I saw a sentence in a paper "Transforming branches into data dependencies to avoid mispredicted branches." (Page 6)
I wonder how to change the code from branches into data dependencies.
This is the paper: http://www.adms-conf.org/p1-SCHLEGEL.pdf
Update: How to transform branches in the binary search?
The basic idea (I would presume) would be to change something like:
if (a>b)
return "A is greater than B";
else
return "A is less than or equal to B";
into:
static char const *strings[] = {
"A is less than or equal to B",
"A is greater than B"
};
return strings[a>b];
For branches in a binary search, let's consider the basic idea of the "normal" binary search, which typically looks (at least vaguely) like this:
while (left < right) {
middle = (left + right)/2;
if (search_for < array[middle])
right = middle;
else
left = middle;
}
We could get rid of most of the branching here in pretty much the same way:
size_t positions[] = {left, right};
while (left < right) {
size_t middle = (left + right)/2;
positions[search_for >= array[middle]] = middle;
}
[For general purpose code use left + (right-left)/2 instead of (left+right)/2.]
We do still have the branching for the loop itself, of course, but we're generally a lot less concerned about that--that branch is extremely amenable to prediction, so even if we did eliminate it, doing so would gain little as a rule.
Removing branches is not always optimal, even (especially) with simple binary conditions like this. I have looked into removing branches in a similar manner in various circumstances before. After running into an instance where code with a conditional branch in a loop ran faster than equivalent, branch-less code, I did some research on processor execution strategies.
I knew that the ARM instruction set had conditional instructions, which could make a conditional branch faster than the kind of branch-less code in the paper, but I was working on an intel (and the branch was not one that cmove could take care of). It turns out that modern CPUs, including intel, will sometimes turn an ordinary instruction into a conditional one: they use eager execution if the two end points of the branch are sufficiently short. That is, the CPU will put both possible paths in the pipeline and execute them both and only keep the correct result once the condition is known. This avoids the possibility of a mis-prediction without having to index an array. See https://en.wikipedia.org/wiki/Speculative_execution#Variants for more information.
It takes a lot of detailed knowledge of a processor to write optimal code for it, even in assembly. This makes it possible for "naive" implementations to sometimes be faster than hand-optimized ones that overlook certain CPU characteristics. The same thing can happen with optimizing compilers: sometimes writing more complicated "optimized" code breaks some of the compiler's optimizations and results in a slower executable than a naive implementation that the compiler can optimize fully.
When in doubt and performance is critical and there is time, it is usually best to try it both ways and see which is faster!
Related
I am trying to learn recursion. for the starting problem of it , which is calculating factorial of a number I have accomplished it using two methods.
the first one being the normal usual approach.
the second one i have tried to do something different.
in the second one i return the value of n at the end rather than getting the starting value as in the first one which uses backtracking.
my question is that does my approach has any advantages over backtracking?
if asked to chose which one would be a better solution?
//first one is ,
ll factorial(int n)
{
if(n==1)
return 1 ;
return n*factorial(n-1) ;
}
int main()
{
factorial(25) ;
return 0 ;
}
// second one is ,
ll fact(ll i,ll n)
{
if(i==0)return n ;
n=(n*i)
i--;
n=fact(i,n);
}
int main()
{
int n ;
cin>>n ;
cout<<fact(n,1) ;
return 0 ;
}
// ll is long long int
First of all I want to point out that premature optimization at the expense of readibility is almost always a mistake, especially when the need for optimization comes from intuition and not from measurements.
"We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%" - Donald Knuth
But let's say we care about the 3% in this case, because all our program ever does is compute lots of factorials. To keep it short: You will never be smarter than the compiler.
If this seems slightly crazy then this definitely applies to you and you should stop thinking about 'micromanaging/optimizing your code'. If you are a very skilled C++ programmer this will still apply to you in most cases, but you will recognize the opportunities to help your compiler out.
