It may be missed optimization

I noted that MSVC generates different disassembly for these two cases: https://godbolt.org/z/bsn93G As far as I know conditional jumps are bad for branch predictor and unnecessary conditional jumps are better to be avoided. In this case I expected that the compiler will not do short-circuit evaluation because it's only float comparisons (shouldn't they be simple enough?). Can someone explain why the compiler didn't get rid of the jumps and didn't make it the same with the second variant? Does the second variant have some disadvantages?

Why shouldnt be different? One of them is “and” other one is “binary and” operator. There is no shortcut evaluation for binary operator.

Why shouldnt be different? One of them is “and” other one is “binary and” operator. There is no shortcut evaluation for binary operator.

They produce the same behaviour with no side effects involved. The code is rather simple. I mean, why the compiler not generated the most efficient version in both cases.

They produce the same behaviour with no side effects involved. The code is rather simple. I mean, why the compiler not generated the most efficient version in both cases.

because bitwise AND works on integral types and outputs an integral type. logical AND works on bool and outputs a bool

Hello

because bitwise AND works on integral types and outputs an integral type. logical AND works on bool and outputs a bool

I guess for assembly bool is also an integral type

I guess for assembly bool is also an integral type

precisely

so we could have a logical AND operator that behaved like bitwise AND. that would be inefficient, so the standard has short-circuit evaluation for logical AND.

Well, so that means I need to use bitwise AND here to make it explicit that I don't care about short-circuit evaluation?

I mean, is the second variant preferable then?

Well, so that means I need to use bitwise AND here to make it explicit that I don't care about short-circuit evaluation?

no. you should use logical AND here because once again—bitwise AND works on integral types, logical AND works on bools

I guess for assembly bool is also an integral type

you almost got the point here.

I guess for assembly bool is also an integral type

but integral types are not bool

Yep, my point is, that it seems to me (but I'm not sure) that the assembly code generated in the second case (with bitwise AND) is more efficient because it doesn't have branching (conditional jumps) in it.

Yep, my point is, that it seems to me (but I'm not sure) that the assembly code generated in the second case (with bitwise AND) is more efficient because it doesn't have branching (conditional jumps) in it.

i think you should benchmark that

i think you should benchmark that

Makes sense

Yep, my point is, that it seems to me (but I'm not sure) that the assembly code generated in the second case (with bitwise AND) is more efficient because it doesn't have branching (conditional jumps) in it.

unless you can guarantee that the three short circuit jumps will fail (in which case just have the fourth check only?) i think the logical operator will be faster

Here I pasted it https://pastebin.com/dyAzW4yd

because it doesn't have conditional jumps in it, and as I heard they bad for branch predictor

Actually I started thinking about that when I found this in UE4 core code checkSlow((ArrayNum >= 0) & (ArrayMax >= ArrayNum)); // & for one branch But in that case as far as I remember assembly code was generated almost identically, so I thought it is some old stuff from the days when compilers wasn't so smart

Btw are there ppl in here who prefer procedural way of handling solutions.. Instead of object oriented ways.. Even while using C++??

Or did I frame the question in a wrong way?

OK, I think I got it, thanks for the explanation @lightness_races_in_orbit, @chandradeepdey

Definitely it would seem it's better to approach some questions in a certain way.. Coz it'd be easier or efficient in the longer run..

Welp.. Maybe I'll ask that question once I dedicate myself to solve problems in a procedural way and then.. After some time.. I'll try doing that in object oriented way..

I guess for assembly bool is also an integral type

Also, I am not sure that a comparison of two float point in this way, might cause such problems without epsilon value that you have to define.

Btw are there ppl in here who prefer procedural way of handling solutions.. Instead of object oriented ways.. Even while using C++??

OOP is not main paradigm in C++

It's not C with classes

In the realm of game dev.. Do ppl prefer OOP as the main paradigm?

In the realm of game dev.. Do ppl prefer OOP as the main paradigm?

As far as I've seen, OOP (well, depends on what meaning you put in it) is used everywhere in the high-level design, more procedural style with Data-oriented design (DOD) approaches used in low-level code.

/adminlist

Heya

Heya

heya!

As far as I've seen, OOP (well, depends on what meaning you put in it) is used everywhere in the high-level design, more procedural style with Data-oriented design (DOD) approaches used in low-level code.

K gotcha

who is owner of this big community🤩

who is owner of this big community🤩

A guy named Mester

A guy named Mester

i thought dima is owner

He went MIA couple of years ago

He went MIA couple of years ago

oh

He went MIA couple of years ago

MIA?

Missing in attendance

Maybe he found something better

Group is full of serious and Focused peps

Yeah barely 7% are alive online.. And from those 7% .. Barely 1% are active

Group is full of serious and Focused peps

Well, it is a c and c++ group, we are only here when we make clean & make -B

Well, it is a c and c++ group, we are only here when we make clean & make -B

😁

such a suspicious name

if I pass a 2d array as an argument to a c++ function like this - void f(char **A), then within the function, can I access A[0][0] as the 0th row and 0th column of char array A?

Yes

okay. thanks!

/get cbook

/get

/get cbook

/get cppbookguide

Thanks !

Thanks you 🙏

Heya😁 Can anyone help? I need to copy the behavior of strlcat But i dont understand how exactly it works. Can anyone enlighten ?

Tq

I noted that MSVC generates different disassembly for these two cases: https://godbolt.org/z/bsn93G As far as I know conditional jumps are bad for branch predictor and unnecessary conditional jumps are better to be avoided. In this case I expected that the compiler will not do short-circuit evaluation because it's only float comparisons (shouldn't they be simple enough?). Can someone explain why the compiler didn't get rid of the jumps and didn't make it the same with the second variant? Does the second variant have some disadvantages?

Update: Benchmarked it with random data. https://ideone.com/CtLqwC The second variant (with bitwise AND) is always faster. The variants get closer in time if the data is prepared in a way when the check always fails or when the check always succeeds. That looks exactly like a branch predictor thing. In the worst case, the difference is more than twice, and surprisingly the worst case happens when data is prepared in a way when the check fails often than succeeds (succeeds in 11% cases). These should be the cases when short-circuit evaluation should stop early more often. So it doesn't look like float comparison is somewhat heavy compared to branch misprediction. Got similar results on MSVC @lightness_races_in_orbit, @chandradeepdey, poke you as you may be interested.

What projects you done?

Heya😁 Can anyone help? I need to copy the behavior of strlcat But i dont understand how exactly it works. Can anyone enlighten ?

https://opensource.apple.com/source/Libc/Libc-825.25/string/strlcat.c.auto.html

I dont get what to do if a get an empty array. It says both dest and src should be terminated. How do i do that?

What projects you done?

Game, chat, website etc

Game, chat, website etc

In which language you create game

That was random DST

I asked a question for us: what projects have you done?

Us?

That was random DST

Yeah

accidentally answered the wrong message

I want to see what people done on c++

You are given 3 integers a,b &c. Can bitwise OR of (b and c )can be calculated if you are provided bitwise OR of (a and b) and(a,b and c) ?

I want to see what people done on c++

Chromium, Tensorflow, LLVM are made with C++

Chromium, Tensorflow, LLVM are made with C++

GTAV, RedDeadRedemption2, Unity, UnrealEngine, Cryengine, Fortnite, FarCry, CounterStrike all C++

Nono, what only one programmer done

Nono, what only one programmer done

Alone programmer is useless

Even alone you can write an engine or game or AI

Nono, what only one programmer done

You want source code?

Or what