Key concepts to learn in Assembly - theory

I am a firm believer in the idea that one of the most important things you get from learning a new language is not how to use a new language, but the knowledge of concepts that you get from it. I am not asking how important or useful you think Assembly is, nor do I care if I never use it in any of my real projects.
What I want to know is what concepts of Assembly do you think are most important for any general programmer to know? It doesn't have to be directly related to Assembly - it can also be something that you feel the typical programmer who spends all their time in higher-level languages would not understand or takes for granted, such as the CPU cache.

Register allocation and management
Assembly gives you a very good idea of how many variables (machine-word-sized integers) the CPU can juggle simultaneously. If you can break your loops down so that they involve only a few temporary variables, they'll all fit in registers. If not, your loop will run slowly as things get swapped out to memory.
This has really helped me with my C coding. I try to make all loops tight and simple, with as little spaghetti as possible.
x86 is dumb
Learning several assembly languages has made me realize how lame the x86 instruction set is. Variable-length instructions? Hard-to-predict timing? Non-orthogonal addressing modes? Ugh.
The world would be better if we all ran MIPS, I think, or even ARM or PowerPC :-) Or rather, if Intel/AMD took their semiconductor expertise and used it to make multi-core, ultra-fast, ultra-cheap MIPS processors instead of x86 processors with all of those redeeming qualities.

I think assembly language can teach you lots of little things, as well as a few big concepts.
I'll list a few things I can think of here, but there is no substitute for going and learning and using both x86 and a RISC instruction set.
You probably think that integer operations are fastest. If you want to find an integer square root of an integer (i.e. floor(sqrt(i))) it's best to use an integer-only approximation routine, right?
Nah. The math coprocessor (on x86 that is) has a fsqrt instruction. Converting to float, taking the square root, and converting to int again is faster than an all-integers algorithm.
Then there are things like accessing memory that you can follow, but not properly apprecatiate, until you've delved into assembly. Say you had a linked list, and the first element in the list contains a variable that you will need to access frequently. The list is reordered rarely. Well, each time you need to access that variable, you need to load the pointer to the first element in the list, then using that, load the variable (assuming you can't keep the address of the variable in a register between uses). If you instead stored the variable outside of the list, you only need a single load operation.
Of course saving a couple of cycles here and there is usually not important these days. But if you plan on writing code that needs to be fast, this kind of knowledge can be applied both with inline assembly and generally in other languages.
How about calling conventions? (Some assemblers take care of this for you - Real Programmers don't use those.) Does the caller or callee clean up the stack? Do you even use the stack? You can pass values in registers - but due to the funny x86 instruction set, it's better to pass certain things in certain registers. And which registers will be preserved? One thing C compilers can't really optimise by themselves is calls.
There are little tricks like PUSHing a return address and then JMPing into a procedure; when the procedure returns it will go to the PUSHed address. This departure from the usual way of thinking about function calls is another one of those "states of enlightenment". If you were ever to design a programming language with innovative features, you ought to know about funny things that the hardware is capable of.
A knowledge of assembly language teaches you architecture-specific things about computer security. How you might exploit buffer overflows, or break into kernel mode, and how to prevent such attacks.
Then there's the ubercoolness of self-modifying code, and as a related issue, mechanisms for things such as relocations and applying patches to code (this needs investigation of machine code as well).
But all these things need the right sort of mind. If you're the sort of person who can put
while(x--)
{
...
}
to good use once you learn what it does, but would find it difficult to work out what it does by yourself, then assembly language is probably a waste of your time.

It's good to know assembly language in order to gain a better appreciation for how the computer works "under the hood," and it helps when you are debugging something and all the debugger can give you is an assembly code listing, which at least gives you fighting chance of figuring out what the problem might be. However, trying to apply low-level knowledge to high-level programming languages, such as trying to take advantage of how the CPU caches instructions and then writing wonky high-level code to force the compiler to produce super-efficient machine code, is probably a sign that you are trying to micro-optimize. In most cases, it's usually better not to try to outsmart the compiler, unless you need the performance gain, in which case, you might as well write those bits in assembly anyway.
So, it's good to know assembly for the sake of better understanding of how things work, but the knowledge gained is not necessarily directly applicable to how you write code in high-level languages. On that note, however, I found that learning how function calls work at the assembly-code level (learning about the stack and related registers, learning about how parameters are passed on the stack, learning how automatic storage works, etc.) made it a lot easier to understand problems I had in higher-level code, such as "out of stack space" errors and "invalid calling convention" errors.

The most important concept is SIMD, and creative use of it. Proper use of SIMD can give enormous performance benefits in a massive variety of applications ranging from everything from string processing to video manipulation to matrix math. This is where you can get over 10x performance boosts over pure C code--this is why assembly is still useful beyond mere debugging.
Some examples from the project I work on (all numbers are clock cycle counts on a Core 2):
Inverse 8x8 H.264 DCT (frequency transform):
c: 1332
mmx: 187
sse2: 127
8x8 Chroma motion compensation (bilinear interpolation filter):
c: 639
mmx: 144
sse2: 110
ssse3: 79
4 16x16 Sum of Absolute Difference operations (motion search):
c: 3948
mmx: 278
sse2: 231
ssse3: 215
(yes, that's right--over 18x faster than C!)
Mean squared error of a 16x16 block:
c: 1013
mmx: 193
sse2: 131
Variance of a 16x16 block:
c: 783
mmx: 171
sse2: 106

Memory, registers, jumps, loops, shifts and the various operations one can perform in assembler. I don't miss the days of debugging my assembly language class programs - they were painful! - but it certainly gave me a good foundation.
We forget (or never knew, perhaps) that all this fancy-pants stuff that we use today (and that I love!) boils down to all this stuff in the end.
Now, we can certainly have a productive and lucrative career without knowing assembler, but I think these concepts are good to know.

I would say that learning recursion and loops in assembly has taught me alot. It made me understand the underlying concept of how the compiler/interpreter of the language i'm using pushes things onto a stack, and pops them off as it needs them. I also learned how to exploit the infamous stack overflow. (which is still surprisingly easy in C with some get- and put- commands).
Other than using asm in every-day situations, i don't think that i would use any of the concepts assembly taught me.

I would say that addressing modes are extremely important.
My alma mater took that to an extreme, and because x86 didn't have enough of them, we studied everything on a simulator of PDP11 that must have had at least 7 of them that I remember. In retrospect, that was a good choice.

timing
fast execution:
parallel processing
simple instructions
lookup tables
branch prediction, pipelining
fast to slow access to storage:
registers
cache, and various levels of cache
memory heap and stack
virtual memory
external I/O

Nowadays, x86 asm is not a direct line to the guts of the CPU, but more of an API. The assembler opcodes you write are themselves are compiled into a completely different instruction-set, rearranged, rewritten, fixed-up and generally mangled beyond recognition.
So it's not like learning assembler gives you a fundamental insight into what's going on inside the CPU. IMHO, more important than learning assembler is to get a good understanding of how the target CPU and the memory hierarchy works.
This series of articles covers the latter topic pretty thoroughly.

Related

what are the steps/strategy to analyze and improve performance of an embedded system

