Arm64

Let me try to check

Arm64

Found here https://github.com/torvalds/linux/blob/master/arch/arm64/include/uapi/asm/setup.h#L25

If I use like 95% of the RAM and let's say the RAM is 32 GB there would be still enough left so the OS don't have to use the swap file

But that depends on your swapiness value in the kernel

AFAIK

But that depends on your swapiness value in the kernel

Yes, swap is also optional too.

Mine is 0 so only when it reaches near 32GB it uses it

Hey people. C Language I'm trying to print the first element of and array (arr[0]) but instead is coming the second element (arr[1]). Does anyone know what might be the problem?

Send the whole code

Send the whole code

Already solved the problem. Do you know how to do concatenation of elements of 1 array of hexadecimal

So basically it is possible to prevent the kernel page the memory to the hard disk by doing that?

Of course, there is also OOM killer which force kills your process.

Already solved the problem. Do you know how to do concatenation of elements of 1 array of hexadecimal

To concatenate asa string?

To concatenate asa string?

For example : array[4] = {0x00, 0x00, 0x03, 0x20} //in hexadecimal Concatenate this and get 00320. Then after print that in decimal. 320 in hex = 800 in decimal

Hi. How to synchronize threads using events? In the main main thread, marker stream instances are created. How to send an event from the current thread to the main thread? And then from the main thread terminate the marker stream instance? https://ru.stackoverflow.com/questions/1127135/%d0%9a%d0%b0%d0%ba-%d1%81%d0%b8%d0%bd%d1%85%d1%80%d0%be%d0%bd%d0%b8%d0%b7%d0%b8%d1%80%d0%be%d0%b2%d0%b0%d1%82%d1%8c-%d0%bf%d0%be%d1%82%d0%be%d0%ba%d0%b8-%d1%81-%d0%bf%d0%be%d0%bc%d0%be%d1%89%d1%8c%d1%8e-%d1%81%d0%be%d0%b1%d1%8b%d1%82%d0%b8%d0%b9

lol

Is that option for including bounds checks on`stack arrays`?

So, if I don't write that then that will only work on dynamic arrays?

upar

not my comment, found on stackoverflow

not my comment, found on stackoverflow

https://stackoverflow.com/questions/36997537/valgrind-missing-error/36997685

Thank you

Does the simple valgrind command check leaks only work for heap allocated arrays and not for stack arrays(that's why that option is needed to include stack-arrays too)?

oh, that was for process going outside the heap and stack limits, I thought you were talking about the out-of-bounds access to heap-allocated arrays and stack-allocated arrays.

👍

Found here https://github.com/torvalds/linux/blob/master/arch/arm64/include/uapi/asm/setup.h#L25

It seems the manufacturer of the board I worked on changed it

Pffff the biggest array I've ever created was a dynamic which reserved the whole RAM left without 100 MB for other programs

Oh I was only talking about static ones, when it comes to dynamic most I reached was a few hundred gigabytes (300 something iirc)

(not counting memory mapped storage here because that would get really out of hand quickly)

@nightmareA

@nightmareA

/report

Not directly on Manjaro Linux it is 0 by default 😉

Oh I was only talking about static ones, when it comes to dynamic most I reached was a few hundred gigabytes (300 something iirc)

I was just joking 😉

https://www.instagram.com/p/CATYg1UA92x/?igshid=4nxn1ynaz7z0

Yarr koi mujhe tree aur graph ke videos dee sakta h coding ninza ke plz ...

void some_func(VeryBigClass &tmp){ this.VeryBigClassVal = tmp; } what will happen when i do something like that? Will it copy tmp to VeryBigClassVall or something different will happen

/rules

/rules

Hello guys i am new for c++ i am making a project music library databse and there is a option to play music in external programme Which in this case i use windows mediaplayer / KMPlayer My code for excuting programme with arguments is this was working in my win10 laptop without errors but when i tried my programme on some other computer it says some server error cannot find --play-and-exit-- . NOTE * This part was not implemented by my self but tried googling over and over to find this , i am wonder what was the error that this programme works on me but not on other computerss. And also file paths are re writed according to other computers. const char *appname = "c:\\Program Files (x86)\\Windows Media Player\\wmplayer.exe"; const char *filename = fullpath.c_str(); STARTUPINFOA si; PROCESS_INFORMATION pi; memset(&si, 0, sizeof(si)); si.cb = sizeof(si); memset(&pi, 0, sizeof(pi)); char buf[MAX_PATH + 300]; wsprintfA(buf, "%s \"%s\" --play-and-exit", appname, filename); CreateProcessA(0, buf, NULL, NULL, FALSE, 0, NULL, NULL, &si, &pi);

stdint.h is not available in ANSI C AFAIK :/

Why you would even use this

Or C99...

