On Friday, December 20, 2019 10:59:52 AM MST Chris Murphy wrote:
> Issuing the command once per week harms no one

Based on what's actual in the Change proposal, this is not the case.

Even if this goes through, in my opinion, it should only affect the GNOME 
Spin, or perhaps even "all graphical" spins at most. 

> It's reasonable to enable fstrim.timer now. *And* conduct parallel
> development to create a kernel facility to do this automatically, if
> it's even possible. I'm not convinced the drives report enough
> information to do this properly automatically, rather than on a
> schedule.

What do you mean by that? It's the filesystem that would make the call, not 
the hardware itself. If the hardware was capable of figuring out when this 
call is useful, and wouldn't impact the user, it'd likely do the call itself, 
in the firmware.

> And to be true, Windows and macOS have used their own white
> listing method to do this, meaning quite a lot of devices aren't
> getting these hints and are left out.

I'm particularly concerned about the handling of the hardware that this will 
negatively affect, as I happen to have an SK Hynix part that is affected by 
momentary I/O freezes whenever TRIM is called, and I don't want users to have 
to deal with that.

The correct time to call TRIM on these devices is when there aren't ongoing I/
O operations.

-- 
John M. Harris, Jr.
Splentity

_______________________________________________
devel mailing list -- devel@lists.fedoraproject.org
To unsubscribe send an email to devel-le...@lists.fedoraproject.org
Fedora Code of Conduct: 
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org

Reply via email to