I will break down this question in to sub questions. I am confused if I should ask them separately or in one question. So I will just stick to one SO question.
What are generally the steps to analyze and improve performance of C applications?
Do these steps change if I am developing for an embedded system?
What tools are out there which can help me?
Recently I have been given a task to improve the performance of our product on ARM11 platform. I am relatively new to this field of embedded systems and need gurus here on SO to help me out.
simply changing compilers can improve your C performance for the same source code by many times over. GCC has not necessarily gotten better for performance over the years, for some programs gcc 3.x produces much tighter code than 4.x. Back when I had access to the tools, ARMs compiler produced significantly better code than gcc. As much as 3 or 4 times faster. LLVM has caught up to GCC 4.x and I suspect will pass gcc by in terms of performance and overall use for cross compiling embedded code. Try different versions of gcc, 3.x and 4.x if you are using gcc. Metaware's compiler and arms adt ran circles around gcc3.x, gcc3.x will give gcc4.x a run for its money with arm code, for thumb code gcc4.x is better and for thumb2 (which doesnt apply to you) gcc4.x also better. Remember I have not said a word about changing a single line of code (yet).
LLVM is capable of full program optimization in addition to infinitely more tuning knobs than gcc. Despite that the code generated (ver 27) is only just catching up to the current gcc 4.x in terms of performance for the few programs I tried. And I didnt try the n factoral number of optimization combinations (optimize on the compile step, different options for each file, or combine two files or three files or all files and optimize those bundles, my theory is do no optimization on the C to bc steps, link all the bc together then do a single optimization pass on the whole program, the allow the default optimization when llc takes it to the target).
By the same token simply knowing your compiler and the optimizations can greatly improve the performance of the code without having to change any of it. You have an ARM11 arr you compiling for arm11 or generic arm? You can gain a few to a dozen percent by telling the compiler specifically which architecture/family (armv6 for example) over the generic armv4 (ARM7) that is often chosen as the default. Knowing to use -O2 or -O3 if you are brave.
It is often not the case but switching to thumb mode can improve performance for specific platforms. Doesnt apply to you but the gameboy advance is a perfect example, loaded with non-zero wait state 16 bit busses. Thumb has a handful of a percent overhead because it takes more instructions to do the same thing, but by increasing the fetch times, and taking advantage of some of the sequential read features of the gba thumb code can run significantly faster than arm code for the same source code.
having an arm11 you probably have an L1 and maybe L2 cache, are they on? Are they configured? Do you have an mmu and is your heavy use memory cached? or are you running zero wait state memory and dont need a cache and should turn it off? In addition to not realizing that you can take the same source code and make it run many times faster by changing compilers or options, folks often dont realize that when you use a cache simply adding a single up to a few nops in your startup code (as a trick to adjust where code lands in memory by one, two, a few words) you can change your codes execution speed by as much as 10 to 20 percent. Where those cache line reads hit in heavily used functions/loops makes a big difference. Even saving one cache line read by adjusting where the code lands is noticeable (cutting it from 3 to 2 or 2 to 1 for example).
Knowing your architecture, both the processor and your memory environment is where the tuning if any would start. Most C libraries if you are high level enough to use one (I often dont use a C library as I run without an operating system and with very limited resources) both in their C code and sometimes add some assembler to make bottleneck routines like memcpy, much faster. If your programs are operating on aligned 32 or even better 64 bit addresses, and you adjust even if it means using a handful of bytes more memory for every structure/array/memcpy to be an integral multiple of 32 bits or 64 bits you will see noticeable improvements (if your code uses structs or copies data in other ways). In addition to getting your structures (if you use them, I certainly dont with embedded code) size aligned, even if you waste memory, getting elements aligned, consider using 32 bit integers for every element instead of bytes or halfwords. Depending on your memory system this can help (it can hurt too btw). As with the GBA example above looking at specific functions that either by profiling or intuition you know are not being implemented in a manner that takes advantage of your processor or platform or libraries you may want to turn to assembler either from scratch or compiling from C initially then disassembling and hand tuning. Memcpy is a good example you may know your systems memory performance and may chose to create your own memcpy specifically for aligned data, copying 64 or 128 or more bits per instruction.
Likewise mixing global and local variables can make a noticeable performance difference. Traditionally folks are told never to use globals, but in embedded this isnt necessarily true, depends on how deeply embedded and how much tuning and speed and other factors you are interested in. This is a touchy subject and I may get flamed for it, so I will leave it at that.
The compiler has to burn and evict registers in order to make function calls, plus if you use local variables a stack frame may be required, so function calls are expensive, but at the same time, depending on the code within a function that has now grown in size by avoiding functions, you may create the problem you were trying to avoid, evicting registers to re-use them. Even a single line of C code can make the difference between all the variables in a function fits in registers to having to start evicting a bunch of registers. For functions or segments of code where you know you need some performance gain compile and disassemble (and look at register usage, how often it fetches memory or writes to memory). You can and will find places where you need to take a well used loop and make it its own function even though the function call has a penalty because by doing that the compiler can better optimize the loop and not evict/reuse registers and you get an overall net gain. Even a single extra instruction in a loop that goes around hundreds of times is a measurable performance hit.
Hopefully you already know to absolutely not compile for debug, turn all of the compile for debug options off. You may already know that code compile for debug that runs without bugs doesnt mean it is debugged, compiling for debug and using debuggers hide bugs leaving them as time bombs in your code for your final compile for release. Learn to always compile for release and test with the release version both for performance and finding bugs in your code.
Most instruction sets do not have a divide function. Avoid using divides or modulo in your code as much as humanly possible they are performance killers. Naturally this is not the case for powers of two, to save the compiler and to mentally avoid divides and modulos try to use shifts and ands. Multplies are easier and more often found in instruction sets, but are still costly. This is a good case to write assembler to do your multiplies instead of letting the C copiler do it. The arm multiply is a 32bit * 32bit = 32 bit so to do accurate math without overflowing there has to be extra C code wrapped around the multiply, if you already know you wont overflow, burn the registers for a function call and do the multiply in assembler (for the arm).
Likewise most instruction sets do not have a floating point unit, with yours you might, even so avoid float if at all possible. If you have to use float that is a whole other pandora's box of performance issues. Most folks dont see the performance problems with code as simple as this:
float a,b;
...
a = b * 7.0;
The rest of the problem is not understanding floating point accuracy and how good or bad the C libraries are just trying to get your constants into floating point form. Again float is a whole other long discussion on performance problems.
I am a product of Michael Abrash (I actually have a print copy of zen of assembly language) and the bottom line is time your code. Come up with an accurate way to time the code, you may think you know where the bottlenecks are and you may think you know your architecture but trying different things even if you think they are wrong, and timing them you may find and eventually have to figure out the error in your thinking. Adding nops to start.S as a final tuning step is a good example of this, all the other work you have done for performance can be instantly erased by not having a good alignment with the cache, this also means re-arranging functions within your source code so that they land in different places in the binary image. I have seen 10 to 20 percent swings of speed increase and decrease as a result of cache line alignments.
Code Review:
What are good code review techniques ?
Static and dynamic analysis of the code.
Tools for static analysis: Sparrow, Prevent, Klockworks
Tools for dynamic analysis : Valgrind, purify
Gprof allows you to learn where your program spent its time and which functions called which other functions while it was executing.
Steps are same
Apart from what is listed is point 1, there are tools like memcheck etc.
There is a big list here based on platform
Phew!! Quite a big question!
What are generally the steps to
analyze and improve performance of C
applications?
As well as other static code analysers mentioned here there is a fairly cheap version called PC-Lint which has been around for ages. Sometimes throws up lots of errors and warnings for one error but by the end of it you'll be happy and know waaaaay more about C/C++ because of it.
With all code analysers some of the issues may be more structural to the code so best to start analysing it from day 1 of coding; running analysis on old software may swamp you with issues which may take a while to untangle, best to keep it clean from the beginning.
But code analysers will not catch all logical errors, i.e. it doesn't do what you want it to do! These are best done by code reviews first, then testing. Performance is often improved by by trying to keep the algorithms as simple as possible, keeping instructions in loops tight, possibly unrolling loops (your compiler optimisations may do this), use of fast caches when accessing data which is slow to get.
Code reviews can raise a lot of issues from lots of other peoples eyes looking at it. Don't get too many people, try to get 3 other people if possible, sometimes junior developers ask the most insightful questions like, "why are we doing this?".
Testing can be roughly split into two sections, automated and manual. Automated testing requires effort producing test handlers for functions/units but once run can be run again and again very quickly. Manual testing requires planning, self-discipline to perform them all to the required, imagination to think up of scenarios that may impair performance and you have to be observant (you may have passed the test but the 'scope trace has a bit of an anomaly before/after the test).
"Do these steps change if I am
developing for an embedded system?"
Performance ananlysis can be different on embedded systems to applications systems; with the very broad brush that "embedded" now covers it depends how hardware-centric you are. It can be done using profilers, if you want a more cheap and chearful method then use test output pins to measure sections of code, or measure them with breakpoints on simulators that come with the development environment.
Make sure that not just a typical length of task is measured but also a maximum, as that is where one task may start impeding on other tasks and your scheduled tasks are not completed in time.
What tools are out there which can
help me?
Simulators on the IDEs, static analysis tools, dynamic analysis tools, but most of all you and other humans getting the requirements right, decent reviewing (of code and testing) and thorough testing (automated and manual).
Good luck!
My experiences.
Function calls are slow, eliminate with macros or inlined methods. Look at the disassembler listing to see.
If using GCC, mark optimized sections with #pragma GCC optimize("O3") or compile them separately.
Play with different combinations of applying the inline attribute (basically find a balance between size and speed).
It is a difficult question to be answered shortly since various techniques have been proposed such as flowchart and state diagram,so you can take a look at some titles:
ARM System-on-Chip Architecture, 2nd Edition -- Steve Furber
ARM System Developer's Guide - Designing and Optimizing System Software -- Andrew N. Sloss, Dominic Symes, Chris Wright & John Rayfield
The Definitive Guide to the ARM Cortex-M3 --Joseph Yiu
C Programming for Embedded Systems --Kirk Zurell
Embedded C -- Michael J. Pont
Programming Embedded Systems in C and C++ --Michael Barr
An Embedded Software Primer --David E, Simon
Embedded Microprocessor Systems 3rd Edition --Stuart Ball
Global Specification and Validation of Embedded Systems - Integrating Heterogeneous Components --G. Nicolescu & A.A Jerraya
Embedded Systems: Modeling, Technology and Applications --Gunter Hommel & Sheng Huanye
Embedded Systems and Computer Architecture --Graham Wilson
Designing Embedded Hardware --John Catsoulis
You have to use a profiler. It will help you identify your application's bottleneck(s). Then focus on improving the functions you spend the most time in and the ones you call the most. Repeat this procedure until you're satisfied with your application performance.
No they don't.
Depending on the platform you're developing onto :
Windows : AMD Code Analyst, VTune, Sleepy
Linux : valgrind / callgrind / cachegrind
Mac : the Xcode profiler is quite good.
Try to find a profiler for the architecture you actually work on.

