inside the loop? What I am trying to get a feel for, is whether this represents a bad choice of assembly in general for this kind of loop.
        tylerhou 3y ago
        Nit: If a variable/object is only being written, that’s fine. There is a data dependency here because there is a read after a write. I don’t have enough compiler knowledge to be able to definitively answer your question, but I think it would be a reasonable heuristic to favor emitting a branch over a cmov in cases where the cmov participates in a loop-carried data dependency. The other advantage of emitting a branch is that branch prediction can better adapt to runtime conditions. Branch prediction tends to struggle when data is perfectly random, but most data is not. So in general it’s also reasonable for compilers to default to emitting branches.
tallanvor 3y ago
No, this is not true. There is branch prediction around the length of loops. This is a case where the processor is not able to accurately predict how long it needs to stay in the loop. The BLT instruction changes the prediction model, causing the processor to be more likely to assume the loop will continue. Honestly, though, worrying about this level of optimization is generally silly. If you're looping through an array often enough that optimizing the code this way is worth your time, you should use a data structure that automatically maintains the max (and min) values for fast retrieval.

by trolan 3y ago

dataflow 3y ago

> The processor only does this prediction with conditional branches This sounds... wrong? Unless ARM64 is designed in an absurd way? I'd love to see the full disassembly; something seems funny here. If it was x86 I would say it's a conditional move causing this, but I don't know what's going on on ARM.

fmstephe 3y ago
It's interesting that you say conditional move here. I am confused by this behaviour, and although I definitely don't know what the answer is here; the non-lol version does have a CSEL (https://developer.arm.com/documentation/dui0802/b/CSEL https://developer.arm.com/documentation/dui0802/b/CSEL) which is totally missing from the lol version. Non-lol https://godbolt.org/z/ds1raTYc9 https://godbolt.org/z/ds1raTYc9 lol https://godbolt.org/z/c3afrb6bG https://godbolt.org/z/c3afrb6bG
- dataflow 3y ago
  Ah there you go, that's the conditional move on ARM. Yeah, those are slow.
  - fmstephe 3y ago
    Yeah, after reading the blog post I felt reasonably confident that this is a compiler bug (in terms of perf). Hopefully someone with a deeper understanding can verify or discredit this here. I'm curious to see the fix for this (I assume it will generate a fix).
    - dataflow 3y ago
      Confused, why do you think this is a compiler bug? You should definitely expect a predictable branch to be faster than a conditional move, and the insertion of a conditional move in the original is totally sensible (albeit not intuitive if you haven't seen it).
      - littlestymaar 3y ago
        > and the insertion of a conditional move in the original is totally sensible (albeit not intuitive if you haven't seen it). Would you mind expanding? If the conditional move isn't needed, and given that it's costly in terms of perfs, how is that “totally sensible” to have one here?
        dataflow 3y ago
        The performance of branches is data-dependent. The performance of conditional moves is data-independent. In this case the branches are predictable, so they perform better here. In general, though, the compiler has no idea whether that's the case, so it makes sense to insert a conditional move to avoid branch misprediction penalties.
        fmstephe 3y ago
        This is a very good answer. Thanks.
        mcv 3y ago
        But still, why the difference? With or without the "lol", the data dependence is identical, the predictability is identical. Yet with the "lol" it optimises differently. My best guess is that it's because of the added cost to the other branch: printing is expensive, so the compiler really doesn't want to do that, and would prefer to incorrectly predict not-printing than to incorrectly predict printing. If that tradeoff is what causes the difference in behaviour here, then I understand. But I don't like it, because it's ridiculous and we shouldn't have to con the compiler into doing what's right. If we're going to have to do this, I'd rather have the language provide a way for us to make explicit what we expect here.
  - tylerhou 3y ago
    "Conditional move is slow" is not the right takeaway. (In fact, conditional move can be much /faster/ than a branch in certain circumstances (e.g. hard-to-predict branches, like binary search).) The reason why this code is slow is because the conditional move is on the critical path for resolving a loop-carried data dependency. In other words, to execute the assignment `maxV = (v if (v > maxV) else maxV)`, the CPU has to test v > maxV. But in order to make this test, the CPU has to wait until the conditional move in the previous loop iteration finishes (to know what the new value of maxV is). So the overall execution time of the loop is worse, because each iteration depends on the previous iteration. (This is analogous to why traversing a linked list is slower than an array.) However, for the branch version, there is no data dependency on the previous loop, so the CPU really can (speculatively) execute multiple loop iterations in parallel. The the next loop iteration can start before the previous finishes. And with the large reorder buffers on modern CPUs, maybe up to 4-8 loop iterations can execute in parallel! On the other hand, conditional moves would be much faster for a loop like: bigs := make([]int, len(numsA)) for idx, v : range numsA { if numsA[idx] > numsB[idx] { bigs[idx] = numsA[idx] } else { bigs[idx] = numsB[idx] } } There is no loop-carried data dependency, so the conditional moves can resolve in parallel. The above code with a conditional move is likely to be much faster than code with an equivalent branch. (And even if branches are perfectly predicted, the conditional move version will only be slightly slower, not 2X slower.)
    - fmstephe 3y ago
      So would you say in the general case, i.e. the compiler obviously doesn't know how predictable a branch may be, using conditional move is a poor choice when writing to an address which is repeatedly written to inside the loop? What I am trying to get a feel for, is whether this represents a bad choice of assembly in general for this kind of loop.
      - tylerhou 3y ago
        Nit: If a variable/object is only being written, that’s fine. There is a data dependency here because there is a read after a write. I don’t have enough compiler knowledge to be able to definitively answer your question, but I think it would be a reasonable heuristic to favor emitting a branch over a cmov in cases where the cmov participates in a loop-carried data dependency. The other advantage of emitting a branch is that branch prediction can better adapt to runtime conditions. Branch prediction tends to struggle when data is perfectly random, but most data is not. So in general it’s also reasonable for compilers to default to emitting branches.

tallanvor 3y ago

No, this is not true. There is branch prediction around the length of loops. This is a case where the processor is not able to accurately predict how long it needs to stay in the loop. The BLT instruction changes the prediction model, causing the processor to be more likely to assume the loop will continue. Honestly, though, worrying about this level of optimization is generally silly. If you're looping through an array often enough that optimizing the code this way is worth your time, you should use a data structure that automatically maintains the max (and min) values for fast retrieval.