To back this up with some fact, we can compile the code (with automatic optimization) and (roughly) compare the assembly output. I will use the wonderful website godbolt.org
Don't be discouraged by the crazy assembler code, we don't need to understand it. But we can see that both methods
are basically the same length when compiled as assembler code
contain almost the same instructions
So to recap, readability should be your number one priority. In case a speed measurement shows that this one part of your code really is a big performance problem, really think about if you can make a change that structurally improves the algorithm (i.e. by decreasing the complexity). Otherwise your compiler will take care of it for you.
The following question is more related to design, rather than actual coding. I don't know if there's a technical term for such problem, so I'll proceed with an example.
I have some openCL code not optimized at all, and in the Kernel there's essentially a switch statement similar to the following
switch(const) {
case const_a : do_something_a(...); break;
case const_b : do_something_b(....); break;
... //etc
}
I cannot write the actual statement since is quite long. As a simple example consider the following switch statement:
int a;
switch(input):
case 13 : {a = 3; break;}
case 1 : {a = 7; break;}
case 23 : {a = 1; break;}
default : {...}
The question is... would it be better to change such switch with an expression like
a = (input == 13)*3 + (input == 1)*7 + (input == 23)
?
If it's not, is it possible to make it more efficient anyway?
You can assume input only takes values in the set of cases of the switch statement.
You've discovered an interesting question that GPU compilers wrestle with. The general advice is try not to branch. Tricks to make that possible are splitting kernels up (as suggested above) and preprocessor (program-time definitions). Research in GPU algorithm development basically works from this axiom.
Branching all over the place won't get great efficiency because of the inherent divergence (channel = work item within the SIMD thread/warp). Remember that all these channels must execute together. So in a switch where all are taking different paths everyone else goes along for the ride silently waiting for their "case" to execute. Now, if input is always always the same value, it can still be a win.
Another popular option is a table indirection.
kernel void foo(const int *t, ...)
...
a = tbl[input];
This case has a few problems too depending on hardware, inputs, and problem size.
Without more specific context, I can conjure up a case where any of these can run well or poorly.
Switching (or big if-then-else chains).
PROS: If all work items generally take the same path (input is mostly the same value), it's going to be efficient. You could also write an if-then-else chain putting the most common cases first. (On GPUs a switch is not necessarily as easy as an indirect jump since there are multiple work items and they may take different paths.)
CONS: Might generate lots of program code and could blow out the instruction cache. Branching all over the place can get a little costly depending on how many cases need to be evaluated. It might just be better to grind through the compute with the predicated code.
Predicated Code (Your (input == 13)*3 ... code).
PROS: This will probably generate smaller programs and stress the I$ less. (Lookup the OpenCL select function to see a more general approach for your case.)
CONS: We've basically punted and decided to evaluate every "case in the switch". If input is usually the same value, we're wasting time here.
Lookup-table based approaches (my example).
PROS: If the switch you are evaluating has a massive number of cases (branches), but can be indexed by integer you might be ahead to just use a lookup table. On some hardware this means a read from global memory (far far away). Other architectures have a dedicated constant cache, but I understand that a vector lookup will serialize (K cycles for each channel). So it might be only marginally better than the global memory table. However, the code table-lookup generated will be short (I$ friendly) and as the number of branches (case statements) grow this will win in the limit. This approach also deals well with uniform/scattered distributions of input's value.
CONS: The read from global memory (or serialized access from the constant cache) has a big latency even compared to branching. In some cases, to eliminate the extra memory traffic I've seen compilers convert lookup tables into if-then-else/switch chains. It's rare that we have 100 element case statements.
I am now inspired to go study this cutoff. :-)
This is a question about performance of code written in Scala.
Consider the following two code snippets, assume that x is some collection containing ~50 million elements:
def process(x: Traversable[T]) = {
processFirst x.head
x reduce processPair
processLast x.last
}
Versus something like this (assume for now we have some way to determine if we're operating on the first element versus the last element):
def isFirstElement[T](x: T) = ???
def isLastElement[T](x: T) = ???
def process(x: Traversable[T]) = {
x reduce {
(left, right) =>
if (isFirstElement(left)
processFirst(left)
else if (isLastElement(right))
processLast(right)
processPair(left, right)
}
}
Which approach is faster? and for ~50 million elements, how much faster?