Is C inefficient compared to Assembly? [duplicate]

This question already has answers here:
When is assembly faster than C? [closed]
(40 answers)
Closed 1 year ago.
This is purely a theory question, so, given an "infinite" time to make a trivial program, and an advanced knowledge of C and Assembly, is it really better to do something in Assembly? is "performance" lost when compiling C into Assembly (to machine code)?
By performance I mean, do modern C compilers do a bad job at certain tasks that programming directly in Assembly speeds up?
Modern C can do a better job than assembly in many cases, because keeping track of which operations can overlap and which will block others is so complex it can only be reasonably be tracked by a computer.
C is not inefficient compared to anything. C is a language, and we don't describe languages in terms of efficiency. We compare programs in terms of efficiency. C doesn't write programs; programmers write programs.
Assembly gives you immense flexibility when comparing with C, and that is at the cost of time programming. If you are a guru C programmer and a guru Assembly programmer, then chances are you might be able to squeeze some more juice with Assembly for writing any given program, but the price for that is virtually certain to be prohibitive.
Most of us aren't gurus in either of these languages. For most of us, giving the responsibility of performance tuning to a C compiler is a double win: you get the wisdom of a number of Assembly gurus, the people who wrote the C compiler, along with an immense amount of time in your hands to further correct and enhance your C program. You also get portability as a bonus.
This question seems to stem from the misconception that higher performance is automatically better. There is too much to be gained from a higher level perspective to make assembly better in the general case. Even if performance is your primary concern, compilers usually do a better job creating efficient assembly than you could write yourself. They have a much broader "understanding" of all of your source code than you could possibly hold in your head. Many optimizations can be had from NOT using well-structured assembly.
Obviously there are exceptions. If you need to access hardware directly, including special processing features of CPUs (e.g. SSE), then assembly is the way to go. However, in that case, you're probably better off using a library that addresses your general problem more directly (e.g. numerics packages).
But you should only worry about things like this if you have a concrete, specific need for the increased performance and you can show that your assembly actually IS faster. Concrete specific needs include: noticed and measure performance problems, embedded systems where performance is a fundamental design concern, etc.
Unless you are an assembly expert and(/or) taking advantage of advanced opcodes not utilized by the compiler, the C compiler will likely win.
Try it for fun ;-)
More realistic solutions are often to let the C compiler do it's bit, then profile and, if needed, tweak specific sections -- many compilers can dump some sort of low-level IL (or even "assembly").
Use C for most tasks, and write inline assembly code for specific ones (for example, to take advantage of SSE, MME, ...)
It depends. C compilers for Intel do a pretty good job nowadays. I wasn't so impressed by compilers for ARM - I could easly write an assembly version of an inner loop that performed twice as fast.
You typically don't need assembly on x86 machines. If you want to gain direct access to SSE instructions, look into compiler intrinsics!
Ignoring how much time it would take to write the code, and assuming you have all the knowledge that is required to do any task most efficiently in both situations, assembly code will, by definition, always be able to either meet or outperform the code generated by a C compiler, because the C compiler has to create the assembly code to do the same task and it cannot optimize everything; and anything the C compiler writes, you could also write (in theory), and unlike the compiler, you can sometimes take a shortcut because you know more about the situation than can be expressed in C code.
However, that doesn't mean they do a bad job and that the code is too slow; just that it's slower than it could be. It may not be by more than a few microseconds, but it can still be slower.
What you have to remember is that some optimizations performed by a compiler are very complex: agressive optimization tends to lead to very unreadable assembly code, and it becomes harder to reason about the code as a result if you were to do them manually. That's why you'd normally write it in C (or some other language) first, then profile it to find problem areas, and then go on to hand-optimize that piece of code until it reaches an acceptable speed - because the cost of writing everything in assembly is much higher, while often providing little or no benefit.
Actually, C might be faster than assembly in many cases, since compilers apply optimizations to your code. Even so, the performance difference (if any) is negligible.
I would focus more on readability & maintainability of the code base, as well as whether what you are trying to do is supported in C. In many cases, assembly will allow you to do more low-level things that C simply cannot do. For example, with assembly you can take advantage of MMX or SSE instructions directly.
So in the end, focus on what you want to accomplish. Remember - assembly language code is terrible to maintain. Use it only when you have no other choice.
No, compilers do not do a bad job at all. The amount of optimization that can be squeezed out by using assembly is insignificant for most programs.
That amount depends on how you define 'modern C compiler'. A brand new compiler (For a chip that has just reached market) may have a large number of inefficiencies that will get ironed out over time. Just compile some simple programs (the string.h functions, for example), and analyze what each line of code does. You may be surprised at some of the wasteful things an untested C compiler does, and recognize the error with a simple read-through of the code. A mature, well-tested, thoroughly optimized compiler (Think x86) will do a great job of generating assembly, though a new one will still do a decent job.
In no case can C do a better job than assembly. You could just benchmark the two, and if your assembly was slower, compile with -S and submit the resulting assembly, and you're guaranteed a tie. C is compiled to assembly, which has a 1:1 correlation with the bytecode. The computer can't do anything that assembly can't do, assuming that the complete instruction set is published.
In some cases, C is not expressive enough to be fully optimized. A programmer may know something about the nature of the data that simply cannot be expressed in C in such a way that the compiler can take advantage of this knowledge. Certainly, C is expressive and close to the metal, and is very good for optimization, but complete optimization is not always possible.
A compiler can't define 'performance' like a human can. I understand that you said trivial programs, but even in the simplest (useful) algorithms, there will be a tradeoff between size and speed. The compiler can't do this at a more fine grained scale than the -Os/-O[1-3] flags, but a human can know what 'best' means in the context of the purpose of a program.
Some architecture-dependent assembly instructions can't be expressed in C. This is where ASM() statements come in. Sometimes, these are not for optimization at all, but simply because there is no way to express in C that this line must use, say, the atomic test-and-set operation, or that we want to issue an SVC interrupt with the encoded parameter X.
The above points notwithstanding, C is orders of magnitude more efficient to program in and to master. If performance is important, analysis of the assembly will be necessary, and optimizations will probably be found, but the tradeoff in developer time and effort is rarely worth the effort for complex programs on a PC. For very simple programs which must be as fast as absolutely possible (like an RTOS), or which have severe memory constraints (like an ATTiny with 1KB of Flash (non-writable) memory and 64Bytes of RAM), assembly may be the only way to go.
Given an infinite time and an extremely deep understanding on how a modern CPU works you can actually write the "perfect" program (i.e. the best performance possible on that machine), but you will have to consider, for any instruction in your program, how CPU behaves in that context, pipelining and caching related optimizations, and many many other things.
A compiler is built to generate the best assembly code possible. You will rarely understand a modern complier generated assembly code because it tends to be really extreme.
At times compliers fail in this task because they can't always foresee what's happening.
Generally they do a great job but they sometimes fail...
Resuming... knowing C and Assembly is absolutely not enough to do a better job than a compiler in 99.99% cases, and considered that programming something in C can be 10000 times faster than programming the same assembly program a nicer way to spend some time is optimizing what the compiler did wrong in the remaining 0.01%, not reinventing the wheel.
This depends on the compiler you use? This is no property of C or any language. Theoretically it's possible to load a compiler with such a sophisticated AI that you can compile prolog to more efficient machine language than GCC can do with C.
This depends 100% on the compiler and 0% on C.
What does matter is that C is written as a language for which it is easy to write an optimizing compiler from C -> assembly, and with assembly this means the instructions of a Von Neumann machine. It depends on the target, some languages like prolog will probably be easier to map on hypothetical 'reduction machines'.
But, given that assembly is your target language for your C compiler (you can technically compile C to brainfuck or to Haskell, there is no theoretical difference) then:
It is possible to write the optimally fast program in that assembly itself (duh)
It is possible to write a C compiler which in every instant shall produce the most optimal assembly. That is to say, there exists a function from every C program to the most optimal way to get the same I/O in assembly, and this function is computable, albeit perhaps not deterministically.
This is also possible with every other programming language in the world.

