I try to find articles, books or anything about programming without jumps (x86 arch). I know that generally it is impossible but I try to avoid jumps but gcc even with inline func uses jumps many times. Coding only in Assembly is some sort of solution, but writing equivalent of 1000 lines in C is like hell party to my eyes..
Unless your jumps are really random, branch prediction should eliminate most of overhead involved.
I would dedicate more effort to optimizing memory access patterns in order to improve locality and reduce cache misses. These days, memory latency is the major bottleneck to performance.
Another good direction is improving parallelism (using both vectorized SIMD instructions and, if possible, more than one core).
Optimize only performance critical code, and only once you really know it is performance critical. Do not try to optimize jumps only because you read they case a performance hit. Everything causes a performance hit, and the fastest possible code is the code which does nothing. There are other things much worse than jumps.
If you will show a particular example of a jump in the generated code, chance is there will be some way to avoid it, but it is more likely the code you will show will still contain more serious issues.
One particular way how to avoid branches is to use "conditional move" instructions. They can be used e.g. to compute max or min. If you allow the compiler to use SSE architecture, it assumes the CPU also supports CMOV/FCOMI/FCOMIP/FUCOMI/FUCOMIP instructions and will use them (beware: sometimes it may be tricky to make the compiler to do what you want, see e.g. this gamedev.net discussion).
I think you may mean branching. In C there are bit twiddling tricks to use to speed up certain operations
See bit hacks:
http://www-graphics.stanford.edu/~seander/bithacks.html
It is not impossible to code without jumps but it seems pointless to try.
In the end if you need to do something more than once then your choices are:
Loop unrolling (i.e. repeating the code instead of looping).
Somehow get the instruction pointer to visit the same code more than once.
The first approach requiers knowing the number of iterations in advance and doesn't scale and the second involves some sort of jump.
Not knowing what your code looks like, it's hard to give any advice. But I will give it a try.
Before you start optimizing, run a profiling tool to locate the problem areas. After optimizing, run the profiling tool again to see if you actually made it faster.
It's hard to actually remove branches, but you can minimize them by doing loop unrolling.
Someone mentioned conditional move instructions, there's plenty of conditional instructions on the ARM architecture, but if they're not executed they will translate to a NOP and take one cycle each. Not sure how they work on x86. It might actually get slower then using a simple branch depending on how long the pipeline is.
There's a lot of other optimizing tricks you could try before removing branches.
Related
I need a better way of profiling numerical code. Assume that I'm using GCC in Cygwin on 64 bit x86 and that I'm not going to purchase a commercial tool.
The situation is this. I have a single function running in one thread. There are no code dependencies or I/O beyond memory accesses, with the possible exception of some math libraries linked in. But for the most part, it's all table look-ups, index calculations, and numerical processing. I've cache aligned all arrays on the heap and stack. Due to the complexity of the algorithm(s), loop unrolling, and long macros, the assembly listing can become quite lengthy -- thousands of instructions.
I have been resorting to using either, the tic/toc timer in Matlab, the time utility in the bash shell, or using the time stamp counter (rdtsc) directly around the function. The problem is this: the variance (which might be as much as 20% of the runtime) of the timing is larger than the size of the improvements I'm making, so I have no way of knowing if the code is better or worse after a change. You might think then it's time to give up. But I would disagree. If you are persistent, many incremental improvements can lead to a two or three times performance increase.
One problem I have had multiple times that is particularly maddening is that I make a change and the performance seems to improve consistently by say 20%. The next day, the gain is lost. Now it's possible I made what I thought was an innocuous change to the code and then completely forgot about it. But I'm wondering if it's possible something else is going on. Like maybe GCC doesn't yield a 100% deterministic output as I believe it does. Or maybe it's something simpler, like the OS moved my process to a busier core.
I have considered the following, but I don't know if any of these ideas are feasible or make any sense. If yes, I would like explicit instructions on how to implement a solution. The goal is to minimize the variance of the runtime so I can meaningfully compare different versions of optimized code.
Dedicate a core of my processor to run only my routine.
Direct control over the cache(s) (load it up or clear it out).
Ensuring my dll or executable always loads to the same place in memory. My thinking here is that maybe the set-associativity of the cache interacts with the code/data location in RAM to alter performance on each run.
Some kind of cycle accurate emulator tool (not commercial).
Is it possible to have a degree of control over context switches? Or does it even matter? My thinking is the timing of the context switches is causing variability, maybe by causing the pipeline to be flushed at an inopportune time.
In the past I have had success on RISC architectures by counting instructions in the assembly listing. This only works, of course, if the number of instructions is small. Some compilers (like TI's Code Composer for the C67x) will give you a detailed analysis of how it's keeping the ALU busy.
I haven't found the assembly listings produced by GCC/GAS to be particularly informative. With full optimization on, code is moved all over the place. There can be multiple location directives for a single block of code dispersed about the assembly listing. Further, even if I could understand how the assembly maps back into my original code, I'm not sure there's much correlation between instruction count and performance on a modern x86 machine anyway.
I made a weak attempt at using gcov for line-by-line profiling, but due to an incompatibility between the version of GCC I built and the MinGW compiler, it wouldn't work.
One last thing you can do is average over many, many trial runs, but that takes forever.
EDIT (RE: Call Stack Sampling)
The first question I have is, practically, how do I do this? In one of your power point slides, you showed using Visual Studio to pause the program. What I have is a DLL compiled by GCC with full optimizations in Cygwin. This is then called by a mex DLL compiled by Matlab using the VS2013 compiler.
The reason I use Matlab is because I can easily experiment with different parameters and visualize the results without having to write or compile any low level code. Further, I can compare my optimized DLL to the high level Matlab code to ensure my optimizations have not broken anything.
The reason I use GCC is that I have a lot more experience with it than with Microsoft's compiler. I'm familiar with many flags and extensions. Further, Microsoft has been reluctant, at least in the past, to maintain and update the native C compiler (C99). Finally, I've seen GCC kick the pants off commercial compilers, and I've looked at the assembly listing to see how it's actually done. So I have some intuition of how the compiler actually thinks.
Now, with regards to making guesses about what to fix. This isn't really the issue; it's more like making guesses about how to fix it. In this example, as is often the case in numerical algorithms, there is really no I/O (excluding memory). There are no function calls. There's virtually no abstraction at all. It's like I'm sitting on top of a piece of saran wrap. I can see the computer architecture below, and there's really nothing in-between. If I re-rolled up all the loops, I could probably fit the code on about one page or so, and I could almost count the resultant assembly instructions. Then I could do a rough comparison to the theoretical number of operations a single core is capable of doing to see how close to optimal I am. The trouble then is I lose the auto-vectorization and instruction level parallelization I got from unrolling. Unrolled, the assembly listing is too long to analyze in this way.
The point is that there really isn't much to this code. However, due to the incredible complexity of the compiler and modern computer architecture, there is quite a bit of optimization to be had even at this level. But I don't know how small changes are going to affect the output of the compiled code. Let me give a couple of examples.
This first one is somewhat vague, but I'm sure I've seen it happen a few times. You make a small change and get a 10% improvement. You make another small change and get another 10% improvement. You undo the first change and get another 10% improvement. Huh? Compiler optimizations are neither linear, nor monotonic. It's possible, the second change required an additional register, which broke the first change by forcing the compiler to alter its register allocation algorithm. Maybe, the second optimization somehow occluded the compiler's ability to do optimizations which was fixed by undoing the first optimization. Who knows. Unless the compiler is introspective enough to dump its full analysis at every level of abstraction, you'll never really know how you ended up with the final assembly.
Here is a more specific example which happened to me recently. I was hand coding AVX intrinsics to speed up a filter operation. I thought I could unroll the outer loop to increase instruction level parallelism. So I did, and the result was that the code was twice as slow. What happened was there were not enough 256 bit registers to go around. So the compiler was temporarily saving results on the stack, which killed performance.
