|
pram posted:doesn't it just do it through mdadm no. it has its own drivers. and i don't think it supports hardware write barriers
|
# ? Nov 12, 2015 22:06 |
|
|
# ? Apr 26, 2024 21:04 |
|
lvm seems kinda lovely idk but then i don't have EnTeRpRiSe StOrAgE rEqUiReMeNtS i just have a laptop ssd with an EFI system partition and an ext4 partition
|
# ? Nov 12, 2015 22:09 |
|
mdadm uses raid from kernel, and lvm uses device-mapper from kernel. they do share a lot of the algorithms though, so e.g. the raid logic implementation is the same. they're configured in the same section in kernel menuconfig.
|
# ? Nov 12, 2015 22:13 |
|
lvm supports thin provisioning and trim/discards, that's nice.
|
# ? Nov 12, 2015 22:15 |
|
Tankakern posted:mdadm uses raid from kernel, and lvm uses device-mapper from kernel. they do share a lot of the algorithms though, so e.g. the raid logic implementation is the same. they're configured in the same section in kernel menuconfig. ahh thx
|
# ? Nov 12, 2015 23:04 |
|
Mr Dog posted:lvm seems kinda lovely idk lvm is really good and cool even if you don't need any enterprisey poo poo e.g. volume snapshots are really good for backups
|
# ? Nov 12, 2015 23:06 |
|
Tankakern posted:mdadm uses raid from kernel, and lvm uses device-mapper from kernel. they do share a lot of the algorithms though, so e.g. the raid logic implementation is the same. they're configured in the same section in kernel menuconfig. hardware write barriers are the important part, and i don't think lvm will enable write barriers except in the trivial case (pass-through on a single device)
|
# ? Nov 12, 2015 23:07 |
|
I'm pretty sure you have to look pretty hard to find a case where LVM does not have write barriers... support for it in device mapper was added in 2010 (2.6.33)
|
# ? Nov 12, 2015 23:50 |
|
easy to test though, just look in dmesg after mounting xfs or ext4 or whatever on lvm with -o barrier=1. Sometimes you have to check if lvm.conf is correctly set up too (e.g. you have to explicitly enable discards/trim in lvm.conf if you want to send those to the device(s)).
|
# ? Nov 12, 2015 23:52 |
|
i wish there was lvm for windows
|
# ? Nov 13, 2015 00:25 |
|
What is with the blote on packages nowadays? To take a smaller package for example, ncurses, took 1 min 4 seconds to compile on Oct 2014 with an i5 2500k, now takes 1 min 20 seconds to compile today on an i7 4790k. Do developers just not give a poo poo?
|
# ? Nov 13, 2015 02:11 |
|
nosl posted:compile
|
# ? Nov 13, 2015 02:16 |
|
nosl posted:What is with the blote on packages nowadays? lol
|
# ? Nov 13, 2015 02:18 |
|
MY COMPILE TIMES
|
# ? Nov 13, 2015 02:18 |
|
Thanks, I almost forgot where I was posting.
|
# ? Nov 13, 2015 02:25 |
|
what if that extra 16 seconds it took to compile was because it was optimizing ncurses to run 16 seconds faster? makes u think
|
# ? Nov 13, 2015 02:28 |
|
2015 year of linux on the desktop of linux system administrators
|
# ? Nov 13, 2015 02:28 |
|
actually its osx
|
# ? Nov 13, 2015 02:31 |
|
Barnyard Protein posted:what if that extra 16 seconds it took to compile was because it was optimizing ncurses to run 16 seconds faster? 16 extra seconds even though it compiled 16 seconds faster on a 2500k than it did a year later on a 4790k was the point how do you manage to manage servers while missing blatantly obvious details like that makes u think
|
# ? Nov 13, 2015 02:33 |
|
|
# ? Nov 13, 2015 02:36 |
|
Barnyard Protein posted:what if that extra 16 seconds it took to compile was because it was optimizing ncurses to run 16 seconds faster? nah, it's definitely not that compilers are doing increasingly more and better optimizations. also can't be that ncurses is more capable. couldn't possibly be. obviously, things were just so much better back in the day
|
# ? Nov 13, 2015 02:36 |
|
im trying to remember the last time i had to compile source code and i think it was in 2001 to get mozilla to run on solaris 8 x86
|
# ? Nov 13, 2015 02:39 |
|
unnecessarily compiling bullshit is the unix way
|
# ? Nov 13, 2015 02:41 |
|
I'm just spergin on Gentoo because it feels like a viable home use hobbyist distro, also capable of supporting my sporadic dev needs while not being a piece of broken, vulnerable poo poo maintained by hipsters. I've used this distro for half a decade distro now, the baby duck is real.