It seems to me that the first example would be faster because there are fewer conditional checks occurring for all but the first and last elements. However for the latter example there is some argument to suggest that the JIT might be clever enough to optimize away those additional head/last conditional checks that would otherwise occur for all but the first/last elements.
Is the JIT clever enough to perform such operations? The obvious advantage of the latter approach is that all business can be placed in the same function body while in the latter case business must be partitioned into three separate function bodies invoked separately.
** EDIT **
Thanks for all the great responses. While I am leaving the second code snippet above to illustrate its incorrectness, I want to revise the first approach slightly to reflect better the problem I am attempting to solve:
// x is some iterator
def process(x: Iterator[T]) = {
if (x.hasNext)
{
var previous = x.next
var current = null
processFirst previous
while(x.hasNext)
{
current = x.next
processPair(previous, current)
previous = current
}
processLast previous
}
}
While there are no additional checks occurring in the body, there is an additional reference assignment that appears to be unavoidable (previous = current). This is also a much more imperative approach that relies on nullable mutable variables. Implementing this in a functional yet high performance manner would be another exercise for another question.
How does this code snippet stack-up against the last of the two examples above? (the single-iteration block approach containing all the branches). The other thing I realize is that the latter of the two examples is also broken on collections containing fewer than two elements.
If your underlying collection has an inexpensive head and last method (not true for a generic Traversable), and the reduction operations are relatively inexpensive, then the second way takes about 10% longer (maybe a little less) than the first on my machine. (You can use a var to get first, and you can keep updating a second far with the right argument to obtain last, and then do the final operation outside of the loop.)
If you have an expensive last (i.e. you have to traverse the whole collection), then the first operation takes about 10% longer (maybe a little more).
Mostly you shouldn't worry too much about it and instead worry more about correctness. For instance, in a 2-element list your second code has a bug (because there is an else instead of a separate test). In a 1-element list, the second code never calls reduce's lambda at all, so again fails to work.
This argues that you should do it the first way unless you're sure last is really expensive in your case.
Edit: if you switch to a manual reduce-like-operation using an iterator, you might be able to shave off up to about 40% of your time compared to the expensive-last case (e.g. list). For inexpensive last, probably not so much (up to ~20%). (I get these values when operating on lengths of strings, for example.)
First of all, note that, depending on the concrete implementation of Traversable, doing something like x.last may be really expensive. Like, more expensive than all the rest of what's going on here.
Second, I doubt the cost of conditionals themselves is going to be noticeable, even on a 50 million collection, but actually figuring out whether a given element is the first or the last, might again, depending on implementation, get pricey.
Third, JIT will not be able to optimize the conditionals away: if there was a way to do that, you would have been able to write your implementation without conditionals to begin with.
Finally, if you are at a point where it starts looking like an extra if statement might affect performance, you might consider switching to java or even "C". Don't get me wrong, I love scala, it is a great language, with lots of power and useful features, but being super-fast just isn't one of them.
I really tried to find something about this kind of operations but I don't find specific information about my question... It's simple: Are boolean operations slower than typical math operations in loops?
For example, this can be seen when working with some kind of sorting. The method will make an iteration and compare X with Y... But is this slower than a summatory or substraction loop?
Example:
Boolean comparisons
for(int i=1; i<Vector.Length; i++) if(Vector[i-1] < Vector[i])
Versus summation:
Double sum = 0;
for(int i=0; i<Vector.Length; i++) sum += Vector[i];
(Talking about big length loops)
Which is faster for the processor to complete?
Do booleans require more operations in order to return "true" or "false" ?
Short version
There is no correct answer because your question is not specific enough (the two examples of code you give don't achieve the same purpose).
If your question is:
Is bool isGreater = (a > b); slower or faster than int sum = a + b;?
Then the answer would be: It's about the same unless you're very very very very very concerned about how many cycles you spend, in which case it depends on your processor and you need to read its documentation.
If your question is:
Is the first example I gave going to iterate slower or faster than the second example?