As I was alluding to in this post, which you commented on, it's best to tell the compiler what you want, but unfortunately, you often have no choice and are forced to hand tweak optimizations, usually via guess and check.
So I guess my question would be, in these scenarios (the code is effectively small until unrolled, each incremental performance change is small, and you're working at a very low level of abstraction), would it be better to have "precision of timing" or is call stack sampling better at telling me which code is superior?
I've faced a similar problem some time ago but that was on Linux which made it easier to tweak. Basically the noise introduced by OS (called "OS jitter") was as big as 5-10% in SPEC2000 tests (I can imagine it's much higher on Windows due to much bigger amount of bloatware).
I was able to bring deviation to below 1% by combination of the following:
disable dynamic frequency scaling (better do this both in BIOS and in Linux kernel as not all kernel versions do this reliably)
disable memory prefetching and other fancy settings like "Turbo boost", etc. (BIOS, again)
disable hyperthreading
enable high-performance process scheduler in kernel
bind process to core to prevent thread migration (use core 0 - for some reason it was more reliable on my kernel, go figure)
boot to single-user mode (in which no services are running) - this isn't as easy in modern systemd-based distros
disable ASLR
disable network
drop OS pagecache
There may be more to it but 1% noise was good enough for me.
I might put detailed instructions to github later today if you need them.
-- EDIT --
I've published my benchmarking script and instructions here.
Am I right that what you're doing is making an educated guess of what to fix, fixing it, and then trying to measure to see if it made any difference?
I do it a different way, which works especially well as the code gets large.
Rather than guess (which I certainly can) I let the program tell me how the time is spent, by using this method.
If the method tells me that roughly 30% is spent doing such-and-so, I can concentrate on finding a better way to do that.
Then I can run it and just time it.
I don't need a lot of precision.
If it's better, that's great.
If it's worse, I can undo the change.
If it's about the same, I can say "Oh well, maybe it didn't save much, but let's do it all again to find another problem,"
I need not worry.
If there's a way to speed up the program, this will pinpoint it.
And often the problem is not just a simple statement like "line or routine X spends Y% of the time", but "the reason it's doing that is Z in certain cases" and the actual fix may be elsewhere.
After fixing it, the process can be done again, because a different problem, which was small before, is now larger (as a percent, because the total has been reduced by fixing the first problem).
Repetition is the key, because each speedup factor multiplies all the previous, like compound interest.
When the program no longer points out things I can fix, I can be sure it is nearly optimal, or at least nobody else is likely to beat it.
And at no point in this process did I need to measure the time with much precision.
Afterwards, if I want to brag about it in a powerpoint, maybe I'll do multiple timings to get smaller standard error, but even then, what people really care about is the overall speedup factor, not the precision.
I am working on a (quite large) existing monothreaded C application. In this context I modified the application to perform some very few additional work consisting in incrementing a counter each time we call a special function (this function is called ~ 80.000 times). The application is compiled on an Ubuntu 12.04 running a 64 bits Linux kernel 3.2.0-31-generic with -O3 option.
Surprisingly the instrumented version of the code is running faster and I am investigating why.I measure execution time with clock_gettime(CLOCK_PROCESS_CPUTIME_ID) and to get representative results, I am reporting an average execution time value over 100 runs. Moreover, to avoid interference from outside world, I tried as much as possible to launch the application in a system without any other applications running (on a side note, because CLOCK_PROCESS_CPUTIME_ID returns process time and not wall clock time, other applications "should" in theory only affect cache and not directly the process execution time)
I was suspecting "instruction cache effects", maybe the instrumented code that is a little bit larger (few bytes) fits differently and better in the cache, is this hypothesis conceivable ? I tried to do some cache investigations with valegrind --tool=cachegrind but unfortunately, the instrumented version has (as it seems logical) more cache misses than the initial version.
Any hints on this subject and ideas that may help to find why instrumented code is running faster are welcomes (some GCC optimizations available in one case and not in the other, why ?, ...)