Why aren't programs written in Assembly more often? [closed]

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
Locked. This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions.
It seems to be a mainstream opinion that assembly programming takes longer and is more difficult to program in than a higher level language such as C. Therefore it seems to be recommend or assumed that it is better to write in a higher level language for these reasons and for the reason of better portability.
Recently I've been writing in x86 assembly and it has dawned on me that perhaps these reasons are not really true, except perhaps portability. Perhaps it is more of a matter of familiarity and knowing how to write assembly well. I also noticed that programming in assembly is quite different than programming in an HLL. Perhaps a good and experienced assembly programmer could write programs just as easily and as quickly as an experienced C programmer writing in C.
Perhaps it is because assembly programming is quite different than HLLs, and so requires different thinking, methods and ways, which makes it seem very awkward to program in for the unfamiliar, and so gives it its bad name for writing programs in.
If portability isn't an issue, then really, what would C have over a good assembler such as NASM?
Edit:
Just to point out. When you are writing in assembly, you don't have to write just in instruction codes. You can use macros and procedures and your own conventions to make various abstractions to make programs more modular, more maintainable and easier to read. This is where being familiar with how to write good assembly comes in.
Hellо, I am a compiler.
I just scanned thousands of lines of code while you were reading this sentence. I browsed through millions of possibilities of optimizing a single line of yours using hundreds of different optimization techniques based on a vast amount of academic research that you would spend years getting at. I won't feel any embarrassment, not even a slight ick, when I convert a three-line loop to thousands of instructions just to make it faster. I have no shame to go to great lengths of optimization or to do the dirtiest tricks. And if you don't want me to, maybe for a day or two, I'll behave and do it the way you like. I can transform the methods I'm using whenever you want, without even changing a single line of your code. I can even show you how your code would look in assembly, on different processor architectures and different operating systems and in different assembly conventions if you'd like. Yes, all in seconds. Because, you know, I can; and you know, you can't.
P.S. Oh, by the way you weren't using half of the code you wrote. I did you a favor and threw it away.
ASM has poor legibility and isn't really maintainable compared to higher-level languages.
Also, there are many fewer ASM developers than for other more popular languages, such as C.
Furthermore, if you use a higher-level language and new ASM instructions become available (SSE for example), you just need to update your compiler and your old code can easily make use of the new instructions.
What if the next CPU has twice as many registers?
The converse of this question would be: What functionality do compilers provide?
I doubt you can/want to/should optimize your ASM better than gcc -O3 can.
I've written shedloads of assembler for the 6502, Z80, 6809 and 8086 chips. I stopped doing so as soon as C compilers became available for the platforms I was addressing, and immediately became at least 10x more productive. Most good programmers use the tools they use for rational reasons.
I love programming in assembly language, but it takes more code to do the same thing as in a high-level languge, and there is a direct correlation between lines of code and bugs. (This was explained decades ago in The Mythical Man-Month.)
It's possible to think of C as 'high level assembly', but get a few steps above that and you're in a different world. In C# you don't think twice about writing this:
foreach (string s in listOfStrings) { /* do stuff */ }
This would be dozens, maybe hundreds of lines of code in assembly, each programmer implementing it would take a different approach, and the next person coming along would have to figure it out. So if you believe (as many do) that programs are written primarily for other people to read, assembly is less readable than the typical HLL.
Edit: I accumulated a personal library of code used for common tasks, and macros for implementing C-like control structures. But I hit the wall in the 90s, when GUIs became the norm. Too much time was being spent on things that were routine.
The last task I had where ASM was essential was a few years ago, writing code to combat malware. No user interface, so it was all the fun parts without the bloat.
In addition to other people's answers of readability, maintainability, shorter code and therefore fewer bugs, and being much easier, I'll add an additional reason:
program speed.
Yes, in assembly you can hand tune your code to make use of every last cycle and make it as fast as is physically possible. However who has the time? If you write a not-completely-stupid C program, the compiler will do a really good job of optimizing for you. Probably making at least 95% of the optimizations you'd do by hand, without you having to worry about keeping track of any of it. There's definitely a 90/10 kind of rule here, where that last 5% of optimizations will end up taking up 95% of your time. So why bother?
If an average production program has say 100k lines of code, and each line is about 8-12 assembler instructions, that would be 1 million of assembler instructions.
Even if you could write all this by hand at a decent speed (remember, its 8 times more code that you have to write), what happens if you want to change some of the functionality? Understanding something you wrote a few weeks ago out of those 1 million instructions is a nightmare! There's no modules, no classes, no object-oriented design, no frameworks, no nothing. And the amount of similar looking code you have to write for even the simplest things is daunting at best.
Besides, you can't optimize your code nearly as well as a high level language. Where C for example performs an insane number of optimizations because you describe your intent, not only your code, in assembler you only write code, the assembler can't really perform any note-worthy optimizations on your code. What you write is what you get, and trust me, you can't reliably optimize 1 million instructions that you patch and patch as you write it.
Well I have been writing a lot of assembly "in the old days", and I can assure you that I am much more productive when I write programs in a high level language.
A reasonable level of assembler competence is a useful skill, especially if you work at any sort of system level or embedded programming, not so much because you have to write that much assembler, but because sometimes it's important to understand what the box is really doing. If you don't have a low-level understanding of assembler concepts and issues, this can be very difficult.
However, as for actually writing much code in assembler, there are several reasons it's not much done.
There's simply no (almost) need. Except for something like the very early system initialization and perhaps a few assembler fragments hidden in C functions or macros, all very low-level code that might once have been written in assembler can be written in C or C++ with no difficulty.
Code in higher-level languages (even C and C++) condenses functionality into far fewer lines, and there is considerable research showing that the number of bugs correlates with the number of lines of source code. Ie, the same problem, solved in assembler and C, will have more bugs in assembler simply because its longer. The same argument motivates the move to higher level languages such as Perl, Python, etc.
Writing in assembler, you have to deal with every single aspect of the problem, from detailed memory layout, instruction selection, algorithm choices, stack management, etc. Higher level languages take all this away from you, which is why are so much denser in terms of LOC.
Essentially, all of the above are related to the level of abstraction available to you in assembler versus C or some other language. Assembler forces you to make all of your own abstractions, and to maintain them through your own self-discipline, where any mid-level language like C, and especially higher level languages, provide you with abstractions out of the box, as well as the ability to create new ones relatively easily.
As a developer who spends most of his time in the embedded programming world, I would argue that assembly is far from a dead/obsolete language. There is a certain close-to-the-metal level of coding (for example, in drivers) that sometimes cannot be expressed as accurately or efficiently in a higher-level language. We write nearly all of our hardware interface routines in assembler.
That being said, this assembly code is wrapped such that it can be called from C code and is treated like a library. We don't write the entire program in assembly for many reasons. First and foremost is portability; our code base is used on several products that use different architectures and we want to maximize the amount of code that can be shared between them. Second is developer familiarity. Simply put, schools don't teach assembly like they used to, and our developers are far more productive in C than in assembly. Also, we have a wide variety of "extras" (things like libraries, debuggers, static analysis tools, etc) available for our C code that aren't available for assembly language code. Even if we wanted to write a pure-assembly program, we would not be able to because several critical hardware libraries are only available as C libs. In one sense, it's a chicken/egg problem. People are driven away from assembly because there aren't as many libraries and development/debug tools available for it, but the libs/tools don't exist because not enough people use assembly to warrant the effort creating them.
In the end, there is a time and a place for just about any language. People use what they are most familiar and productive with. There will probably always be a place in a programmer's repertoire for assembly, but most programmers will find that they can write code in a higher-level language that is almost as efficient in far less time.
When you are writing in assembly, you don't have to write just in instruction codes. You can use macros and procedures and your own conventions to make various abstractions to make programs more modular, more maintainable and easier to read.
So what you're basically saying is, that with skilled use of a sophisticated assembler, you can make your ASM code closer and closer to C (or anyway another low-ish-level language of your own invention), until eventually you are just as productive as a C programmer.
Does that answer your question? ;-)
I don't say this idly: I have programmed using exactly such an assembler and system. Even better, the assembler could target a virtual processor, and a separate translator compiled the output of the assembler for a target platform. Much as happens with LLVM's IF, but in its early forms pre-dating it by about 10 years. So there was portability, plus the ability to write routines for a specific target asssembler where required for efficiency.
Writing using that assembler was about as productive as C, and with by comparison with GCC-3 (which was around by the time I was involved) the assembler/translator produced code that was roughly as fast and usually smaller. Size was really important, and the company had few programmers and was willing to teach new hires a new language before they could do anything useful. And we had the back-up that people who didn't know the assembler (e.g. customers) could write C and compile it for the same virtual processor, using the same calling convention and so on, so that it interfaced neatly. So it felt like a marginal win.
That was with multiple man-years of work in the bag developing the assembler technology, libraries, and so on. Admittedly much of which went into making it portable, if it had only ever been targeting one architecture then the all-singing all-dancing assembler would have been much easier.
In summary: you may not like C, but it doesn't mean that the effort of using C is greater than the effort of coming up with something better.
Assembly is not portable between different microprocessors.
The same reason we don't go to the bathroom outside anymore, or why we don't speak Latin or Aramaic.
Technology comes along and makes things easier and more accessible.
EDIT - to cease offending people, I've removed certain words.
Why? Simple.
Compare this :
for (var i = 1; i <= 100; i++)
{
if (i % 3 == 0)
Console.Write("Fizz");
if (i % 5 == 0)
Console.Write("Buzz");
if (i % 3 != 0 && i % 5 != 0)
Console.Write(i);
Console.WriteLine();
}
with
.locals init (
[0] int32 i)
L_0000: ldc.i4.1
L_0001: stloc.0
L_0002: br.s L_003b
L_0004: ldloc.0
