Bug 252358 - cp(1) of large files is causing 100% CPU utilization and poor transfer of ~168M/minute

[Rick Macklem]

Well, it depends on what "stressdisk" (never heard of it) does.
If it:
1 - Reads a large non-sparse file on a UFS file system.
2 - Uses lseek(SEEK_DATA/SEEK_HOLE) with a small blocksize
     or copy_file_range(2) with a small length. (copy_file_range(2)
     does SEEK_DATA/SEEK_HOLE internally.)
then maybe. Otherwise, not.

rick

________________________________________
From: owner-freebsd-current at freebsd.org <owner-freebsd-current at freebsd.org> on behalf of Graham Perrin <grahamperrin at gmail.com>
Sent: Sunday, January 3, 2021 4:21 AM
To: freebsd-current at freebsd.org
Subject: Bug 252358 - cp(1) of large files is causing 100% CPU utilization and poor transfer of ~168M/minute

CAUTION: This email originated from outside of the University of Guelph. Do not click links or open attachments unless you recognize the sender and know the content is safe. If in doubt, forward suspicious emails to IThelp at uoguelph.ca


On 02/01/2021 20:05, Matthias Apitz wrote:

> Here we go:https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=252358


Thank you.

I plan to build and install 82397d791966b09d344251bc709cd9db2b3a1902;
<https://cgit.freebsd.org/src/log/?qt=range&q=82397d791966b09d344251bc709cd9db2b3a1902>

----

In the meantime: might 252358 be the cause of a peculiarity that I
encountered, a week or so ago, whilst using sysutils/stressdisk with a
particular (or particular type of) USB thumb drive?

<https://github.com/ncw/stressdisk/issues/14> tl;dr whilst trying to
test the flash device on USB at da0, there was _unexpected thrashing_ of
the hard disk drive at ada0

_______________________________________________
freebsd-current at freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"