Since there are not many details in the question, I can only recommend some factors to consider while investigating the problem.
Very few additional work (such as incrementing a counter) might alter compiler's decision on whether to apply some optimizations or not. Compiler has not always enough information to make perfect choice. It may try to optimize for speed where bottleneck is code size. It may try to auto-vectorize computations when there is not too much data to process. Compiler may not know what kind of data is to be processed or what is the exact model of CPU, that will execute the code.
Incrementing a counter may increase size of some loop and prevent loop unrolling. This may decrease code size (and improve code locality, which is good for instruction or microcode caches or for loop buffer and allows CPU to fetch/decode instructions quickly).
Incrementing a counter may increase size of some function and prevent inlining. This also may decrease code size.
Incrementing a counter may prevent auto-vectorization, which again may decrease code size.
Even if this change does not affect compiler optimization, it may alter the way how the code is executed by CPU.
If you insert counter-incrementing code in place, full of branch targets, this may make branch targets less dense and improve branch prediction.
If you insert counter-incrementing code in front of some particular branch target, this may make branch target's address better aligned and make code fetch faster.
If you place counter-incrementing code after some data is written but before the same data is loaded again (and store-to-load forwarding did not work for some reason), the load operation may be completed earlier.
Insertion of counter-incrementing code may prevent two conflicting load attempts to the same bank in L1 data cache.
Insertion of counter-incrementing code may alter some CPU scheduler decision and make some execution port available just in time for some performance-critical instruction.
To investigate effects of compiler optimization, you can compare generated assembler code before and after addition of counter-incrementing code.
To investigate CPU effects, use a profiler allowing to inspect processor performance counters.
Just guessing from my experience with embedded compilers, Optimization tools in compilers look for recursive tasks. Perhaps the additional code forced the compiler to see something more recursive and it structured the machine code differently. Compilers do some weird things for optimization. In some languages (Perl I think?) a "not not" conditional is faster to execute than a "true" conditional. Does your debugging tool allow you to single step through a code/assembly comparison? This could add some insight as to what the compiler decided to do with the extra tasks.
I need to optimize some C code, which does lots of physics computations, using SIMD extensions on the SPE of the Cell Processor. Each vector operator can process 4 floats at the same time. So ideally I would expect a 4x speedup in the most optimistic case.
Do you think the use of vector operators could give bigger speedups?
Thanks
The best optimization occurs in rethinking the algorithm. Eliminate unnecessary steps. Find more a direct way of accomplishing the same result. Compute the solution in a domain more relevant to the problem.
For example, if the vector array is a list of n which are all on the same line, then it is sufficient to transform the end points only and interpolate the intermediate points.
It CAN give better speeds up than 4 times over straight floating point as the SIMD instructions could be less exact (Not so much as to give too many problems though) and so take fewer cycles to execute. It really depends.
Best plan is to learn as much about the processor you are optimising for as possible. You may find it can give you far better than 4x improvements. You may find out you can't. We can't say though without knowing more about the algorithm you are optimising and what CPU you are targetting.
On their own, no. But if the process of re-writing your algorithms to support them also happens to improve, say, cache locality or branching behaviour, then you could find unrelated speed-ups. However, this is true of any re-write...
This is entirely possible.
You can do more clever instruction-level micro optimizations than a compiler, if you know what you're doing.
Most SIMD instruction sets offers several powerful operations that don't have any equivalent in normal scalar FPU/ALU code (e.g. PAVG/PMIN etc. in SSE2). Even if these don't fit your problem exactly, you can often combine these instructions for great effect.
Not sure about Cell, but most SIMD instruction sets have features to optimize memory access, for example to prefetch data into cache. I've had very good results with these.
Now this isn't Cell or PPC at all, but a simple image convolution filter of mine got a 20x speedup (C vs. SSE2) on Atom, which is higher than the level of parallelity (16 pixels at a time).
It depends on the architecture.. For the moment I assume x86 architecture (aka SSE).
You can get factor four on tight loops easily. Just replace your existing math with SSE instruction and you're done.