L_0005: ldc.i4.3
L_0006: rem
L_0007: brtrue.s L_0013
L_0009: ldstr "Fizz"
L_000e: call void [mscorlib]System.Console::Write(string)
L_0013: ldloc.0
L_0014: ldc.i4.5
L_0015: rem
L_0016: brtrue.s L_0022
L_0018: ldstr "Buzz"
L_001d: call void [mscorlib]System.Console::Write(string)
L_0022: ldloc.0
L_0023: ldc.i4.3
L_0024: rem
L_0025: brfalse.s L_0032
L_0027: ldloc.0
L_0028: ldc.i4.5
L_0029: rem
L_002a: brfalse.s L_0032
L_002c: ldloc.0
L_002d: call void [mscorlib]System.Console::Write(int32)
L_0032: call void [mscorlib]System.Console::WriteLine()
L_0037: ldloc.0
L_0038: ldc.i4.1
L_0039: add
L_003a: stloc.0
L_003b: ldloc.0
L_003c: ldc.i4.s 100
L_003e: ble.s L_0004
L_0040: ret
They're identical feature-wise.
The second one isn't even assembler but .NET IL (Intermediary Language, similar to Java's bytecode). The second compilation transforms the IL into native code (i.e. almost assembler), making it even more cryptical.
I'd guess ASM on even x86(_64) makes sense in cases where you gain a lot by utilizing instructions that are difficult for a compiler to optimize for. x264 for example uses a lot of asm for its encoding, and the speed gains are huge.
I'm sure there are many reasons, but two quick reasons I can think of are
Assembly code is definitely harder to read (I'm positive its more time-consuming to write as well)
When you have a huge team of developers working on a product, it is helpful to have your code divided into logical blocks and protected by interfaces.
One of the early discoveries (you'll find it in Brooks' Mythical Man-Month, which is from experience in the 1960s) was that people were more or less as productive in one language as another, in debugged lines of code per day. This obviously isn't universally true, and can breaks when pushed too far, but it was generally true of the high-level languages of Brooks' time.
Therefore, the fastest way to get productivity would be to use languages where one individual line of code did more, and indeed this works, at least for languages of complexity like FORTRAN and COBOL, or to give a more modern example C.
Portability is always an issue -- if not now, at least eventually. The programming industry spends billions every year to port old software which, at the time it was written, had "obviously" no portability issue whatsoever.
There was a vicious cycle as assembly became less commonplace: as higher level languages matured, assembly language instruction sets were built less for programmer convenience and more for the convenience of compilers.
So now, realistically, it may be very hard to make the right decisions on, say, which registers you should use or which instructions are slightly more efficient. Compilers can use heuristics to figure out which tradeoffs are likely to have the best payoff. We can probably think through smaller problems and find local optimizations that might beat our now pretty sophisticated compilers, but odds are that in the average case, a good compiler will do a better job on the first try than a good programmer probably will. Eventually, like John Henry, we might beat the machine, but we might seriously burn ourselves out getting there.
Our problems are also now quite different. In 1986 I was trying to figure out how to get a little more speed out of small programs that involved putting a few hundred pixels on the screen; I wanted the animation to be less jerky. A fair case for assembly language. Now I'm trying to figure out how to represent abstractions around contract language and servicer policy for mortgages, and I'd rather read something that looks close to the language that the business folks speak. Unlike LISP macros, Assembly macros don't enforce much in the way of rules, so even though you might be able to get something reasonably close to a DSL in a good assembler, it'll be prone to all sorts of quirks that won't cause me problems if I wrote the same code in Ruby, Boo, Lisp, C# or even F#.
If your problems are easy to express in efficient assembly language, though, more power to you.
Ditto most of what others have said.
In the good old days before C was invented, when the only high level languages were things like COBOL and FORTRAN, there were lots of things that just weren't possible to do without resorting to assembler. It was the only way to get the full breadth of flexibility, to be able to access all the devices, etc. But then C was invented, and almost anything that was possible in assembly was possible in C. I have written very little assembly since then.
That said, I think it is a very useful exercise for new programmers to learn to write in assembler. Not because they would actually use it much, but because then you understand what is really happening inside the computer. I've seen lots of programming errors and inefficient code from programmers who clearly have no idea what's really happening with the bits and bytes and registers.
I've been programming in assembly now for about a month. I often write a piece of code in C and then compile it to assembly to assist me. Perhaps I am not utilizing the full optimizing power of the C compiler but it appears that my C asm source is including unnecessary operations. So I am beginning to see that the talk of a good C compiler outperforming a good assembly coder is not always true.
Anyways, my assembly programs are so fast. And the more I use assembly the less time it takes me to write out my code because it's really not that hard. Also the comment about assembly having poor legibility is not true. If you label your programs correctly and make comments when there is additional elaboration needed you should be all set. In fact in ways assembly is more clear to the programmer because they are seeing what is happening at the level of the processor. I don't know about other programmers but for me I like knowing what's happening, rather than things being in a sort of black box.
With that said the real advantage of compilers is that a compiler can understand patterns and relationships and then automatically code them in the appropriate locations in the source. One popular example are virtual functions in C++ which requires the compiler to optimally map function pointers. However a compiler is limited to doing what the maker of the compiler allows the compiler to do. This leads to programmers sometimes having to resort to doing bizarre things with their code , adding coding time, when they could have been done trivially with assembly.
Personally I think the marketplace heavily supports high level languages. If assembly language was the only language in existence today then their would be about 70% less people programming and who knows where our world would be, probably back in the 90's. Higher level languages appeal to a broader range of people. This allows a higher supply of programmers to build the needed infrastructure of our world. Developing nations like China and India benefit heavily from languages like Java. These countries will fast develop their IT infrastructure and people will become more interconnected. So my point is that high level languages are popular not because they produce superior code but because they help to meet demand in the world's marketplaces.
I'm learning assembly in comp org right now, and while it is interesting, it is also very inefficient to write in. You have to keep alot more details in your head to get things working, and its also slower to write the same things. For example, a simple 6 line for loop in C++ can equal 18 lines or more of assembly.
Personally, its alot of fun learning how things work down at the hardware level, and it gives me greater appreciation for how computing works.
What C has over a good macro assembler is the language C. Type checking. Loop constructs. Automatic stack management. (Nearly) automatic variable management. Dynamic memory techniques in assembler are a massive pain in the butt. Doing a linked list properly is just down right scary compared to C or better yet list foo.insert(). And debugging - well, there's no contest on what is easier to debug. HLLs win hands down there.
I've coded nearly half my career in assembler which makes it very easy for me to think in assmebler. it helps me to see what the C compiler is doing which again helps me write code that the C compiler can efficiently handle. A well thought out routine written in C can be written to output exactly what you want in assembler with a little work - and it's portable! I've already had to rewrite a few older asm routines back to C for cross platform reasons and it's no fun.
No, I'll stick with C and deal with the occasional slight slowdown in performance against the productivity time I gain with HLL.
I can only answer why I personally don't write programs in assembly more often, and the main reason is that it's more tedious to do. Also, I think that it is easier to get things subtly wrong without noticing immediately. E.g., you might change the way you use a register in one routine but forget to change this in one place. It'll assemble fine and you may not notice until much later.
That said, I do think there are still valid uses for assembly. For instance, I have a number of pretty optimised assembly routines for processing large amounts of data, using SIMD and following the paranoid "every bit is sacred"[quote V.Stob] approach. (But note that naive assembly implementations are often a lot worse than what a compiler would generate for you.)
C is a macro assembler! And it's the best one!
It can do nearly everything assembly can, it can be portable and in most of the rare cases where it can't do something you can still use embedded assembly code. This leaves only a small fraction of programs that you absolutely need to write in assembly and nothing but assembly.
And the higher level abstractions and the portability make it more worthwhile for most people to write system software in C. And although you might not need portability now if you invest a lot of time and money in writing some program you might not want to limit yourself in what you'll be able to use it for in the future.
People seem to forget that there is also the other direction.
Why are you writing in Assembler in the first place? Why not write the program in a truly low level language?
Instead of
mov eax, 0x123
add eax, 0x456
push eax
call printInt
you could just as well write
B823010000
0556040000
50
FF15.....
That has so many advantages, you know the exact size of your program, you can reuse the value of instructions as input for other instructions and you do not even need an assembler to write it, you can use any text editor...
And the reason you still prefer Assembler about this, is the reason other people prefer C...
Because it's always that way: time pass and good things pass away too :(
But when you write asm code it's totally different feeling than when you code high-level langs, though you know it's much less productive. It's like you're a painter: you are free to draw anything you like the way you like with absolutely no restrictions(well, only by CPU features)... That is why I love it. It's a pity this language goes away. But while somebody still remembers it and codes it, it will never die!
$$$
A company hires a developer to help turn code into $$$. The faster that useful code can be produced, the faster the company can turn that code into $$$.
Higher level languages are generally better at churning out larger volumes of useful code. This is not to say that assembly does not have its place, for there are times and places where nothing else will do.
The advantage of HLL's is even greater when you compare assembly to a higher level language than C, e.g. Java or Python or Ruby. For instance, these languages have garbage collection: no need to worry about when to free a chunk of memory, and no memory leaks or bugs due to freeing too early.
As others mentioned before, the reason for any tool to exist is how efficiently it can work. As HLLs can accomplish the same jobs as many lines of asm code I guess it's natural for assembly to be superseded by other languages. And for the close-to-hardware fiddling - there's inline assembly in C and other variants as per language.
Dr. Paul Carter in says in the PC Assembly Language
"...a better understanding of how
computers really work at a lower level
than in programming languages like
Pascal. By gaining a deeper
understanding of how computers work,
the reader can often be much more
productive developing software in
higher level languages such as C and
C++. Learning to program in assembly
language is an excellent way to
achieve this goal."
We've got introduction to assembly in my college courses. It'll help to clear concepts. However I doubt any of us would write 90% of code in assembly. How relevant is in-depth assembly knowledge today?
Flipping through these answers, I'd bet 9/10 of the responders have never worked with assembly.
This is an ages old question that comes up every so often and you get the same, mostly misinformed answers. If it weren't for portability, I'd still do everything in assembly myself. Even then, I code in C almost like I did in assembly.

