On Sun, May 22, 2022 at 09:17:48AM +1000, Dave Chinner wrote:
> On Sat, May 21, 2022 at 01:43:49PM +0800, Zorro Lang wrote:
> > The mkgroupfile doesn't sort cases in number order, that cause
> > fstests' ./new script can't find a proper number which isn't taken.
> > The unsorted case number might cause more problems, so fix it.
>
> Yet more reason for these group lists to die and get replaced with a
> common generation function.
>
> > Fixes: 441606d28681 ("fstests: faster group file creation")
> > Signed-off-by: Zorro Lang <zlang@xxxxxxxxxx>
> > ---
> >
> > Hi,
> >
> > If you'd like to fix this issue seperately, I can merge this patch with Dave's
> > "[PATCH 04/12] fstests: fix group list generation for whacky test names"
> > together.
>
> It needs documentation updates, too. I had no idea that the group
I've merged below changes[1]. As other two patches[2] need more changes, so how
about we add it into doc by "[PATCH 05/12] README: document _begin_fstests
better"? Or I can send another documentation patch to do that next week :) I
hope to fix this regression issue today, due to it's been 2 weeks since we
brought in it.
[1]
68497440 tools/mkgroupfile: help the cases of group.list in number order
1a04a7c9 fstests: fix group list generation for whacky test names
e6d4d963 fstests: remove xfs deprecated test
c1941d6f xfs/191: remove broken test
6812c3b2 xfs/148: fix failure from bad shortform size assumptions
460c4604 xfs/148: make test debuggable
dbcc5493 xfs/348: golden output is not correct
9a392f8d xfs/122: add attribute log formats to test output.
307f0514 fstests: filter quota warnings
[2]
[PATCH 05/12] README: document _begin_fstests better
[PATCH 08/12] generic/081: don't run on DAX capable devices
> list files were used by the new script or that there was a
> requirement for them to be sorted into numerical order.
Sorry, I didn't notice that either, until I tried to write a new case.
When you have any better idea to replace the current group list method,
feel free to share :)
Thanks,
Zorro
>
> Cheers,
>
> Dave.
> --
> Dave Chinner
> david@xxxxxxxxxxxxx
>
