Hi all, > Hi Petr, All, > Petr Vorel <pvorel@xxxxxxx> wrote: > > Hi all, > > Could we in the end accept this patch? > > It'd fix the issue for now and I could set size of the XFS loop device > > smaller > > than 300 MB (better for embedded). i.e. 16 MB (or 32 or 64 MB or anything > > higher > > if XFS developers are convinced it's needed). > I personally think YES! > (sorry for replying so late, I was on vacation last week) I still plan to implement Old thread thus recap Cyril's original proposal [1] I'm starting to wonder if we should start tracking minimal FS size per filesystem since btrfs and xfs will likely to continue to grow and with that we will end up disabling the whole fs related testing on embedded boards with a little disk space. If we tracked that per filesystem we would be able to skip a subset of filesystems when there is not enough space. The downside is obviously that we would have to add a bit more complexity to the test library. I'm willing to implement this Cyril's proposal, but what puts me off is that Btrfs and XFS developers are already complaining that testing on minimal size is not a realistic testing. fstests are testing on realistic sizes, Darrick J. Wong [2]: ... In the ideal world we'll some day get around to restructuring all the xfstests that do tricky things with sub-500M filesystems, but that's the unfortunate part of removing support for small disks. Most of the fstests don't care about the fs size and so they'll run with the configured storage (some tens or millions of gigabytes) so we're mostly using the same fs sizes that users are expected to have. Geert Uytterhoeven suggested [3]: Yeah, we used to have ext2 root file systems that fit on 1440 KiB floppies. IIRC, ext3 does have a minimum size of 32 MiB or so. > > As I wrote before I plan to suggest sizes: > > btrfs 110 MB > > the rest (ext[234], xfs, ntfs, vfat, exfat, tmpfs): 16 MB Maybe safer would be (from more realistic point of testing) * Btrfs, XFS: 300 MB (keep the current size for Btrfs) * Maybe raise ext4 to 32 MB (more realistic?) * others: 16 MB > +1 thanks for finding the minimal size. Also Amir noted [4] that we have squashfs01.c asks for .dev_min_size = 1, but we do the minimal default (300 MB now, it would be 16 MB). Do we want to force .dev_min_size in this case or use the default minimal? Kind regards, Petr [1] https://lore.kernel.org/ltp/Yv4MBF79PnJKJbwm@yuki/ [2] https://lore.kernel.org/ltp/Yv2A9Ggkv%2FNBrTd4@magnolia/ [3] https://lore.kernel.org/ltp/CAMuHMdUMBjCTwPu7wxrnagXnbyVxxmXN+vHmML0Lr=SyrTw0nQ@xxxxxxxxxxxxxx/ [4] https://lore.kernel.org/ltp/CAOQ4uxjMEHYQwO25dhs5WtzbOkJcee0HofQDTT3cD-qXJn7xQw@xxxxxxxxxxxxxx/
