On 9/11/25 5:33 PM, Olga Kornievskaia wrote: > When I have this patch it causes the following panic just from doing > "systemctl start nfs-server" followed by > "systemctl stop nfs-server" > > [ 69.705986] Unable to handle kernel paging request at virtual > address dfff800000000001 > [ 69.706706] KASAN: null-ptr-deref in range > [0x0000000000000008-0x000000000000000f] > [ 69.707299] Mem abort info: > [ 69.707514] ESR = 0x0000000096000005 > [ 69.707787] EC = 0x25: DABT (current EL), IL = 32 bits > [ 69.708186] SET = 0, FnV = 0 > [ 69.708415] EA = 0, S1PTW = 0 > [ 69.708637] FSC = 0x05: level 1 translation fault > [ 69.708986] Data abort info: > [ 69.709203] ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000 > [ 69.709604] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 > [ 69.709986] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 > [ 69.710345] [dfff800000000001] address between user and kernel address ranges > [ 69.710842] Internal error: Oops: 0000000096000005 [#1] SMP Yep, I've hit this. I'm dropping the patch from nfsd-testing until the issue is addressed. Thanks for the report. -- Chuck Lever
