Wind River Support Network

Description

In the Linux kernel, the following vulnerability has been resolved:[EOL][EOL]net: enetc: fix the deadlock of enetc_mdio_lock[EOL][EOL]After applying the workaround for err050089, the LS1028A platform[EOL]experiences RCU stalls on RT kernel. This issue is caused by the[EOL]recursive acquisition of the read lock enetc_mdio_lock. Here list some[EOL]of the call stacks identified under the enetc_poll path that may lead to[EOL]a deadlock:[EOL][EOL]enetc_poll[EOL]  -> enetc_lock_mdio[EOL]  -> enetc_clean_rx_ring OR napi_complete_done[EOL]     -> napi_gro_receive[EOL]        -> enetc_start_xmit[EOL]           -> enetc_lock_mdio[EOL]           -> enetc_map_tx_buffs[EOL]           -> enetc_unlock_mdio[EOL]  -> enetc_unlock_mdio[EOL][EOL]After enetc_poll acquires the read lock, a higher-priority writer attempts[EOL]to acquire the lock, causing preemption. The writer detects that a[EOL]read lock is already held and is scheduled out. However, readers under[EOL]enetc_poll cannot acquire the read lock again because a writer is already[EOL]waiting, leading to a thread hang.[EOL][EOL]Currently, the deadlock is avoided by adjusting enetc_lock_mdio to prevent[EOL]recursive lock acquisition.