You can even get a little more than that because if you use SSE you do the math in registers which are usually not used by the compiler. This frees up the general purpose register for other task such as loop control and address calculation. In short the code that surrounds the SSE instruction will be more compact and execute faster.
And then there is the option to hint the memory controller how you want to access the memory, e.g. if you want to store data in a way that it bypasses the cache or not. For bandwidth hungry algorithms that may give you some more extra speed ontop of that.
I have programmed an embedded software (using C of course) and now I'm considering ways to improve the running time of the system. The most important single module in my system is one very large nested for loop module.
That module consists of two nested for loops that loops max 122500 times. That's not very much yet, but the problem is that inside that nested for loop I have a function call to a function that is in another source file. That specific function consists mostly of two another nested for loops which loops always 22500 times. So now I have to make a function call 122500 times.
I have made that function that is to be called a lot lighter and shorter (yet still works as it should) and now I started to think that would it be faster to rip off that function call and write that process directly inside those first two for loops?
The processor in that system is ARM7TDMI and its frequency is 55MHz. The system itself isn't very time critical so it doesn't have to be real time capable. However the faster it can process its duties the better.
Also would it be also faster to use while loops instead of fors? And any piece of advice about how to improve the running time is appreciated.
-zaplec
TRY IT AND SEE!!
It'll almost certainly make a difference. Function call overhead isn't usually that much of an issue, but at over 100K repetitions it starts to add up.
...But whether or not it makes any real-world difference is something only you can answer, after trying it and timing the results.
As for for vs while... it shouldn't matter unless you actually change the behavior when changing the loop. If in doubt, make your compiler spit out assembler code for both and compare... or just change it and time it.
You need to be careful in the optimizations you make because you aren't always clear on which optimizations the compiler is making for you. Pre-optimization is a common mistake people make. Is it important that your code is readable and easily maintained or slightly faster? Like others have suggested, the best approach is to benchmark the different ways and see if there is a noticeable difference.
If you don't believe your compiler does much in the way of optimization I would look at some older concepts in optimizing C (searches on SO or google should provide some good links).
The ARM processor has an instruction pipeline (cache). When the processor encounters a branch (call) instruction, it must clear the pipeline and reload, thus wasting some time. One objective when optimizing for speed is to reduce the number of reloads to the instruction pipeline. This means reducing branch instructions.
As others have stated in SO, compile your code with optimization set for speed, and profile. I prefer to look at the assembly language listing as well (either printed from the compiler or displayed interwoven in the debugger). Use this as a baseline. If you can't profile, you can use assembly instruction counting as a rough estimate.
The next step is to reduce the number of branches; or the number times a branch is taken. Unrolling loops helps to reduce the number of times a branch is taken. Inlining helps reduce the number of branches. Before applying this fine-tuning techniques, review the design and code implementation to see if branches can be reduced. For example, reduce the number of "if" statements by using Boolean arithmetic or using Karnaugh Maps. My favorite is reducing requirements and eliminating code that doesn't need to be executed.
In the code implementation, move code that doesn't change outside of the for or while loops. Some loops may be reduce to equations (example, replacing a loop of additions with a multiplication). Also, reduce the quantity of iterations, by asking "does this loop really need to be executed this many times").
Another technique is to optimize for Data Oriented Design. Also check this reference.
Just remember to set a limit for optimizing. This is where you decide any more optimization is not generating any ROI or customer satisfaction. Also, apply optimizations in stages; which will allow you to have a deliverable when your manager asks for one.
Run a profiler on your code. If you are just guessing at where you are spending your time, you are probably wrong. A profiler will show what function is taking the most time and you can focus on that. You could be doing something in the function that takes longer than the function call itself. Did you look to see if you can change floating operations to integer, or integer math to shifts? You can spend a lot of time fiddling with things that don't make much difference. Run a profiler on your code and know for sure that the things you are changing will make a difference.
For function vs. inline, unfortunately there is no easy answer. I.e. it depends. See this FAQ. For "for" vs. "while", I wouldn't think there is any significant difference in performance.
