From b50cac55bf859d5b2fdcc1803a553a251b703456 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 6 Oct 2011 14:08:18 -0400 Subject: [PATCH 01/10] PCI/sysfs: add per pci device msi[x] irq listing (v5) This patch adds a per-pci-device subdirectory in sysfs called: /sys/bus/pci/devices//msi_irqs This sub-directory exports the set of msi vectors allocated by a given pci device, by creating a numbered sub-directory for each vector beneath msi_irqs. For each vector various attributes can be exported. Currently the only attribute is called mode, which tracks the operational mode of that vector (msi vs. msix) Acked-by: Greg Kroah-Hartman Signed-off-by: Jesse Barnes --- Documentation/ABI/testing/sysfs-bus-pci | 18 ++++ drivers/pci/msi.c | 111 ++++++++++++++++++++++++ include/linux/msi.h | 3 + include/linux/pci.h | 1 + 4 files changed, 133 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-bus-pci b/Documentation/ABI/testing/sysfs-bus-pci index 349ecf26ce10..34f51100f029 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci +++ b/Documentation/ABI/testing/sysfs-bus-pci @@ -66,6 +66,24 @@ Description: re-discover previously removed devices. Depends on CONFIG_HOTPLUG. +What: /sys/bus/pci/devices/.../msi_irqs/ +Date: September, 2011 +Contact: Neil Horman +Description: + The /sys/devices/.../msi_irqs directory contains a variable set + of sub-directories, with each sub-directory being named after a + corresponding msi irq vector allocated to that device. Each + numbered sub-directory N contains attributes of that irq. + Note that this directory is not created for device drivers which + do not support msi irqs + +What: /sys/bus/pci/devices/.../msi_irqs//mode +Date: September 2011 +Contact: Neil Horman +Description: + This attribute indicates the mode that the irq vector named by + the parent directory is in (msi vs. msix) + What: /sys/bus/pci/devices/.../remove Date: January 2009 Contact: Linux PCI developers diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 0e6d04d7ba4f..e6b6b9c67023 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -323,6 +323,8 @@ static void free_msi_irqs(struct pci_dev *dev) if (list_is_last(&entry->list, &dev->msi_list)) iounmap(entry->mask_base); } + kobject_del(&entry->kobj); + kobject_put(&entry->kobj); list_del(&entry->list); kfree(entry); } @@ -403,6 +405,98 @@ void pci_restore_msi_state(struct pci_dev *dev) } EXPORT_SYMBOL_GPL(pci_restore_msi_state); + +#define to_msi_attr(obj) container_of(obj, struct msi_attribute, attr) +#define to_msi_desc(obj) container_of(obj, struct msi_desc, kobj) + +struct msi_attribute { + struct attribute attr; + ssize_t (*show)(struct msi_desc *entry, struct msi_attribute *attr, + char *buf); + ssize_t (*store)(struct msi_desc *entry, struct msi_attribute *attr, + const char *buf, size_t count); +}; + +static ssize_t show_msi_mode(struct msi_desc *entry, struct msi_attribute *atr, + char *buf) +{ + return sprintf(buf, "%s\n", entry->msi_attrib.is_msix ? "msix" : "msi"); +} + +static ssize_t msi_irq_attr_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct msi_attribute *attribute = to_msi_attr(attr); + struct msi_desc *entry = to_msi_desc(kobj); + + if (!attribute->show) + return -EIO; + + return attribute->show(entry, attribute, buf); +} + +static const struct sysfs_ops msi_irq_sysfs_ops = { + .show = msi_irq_attr_show, +}; + +static struct msi_attribute mode_attribute = + __ATTR(mode, S_IRUGO, show_msi_mode, NULL); + + +struct attribute *msi_irq_default_attrs[] = { + &mode_attribute.attr, + NULL +}; + +void msi_kobj_release(struct kobject *kobj) +{ + struct msi_desc *entry = to_msi_desc(kobj); + + pci_dev_put(entry->dev); +} + +static struct kobj_type msi_irq_ktype = { + .release = msi_kobj_release, + .sysfs_ops = &msi_irq_sysfs_ops, + .default_attrs = msi_irq_default_attrs, +}; + +static int populate_msi_sysfs(struct pci_dev *pdev) +{ + struct msi_desc *entry; + struct kobject *kobj; + int ret; + int count = 0; + + pdev->msi_kset = kset_create_and_add("msi_irqs", NULL, &pdev->dev.kobj); + if (!pdev->msi_kset) + return -ENOMEM; + + list_for_each_entry(entry, &pdev->msi_list, list) { + kobj = &entry->kobj; + kobj->kset = pdev->msi_kset; + pci_dev_get(pdev); + ret = kobject_init_and_add(kobj, &msi_irq_ktype, NULL, + "%u", entry->irq); + if (ret) + goto out_unroll; + + count++; + } + + return 0; + +out_unroll: + list_for_each_entry(entry, &pdev->msi_list, list) { + if (!count) + break; + kobject_del(&entry->kobj); + kobject_put(&entry->kobj); + count--; + } + return ret; +} + /** * msi_capability_init - configure device's MSI capability structure * @dev: pointer to the pci_dev data structure of MSI device function @@ -454,6 +548,13 @@ static int msi_capability_init(struct pci_dev *dev, int nvec) return ret; } + ret = populate_msi_sysfs(dev); + if (ret) { + msi_mask_irq(entry, mask, ~mask); + free_msi_irqs(dev); + return ret; + } + /* Set MSI enabled bits */ pci_intx_for_msi(dev, 0); msi_set_enable(dev, pos, 1); @@ -574,6 +675,12 @@ static int msix_capability_init(struct pci_dev *dev, msix_program_entries(dev, entries); + ret = populate_msi_sysfs(dev); + if (ret) { + ret = 0; + goto error; + } + /* Set MSI-X enabled bits and unmask the function */ pci_intx_for_msi(dev, 0); dev->msix_enabled = 1; @@ -732,6 +839,8 @@ void pci_disable_msi(struct pci_dev *dev) pci_msi_shutdown(dev); free_msi_irqs(dev); + kset_unregister(dev->msi_kset); + dev->msi_kset = NULL; } EXPORT_SYMBOL(pci_disable_msi); @@ -830,6 +939,8 @@ void pci_disable_msix(struct pci_dev *dev) pci_msix_shutdown(dev); free_msi_irqs(dev); + kset_unregister(dev->msi_kset); + dev->msi_kset = NULL; } EXPORT_SYMBOL(pci_disable_msix); diff --git a/include/linux/msi.h b/include/linux/msi.h index 05acced439a3..ce93a341337d 100644 --- a/include/linux/msi.h +++ b/include/linux/msi.h @@ -1,6 +1,7 @@ #ifndef LINUX_MSI_H #define LINUX_MSI_H +#include #include struct msi_msg { @@ -44,6 +45,8 @@ struct msi_desc { /* Last set MSI message */ struct msi_msg msg; + + struct kobject kobj; }; /* diff --git a/include/linux/pci.h b/include/linux/pci.h index 7cda65b5f798..84225c756bd1 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -336,6 +336,7 @@ struct pci_dev { struct bin_attribute *res_attr_wc[DEVICE_COUNT_RESOURCE]; /* sysfs file for WC mapping of resources */ #ifdef CONFIG_PCI_MSI struct list_head msi_list; + struct kset *msi_kset; #endif struct pci_vpd *vpd; #ifdef CONFIG_PCI_ATS From 69166fbf02c7a21745013f2de037bf7af26e4279 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Wed, 2 Nov 2011 14:07:15 -0600 Subject: [PATCH 02/10] PCI: Fix PRI and PASID consistency These are extended capabilities, rename and move to proper group for consistency. Signed-off-by: Alex Williamson Signed-off-by: Jesse Barnes --- drivers/pci/ats.c | 20 ++++++++++---------- include/linux/pci_regs.h | 4 ++-- 2 files changed, 12 insertions(+), 12 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 7ec56fb0bd78..831e1920386c 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -174,7 +174,7 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) u32 max_requests; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return -EINVAL; @@ -205,7 +205,7 @@ void pci_disable_pri(struct pci_dev *pdev) u16 control; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return; @@ -226,7 +226,7 @@ bool pci_pri_enabled(struct pci_dev *pdev) u16 control; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return false; @@ -248,7 +248,7 @@ int pci_reset_pri(struct pci_dev *pdev) u16 control; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return -EINVAL; @@ -281,7 +281,7 @@ bool pci_pri_stopped(struct pci_dev *pdev) u16 control, status; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return true; @@ -310,7 +310,7 @@ int pci_pri_status(struct pci_dev *pdev) u16 status, control; int pos; - pos = pci_find_ext_capability(pdev, PCI_PRI_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return -EINVAL; @@ -341,7 +341,7 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) u16 control, supported; int pos; - pos = pci_find_ext_capability(pdev, PCI_PASID_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return -EINVAL; @@ -375,7 +375,7 @@ void pci_disable_pasid(struct pci_dev *pdev) u16 control = 0; int pos; - pos = pci_find_ext_capability(pdev, PCI_PASID_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return; @@ -399,7 +399,7 @@ int pci_pasid_features(struct pci_dev *pdev) u16 supported; int pos; - pos = pci_find_ext_capability(pdev, PCI_PASID_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return -EINVAL; @@ -425,7 +425,7 @@ int pci_max_pasids(struct pci_dev *pdev) u16 supported; int pos; - pos = pci_find_ext_capability(pdev, PCI_PASID_CAP); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return -EINVAL; diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index b5d9657f3100..090d3a9f5b26 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -537,7 +537,9 @@ #define PCI_EXT_CAP_ID_ARI 14 #define PCI_EXT_CAP_ID_ATS 15 #define PCI_EXT_CAP_ID_SRIOV 16 +#define PCI_EXT_CAP_ID_PRI 19 #define PCI_EXT_CAP_ID_LTR 24 +#define PCI_EXT_CAP_ID_PASID 27 /* Advanced Error Reporting */ #define PCI_ERR_UNCOR_STATUS 4 /* Uncorrectable Error Status */ @@ -664,7 +666,6 @@ #define PCI_ATS_MIN_STU 12 /* shift of minimum STU block */ /* Page Request Interface */ -#define PCI_PRI_CAP 0x13 /* PRI capability ID */ #define PCI_PRI_CONTROL_OFF 0x04 /* Offset of control register */ #define PCI_PRI_STATUS_OFF 0x06 /* Offset of status register */ #define PCI_PRI_ENABLE 0x0001 /* Enable mask */ @@ -676,7 +677,6 @@ #define PCI_PRI_ALLOC_REQ_OFF 0x0c /* Cap offset for max reqs allowed */ /* PASID capability */ -#define PCI_PASID_CAP 0x1b /* PASID capability ID */ #define PCI_PASID_CAP_OFF 0x04 /* PASID feature register */ #define PCI_PASID_CONTROL_OFF 0x06 /* PASID control register */ #define PCI_PASID_ENABLE 0x01 /* Enable/Supported bit */ From 3c076351c4027a56d5005a39a0b518a4ba393ce2 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Thu, 10 Nov 2011 16:38:33 -0500 Subject: [PATCH 03/10] PCI: Rework ASPM disable code Right now we forcibly clear ASPM state on all devices if the BIOS indicates that the feature isn't supported. Based on the Microsoft presentation "PCI Express In Depth for Windows Vista and Beyond", I'm starting to think that this may be an error. The implication is that unless the platform grants full control via _OSC, Windows will not touch any PCIe features - including ASPM. In that case clearing ASPM state would be an error unless the platform has granted us that control. This patch reworks the ASPM disabling code such that the actual clearing of state is triggered by a successful handoff of PCIe control to the OS. The general ASPM code undergoes some changes in order to ensure that the ability to clear the bits isn't overridden by ASPM having already been disabled. Further, this theoretically now allows for situations where only a subset of PCIe roots hand over control, leaving the others in the BIOS state. It's difficult to know for sure that this is the right thing to do - there's zero public documentation on the interaction between all of these components. But enough vendors enable ASPM on platforms and then set this bit that it seems likely that they're expecting the OS to leave them alone. Measured to save around 5W on an idle Thinkpad X220. Signed-off-by: Matthew Garrett Signed-off-by: Jesse Barnes --- drivers/acpi/pci_root.c | 7 +++++ drivers/pci/pci-acpi.c | 1 - drivers/pci/pcie/aspm.c | 58 +++++++++++++++++++++++++--------------- include/linux/pci-aspm.h | 4 +-- 4 files changed, 46 insertions(+), 24 deletions(-) diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index 2672c798272f..7aff6312ce7c 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -596,6 +596,13 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) if (ACPI_SUCCESS(status)) { dev_info(root->bus->bridge, "ACPI _OSC control (0x%02x) granted\n", flags); + if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) { + /* + * We have ASPM control, but the FADT indicates + * that it's