What specific examples are there of knowing C making you a better high level programmer?

I know about the existance of question such as this one and this one. Let me explain.
Afet reading Joel's article Back to Basics and seeing many similar questions on SO, I've begun to wonder what are specific examples of situations where knowing stuff like C can make you a better high level programmer.
What I want to know is if there are many examples of this. Many times, the answer to this question is something like "Knowing C gives you a better feel of what's happening under the covers" or "You need a solid foundation for your program", and these answers don't have much meaning. I want to understand the different specific ways in which you will benefit from knowing low level concepts,
Joel gave a couple of examples: Binary databases vs XML, and strings. But two examples don't really justify learning C and/or Assembly. So my question is this: What specific examples are there of knowing C making you a better high level programmer?
My experience with teaching students and working with people who only studied high-level languages is that they tend to think at a certain high level of abstraction, and they assume that "everything comes for free". They can become very competent programmers, but eventually they have to deal with some code that has performance issues and then it comes to bite them.
When you work a lot with C, you do think about memory allocation. You often think about memory layout (and cache locality if that's an issue). You understand how and why certain graphics operations just cost a lot. How efficient or inefficient certain socket behaviors are. How buffers work, etc. I feel that using the abstractions in a higher level language when you do know how it is implemented below the covers sometimes gives you "that extra secret sauce" when thinking about performance.
For example, Java has a garbage collector and you can't directly assign things to memory directly. And yet, you can make certain design choices (e.g., with custom data structures) that affect performance because of the same reasons this would be an issue in C.
Also, and more generally, I feel that it is important for a power programmer to not only know big-O notation (which most schools teach), but that in real-life applications the constant is also important (which schools try to ignore). My anecdotal experience is that people with skills in both language levels tend to have a better understanding of the constant, perhaps because of what I described above.
In addition, many higher level systems that I have seen interface with lower level libraries and infrastructures. For instance, some communications, databases or graphics libraries. Some drivers for certain devices, etc. If you are a power programmer, you may eventially have to venture out there and it helps to at least have an idea of what is going on.
Knowing low level stuff can help a lot.
To become a racing driver, you have to learn and understand the basic physics of how tyres grip the road. Anyone can learn to drive pretty fast, but you need a good understanding of the "low level" stuff (forces and friction, racing lines, fine throttle and brake control, etc) to get those last few percent of performance that will allow you to win the race.
For example, if you understand how the CPU architecture works in your computer, you can write code that works better with it (e.g. if you know you have a certain CPU cache size or a certain number of bytes in each CPU cache line, you can arrange your data structures and the way that you access them to make the best use of the cache - for example, processing many elements of an array in order is often faster than processing random elements, due to the CPU cache). If you have a multi-core computer, then understanding how low level techniques like threading work can gave huge benefits (just as not understanding the low level can lead to disaster in threading).
If you understand how Disk I/O and caching works, you can modify file operations to work well with it (e.g. if you read from one file and write to another, working on large batches of data in RAM can help reduce I/O contention between the reading and writing phases of your code, and vastly improve throughput)
If you understand how virtual functions work, you can design high-level code that uses virtual functions well. If used incorrectly they can severely hamper performance.
If you understand how drawing is handled, you can use clever tricks to improve drawing speed. e.g. You can draw a chessboard by alternately drawing 64 white and black squares. But it is often faster to draw 32 white sqares and then 32 black ones (because you only have to change the drawing colour twice instead of 64 times). But you can actually draw the whole board black, then XOR 4 stripes across the board and 4 stripes down the board in white, and this can be much faster still (2 colour changes, and only 9 rectangles to draw instead of 64). This chessboard trick teaches you a very important programming skill: Lateral thinking. By designing your algorithm well, you can often make a big difference to how well your program operates.
Understanding C, or for that matter, any low level programming language, gives you an opportunity to understand things like memory usage (i.e. why is it a bad thing to create several million heavy objects), how pointers/object references work, etc.
The problem is that as we've created ever increasing levels of abstraction, we find ourselves doing a lot of 'lego block' programming, without understanding how the legos actually function. And by having almost infinite resources, we start treating memory and resources like water, and tend to solve problems by throwing more iron at the situation.
While not limited to C, there's a tremendous benefit to working at a low level with much smaller, memory constrained systems like the Arduino or old-school 8-bit processors. It lets you experience close to the metal coding in a much more approachable package, and after spending time squeezing apps into 512K, you will find yourself applying these skills at a larger level within your day to day programming.
So the language itself is not important, but having a deeper appreciation for how all of the bits come together, and how to work effectively at a level closer to the hardware is a set of skills beneficial to any software developer.
For one, knowing C helps you understand how memory works in the OS and in other high level languages. When your C# or Java program balloons on memory usage, understanding that references (which are basically just pointers) take memory too, and understand how many of the data structures are implemented (which you get from making your own in C) helps you understand that your dictionary is reserving huge amounts of memory that aren't actually used.
For another, knowing C can help you understand how to make use of lower level operating system features. You don't need this often, but sometimes you may need memory mapped files, or to use marshalling in C#, and C will greatly help understand what you're doing when that happens.
I think C has also helped my understanding of network protocols, but I can't put my finger on specific examples. I was reading another SO question the other day where someone was complaining about how C's bit-fields are 'basically useless' and I was thinking how elegantly C bit fields represent low-level network protocols. High level languages dealing with structures of bits always end up a mess!
In general, the more you know, the better programmer you will be.
However, sometimes knowing another language, such as C, can make you do the wrong thing, because there might be an assumption that is not true in a higher-level language (such as Python, or PHP). For example, one might assume that finding the length of a list might be O(N) where N is the length of the list. However, this is probably not the case in many high-level language instances. In Python, for most list-like things the cost is O(1).
Knowing more about the specifics of a language will help, but knowing more in general might lead one to make incorrect assumptions.
Just "knowing" C would not make you better.
But, if you understand the whole thing, how native binaries work, how does CPU work with it, what are architecture limitations, you may write a code which is easier for CPU.
For example, how L1/L2 caches affect your work, and how should you write your code to have more hits in L1/L2 caches. When working with C/C++ and doing heavy optimizations, you will have to go down to that kind of things.
It isn't so much knowing C as it is that C is closer to the bare metal than many other languages. You need to be more aware of how to allocate/deallocate memory because you have to do it yourself. Doing it yourself helps you understand the implications of many decisions that you make.
To me any language is acceptable as long as you understand how the compiler/interpreter (basically) maps your code onto the machine. It's a bit easier to do in a language that exposes this directly, but you should be able to, with a bit of reading, figure out how memory is allocated and organized, what sort of indexing patterns are more optimal than others, what constructs are more efficient for particular applications, etc.
More important, I think, is a good understanding of operating systems, memory architectures, and algorithms. If you understand how your algorithm works, why it would be better to choose one algorithm or data structure over another (e.g., HashSet vs. List), and how your code maps onto the machine, it shouldn't matter what language you are using.
This is my experience of how I learnt and taught myself programming, specifically, understanding C, this is going back to early 1990's so may be a bit antique, but the passion and the drive is important:
Learn to understand the low level principles of the computer, such as EGA/VGA programming, here's a link to the Simtel archive on the C programmer's guide to the PC.
Understanding how TSR's work
Download the whole archive of Bob Stout's snippets which is a big collection of C code that does one thing only - study them and understand it, not alone that, the collection of snippets strives to be portable.
Browse at the International Obfuscated C Code Contest (IOCCC) online, and see how the C code can be abused and understand the intracies of the language. The worst code abuse is the winner! Download the archives and study them.
Like myself, I loved the infamous Ponzo's C Tutorial which helped me immensely, unfortunately, the archive is very hard to find. If anyone knows of where to obtain them, please leave a comment and I will amend this answer to include the link. There is another one that I can remember - Coronado's [Generic?] C Tutorial, again, my memory on this one is hazy...
Look at Dr. Dobb's journal and C User Journal here - I do not know if you can still get them in print but they were a classic, can remember the feeling of holding a printed copy in my hand and tearing off home to type in the code to see what happens!
Grab an ancient copy of Turbo C v2 which I believe you can get from borland.com and just play with 16bit C programming to get a feel and mess with the pointers...sure it is ancient and old but playing with pointers on it is fine.
Understand and learn Pointers, link here to the legacy Simtel.net - a crucial link to achieving C Guru'ship for want of a better word, also you will find a host of downloads pertaining to the C programming language - I remember actually ordering the Simtel CD Archive and looking for the C stuff...
A couple of things that you have to deal directly with in C that other languages abstract away from you include explicit memory management (malloc) and dealing directly with pointers.
My girlfriend is one semester from graduating MIT (where they mainly use Java, Scheme, and Python) with a Computer Science degree, and she is currently working at a company whose codebase is in C++. For the first few days she had a difficult time understanding all the pointers/references/etc.
On the other hand, I found moving from C++ to Java very easy, because I was never confused about pass-references-by-value vs pass-by-reference.
Similarly, in C/C++ it is much more apparent that primitives are just the compiler treating the same sets of bits in different ways, as opposed to a language like Python or Ruby where everything is an object with its own distinct properties.
A simple (not entirely realistic) example to illustrate some of the advice above. Consider the seemingly harmless
while(true)
for(Iterator iter = foo.iterator(); iter.hasNext();)
bar.doSomething( iter.next() )
or the even higher level
while(true)
for(Baz b: foo)
bar.doSomething(b)
A possible problem here is that each time round the while loop a new object (the iterator) is created. If all you care about is programmer convenience, then the latter is definitely better. But if the loop has to be efficient or the machine is resource constrained then you are pretty much at the mercy of the designers of your high level language.
For example, a typical complaint for doing high-performance Java is having execution stop while garbage (such as all those allocated Iterator objects) is reclaimed. Not very good if your software is charged with tracking incoming missiles, auto-piloting a passenger jet, or just not leaving the user wondering why the GUI has stopped responding.
One possible solution (still in the higher-level language) would be to weaken the convenience of the iterator to something like
Iterator iter = new Iterator();
while(true)
for(foo.initAlreadyAllocatedIterator(iter); iter.hasNext();)
bar.doSomething(iter.next())
But this would only make sense if you had some idea about memory allocation...otherwise it just looks like a nasty API. Convenience always costs somewhere, and knowing lower-level stuff can help you identify and mitigate those costs.

