From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga12.intel.com ([192.55.52.136]:14083 "EHLO mga12.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727405AbeIABep (ORCPT ); Fri, 31 Aug 2018 21:34:45 -0400 From: Keith Busch To: Linux PCI , Bjorn Helgaas Cc: Benjamin Herrenschmidt , Sinan Kaya , Thomas Tai , poza@codeaurora.org, Lukas Wunner , Keith Busch Subject: [PATCH 06/16] PCI/ERR: Remove devices on recovery failure Date: Fri, 31 Aug 2018 15:26:29 -0600 Message-Id: <20180831212639.10196-7-keith.busch@intel.com> In-Reply-To: <20180831212639.10196-1-keith.busch@intel.com> References: <20180831212639.10196-1-keith.busch@intel.com> Sender: linux-pci-owner@vger.kernel.org List-ID: This patch removes devices connected through a bus that can't recover from an error. Signed-off-by: Keith Busch --- drivers/pci/pcie/err.c | 42 ++++++++++++++++++++++++++++++++++++++---- 1 file changed, 38 insertions(+), 4 deletions(-) diff --git a/drivers/pci/pcie/err.c b/drivers/pci/pcie/err.c index 44c55f7ceb39..45f574954fd6 100644 --- a/drivers/pci/pcie/err.c +++ b/drivers/pci/pcie/err.c @@ -166,6 +166,15 @@ static int report_resume(struct pci_dev *dev, void *data) return 0; } +static int report_disconnect(struct pci_dev *dev, void *data) +{ + device_lock(&dev->dev); + pci_dev_set_disconnected(dev, NULL); + pci_uevent_ers(dev, PCI_ERS_RESULT_DISCONNECT); + device_unlock(&dev->dev); + return 0; +} + /** * default_reset_link - default reset function * @dev: pointer to pci_dev data structure @@ -271,6 +280,34 @@ static pci_ers_result_t broadcast_error_message(struct pci_dev *dev, return result_data.result; } +/** + * pcie_disconnect_device - Called when error handling ends with + * PCI_ERS_RESULT_DISCONNECT status. + * + * Reaching here means error handling has irrevocably failed. This function + * will ungracefully disconnect all the devices below the bus that has + * experienced the unrecoverable error. + * + * If the link is active after the removing all devices on the bus, this will + * attempt to re-enumerate the bus from scratch. + */ +static void pcie_disconnect_device(struct pci_dev *dev) +{ + struct pci_bus *bus = dev->subordinate; + struct pci_dev *child, *tmp; + + broadcast_error_message(dev, PCI_ERS_RESULT_DISCONNECT, + "disconnect", report_disconnect); + pci_lock_rescan_remove(); + list_for_each_entry_safe(child, tmp, &bus->devices, bus_list) + pci_stop_and_remove_bus_device(child); + + pci_bridge_secondary_bus_reset(dev); + if (pcie_wait_for_link(dev, true)) + pci_rescan_bus(bus); + pci_unlock_rescan_remove(); +} + static void pcie_do_recovery(struct pci_dev *dev, enum pci_channel_state state, u32 service) { @@ -313,12 +350,9 @@ static void pcie_do_recovery(struct pci_dev *dev, enum pci_channel_state state, pci_info(dev, "AER: Device recovery successful\n"); return; - failed: - pci_uevent_ers(dev, PCI_ERS_RESULT_DISCONNECT); - - /* TODO: Should kernel panic here? */ pci_info(dev, "AER: Device recovery failed\n"); + pcie_disconnect_device(dev); } void pcie_do_fatal_recovery(struct pci_dev *dev, u32 service) -- 2.14.4