Message ID | 1468316605-788-1-git-send-email-wangkefeng.wang@huawei.com (mailing list archive) |
---|---|
State | Accepted, archived |
Headers | show |
>>>>> "Kefeng" == Kefeng Wang <wangkefeng.wang@huawei.com> writes:
Kefeng> The _scsih_pci_mmio_enabled called if scsih_pci_error_detected
Kefeng> returns PCI_ERS_RESULT_CAN_RECOVER, at this point, read/write to
Kefeng> the device still works, no need to reset slot.
Applied to 4.8/scsi-queue.
diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c index 6bff13e..eedd62e3 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c +++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c @@ -9033,8 +9033,11 @@ scsih_pci_mmio_enabled(struct pci_dev *pdev) /* TODO - dump whatever for debugging purposes */ - /* Request a slot reset. */ - return PCI_ERS_RESULT_NEED_RESET; + /* This called only if scsih_pci_error_detected returns + * PCI_ERS_RESULT_CAN_RECOVER, read/write to the device + * still works, not need to reset slot. + */ + return PCI_ERS_RESULT_RECOVERED; } /*
The _scsih_pci_mmio_enabled called if scsih_pci_error_detected returns PCI_ERS_RESULT_CAN_RECOVER, at this point, read/write to the device still works, no need to reset slot. Or the mpt3sas_base_map_resources in scsih_pci_slot_reset will fail, and iounamp ioc->chip, then we will meet issue when read ioc->chip in mpt3sas_base_get_iocstate from _base_fault_reset_work. Cc: Sathya Prakash <sathya.prakash@broadcom.com> Cc: Chaitra P B <chaitra.basappa@broadcom.com> Cc: Suganath Prabu Subramani <suganath-prabu.subramani@broadcom.com> Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com> --- NOTE: I found this with an earlier kernel version, but the logic is not changed. drivers/scsi/mpt3sas/mpt3sas_scsih.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)