Then the answer is: It's going to depend primarily on the values the array contains, but also on the compiler, the processor, and plenty of other factors.
Longer version
On most processors a boolean operation has no reason to significantly be slower or faster than an addition: both are basic instructions, even though comparison may take two of them (subtracting, then comparing to zero). The number of cycles it takes to decode the instruction depends on the processor and might be different, but a few cycles won't make a lot of difference unless you're in a critical loop.
In the example you give though, the if condition could potentially be harmful, because of instruction pipelining. Modern processors try very hard to guess what the next bunch of instructions are going to be so they can pre-fetch them and treat them in parallel. If there is branching, the processor doesn't know if it will have to execute the then or the else part, so it guesses based on the previous times.
If the result of your condition is the same most of the time, the processor will likely guess it right and this will go well. But if the result of the condition keeps changing, then the processor won't guess correctly. When such a branch misprediction happens, it means it can just throw away the content of the pipeline and do it all over again because it just realized it was moot. That. does. hurt.
You can try it yourself: measure the time it takes to run your loop over a million elements when they are of same, increasing, decreasing, alternating, or random value.
Which leads me to the conclusion: processors have become some seriously complex beasts and there is no golden answers, just rules of thumb, so you need to measure and profile. You can read what other people did measure though to get an idea of what you should or should not do.
Have fun experimenting. :)
Recently I realized I have been doing too much branching without caring the negative impact on performance it had, therefore I have made up my mind to attempt to learn all about not branching. And here is a more extreme case, in attempt to make the code to have as little branch as possible.
Hence for the code
if(expression)
A = C; //A and C have to be the same type here obviously
expression can be A == B, or Q<=B, it could be anything that resolve to true or false, or i would like to think of it in term of the result being 1 or 0 here
I have come up with this non branching version
A += (expression)*(C-A); //Edited with thanks
So my question would be, is this a good solution that maximize efficiency?
If yes why and if not why?
Depends on the compiler, instruction set, optimizer, etc. When you use a boolean expression as an int value, e.g., (A == B) * C, the compiler has to do the compare, and the set some register to 0 or 1 based on the result. Some instruction sets might not have any way to do that other than branching. Generally speaking, it's better to write simple, straightforward code and let the optimizer figure it out, or find a different algorithm that branches less.
Jeez, no, don't do that!
Anyone who "penalize[s] [you] a lot for branching" would hopefully send you packing for using something that awful.
How is it awful, let me count the ways:
There's no guarantee you can multiply a quantity (e.g., C) by a boolean value (e.g., (A==B) yields true or false). Some languages will, some won't.
Anyone casually reading it is going observe a calculation, not an assignment statement.
You're replacing a comparison, and a conditional branch with two comparisons, two multiplications, a subtraction, and an addition. Seriously non-optimal.
It only works for integral numeric quantities. Try this with a wide variety of floating point numbers, or with an object, and if you're really lucky it will be rejected by the compiler/interpreter/whatever.
You should only ever consider doing this if you had analyzed the runtime properties of the program and determined that there is a frequent branch misprediction here, and that this is causing an actual performance problem. It makes the code much less clear, and its not obvious that it would be any faster in general (this is something you would also have to measure, under the circumstances you are interested in).
After doing research, I came to the conclusion that when there are bottleneck, it would be good to include timed profiler, as these kind of codes are usually not portable and are mainly used for optimization.
An exact example I had after reading the following question below
Why is it faster to process a sorted array than an unsorted array?
I tested my code on C++ using that, that my implementation was actually slower due to the extra arithmetics.
HOWEVER!
For this case below
if(expression) //branched version
A += C;
//OR
A += (expression)*(C); //non-branching version
The timing was as of such.
Branched Sorted list was approximately 2seconds.
Branched unsorted list was aproximately 10 seconds.
My implementation (whether sorted or unsorted) are both 3seconds.
This goes to show that in an unsorted area of bottleneck, when we have a trivial branching that can be simply replaced by a single multiplication.
It is probably more worthwhile to consider the implementation that I have suggested.
** Once again it is mainly for the areas that is deemed as the bottleneck **