I use pv(1) all the time. Not a standard command, but everyone should install it and start using it. It replaces dd for most use cases and Just Does The Right Thing by default. It's part of my required packages list on every system, along with strace, htop, iotop, and vim.https://twitter.com/Yeah1Whatever/status/991568978479910913 …
-
WHY THE FUCK HAVE I BEEN DOING THIS FOR SO LONG: dd if=/dev/bullshit1 | pv | dd of=/dev/bullshit2 I SAW THAT RECOMMENDED SO MANY PLACES. PV TOTALLY REPLACES DD? WHY IN THE HELL WOULD SOMEONE RECOMMEND THIS
You can just use cat (or nothing if just redirecting around another command). dd is total and utter cargo cult unless you need the seek/skip options. And even for that it's crap, because there's no reason for seek/skip to be in units of the transfer blocksize. And it's slow.
-
-
dd started as a joke about the syntax of IBM mainframe JCL program definitions (hence the silly a=b format). It became obsolete on OSes like Linux where block devices can be accessed byte-wise just fine and behave totally normally. But everyone still thinks "dd for blockdevs".
-
knowing this makes me so upset on so many levels.
- Show replies
New conversation -
-
-
Is there a faster alternative for offset? TBH it is slow but it's 30s Vs 20s on an SSD.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I'm using dd because it allows me to set sync options so I get real throughput info instead of how fast my cache in memory is. Can this easily be done directly with pv as well?
-
No, but direct I/O is a bad idea anyway, because it isn't actually how filesystems are used in practice. If you want to do a cache-cold test then just drop caches with echo 3 > /proc/sys/vm/drop_caches.
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.