qvector replace unnecessary code

[Hojun-Cho]

I add commit about replaced unnecessary code.

[wolkykim]

I think the behavior of the qvector_reverse() is to reverse the order of the elements, not the byte order inside the elements.

[wolkykim]

My bad. Plz ignore the above comment. I misread.
But performance-wise, how will it compare?

[Hojun-Cho]

don't need call malloc

[wolkykim]

Hi Hojun,

Thanks for your PR.
I like the update of replace for-loop to memcpy.
But have a concern about the 2nd commit of malloc removal as you might be misunderstanding the behavior of the function.
Please take a look and let me know what you think.

[Hojun-Cho]

First, thank you for time. I agree with you.
But I was thinking that it would be better to avoid dynamic allocation. If we can predict that the objsize will not be realistically large, how about use stack?

[wolkykim]

Hi Hojun,

Oh sorry, I misread your code. Yeah, your update doesn't change the behavior.
But still not comfortable with doing for-loop in the byte level.

[Hojun-Cho]

I agree with you. How about discard for-loop commit?

[wolkykim]

I Agree with you. How about discard for-lopp commit?

I'm good with that. We're good to go with 20d7435
Looks like the existing qvector test cases cover the changes but can you double-check?

[wolkykim]

Copying bytes in for loop. It feels that's too much for removing a temporary malloc.

…

On Wed, May 15, 2024, 2:37 AM O_WRONLY ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In src/containers/qvector.c <#114 (comment)>: > - memcpy(tmp, data1, vector->objsize); - memcpy(data1, data2, vector->objsize); - memcpy(data2, tmp, vector->objsize); + for (int k = 0; k < vector->objsize; k++) { don't need call malloc — Reply to this email directly, view it on GitHub <#114 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABUKQMKR7NUZPPXXNIFVL7DZCJDUBAVCNFSM6AAAAABHTMNDCKVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDANJWGAZDQNJWGA> . You are receiving this because you commented.Message ID: ***@***.***>

[Hojun-Cho]

It feels that's too much for removing a
temporary malloc.

I didn't think of that. Thank you for your feedback.

[wolkykim]

Thank you again for a great PR!!!