Or C18

Or C18

Naaa C 18 is too new for me and not supported enough

Why you would even use this

The only variable size types I use is size_t and intptr_t because I really hate variable size induced bug or non-cross-platform bugs

The only variable size types I use is size_t and intptr_t because I really hate variable size induced bug or non-cross-platform bugs

I was talking about ANSI C

I was talking about ANSI C

Exactly why I hate ANSI C boy, it somehow wants you to fail

Exactly why I hate ANSI C boy, it somehow wants you to fail

Yeah stdbool.h is a nice thing

Yeah stdbool.h is a nice thing

It could be... if there was any type safety in C

I don't need type safety I have Rust for that

printf("hello, world!");

void some_func(VeryBigClass &tmp){ this.VeryBigClassVal = tmp; } what will happen when i do something like that? Will it copy tmp to VeryBigClassVall or something different will happen

Copy will happen

The only variable size types I use is size_t and intptr_t because I really hate variable size induced bug or non-cross-platform bugs

Why intptr_t and not uintptr_t?

intptr_t is bogus imo. Why would I want to store address as a signed integer?

intptr_t is bogus imo. Why would I want to store address as a signed integer?

You are totally right (but I don't do operations on them anyway) but a very valid usecase is pointer subtraction to get byte size. A negative result means that the operands are in wrong order, a positive that they are in correct order, and clearing the sign bit (on the architectures where it is possible) gives you the result whichever they are

Legit

void some_func(VeryBigClass &tmp){ this.VeryBigClassVal = tmp; } what will happen when i do something like that? Will it copy tmp to VeryBigClassVall or something different will happen

Unless VeryBigClassVal is, for example, an std::reference_wrapper<VeryBigClass>, it will copy

Unless VeryBigClassVal is, for example, an std::reference_wrapper<VeryBigClass>, it will copy

But std::reference_wrapper the urumi sword of C++, don't use it if you don't know what you are doing because it could come and bite you back in very confusing ways

Casting to intptr_t is semantically correct, putting the difference in a ptrdiff_t also is, but converting a pointer to a ptrdiff_t is not semantically correct

You are, the idea is to obtain an explicit memory address instead or a typed memory address (also void* is cursed)

Hence I am storing a memory address, not a pointer

I write code that may be used on avr CPUs just as much as on x86_64 so I tend to write things as in line with the standard as I can and leave as little undefined as possible

i need a tutorial or document about mariadb connection with c programming language? If anyone help me, i will be happy

/get ide

/get cbook

you create the filter with (for example): gp::quotient_filter<uint16_t, 10,6,stepcache_skipstone>

It is built in the code of the skipstone itself

size_t ret = 0; void next() { ++ret; } operator size_t() { [[unlikely]] if ( ret >= 128 ) return ret+list[127]; return list[ret]; }

when ret is above 128, it takes the last tested element+128, then the next, then the next etc etc

0, 1, 2, 3, those are the first offsets from the original hash position that are explored

all of those are already in L1 as cache line length is assumed to be 64bytes

then we skip 4 offsets and go to 8, 9, 10, 11,

another probable cache line

so for 8 attempts, we hit the ram twice

That is if you assume you are reading 32 bits integer

that skipstone is for 64bit integers

yes

The reason I went with the 64bit one first in the lib is because it is generic and I plan to move them out of here to use them more generically

It is not only posix but for OS specifics, it assumes posix

Most of it is freestanding, so OS-less

This quotient filter for example can be used on an arduino board without an OS

I tried to be in between cache friendliness and cluster avoidance on that one, so for a fair evaluation, I advise to try at multiple degrees of "fullness", like 10% filled, 25% filled, 50% filled, 75% filled, 90% filled and 95% filled