ice: Avoid RTNL lock when re-creating auxiliary device
Subsystems (2)
PATCH 1/1 ice: Avoid RTNL lock when re-creating auxiliary device
If a call to re-create the auxiliary device happens in a context that has already taken the RTNL lock, then the call flow that recreates auxiliary device can hang if there is another attempt to claim the RTNL lock by the auxiliary driver.
To avoid this, any call to re-create auxiliary devices that comes from an source that is holding the RTNL lock (e.g. netdev notifier when interface exits a bond) should execute in a separate thread. To accomplish this, add a flag to the PF that will be evaluated in the service task and dealt with there.
--- drivers/net/ethernet/intel/ice/ice.h | 3 ++- drivers/net/ethernet/intel/ice/ice_main.c | 3 +++ 2 files changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/ice/ice.h +2 -1
@@ -483,6 +483,7 @@ enum ice_pf_flags {ICE_FLAG_VF_TRUE_PROMISC_ENA ,ICE_FLAG_MDD_AUTO_RESET_VF ,ICE_FLAG_LINK_LENIENT_MODE_ENA , +ICE_FLAG_PLUG_AUX_DEV , ICE_PF_FLAGS_NBITS /* must be last */ };
@@ -887,7 +888,7 @@ static inline void ice_set_rdma_cap(struct ice_pf *pf)if (pf->hw.func_caps.common_cap.rdma && pf->num_rdma_msix) {set_bit (ICE_FLAG_RDMA_ENA , pf->flags);set_bit (ICE_FLAG_AUX_ENA , pf->flags); -ice_plug_aux_dev (pf); +set_bit (ICE_FLAG_PLUG_AUX_DEV , pf->flags); } }
drivers/net/ethernet/intel/ice/ice_main.c +3 -0
@@ -2253,6 +2253,9 @@ static void ice_service_task(struct work_struct *work)return ; } +if (test_and_clear_bit (ICE_FLAG_PLUG_AUX_DEV , pf->flags)) +ice_plug_aux_dev (pf); +ice_clean_adminq_subtask (pf);ice_check_media_subtask (pf);ice_check_for_hang_subtask (pf);
Signature · 1 line
2.33.1
PATCH 1/1 ice: Avoid RTNL lock when re-creating auxiliary device
Quoted 5 lines · Dave Ertman
If a call to re-create the auxiliary device happens in a context that has already taken the RTNL lock, then the call flow that recreates auxiliary device can hang if there is another attempt to claim the RTNL lock by the auxiliary driver.
Quoted 8 lines
To avoid this, any call to re-create auxiliary devices that comes from an source that is holding the RTNL lock (e.g. netdev notifier when interface exits a bond) should execute in a separate thread. To accomplish this, add a flag to the PF that will be evaluated in the service task and dealt with there.
Fixes f9f5301e7e2d ("ice: Register auxiliary device to provide RDMA")Signed-off-by Dave Ertman <[email protected]>
---
This looks like it will fix the issue and we tested internally and did not see the issue.
Quoted 40 lines
--- drivers/net/ethernet/intel/ice/ice.h | 3 ++- drivers/net/ethernet/intel/ice/ice_main.c | 3 +++ 2 files changed, 5 insertions(+), 1 deletion(-)drivers/net/ethernet/intel/ice/ice.h +2
-1@@ -483,6 +483,7 @@ enum ice_pf_flags {ICE_FLAG_VF_TRUE_PROMISC_ENA ,ICE_FLAG_MDD_AUTO_RESET_VF ,ICE_FLAG_LINK_LENIENT_MODE_ENA , +ICE_FLAG_PLUG_AUX_DEV , ICE_PF_FLAGS_NBITS /* must be last */ };@@ -887,7 +888,7 @@ static inline void ice_set_rdma_cap(struct ice_pf *pf)if (pf->hw.func_caps.common_cap.rdma && pf->num_rdma_msix) {set_bit (ICE_FLAG_RDMA_ENA , pf->flags);set_bit (ICE_FLAG_AUX_ENA , pf->flags); -ice_plug_aux_dev (pf); +set_bit (ICE_FLAG_PLUG_AUX_DEV , pf->flags); } }drivers/net/ethernet/intel/ice/ice_main.c +3
-0@@ -2253,6 +2253,9 @@ static void ice_service_task(struct work_struct *work)return ; } +if (test_and_clear_bit (ICE_FLAG_PLUG_AUX_DEV , pf->flags)) +ice_plug_aux_dev (pf); +ice_clean_adminq_subtask (pf);ice_check_media_subtask (pf);ice_check_for_hang_subtask (pf);
G
PATCH 1/1 ice: Avoid RTNL lock when re-creating auxiliary device
In reply to Dave ErtmanQuoted 19 lines
-----Original Message-----
From: Intel-wired-lan <[email protected]> On Behalf Of Dave Ertman
Sent: Friday, January 21, 2022 5:58 AM
To: intel-wired-lan at lists.osuosl.org
Subject: [Intel-wired-lan] [PATCH net] ice: Avoid RTNL lock when re-creating auxiliary deviceIf a call to re-create the auxiliary device happens in a context that has already taken the RTNL lock, then the call flow that recreates auxiliary device can hang if there is another attempt to claim the RTNL lock by the auxiliary driver.
To avoid this, any call to re-create auxiliary devices that comes from an source that is holding the RTNL lock (e.g. netdev notifier when interface exits a bond) should execute in a separate thread. To accomplish this, add a flag to the PF that will be evaluated in the service task and dealt with there.
Quoted 7 lines
Fixes f9f5301e7e2d ("ice: Register auxiliary device to provide RDMA")Signed-off-by Dave Ertman <[email protected]>--- drivers/net/ethernet/intel/ice/ice.h | 3 ++- drivers/net/ethernet/intel/ice/ice_main.c | 3 +++ 2 files changed, 5 insertions(+), 1 deletion(-)