In general, a function call should have more overhead than inlining. You really should profile however, as this can be affected quite a bit by your compiler (especially the compile/optimization settings). Some compilers will automatically inline code for example.
I'm looking to see what can a programmer do in C, that can determine the performance and/or the size of the generated object file.
For e.g,
1. Declaring simple get/set functions as inline may increase performance (at the cost of a larger footprint)
2. For loops that do not use the value of the loop variable itself, count down to zero instead of counting up to a certain value
etc.
It looks like compilers now have advanced to a level where "simple" tricks (like the two points above) are not required at all. Appropriate options during compilation do the job anyway. Heck, I also saw posts here on how compilers handle recursion - that was very interesting! So what are we left to do at a C level then? :)
My specific environment is: GCC 4.3.3 re-targeted for ARM architecture (v4). But responses on other compilers/processors are also welcome and will be munched upon.
PS: This approach of mine goes against the usual "code first!, then benchmark, and finally optimize" approach.
Edit: Just like it so happens, I found a similar post after posting the question: Should we still be optimizing "in the small"?
One thing I can think of that a compiler probably won't optimize is "cache-friendliness": If you're iterating over a two-dimensional array in row-major order, say, make sure your inner loop runs across the column index to avoid cache thrashing. Having the inner loop run over the wrong index can cause a huge performance hit.
This applies to all programming languages, but if you're programming in C, performance is probably critical to you, so it's especially relevant.
"Always" know the time and space complexity of your algorithms. The compiler will never be able to do that job as well as you can. :)
Compilers these days still aren't very good at vectorizing your code so you'll still want to do the SIMD implementation of most algorithms yourself.
Choosing the right datastructures for your exact problem can dramatically increase performance (I've seen cases where moving from a Kd-tree to a BVH would do that, in that specific case).
Compilers might pad some structs/ variables to fit into the cache but other cache optimizations such as the locality of your data are still up to you.
Compilers still don't automatically make your code multithreaded and using openmp, in my experience, doesn't really help much. (You really have to understand openmp anyway to dramatically increase performance). So currently, you're on your own doing multithreading.
To add to what Martin says above about cache-friendliness:
reordering your structures such that fields which are commonly accessed together are in the same cache line can help (for instance by loading just one cache line rather than two.) You are essentially increasing the density of useful data in your data cache by doing this. There is a linux tool which can help you in doing this: dwarves 1. http://www.linuxinsight.com/files/ols2007/melo-reprint.pdf
you can use a similar strategy for increasing density of your code. In gcc you can mark hot and cold branches using likely/unlikely tags. That enables gcc to keep the cold branches separately which helps in increasing the icache density.
And now for something completely different:
for fields that might be accessed (read and written) across CPUs, the opposite strategy makes sense. The trouble is that for coherence purposes only one CPU can be allowed to write to the same address (in reality the same cacheline.) This can lead to a condition called cache-line ping pong. This is pretty bad and could be worse if that cache-line contains other unrelated data. Here, padding this contended data to a cache-line length makes sense.
Note: these clearly are micro-optimizations, to be done only at later stages when you are trying to wring the last bits of performance from your code.
PreComputation where possible... (sorry but its not always possible... I did extensive precomputation on my chess engine.) Store those results in memory, keeping cache in mind.. the bigger the size of precomputation data in memory the lesser is the chance of doing a cache hit. Since most of recent hardware is multicore you can design your application to target it.
if you are using several big arrays make sure you group them close to each other on where they would be used, boosting cache hits
Many people are not aware of this: Define an inline label (varies by compiler) which means inline, in its intent - many compilers place the keyword in an entirely different context from the original meaning. There are also ways to increase the inline size limits, before the compiler begins popping trivial things out of line. Human directed inlining can produce much faster code (compilers are often conservative, or do not account for enough of the program), but you need to learn to use it correctly, because it can (easily) be counterproductive. And yes, this absolutely applies to code size as well as speed.