Most active commenters
  • grandempire(9)
  • pjmlp(9)

←back to thread

Hacktical C: practical hacker's guide to the C programming language

(github.com)
218 points signa11 | 35 comments | 14 Apr 25 10:20 UTC | HN request time: 0.454s | source | bottom
Show context
pjmlp ◴[14 Apr 25 13:43 UTC] No.43681194[source]▶
> The reason I believe C is and always will be important is that it stands in a class of its own as a mostly portable assembler language, offering similar levels of freedom.

When your computer is a PDP-11, otherwise it is a high level systems language like any other.

replies(7): >>43682173 #>>43682246 #>>43682554 #>>43682928 #>>43683332 #>>43683914 #>>43701871 #
1. grandempire ◴[14 Apr 25 15:17 UTC] No.43682246[source]▶
Which other popular language more accurately represents a random access machine of fixed word length?
replies(3): >>43682309 #>>43682371 #>>43682763 #
2. alexvitkov ◴[14 Apr 25 15:22 UTC] No.43682309[source]▶
None, but that'a not what computers are. C assumes that in a few places, e.g. variadic functions, and those are the worst parts of the language.
replies(1): >>43682322 #
3. grandempire ◴[14 Apr 25 15:23 UTC] No.43682322[source]▶
> but that'a not what computers are

Which language more accurately represents hardware then?

replies(3): >>43682445 #>>43682466 #>>43682491 #
4. pjmlp ◴[14 Apr 25 15:28 UTC] No.43682371[source]▶
I don't know, Ada, Modula-2, Object Pascal, PL/I, NEWP, PL.8, D, Zig, Mesa, ATS,....

But then again, you booby trapped the question with popular language.

replies(2): >>43682583 #>>43682636 #
5. pjmlp ◴[14 Apr 25 15:35 UTC] No.43682466{3}[source]▶
Assembly language from the hardware vendor.
replies(1): >>43682533 #
6. alexvitkov ◴[14 Apr 25 15:36 UTC] No.43682491{3}[source]▶
C++ for one - it has atomics with well defined memory barriers, and guarentees for what happens around them.

The real answer is obviously Assembly - pick a random instruction from any random modern CPU and I'd wager there's a 95% chance it's something you can't express in C at all. If the goal is to model hardware (it's not), it's doing a terrible job.

replies(2): >>43682642 #>>43683141 #
7. oguz-ismail ◴[14 Apr 25 15:40 UTC] No.43682533{4}[source]▶
isn't it translated to microcode before being executed?
replies(1): >>43682997 #
8. guywithahat ◴[14 Apr 25 15:43 UTC] No.43682583[source]▶
If a language is unpopular, people won't want to work for you and you'll run into poor support. Rewriting a library may take months of dev time, whereas C has an infinite number of libraries to work with and examples to look at.
replies(2): >>43682910 #>>43701784 #
9. grandempire ◴[14 Apr 25 15:49 UTC] No.43682636[source]▶
Many of those languages do not have pointers - which are fundamental to how modern instruction sets work.
replies(1): >>43682903 #
10. sugrado ◴[14 Apr 25 15:49 UTC] No.43682638{4}[source]▶
bro just quoted a chatbot
11. grandempire ◴[14 Apr 25 15:49 UTC] No.43682642{4}[source]▶
C++ better represents the machine?
12. kryptiskt ◴[14 Apr 25 16:01 UTC] No.43682763[source]▶
C lacks sympathy with nearly all additions to hardware capabilities since the late 80s. And it's only with the addition of atomics that it earns the qualification of "nearly". The only thing that makes it appear as lower level than other languages is the lack of high-level abstraction capabilities, not any special affinity for the hardware.

For one, would expect that a low level language wouldn't be so completely worthless at bit twiddling. Another thing, if C is so low level, why can't I define a new calling convention optimized for my use case? Why doesn't C have a rich library for working with SIMD types that has been ubiquitous in processors for 25 years?

