you need to copy to each memory position once

I don't understand why. it's vector, one contiguous memory block of pod objects. should be memcpyble as I understand (even without allocation if destination vector is big enough)

you want to copy one POD to multiple memory locations to have it in the vector multiple times, right?

you want to copy one POD to multiple memory locations to have it in the vector multiple times, right?

pretty sure he wants to copy a vector (e.g copy construct/assign)

pretty sure he wants to copy a vector (e.g copy construct/assign)

yes

oh

you can do vector::data() and memcpy from that

is that what you want?

is that what you want?

hm, probably, will try it too if std::copy don't help

I don't really know what you want to do

optimize that vector copying, as it looks too expensive in the profiler

copying memory is not fast

Can anyone help the fundamental procedures for declaring functions

optimize that vector copying, as it looks too expensive in the profiler

I assume you're compiling using optimizations?

if you want to speed it up a bit, use DMA

will speed it up, help with heat spread and open the CPU up for something else

Can anyone help the fundamental procedures for declaring functions in cpp

Can anyone help the fundamental procedures for declaring functions in cpp

/warn spam

why would they need help

ah right, I forgot

how come it is a spam 😂 it’s just the silly question

he asked it twice

Can anyone help the fundamental procedures for declaring functions

.

Can anyone help the fundamental procedures for declaring functions in cpp

.

well two messages aren’t spam

he tries very hard asking the question

😉

they are two times the same message

I assume you're compiling using optimizations?

well, now when you said it, I looked at the flags and it seems like I'm not 😅

I mean

well, now when you said it, I looked at the flags and it seems like I'm not 😅

it won't help too much

well, now when you said it, I looked at the flags and it seems like I'm not 😅

Profiling unoptimized code is not a good idea... :)

It depends on what you want

for learning its probably best

well, now when you said it, I looked at the flags and it seems like I'm not 😅

In my case this will be emitted mov rdi, r12 mov rsi, r15 mov rdx, rbx call memcpy

whats in rbx?

number of bytes to copy

no shit

but what does the compiler generate

just for curiosity, no bashing this time 😉

Hi guys! I’m currently working with Verilator (a tool for simulation of circuit boards) and it generates some c++ code. And the thing it does is quite weird and I don’t really get what it is, i mean the VL_Module thing.. what is it? Is this some kind of wrapper around VTop call or something? It includes methods that i need, but they are not really methods, cause they have no return type and parameters are all weird..

looks a lot like a macro

look into "verilated.,h"

you probably have a #define VL_MODULE <something>

yeah

It really is!

Hehe, there you have your answer 😉

In my case this will be emitted mov rdi, r12 mov rsi, r15 mov rdx, rbx call memcpy

For some reason, it doesn't give me better result with -O2, but std::copy gave a huge improvement here (at least I don't see it anymore in the profiler, before that it was 11% of run time)

thanks for the help

Thanks anyways! Now it's a lot clearer!

But still seems kinda strange

Thanks anyways! Now it's a lot clearer!

Sure 😉

But still seems kinda strange

a #define macro just tells the compiler "whenever you see this, replace it with that"

you can collapse bigger codes into just one word

I get it now.. thanks a lot @BinaryByter!

Sure 😊

but what does the compiler generate

basically std::vector<T>::end - std::vector<T>::begin.

okay

thanks 😊

no worries (:

Hi Everyone.

Hi everyone, this is shubham here

Is it so all floats are treated as double by modern compilers ?

Is it so all floats are treated as double by modern compilers ?

Depends on the machine

And compiler

Try to compile a code using sizeof operator and see for yourself what's going on under the hood in a specific machine

then why it is official language of Android?🙄

It is not?

/me wonders why some people typedef existing types xD

/me wonders why some people typedef existing types xD

If you're talking about the fixed width integers then it actually makes a lot of sense

If you're talking about the fixed width integers then it actually makes a lot of sense

Yes. Can you elaborate ?

Yes. Can you elaborate ?

Those typedef's are different per every platform. So a uint32_t might be a unsigned int on one platform, but an unsigned long on another. And with the fixed with integers the standard library implementors set them properly, so you can be assured that it'll be an unsigned integer type with 4bytes

Makes your code less platform dependant

And also easier to read imo

https://en.cppreference.com/w/cpp/types/integer

There you can read more about them

Those typedef's are different per every platform. So a uint32_t might be a unsigned int on one platform, but an unsigned long on another. And with the fixed with integers the standard library implementors set them properly, so you can be assured that it'll be an unsigned integer type with 4bytes

Okey.. so it can be a mitigation to portable fixed width integers.

There you can read more about them

Thanks :)

CTF doesn’t teach you, real wold problems grab some CVE and reverse them and create the exploit . You will see what I am talking about.

... ok fine do you have cves?

Or do you have examples of cves where they just use brute force to completely break aslr, stack cookies, pie

Try to compile a code using sizeof operator and see for yourself what's going on under the hood in a specific machine

Ok tnx