What's the copy algorithm, Sup Forums?

What's the copy algorithm, Sup Forums?

Other urls found in this thread:

coreutils.sourcearchive.com/documentation/7.4/cp_8c-source.html
twitter.com/NSFWRedditImage

Dont know

while queue not empty:
pop
copy

Apparently thermal throttling is the answer.

>2016+1
>Still no proper DMA on x64
IBM is laughing.

void copy[file from[], file to[]] {
int i;
while((to[i] = from[i]) != NULL)
++i;
}

mov ecx,[len]
mov esi,source
mov edi,destination
rep movsd


?

Why not
while((to[i++] = from[i++] != NULL)
;

Why would you increase i twice? wouldn't it skip every other item in to and from?

i is incremented twice here
to[0] = from[1]
to[2] = from[3]

I must be retarded. Thought it was in K&R that way

run it twice offset by 1 second time and have from = from - 1 at the top

ld hl, source
ld de, destination
ldir

fug i forgot
ld bc, len

>C-turds actually believe you copy byte by byte using the CPU

Figures.

for file in list:
Enqueue read operation.
Enqueue write operation.

with open(destination, users_selected_write_method) as dest:
for item in copy_queue:
dest.write(item)

>Windows
>Anime in 720p, instead of glorious 1080p

That doujin herneated my scrotum

>watching upscales

>pop
>LIFO
>2017

good luck with your race conditions

install teracopy

>Not watching the original broadcast
>Not buying the BDs
>Not Watching at 1552x873

You're just watching downscales of upscales, so you're winning either.

Well, yeah.

How else?

DMA.

by randomly generating data and checking the hash against whatever was originally being copied until they match obviously.

It copies blocks to another location.

I'm definitely not paying 40 dollars for 2 seasons of some anime when I can watch it for free online

It's not byte by byte. Not even a good memcpy is byte by byte. They go off the size of your register so likely it's copying 64-bits at a time or whatever the width of your register is.

Doesn't matter. Copying data from any peripheral directly using the CPU is fucking retarded.

More like 90$ for one season, more for bigger shows.

There is some source anime in 1080p and even 4k, so it's not all upscales.


Though 90% are

coreutils.sourcearchive.com/documentation/7.4/cp_8c-source.html

this is how gnutards do it.

W-w-what doujin is it

shit

you'd think microshaft would eventually fix shit
even stuff like robocopy still is a total piece of shit compared to rsync, I mean for one not even checking the data it writes

I assume that start at 2.5GB/s has something to do with caching or some shit, but 450MB/s is decent/normal for what I'd expect from SATA SSD real-world performance, seems to work fine.

ctrl+c
ctrl+v

microshit copy is pretty shit. It has a few better ones but require command line. Even then it doesn't compare to third party tools made just for copying/moving files as fast as possible.

>some anime in 4k

Name one that's out and isn't Thunderbolt.

Why/how is it caching gigs worth of data of that file?

you meant
while((to[i++] = from[i] != NULL)

While not done:
Read from source
Write to destination

>executing instructions using the CPU is retarded
incredible argument; i eagerly await your revolutionary computer specifications. what journal can i expect them to be published in?

>name anime that is 4k source besides the anime that is 4k source
lol good job

s-source?

As you may already know harddrives and SSDs have their own microcontroller so copying or moving a file is handled by that, issued by the CPU of course. You essentially tell it to move blocks around and it does it for you. Eliminating any need to transfer that memory into memory/cache and then back into the harddrive or ssd.

>720p
>FLAC
jesus christ, maybe audio and video size are the same

it still goes into cache, faggot

...

While (!done) {
transfer_speed = math.rand()
}

[spoiler]Lady Yupiel[/spoiler]

Pajeet spaghetti code no doubt

memcpy()

what exactly are the axis on that graph supposed to represent?

rate over time

then it's probably something to do with how the OS is choosing to schedule the thread

right, so you don't understand how instructions are executed by the cpu
take a 200-level CS hardware course before daring to reply to my posts again

while (*dst++ = *src++);

>copy algorithm
>try DMA

>set up transaction to copy from START_HDD to END_HDD to destination START_MEMORY to END_MEMORY.
>???
>profit

Damn fucking brainlet

>pop
>not dup