unsupported. Clear it. + */ + pcie_clear_aspm(root->bus); + } } else { dev_info(root->bus->bridge, "ACPI _OSC request failed (%s), " diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index 4ecb6408b0d6..c8e75851a314 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -395,7 +395,6 @@ static int __init acpi_pci_init(void) if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) { printk(KERN_INFO"ACPI FADT declares the system doesn't support PCIe ASPM, so disable it\n"); - pcie_clear_aspm(); pcie_no_aspm(); } diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index cbfbab18be91..1cfbf228fbb1 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -68,7 +68,7 @@ struct pcie_link_state { struct aspm_latency acceptable[8]; }; -static int aspm_disabled, aspm_force, aspm_clear_state; +static int aspm_disabled, aspm_force; static bool aspm_support_enabled = true; static DEFINE_MUTEX(aspm_lock); static LIST_HEAD(link_list); @@ -500,9 +500,6 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev) int pos; u32 reg32; - if (aspm_clear_state) - return -EINVAL; - /* * Some functions in a slot might not all be PCIe functions, * very strange. Disable ASPM for the whole slot @@ -574,9 +571,6 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev) pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) return; - if (aspm_disabled && !aspm_clear_state) - return; - /* VIA has a strange chipset, root port is under a bridge */ if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT && pdev->bus->self) @@ -608,7 +602,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev) * the BIOS's expectation, we'll do so once pci_enable_device() is * called. */ - if (aspm_policy != POLICY_POWERSAVE || aspm_clear_state) { + if (aspm_policy != POLICY_POWERSAVE) { pcie_config_aspm_path(link); pcie_set_clkpm(link, policy_to_clkpm_state(link)); } @@ -649,8 +643,7 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev) struct pci_dev *parent = pdev->bus->self; struct pcie_link_state *link, *root, *parent_link; - if ((aspm_disabled && !aspm_clear_state) || !pci_is_pcie(pdev) || - !parent || !parent->link_state) + if (!pci_is_pcie(pdev) || !parent || !parent->link_state) return; if ((parent->pcie_type != PCI_EXP_TYPE_ROOT_PORT) && (parent->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)) @@ -734,13 +727,18 @@ void pcie_aspm_powersave_config_link(struct pci_dev *pdev) * pci_disable_link_state - disable pci device's link state, so the link will * never enter specific states */ -static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) +static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem, + bool force) { struct pci_dev *parent = pdev->bus->self; struct pcie_link_state *link; - if (aspm_disabled || !pci_is_pcie(pdev)) + if (aspm_disabled && !force) return; + + if (!pci_is_pcie(pdev)) + return; + if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT || pdev->pcie_type == PCI_EXP_TYPE_DOWNSTREAM) parent = pdev; @@ -768,16 +766,31 @@ static void __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) void pci_disable_link_state_locked(struct pci_dev *pdev, int state) { - __pci_disable_link_state(pdev, state, false); + __pci_disable_link_state(pdev, state, false, false); } EXPORT_SYMBOL(pci_disable_link_state_locked); void pci_disable_link_state(struct pci_dev *pdev, int state) { - __pci_disable_link_state(pdev, state, true); + __pci_disable_link_state(pdev, state, true, false); } EXPORT_SYMBOL(pci_disable_link_state); +void pcie_clear_aspm(struct pci_bus *bus) +{ + struct pci_dev *child; + + /* + * Clear any ASPM setup that the firmware has carried out on this bus + */ + list_for_each_entry(child, &bus->devices, bus_list) { + __pci_disable_link_state(child, PCIE_LINK_STATE_L0S | + PCIE_LINK_STATE_L1 | + PCIE_LINK_STATE_CLKPM, + false, true); + } +} + static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp) { int i; @@ -935,6 +948,7 @@ void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev) static int __init pcie_aspm_disable(char *str) { if (!strcmp(str, "off")) { + aspm_policy = POLICY_DEFAULT; aspm_disabled = 1; aspm_support_enabled = false; printk(KERN_INFO "PCIe ASPM is disabled\n"); @@ -947,16 +961,18 @@ static int __init pcie_aspm_disable(char *str) __setup("pcie_aspm=", pcie_aspm_disable); -void pcie_clear_aspm(void) -{ - if (!aspm_force) - aspm_clear_state = 1; -} - void pcie_no_aspm(void) { - if (!aspm_force) + /* + * Disabling ASPM is intended to prevent the kernel from modifying + * existing hardware state, not to clear existing state. To that end: + * (a) set policy to POLICY_DEFAULT in order to avoid changing state + * (b) prevent userspace from changing policy + */ + if (!aspm_force) { + aspm_policy = POLICY_DEFAULT; aspm_disabled = 1; + } } /** diff --git a/include/linux/pci-aspm.h b/include/linux/pci-aspm.h index 7cea7b6c1413..c8320144fe79 100644 --- a/include/linux/pci-aspm.h +++ b/include/linux/pci-aspm.h @@ -29,7 +29,7 @@ extern void pcie_aspm_pm_state_change(struct pci_dev *pdev); extern void pcie_aspm_powersave_config_link(struct pci_dev *pdev); extern void pci_disable_link_state(struct pci_dev *pdev, int state); extern void pci_disable_link_state_locked(struct pci_dev *pdev, int state); -extern void pcie_clear_aspm(void); +extern void pcie_clear_aspm(struct pci_bus *bus); extern void pcie_no_aspm(void); #else static inline void pcie_aspm_init_link_state(struct pci_dev *pdev) @@ -47,7 +47,7 @@ static inline void pcie_aspm_powersave_config_link(struct pci_dev *pdev) static inline void pci_disable_link_state(struct pci_dev *pdev, int state) { } -static inline void pcie_clear_aspm(void) +static inline void pcie_clear_aspm(struct pci_bus *bus) { } static inline void pcie_no_aspm(void) From 027e8d52abdd44bc00e405af83cd2fbfb96c0824 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Mon, 7 Nov 2011 20:55:46 +0900 Subject: [PATCH 04/10] PCI: pciehp: Fix wrong workqueue cleanup Fix improper workqueue cleanup. In the current pciehp, pcied_cleanup() calls destroy_workqueue() before calling pcie_port_service_unregister(). This causes kernel oops because flush_workqueue() is called in the pcie_port_service_unregister() code path after the workqueue was destroyed. So pcied_cleanup() must call pcie_port_service_unregister() first before calling destroy_workqueue(). Signed-off-by: Kenji Kaneshige Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/pciehp_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 7ac8358df8fd..a13ad1308adb 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -366,9 +366,9 @@ static int __init pcied_init(void) static void __exit pcied_cleanup(void) { dbg("unload_pciehpd()\n"); + pcie_port_service_unregister(&hpdriver_portdrv); destroy_workqueue(pciehp_ordered_wq); destroy_workqueue(pciehp_wq); - pcie_port_service_unregister(&hpdriver_portdrv); info(DRIVER_DESC " version: " DRIVER_VERSION " unloaded\n"); } From 486b10b9f43500741cd63a878d0ef23cd87fc66d Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Mon, 7 Nov 2011 20:56:50 +0900 Subject: [PATCH 05/10] PCI: pciehp: Handle push button event asynchronously Use non-ordered workqueue for attention button events. Attention button events on each slot can be handled asynchronously. So we should use non-ordered workqueue. This patch also removes ordered workqueue in pciehp as a result. Signed-off-by: Kenji Kaneshige Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/pciehp.h | 1 - drivers/pci/hotplug/pciehp_core.c | 9 --------- drivers/pci/hotplug/pciehp_ctrl.c | 4 ++-- drivers/pci/hotplug/pciehp_hpc.c | 1 - 4 files changed, 2 insertions(+), 13 deletions(-) diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 838f571027b7..9a33fdde2d16 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -45,7 +45,6 @@ extern int pciehp_poll_time; extern int pciehp_debug; extern int pciehp_force; extern struct workqueue_struct *pciehp_wq; -extern struct workqueue_struct *pciehp_ordered_wq; #define dbg(format, arg...) \ do { \ diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index a13ad1308adb..b8c99d35ac97 