On Sun, 31 May 2026 at 02:29, Eric Biggers <[email protected]> wrote: > > On Fri, May 22, 2026 at 09:00:46AM +0200, Daniel Vacek wrote: > > Hi Eric, > > > > This is just a gentle ping. > > I was wondering if you had a chance to look at this version? > > I believe all your previous feedback has been addressed and this > > version is solid. > > Please, let me know your thoughts. > > > > Regards, > > Daniel > > It's been really hard to find time to review this huge patchset. I've > started going through it and will try to leave comments next week.
Hi Eric, First, thank you for looking into this. > In the mean time it would be really helpful if you went through the > Sashiko reviews > (https://sashiko.dev/#/patchset/20260513085340.3673127-1-neelx%40suse.com) > and address the ones that make sense to. It found 93 issues including > 16 critical ones, which is kind of a lot. Some of them are the same > things I'm noticing already. Same for the issue that Christoph noticed > where new devices can be added; Sashiko had already found that too. > > If I'm going to have to use my limited human review time to point out > issues that were already found, that's not a great use of time. I already went through some of the Sashiko reviews (they were slowly coming up one by one) before leaving for my vacation. And I found them mostly confusing or misleading. I'm gonna have a look into the rest I haven't seen yet to see if there are any useful points. And I'll compare it to your notes to see if it was Sashiko being confused or if it was me. > I also don't see any information about how this was tested (and will > continue to be tested in the future). I'm using the `encrypt` group of xfstests as the acceptance criteria and the full test run to ensure no regressions. General support for btrfs had to be added. And since we only support the v2 fscrypt policy, some tests had to be split into two tests - one testing v1 and the other testing v2 policy. There are also additional new btrfs specific tests (reflinks, snapshots). xfstests also need fscrypt support in btrfs-progs with related changes to export nonces and other metadata like device offsets of encrypted blocks for validation. I'll send the fscrypt updated btrfs-progs and xfstests next. I just need to clean them up a bit first. I'm sorry about the delay, but I only managed to post the kernel part before my vacation. I was hoping it would give you all more time to review. Thanks, Daniel > - Eric