Why do you program in assembly? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 4 years ago.
Improve this question
I have a question for all the hardcore low level hackers out there. I ran across this sentence in a blog. I don't really think the source matters (it's Haack if you really care) because it seems to be a common statement.
For example, many modern 3-D Games have their high performance core engine written in C++ and Assembly.
As far as the assembly goes - is the code written in assembly because you don't want a compiler emitting extra instructions or using excessive bytes, or are you using better algorithms that you can't express in C (or can't express without the compiler mussing them up)?
I completely get that it's important to understand the low-level stuff. I just want to understand the why program in assembly after you do understand it.
I think you're misreading this statement:
For example, many modern 3-D Games have their high performance core engine written in C++ and Assembly.
Games (and most programs these days) aren't "written in assembly" the same way they're "written in C++". That blog isn't saying that a significant fraction of the game is designed in assembly, or that a team of programmers sit around and develop in assembly as their primary language.
What this really means is that developers first write the game and get it working in C++. Then they profile it, figure out what the bottlenecks are, and if it's worthwhile they optimize the heck out of them in assembly. Or, if they're already experienced, they know which parts are going to be bottlenecks, and they've got optimized pieces sitting around from other games they've built.
The point of programming in assembly is the same as it always has been: speed. It would be ridiculous to write a lot of code in assembler, but there are some optimizations the compiler isn't aware of, and for a small enough window of code, a human is going to do better.
For example, for floating point, compilers tend to be pretty conservative and may not be aware of some of the more advanced features of your architecture. If you're willing to accept some error, you can usually do better than the compiler, and it's worth writing that little bit of code in assembly if you find that lots of time is spent on it.
Here are some more relevant examples:
Examples from Games
Article from Intel about optimizing a game engine using SSE intrinsics. The final code uses intrinsics (not inline assembler), so the amount of pure assembly is very small. But they look at the assembler output by the compiler to figure out exactly what to optimize.
Quake's fast inverse square root. Again, the routine doesn't have assembler in it, but you need to know something about architecture to do this kind of optimization. The authors know what operations are fast (multiply, shift) and which are slow (divide, sqrt). So they come up with a very tricky implementation of square root that avoids the slow operations entirely.
High-Performance Computing
Outside the domain of games, people in scientific computing frequently optimize the crap out of things to get them to run fast on the latest hardware. Think of this as games where you can't cheat on the physics.
A great recent example of this is Lattice Quantum Chromodynamics (Lattice QCD). This paper describes how the problem pretty much boils down to one very small computational kernel, which was optimized heavily for PowerPC 440's on an IBM Blue Gene/L. Each 440 has two FPUs, and they support some special ternary operations that are tricky for compilers to exploit. Without these optimizations, Lattice QCD would've run much slower, which is costly when your problem requires millions of CPU hours on expensive machines.
If you are wondering why this is important, check out the article in Science that came out of this work. Using Lattice QCD, these guys calculated the mass of a proton from first principles, and showed last year that 90% of the mass comes from strong force binding energy, and the rest from quarks. That's E=mc2 in action. Here's a summary.
For all of the above, the applications are not designed or written 100% in assembly -- not even close. But when people really need speed, they focus on writing the key parts of their code to fly on specific hardware.
I have not coded in assembly language for many years, but I can give several reasons that I frequently saw:
Not all compilers can make use of certain CPU optimizations and instruction set (e.g., the new instruction sets that Intel adds once in a while). Waiting for compiler writers to catch up means losing a competitive advantage.
Easier to match actual code to known CPU architecture and optimization. For example, things you know about the fetching mechanism, caching, etc. This is supposed to be transparent to the developer, but the fact is that it is not, that's why compiler writers can optimize.
Certain hardware level accesses are only possible/practical via assembly language (e.g., when writing device driver).
Formal reasoning is sometimes actually easier for the assembly language than for the high-level language since you already know what the final or almost final layout of the code is.
Programming certain 3D graphic cards (circa late 1990s) in the absence of APIs was often more practical and efficient in assembly language, and sometimes not possible in other languages. But again, this involved really expert-level games based on the accelerator architecture like manually moving data in and out in certain order.
I doubt many people use assembly language when a higher-level language would do, especially when that language is C. Hand-optimizing large amounts of general-purpose code is impractical.
There is one aspect of assembler programming which others have not mentioned - the feeling of satisfaction you get knowing that every single byte in an application is the result of your own effort, not the compiler's. I wouldn't for a second want to go back to writing whole apps in assembler as I used to do in the early 80s, but I do miss that feeling sometimes...
Usually, a layman's assembly is slower than C (due to C's optimization) but many games (I distinctly remember Doom) had to have specific sections of the game in Assembly so it would run smoothly on normal machines.
Here's the example to which I am referring.
I started professional programming in assembly language in my very first job (80's). For embedded systems the memory demands - RAM and EPROM - were low. You could write tight code that was easy on resources.
By the late 80's I had switched to C. The code was easier to write, debug and maintain. Very small snippets of code were written in assembler - for me it was when I was writing the context switching in an roll-your-own RTOS. (Something you shouldn't do anymore unless it is a "science project".)
You will see assembler snippets in some Linux kernel code. Most recently I've browsed it in spinlocks and other synchronization code. These pieces of code need to gain access to atomic test-and-set operations, manipulating caches, etc.
I think you would be hard pressed to out-optimize modern C compilers for most general programming.
I agree with #altCognito that your time is probably better spent thinking harder about the problem and doing things better. For some reason programmers often focus on micro-efficiencies and neglect the macro-efficiencies. Assembly language to improve performance is a micro-efficiency. Stepping back for a wider view of the system can expose the macro problems in a system. Solving the macro problems can often yield better performance gains.
Once the macro problems are solved then collapse to the micro level.
I guess micro problems are within the control of a single programmer and in a smaller domain. Altering behavior at the macro level requires communication with more people - a thing some programmers avoid. That whole cowboy vs the team thing.
"Yes". But, understand that for the most part the benefits of writing code in assembler are not worth the effort. The return received for writing it in assembly tends to be smaller than the simply focusing on thinking harder about the problem and spending your time thinking of a better way of doing thigns.
John Carmack and Michael Abrash who were largely responsible for writing Quake and all of the high performance code that went into IDs gaming engines go into this in length detail in this book.
I would also agree with Ólafur Waage that today, compilers are pretty smart and often employ many techniques which take advantage of hidden architectural boosts.
These days, for sequential codes at least, a decent compiler almost always beats even a highly seasoned assembly-language programmer. But for vector codes it's another story. Widely deployed compilers don't do such a great job exploiting the vector-parallel capabilities of the x86 SSE unit, for example. I'm a compiler writer, and exploiting SSE tops my list of reasons to go on your own instead of trusting the compiler.
SSE code works better in assembly than compiler intrinsics, at least in MSVC. (i.e. does not create extra copies of data )
I've three or four assembler routines (in about 20 MB source) in my sources at work. All of them are SSE(2), and are related to operations on (fairly large - think 2400x2048 and bigger) images.
For hobby, I work on a compiler, and there you have more assembler. Runtime libraries are quite often full of them, most of them have to do with stuff that defies the normal procedural regime (like helpers for exceptions etc.)
I don't have any assembler for my microcontroller. Most modern microcontrollers have so much peripheral hardware (interrupt controled counters, even entire quadrature encoders and serial building blocks) that using assembler to optimize the loops is often not needed anymore. With current flash prices, the same goes for code memory. Also there are often ranges of pin-compatible devices, so upscaling if you systematically run out of cpu power or flash space is often not a problem
Unless you really ship 100000 devices and programming assembler makes it possible to really make major savings by just fitting in a flash chip a category smaller. But I'm not in that category.
A lot of people think embedded is an excuse for assembler, but their controllers have more CPU power than the machines Unix was developed on. (Microchip coming
with 40 and 60 MIPS microcontrollers for under USD 10).
However a lot people are stuck with legacy, since changing microchip architecture is not easy. Also the HLL code is very architecture dependent (because it uses the hardware periphery, registers to control I/O, etc). So there are sometimes good reasons to keep maintaining a project in assembler (I was lucky to be able to setup affairs on a new architecture from scratch). But often people kid themselves that they really need the assembler.
I still like the answer a professor gave when we asked if we could use GOTO (but you could read that as ASSEMBLER too): "if you think it is worth writing a 3 page essay on why you need the feature, you can use it. Please submit the essay with your results. "
