[Intel-wired-lan] [PATCH iwl-net] ice: Fix ice module unload
Jakub Buchocki
jakubx.buchocki at intel.com
Wed May 10 09:25:30 UTC 2023
Clearing interrupt scheme before PFR reset, during the removal routine,
could cause the hardware errors and possibly lead to system reboot, as
the PF reset can cause the interrupt to be generated.
Move clearing interrupt scheme from device deinitialization subprocedure,
and call it directly in particular routines. In ice_remove(), call the
ice_clear_interrupt_scheme() after the PFR is complete and all pending
transactions are done.
Error example:
[ 75.229328] ice 0000:ca:00.1: Failed to read Tx Scheduler Tree - User Selection data from flash
[ 77.571315] {1}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 1
[ 77.571418] {1}[Hardware Error]: event severity: recoverable
[ 77.571459] {1}[Hardware Error]: Error 0, type: recoverable
[ 77.571500] {1}[Hardware Error]: section_type: PCIe error
[ 77.571540] {1}[Hardware Error]: port_type: 4, root port
[ 77.571580] {1}[Hardware Error]: version: 3.0
[ 77.571615] {1}[Hardware Error]: command: 0x0547, status: 0x4010
[ 77.571661] {1}[Hardware Error]: device_id: 0000:c9:02.0
[ 77.571703] {1}[Hardware Error]: slot: 25
[ 77.571736] {1}[Hardware Error]: secondary_bus: 0xca
[ 77.571773] {1}[Hardware Error]: vendor_id: 0x8086, device_id: 0x347a
[ 77.571821] {1}[Hardware Error]: class_code: 060400
[ 77.571858] {1}[Hardware Error]: bridge: secondary_status: 0x2800, control: 0x0013
[ 77.572490] pcieport 0000:c9:02.0: AER: aer_status: 0x00200000, aer_mask: 0x00100020
[ 77.572870] pcieport 0000:c9:02.0: [21] ACSViol (First)
[ 77.573222] pcieport 0000:c9:02.0: AER: aer_layer=Transaction Layer, aer_agent=Receiver ID
[ 77.573554] pcieport 0000:c9:02.0: AER: aer_uncor_severity: 0x00463010
[ 77.691273] {2}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 1
[ 77.691738] {2}[Hardware Error]: event severity: recoverable
[ 77.691971] {2}[Hardware Error]: Error 0, type: recoverable
[ 77.692192] {2}[Hardware Error]: section_type: PCIe error
[ 77.692403] {2}[Hardware Error]: port_type: 4, root port
[ 77.692616] {2}[Hardware Error]: version: 3.0
[ 77.692825] {2}[Hardware Error]: command: 0x0547, status: 0x4010
[ 77.693032] {2}[Hardware Error]: device_id: 0000:c9:02.0
[ 77.693238] {2}[Hardware Error]: slot: 25
[ 77.693440] {2}[Hardware Error]: secondary_bus: 0xca
[ 77.693641] {2}[Hardware Error]: vendor_id: 0x8086, device_id: 0x347a
[ 77.693853] {2}[Hardware Error]: class_code: 060400
[ 77.694054] {2}[Hardware Error]: bridge: secondary_status: 0x0800, control: 0x0013
[ 77.719115] pci 0000:ca:00.1: AER: can't recover (no error_detected callback)
[ 77.719140] pcieport 0000:c9:02.0: AER: device recovery failed
[ 77.719216] pcieport 0000:c9:02.0: AER: aer_status: 0x00200000, aer_mask: 0x00100020
[ 77.719390] pcieport 0000:c9:02.0: [21] ACSViol (First)
[ 77.719557] pcieport 0000:c9:02.0: AER: aer_layer=Transaction Layer, aer_agent=Receiver ID
[ 77.719723] pcieport 0000:c9:02.0: AER: aer_uncor_severity: 0x00463010
Fixes: 5b246e533d01 ("ice: split probe into smaller functions")
Signed-off-by: Jakub Buchocki <jakubx.buchocki at intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski at linux.intel.com>
---
drivers/net/ethernet/intel/ice/ice_main.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 0d8b8c6f9bd3..42435d124822 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -4794,7 +4794,6 @@ static int ice_init_dev(struct ice_pf *pf)
static void ice_deinit_dev(struct ice_pf *pf)
{
ice_free_irq_msix_misc(pf);
- ice_clear_interrupt_scheme(pf);
ice_deinit_pf(pf);
ice_deinit_hw(&pf->hw);
}
@@ -5063,6 +5062,7 @@ static int ice_init(struct ice_pf *pf)
ice_dealloc_vsis(pf);
err_alloc_vsis:
ice_deinit_dev(pf);
+ ice_clear_interrupt_scheme(pf);
return err;
}
@@ -5090,6 +5090,8 @@ int ice_load(struct ice_pf *pf)
if (err)
return err;
+ ice_clear_interrupt_scheme(pf);
+
err = ice_init_dev(pf);
if (err)
return err;
@@ -5124,6 +5126,7 @@ int ice_load(struct ice_pf *pf)
ice_vsi_decfg(ice_get_main_vsi(pf));
err_vsi_cfg:
ice_deinit_dev(pf);
+ ice_clear_interrupt_scheme(pf);
return err;
}
@@ -5243,6 +5246,7 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent)
ice_deinit_eth(pf);
err_init_eth:
ice_deinit(pf);
+ ice_clear_interrupt_scheme(pf);
err_init:
pci_disable_device(pdev);
return err;
@@ -5352,6 +5356,7 @@ static void ice_remove(struct pci_dev *pdev)
*/
ice_reset(&pf->hw, ICE_RESET_PFR);
pci_wait_for_pending_transaction(pdev);
+ ice_clear_interrupt_scheme(pf);
pci_disable_device(pdev);
}
--
2.25.1
More information about the Intel-wired-lan
mailing list