Nine minutes to copy a 2.2 gigabyte file tracks to about 4MB/s which
wouldn't surprise me for a cheap USB memory stick or using a slowish USB
port or copying from a fragmented file from a spinning-metal drive.

My wild guess is that the copy operation is the usual "read a megabyte
from source, write a megabyte to destination, over and over again; once
that is finished, issue an fsync() on the destination".

Probably the progress bar is updated in the loop. That tracks the read
speed of the source, not the write speed of the destination.

iotop or biotop could probably show if the writes are on-going or not.

A "fix" would be to add fsync() calls within the loop. Whether or not
that's feasible depends on much more gnome IO knowledge than I've got.
Whether or not that would have other horrible consequences would also
need to be investigated. (If the source and destination are slow, it
might increase the total operation time uncomfortably.) There might be
better fixes if a process can learn from the OS how much of the file is
actually on the storage medium -- I don't know off-hand a way to do
this, but that doesn't mean it doesn't exist.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1871869

Title:
  Nautilus copy task status is not accurate.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1871869/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to