I've used that as a guiding principle for lowlevel features. Don't be too cramped to use it, but make sure you motivate it properly. Even throw up an artificial barrier or two (like the essay) to avoid convoluted reasoning as justification.
Some instructions/flags/control simply aren't there at the C level.
For example, checking for overflow on x86 is the simple overflow flag. This option is not available in C.
Defects tend to run per-line (statement, code point, etc.); while it's true that for most problems, assembly would use far more lines than higher level languages, there are occasionally cases where it's the best (most concise, fewest lines) map to the problem at hand. Most of these cases involve the usual suspects, such as drivers and bit-banging in embedded systems.
If you were around for all the Y2K remediation efforts, you could have made a lot of money if you knew Assembly. There's still plenty of legacy code around that was written in it, and that code occasionally needs maintenance.
Another reason could be when the available compiler just isn't good enough for an architecture and the amount of code needed in the program is not that long or complex as for the programmer to get lost in it. Try programming a microcontroller for an embedded system, usually assembly will be much easier.
Beside other mentioned things, all higher languages have certain limitations. Thats why some people choose to programm in ASM, to have full control over their code.
Others enjoy very small executables, in the range of 20-60KB, for instance check HiEditor, which is implemented by author of the HiEdit control, superb powerfull edit control for Windows with syntax highlighting and tabs in only ~50kb). In my collection I have more then 20 such gold controls from Excell like ssheets to html renders.
I think a lot of game developers would be surprised at this bit of information.
Most games I know of use as little assembly as at all possible. In some cases none at all, and at worst, one or two loops or functions.
That quote is over-generalized, and nowhere near as true as it was a decade ago.
But hey, mere facts shouldn't hinder a true hacker's crusade in favor of assembly. ;)
If you are programming a low end 8 bit microcontroller with 128 bytes of RAM and 4K of program memory you don't have much choice about using assembly. Sometimes though when using a more powerful microcontroller you need a certain action to take place at an exact time. Assembly language comes in useful then as you can count the instructions and so measure the clock cycles used by your code.
Games are pretty performance hungry and although in the meantime the optimizers are pretty good a "master programmer" is still able to squeeze out some more performance by hand coding the right parts in assembly.
Never ever start optimizing your program without profiling it first. After profiling should be able to identify bottlenecks and if finding better algorithms and the like don't cut it anymore you can try to hand code some stuff in assembly.
Aside from very small projects on very small CPUs, I would not set out to ever program an entire project in assembly. However, it is common to find that a performance bottleneck can be relieved with the strategic hand coding of some inner loops.
In some cases, all that is really required is to replace some language construct with an instruction that the optimizer cannot be expected to figure out how to use. A typical example is in DSP applications where vector operations and multiply-accumulate operations are difficult for an optimizer to discover, but easy to hand code.
For example certain models of the SH4 contain 4x4 matrix and 4 vector instructions. I saw a huge performance improvement in a color correction algorithm by replacing equivalent C operations on a 3x3 matrix with the appropriate instructions, at the tiny cost of enlarging the correction matrix to 4x4 to match the hardware assumption. That was achieved by writing no more than a dozen lines of assembly, and carrying matching adjustments to the related data types and storage into a handful of places in the surrounding C code.
It doesn't seem to be mentioned, so I thought I'd add it: in modern games development, I think at least some of the assembly being written isn't for the CPU at all. It's for the GPU, in the form of shader programs.
This might be needed for all sorts of reasons, sometimes simply because whatever higher-level shading language used doesn't allow the exact operation to be expressed in the exact number of instructions wanted, to fit some size-constraint, speed, or any combination. Just as usual with assembly-language programming, I guess.
Almost every medium-to-large game engine or library I've seen to date has some hand-optimized assembly versions available for matrix operations like 4x4 matrix concatenation. It seems that compilers inevitably miss some of the clever optimizations (reusing registers, unrolling loops in a maximally efficient way, taking advantage of machine-specific instructions, etc) when working with large matrices. These matrix manipulation functions are almost always "hotspots" on the profile, too.
I've also seen hand-coded assembly used a lot for custom dispatch -- things like FastDelegate, but compiler and machine specific.
Finally, if you have Interrupt Service Routines, asm can make all the difference in the world -- there are certain operations you just don't want occurring under interrupt, and you want your interrupt handlers to "get in and get out fast"... you know almost exactly what's going to happen in your ISR if it's in asm, and it encourages you to keep the bloody things short (which is good practice anyway).
I have only personally talked to one developer about his use of assembly.
He was working on the firmware that dealt with the controls for a portable mp3 player.
Doing the work in assembly had 2 purposes:
Speed: delays needed to be minimal.
Cost: by being minimal with the code, the hardware needed to run it could be slightly less powerful. When mass-producing millions of units, this can add up.
The only assembler coding I continue to do is for embedded hardware with scant resources. As leander mentions, assembly is still well suited to ISRs where the code needs to be fast and well understood.
A secondary reason for me is to keep my knowledge of assembly functional. Being able to examine and understand the steps which the CPU is taking to do my bidding just feels good.
Last time I wrote in assembler was when I could not convince the compiler to generate libc-free, position independent code.
Next time will probably be for the same reason.
Of course, I used to have other reasons.
A lot of people love to denigrate assembly language because they've never learned to code with it and have only vaguely encountered it and it has left them either aghast or somewhat intimidated. True talented programmers will understand that it is senseless to bash C or Assembly because they are complimentary. in fact the advantage of one is the disadvantage of the other. The organized syntaxic rules of C improves clarity but at the same gives up all the power assembly has from being free of any structural rules ! C code instruction are made to create non-blocking code which could be argued forces clarity of programming intent but this is a power loss. In C the compiler will not allow a jump inside an if/elseif/else/end. Or you are not allowed to write two for/end loops on diferent variables that overlap each other, you cannot write self modifying code (or cannot in an seamless easy way), etc.. conventional programmers are spooked by the above, and would have no idea how to even use the power of these approaches as they have been raised to follow conventional rules.
Here is the truth : Today we have machine with the computing power to do much more that the application we use them for but the human brain is too incapable to code them in a rule free coding environment (= assembly) and needs restrictive rules that greatly reduce the spectrum and simplifies coding.
I have myself written code that cannot be written in C code without becoming hugely inefficient because of the above mentionned limitations. And i have not yet talked about speed which most people think is the main reason for writting in assembly, well it is if you mind is limited to thinking in C then you are the slave of you compiler forever. I always thought chess players masters would be ideal assembly programmers while the C programmers just play "Dames".
No longer speed, but Control. Speed will sometimes come from control, but it is the only reason to code in assembly. Every other reason boils down to control (i.e. SSE and other hand optimization, device drivers and device dependent code, etc.).
If I am able to outperform GCC and Visual C++ 2008 (known also as Visual C++ 9.0) then people will be interested in interviewing me about how it is possible.
This is why for the moment I just read things in assembly and just write __asm int 3 when required.
I hope this help...
I've not written in assembly for a few years, but the two reasons I used to were:
The challenge of the thing! I went through a several-month period years
ago when I'd write everything in x86 assembly (the days of DOS and Windows
3.1). It basically taught me a chunk of low level operations, hardware I/O, etc.
For some things it kept size small (again DOS and Windows 3.1 when writing TSRs)
I keep looking at coding assembly again, and it's nothing more than the challenge and joy of the thing. I have no other reason to do so :-)
I once took over a DSP project which the previous programmer had written mostly in assembly code, except for the tone-detection logic which had been written in C, using floating-point (on a fixed-point DSP!). The tone detection logic ran at about 1/20 of real time.
I ended up rewriting almost everything from scratch. Almost everything was in C except for some small interrupt handlers and a few dozen lines of code related to interrupt handling and low-level frequency detection, which runs more than 100x as fast as the old code.
An important thing to bear in mind, I think, is that in many cases, there will be much greater opportunities for speed enhancement with small routines than large ones, especially if hand-written assembler can fit everything in registers but a compiler wouldn't quite manage. If a loop is large enough that it can't keep everything in registers anyway, there's far less opportunity for improvement.
The Dalvik VM that interprets the bytecode for Java applications on Android phones uses assembler for the dispatcher. This movie (about 31 minutes in, but its worth watching the whole movie!) explains how
"there are still cases where a human can do better than a compiler".
I don't, but I've made it a point to at least try, and try hard at some point in the furture (soon hopefully). It can't be a bad thing to get to know more of the low level stuff and how things work behind the scenes when I'm programming in a high level language. Unfortunately time is hard to come by with a full time job as a developer/consultant and a parent. But I will give at go in due time, that's for sure.

Resources