|
# ? Nov 13, 2015 02:43 |
|
yeah man, hipsters are the bane of any linux distro
|
# ? Nov 13, 2015 02:43 |
|
Barnyard Protein posted:im trying to remember the last time i had to compile source code and i think it was in 2001 to get mozilla to run on solaris 8 x86 i compiled a hosed up version of KDEN Live from source and it was epic ftw
|
# ? Nov 13, 2015 02:47 |
|
spergin is great. srsly though i think that kind of time delta could be expected to be seen on the same system doing multiple back-to-back compiles. it'll change depending on the load of the system and the phase of the moon and stuff. i get paid earth dollars to write software, so i'm hitting the compile button all day long. my current project builds 30s+/-5s.
|
# ? Nov 13, 2015 02:47 |
|
i actually had to custom compile ncurses last night to get utf8 support for nano on my synology i think it took around who-gives-a-gently caress minutes but i didnt grab the output
|
# ? Nov 13, 2015 02:50 |
|
nosl posted:I'm just spergin on Gentoo because it feels like a viable home use hobbyist distro, also capable of supporting my sporadic dev needs while not being a piece of broken, vulnerable poo poo maintained by hipsters. I've used this distro for half a decade distro now, the baby duck is real. post your sikk compile flags itt --funroll-loops etc
|
# ? Nov 13, 2015 02:52 |
|
16 seconds is there nothing else in your life
|
# ? Nov 13, 2015 02:53 |
|
bobbilljim posted:16 seconds
|
# ? Nov 13, 2015 02:55 |
|
this morning i exported a nfs mount on the NAS and mounted it on my ~*linux box*~ and tried to test it with dd if=/dev/urandom of=./test.dat count=500 size=500k and about 3 seconds later my machine no longer responds to pings hoping i dont come home to a house on fire
|
# ? Nov 13, 2015 02:56 |
|
bobbilljim posted:16 seconds my compile flags are quite normal and not riced I don't often do a system update unless it's been a few months, and just upgrade important packages until then. The only reason I checked the difference was because I installed a brand new processor that is supposedly twice as fast at compiling. While this is apparent for larger packages blatantly, smaller ones seem to have no major difference. I wasn't trying to divert the ambiguous nonexistent topic, sorry for ruining your day man. Next time I'll just cut my dick off and install Ubuntu while listening to boy bands through mpd before I even consider posting something related to linux on the linux thread.
|
# ? Nov 13, 2015 02:56 |
|
|
# ? Nov 13, 2015 02:56 |
|
nosl posted:my compile flags are quite normal and not riced wait you were actually serious about "compile time bloat" lol just fuckin lol
|
# ? Nov 13, 2015 02:57 |
|
gentoo is okay to use for about 2 years maximum at which point youve learned all you will ever learn from a lovely compiled system and its time to move on to something that doesnt actively waste your time for no gain whatsoever, like forums posting or reading NBSD posts
|
# ? Nov 13, 2015 02:58 |
|
Use arch If you must be a sperg then at least don't pointlessly waste your time in the process
|
# ? Nov 13, 2015 03:28 |
|
nosl posted:my compile flags are quite normal and not riced im going to post at u in all seriousness. theres many factors that contribute to compile speed. mostly disk in my experience, which could be the bottleneck in a small package like that. as other posters have mentioned, it might now run faster or have more features. you can understand though, right, that noone gives a poo poo abotu waiting an extra 16 seconds. finally, if you want to compile poo poo arch linux is a good distro. there's no reason to use gentoo
|
# ? Nov 13, 2015 03:30 |
|
im compiling. a collection of your moms pubes!!
|
# ? Nov 13, 2015 03:38 |
|
|
# ? Apr 26, 2024 21:04 |
|
thats gross lol
|
# ? Nov 13, 2015 03:39 |