replies(4): >>43682820 #>>43682884 #>>43683999 #>>43684237 #
13. kryptiskt ◴[14 Apr 25 16:05 UTC] No.43682820[source]▶
Like, say I have a data structure that is four bits wide (consisting of a couple of flags or something) and I want to make an array of them and access them randomly. What help do I get from C to do this? C says "fuck you".
replies(2): >>43684008 #>>43684308 #
14. grandempire ◴[14 Apr 25 16:12 UTC] No.43682884[source]▶
It also has pointers which are absent from most languages but essential to instruction sets.
replies(1): >>43682988 #
15. pjmlp ◴[14 Apr 25 16:15 UTC] No.43682903{3}[source]▶
Yes they do, point an example from that group, and I will gladly prove you wrong.
replies(2): >>43682999 #>>43683399 #
16. pjmlp ◴[14 Apr 25 16:15 UTC] No.43682910{3}[source]▶
Moving goalposts regarding systems programming languages features, some on the group predate C by a decade.
replies(1): >>43683148 #
17. pjmlp ◴[14 Apr 25 16:22 UTC] No.43682988{3}[source]▶
Lots of languages since the 1950's have pointers.
18. pjmlp ◴[14 Apr 25 16:22 UTC] No.43682997{5}[source]▶
Depends on the hardware design.
19. ◴[14 Apr 25 16:23 UTC] No.43682999{4}[source]▶
20. uecker ◴[14 Apr 25 16:34 UTC] No.43683141{4}[source]▶
C has the same atomics and concurrency model as C++.
21. guywithahat ◴[14 Apr 25 16:35 UTC] No.43683148{4}[source]▶
Being old doesn't mean anyone knows the language. I mean if the language predates C significantly and nobody uses is then there's probably a really good for it. The goalposts aren't moving they're just missing the shot
replies(1): >>43684997 #
22. grandempire ◴[14 Apr 25 17:00 UTC] No.43683399{4}[source]▶
Well sounds like you are confident and we are going to get into a semantic argument about what qualifies as a pointer.

So which of these languages do you think is a better representation of hardware and not a PDP-11?

replies(1): >>43689392 #
23. codr7 ◴[14 Apr 25 17:42 UTC] No.43683999[source]▶
It puts less obstacles in the way of dealing with hardware than almost any other language for sure.

What's standardized was never as important in C land, at least traditionally, which I guess partly explains why it's trailing so far behind. But the stability of the language is also one of its features.

24. codr7 ◴[14 Apr 25 17:43 UTC] No.43684008{3}[source]▶
Only if you don't know C.

Otherwise is says, do whatever you feel like.

25. cv5005 ◴[14 Apr 25 17:59 UTC] No.43684237[source]▶
simd doesnt make much sense as a standard feature/library for a general purpose language. If you're doing simd its because you're doing something particular for a particular machine and you want to leverage platform specific instructions, so thats why intrinsics (or hell, even externally linked blobs written in asm) is the way to go and C supports that just fine.

But sure, if all youre doing is dot products I guess you can write a standard function that will work on most simd platforms, but who cares, use a linalg library instead.

26. cv5005 ◴[14 Apr 25 18:05 UTC] No.43684308{3}[source]▶
Pick an appropriate base type (uintN_t) for a bitset, make an array of those (K * N/4) and write a couple inline functions or macros to set and clear those bits.
27. OCASMv2 ◴[14 Apr 25 19:12 UTC] No.43684997{5}[source]▶
Popularity isn't a measure of quality. Never has been and certainly not in the case of programming languages.
replies(1): >>43685645 #
28. grandempire ◴[14 Apr 25 19:59 UTC] No.43685645{6}[source]▶
There is unpopular - and then there is can I get a working toolchain for modern OS that’s not emulated.
replies(1): >>43686764 #
29. OCASMv2 ◴[14 Apr 25 21:57 UTC] No.43686764{7}[source]▶
Still not a measure of quality.
replies(1): >>43687227 #
30. grandempire ◴[14 Apr 25 23:00 UTC] No.43687227{8}[source]▶
Are we having a discussion about the greatest language of all time? What’s your context here.
31. pjmlp ◴[15 Apr 25 05:53 UTC] No.43689392{5}[source]▶
Better representation of the hardware?

None of them, you use Assembly if you want the better representation of hardware.

Yes, I am quite confident, because I have been dispelling the C myth of the true and only systems programming language since the 1990's.

replies(1): >>43695517 #
32. grandempire ◴[15 Apr 25 16:56 UTC] No.43695517{6}[source]▶
So then your comment about C being an outdated PDP-11 must be equally true of other languages. So it says nothing.
replies(1): >>43703827 #
33. ryao ◴[16 Apr 25 05:24 UTC] No.43701784{3}[source]▶
wears math hat

C does not have an infinite number of libraries and examples. The number of libraries and examples C has is quite large, and there are an infinite number of theoretically possible libraries and examples, but the number of libraries and examples that exist are finite.

replies(1): >>43711414 #
34. pjmlp ◴[16 Apr 25 11:02 UTC] No.43703827{7}[source]▶
Not really, some of those languages predate the very existence of C and PDP-11.
35. Koshkin ◴[16 Apr 25 23:26 UTC] No.43711414{4}[source]▶
The infinite is a convenient abstraction of the finite.