I saw the following warning about workqueues when running xfstests with multichannel to Windows server target (at test generic/048, which failed due to umount busy). Any thoughts about whether WQ_UNBOUND would help? [ 4525.116583] run fstests generic/048 at 2024-02-03 14:17:11 [ 4525.290294] CIFS: Attempting to mount //172.30.144.1/test [ 4525.305238] CIFS: VFS: Error connecting to socket. Aborting operation. [ 4525.305244] CIFS: VFS: failed to open extra channel on iface:fe80:0000:0000:0000:f196:776f:4b2c:fa3b rc=-22 [ 4525.313835] CIFS: VFS: successfully opened new channel on iface:10.20.15.198 [ 4525.364120] CIFS: Attempting to mount //172.30.144.1/scratch [ 4525.377185] TCP: eth0: Driver has suspect GRO implementation, TCP performance may be compromised. [ 4526.365222] CIFS: Attempting to mount //172.30.144.1/scratch [ 4526.376402] CIFS: VFS: shut down requested (1) [ 4526.467346] CIFS: Attempting to mount //172.30.144.1/scratch [ 4526.495301] CIFS: Attempting to mount //172.30.144.1/scratch [ 4530.882149] workqueue: smb2_deferred_work_close [cifs] hogged CPU for >10000us 8 times, consider switching to WQ_UNBOUND [ 4542.382103] workqueue: smb2_cached_lease_break [cifs] hogged CPU for >10000us 4 times, consider switching to WQ_UNBOUND [ 4543.650093] workqueue: smb2_deferred_work_close [cifs] hogged CPU for >10000us 16 times, consider switching to WQ_UNBOUND [ 4550.478063] workqueue: smb2_cached_lease_break [cifs] hogged CPU for >10000us 8 times, consider switching to WQ_UNBOUND [ 4565.381997] workqueue: smb2_deferred_work_close [cifs] hogged CPU for >10000us 32 times, consider switching to WQ_UNBOUND [ 4569.933971] workqueue: smb2_cached_lease_break [cifs] hogged CPU for >10000us 16 times, consider switching to WQ_UNBOUND [ 4587.181886] workqueue: smb2_cached_lease_break [cifs] hogged CPU for >10000us 32 times, consider switching to WQ_UNBOUND [ 4601.121822] workqueue: smb2_deferred_work_close [cifs] hogged CPU for >10000us 64 times, consider switching to WQ_UNBOUND -- Thanks, Steve
