Any new insights on large folio issues causing data loss?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,

I’ve been trying to hunt down an XFS issue last year that Daniel was (and is) also experiencing:
https://lore.kernel.org/lkml/CA+wXwBS7YTHUmxGP3JrhcKMnYQJcd6=7HE+E1v-guk01L2K3Zw@xxxxxxxxxxxxxx/T/

I did also register a kernel bug back then:
https://bugzilla.kernel.org/show_bug.cgi?id=217572#c7

The overall impression was that this is related to large folios and Daniel told me he’s currently running 6.1/6.6 with large folios disabled, by applying a revert for this patch: https://github.com/torvalds/linux/commit/6795801366da0cd3d99e27c37f020a8f16714886

We’re currently still running on 5.15 but this is getting long in the tooth and we’re now experiencing issues where backports are starting to slow down so that we get bitten by other issues.

I’m generally rather happy to upgrade, but we’ve had multiple instances of data loss with 6.1 last year and we’d like to avoid running into that again … but I also don’t feel so good about reverting the folio support indiscriminately … 

As the issue was quite fuzzy I wonder whether someone has a gut feeling whether this should have improved over time or maybe this issue triggers someones memory where it might be related to a fix that isn’t obvious?

Thanks!
Christian

-- 
Christian Theune · ct@xxxxxxxxxxxxxxx · +49 345 219401 0
Flying Circus Internet Operations GmbH · https://flyingcircus.io
Leipziger Str. 70/71 · 06108 Halle (Saale) · Deutschland
HR Stendal HRB 21169 · Geschäftsführer: Christian Theune, Christian Zagrodnick






[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux