/g/ - syscall of the day: (f)utime(n)(s)(at) - Technology


08/21/20	New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17	New trial board added: /bant/ - International/Random
10/04/16	New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]

Anonymous
syscall of the day: (f)utime(n(...) 04/03/26(Fri)07:48:26 No.108515827

File: 1624744725170.jpg (203 KB, 832x1200)

syscall of the day: (f)utime(n)(s)(at) Anonymous 04/03/26(Fri)07:48:26 No.108515827

previous: >>108507823
#define __NR_utime                132
#define __NR_utimes                    235
#define __NR_futimesat                261
#define __NR_utimensat                280
https://man7.org/linux/man-pages/man2/utimes.2.html
https://man7.org/linux/man-pages/man2/futimesat.2.html
https://man7.org/linux/man-pages/man2/utimensat.2.html

tl;dr:
manage file timestamps

god, four fucking syscalls for this, lol. at this point it's honestly just comical.
this is interesting, at least
>Linux does not allow ... setting the timestamps to something other than the current time on an append-only file.
and this is just plain absurd
>If both tv_nsec fields are specified as UTIME_OMIT, then the Linux implementation of utimensat() succeeds even if the file referred to by dirfd and path does not exist.
like, lmfao
and all this for what? to change file access timestamps??? fucking why? who even wants to do this? the more i go through all of these syscalls, the more i have to question just what the fuck the linux devs were smoking

relevant resources:
man man
man syscalls
https://man7.org/linux/man-pages/
https://linux.die.net/man/
https://elixir.bootlin.com/linux/
https://elixir.bootlin.com/musl/
https://elixir.bootlin.com/glibc/

Anonymous
04/03/26(Fri)08:53:32 No.108516217

Anonymous 04/03/26(Fri)08:53:32 No.108516217

>god, four fucking syscalls for this, lol. at this point it's honestly just comical.
When I do stuff like rsyncing my codebase, sometime the build system would sperg out when the timestamp of the of the sources was updated, triggering full rebuilds at every invocation.
Especially for big codebases, I see why you would want to update them all efficiently.
also, I usually set noatime for all my filesystems. I don't know why atime is useful, but I'm sure there's a usecase, so having options to set it only when needed seems useful to me

Anonymous
04/03/26(Fri)10:42:33 No.108516835

Anonymous 04/03/26(Fri)10:42:33 No.108516835

bampu

Anonymous
04/03/26(Fri)11:08:54 No.108517037

Anonymous 04/03/26(Fri)11:08:54 No.108517037

>>108515827
>the more i go through all of these syscalls, the more i have to question just what the fuck the linux devs were smoking
Is this specifically a linux thing or a posix thing?

Anonymous
04/03/26(Fri)17:29:59 No.108519835

Anonymous 04/03/26(Fri)17:29:59 No.108519835

>>108517037
i guess posix, probably

Anonymous
04/03/26(Fri)17:40:29 No.108519905

Anonymous 04/03/26(Fri)17:40:29 No.108519905

>>108515827
love u syscall anon

Anonymous
04/03/26(Fri)18:52:56 No.108520371

Anonymous 04/03/26(Fri)18:52:56 No.108520371

>>108519905
l-love you too....

Anonymous
04/03/26(Fri)19:31:00 No.108520569

Anonymous 04/03/26(Fri)19:31:00 No.108520569

What is a system call?

Anonymous
04/03/26(Fri)19:41:14 No.108520634

Anonymous 04/03/26(Fri)19:41:14 No.108520634

>>108520569
lowest level of the operating system's user mode accessible API, how you delegate function calls between kernel mode and user mode
windows doesn't really call them in most documentation, it calls its syscalls the NT API, because they do things a bit different, especially with regards to access and interface stability as well as context switching since some of their system calls don't switch modes and are instead directly accessible, clones of the ones that do, meant for things like user mode drivers or NT API applications

Name
Options
Comment
Verification	4chan Pass users can bypass this verification. [Learn More] [Login]
File
Please read the Rules and FAQ before posting. You may highlight syntax and preserve whitespace by using [code] tags.