On Sun, Apr 27, 2025 at 08:10:03AM -0700, syzbot wrote: > syzbot suspects this issue was fixed by commit: > > commit 71f8e806a5e4edada72456ee3b2e2d7eab6fadee > Author: Kent Overstreet <[email protected]> > Date: Fri Apr 18 16:49:00 2025 +0000 > > bcachefs: Stricter checks on "key allowed in this btree" > > bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=16b9e270580000 > start commit: ccb98ccef0e5 Merge tag 'platform-drivers-x86-v6.13-4' of g.. > git tree: upstream > kernel config: https://syzkaller.appspot.com/x/.config?x=ba7cde9482d6bb6 > dashboard link: https://syzkaller.appspot.com/bug?extid=f57d864a5d6eff90da3b > syz repro: https://syzkaller.appspot.com/x/repro.syz?x=131666df980000 > C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10556ac4580000 > > If the result looks correct, please mark the issue as fixed by replying with:
#syz fix: bcachefs: Stricter checks on "key allowed in this btree" We had a similar one where syzbot manager to come up with a reflink pointer in the inodes btree - triggers and validation get very confused at that sort of thing, in particular because whether or not k->size should be zero is btree specific