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -43,7 +43,6 @@ int pciehp_poll_mode; int pciehp_poll_time; int pciehp_force; struct workqueue_struct *pciehp_wq; -struct workqueue_struct *pciehp_ordered_wq; #define DRIVER_VERSION "0.4" #define DRIVER_AUTHOR "Dan Zink , Greg Kroah-Hartman , Dely Sy " @@ -345,18 +344,11 @@ static int __init pcied_init(void) if (!pciehp_wq) return -ENOMEM; - pciehp_ordered_wq = alloc_ordered_workqueue("pciehp_ordered", 0); - if (!pciehp_ordered_wq) { - destroy_workqueue(pciehp_wq); - return -ENOMEM; - } - pciehp_firmware_init(); retval = pcie_port_service_register(&hpdriver_portdrv); dbg("pcie_port_service_register = %d\n", retval); info(DRIVER_DESC " version: " DRIVER_VERSION "\n"); if (retval) { - destroy_workqueue(pciehp_ordered_wq); destroy_workqueue(pciehp_wq); dbg("Failure to register service\n"); } @@ -367,7 +359,6 @@ static void __exit pcied_cleanup(void) { dbg("unload_pciehpd()\n"); pcie_port_service_unregister(&hpdriver_portdrv); - destroy_workqueue(pciehp_ordered_wq); destroy_workqueue(pciehp_wq); info(DRIVER_DESC " version: " DRIVER_VERSION " unloaded\n"); } diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 085dbb5fc168..27f44295a657 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -344,7 +344,7 @@ void pciehp_queue_pushbutton_work(struct work_struct *work) kfree(info); goto out; } - queue_work(pciehp_ordered_wq, &info->work); + queue_work(pciehp_wq, &info->work); out: mutex_unlock(&p_slot->lock); } @@ -439,7 +439,7 @@ static void handle_surprise_event(struct slot *p_slot) else p_slot->state = POWERON_STATE; - queue_work(pciehp_ordered_wq, &info->work); + queue_work(pciehp_wq, &info->work); } static void interrupt_event_handler(struct work_struct *work) diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 7b1414810ae3..bcdbb1643621 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -806,7 +806,6 @@ static void pcie_cleanup_slot(struct controller *ctrl) struct slot *slot = ctrl->slot; cancel_delayed_work(&slot->work); flush_workqueue(pciehp_wq); - flush_workqueue(pciehp_ordered_wq); kfree(slot); } From d90116ea38f7768dac0349f01ffbc2663d63b7e9 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 6 Nov 2011 23:11:28 +0100 Subject: [PATCH 06/10] PCI/ACPI: Make acpiphp ignore root bridges using SHPC native hotplug If the kernel has requested control of the SHPC native hotplug feature for a given root bridge, the acpiphp driver should not try to handle that root bridge and it should leave it to shpchp. Failing to do so causes problems to happen if shpchp is loaded and unloaded before loading acpiphp (ACPI-based hotplug won't work in that case anyway). To address this issue make find_root_bridges() ignore PCI root bridges with SHPC native hotplug enabled and make add_bridge() return error code if SHPC native hotplug is enabled for the given root bridge. This causes acpiphp to refuse to load if SHPC native hotplug is enabled for all root bridges and to refuse binding to the root bridges with SHPC native hotplug enabled. Reviewed-by: Kenji Kaneshige Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/acpiphp_glue.c | 4 ++-- include/linux/acpi.h | 4 ++++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index fce1c54a0c8d..ba43c037de80 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -467,7 +467,7 @@ static int add_bridge(acpi_handle handle) * granted by the BIOS for it. */ root = acpi_pci_find_root(handle); - if (root && (root->osc_control_set & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL)) + if (root && (root->osc_control_set & OSC_PCI_NATIVE_HOTPLUG)) return -ENODEV; /* if the bridge doesn't have _STA, we assume it is always there */ @@ -1395,7 +1395,7 @@ find_root_bridges(acpi_handle handle, u32 lvl, void *context, void **rv) if (!root) return AE_OK; - if (root->osc_control_set & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL) + if (root->osc_control_set & OSC_PCI_NATIVE_HOTPLUG) return AE_OK; (*count)++; diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 6001b4da39dd..627a3a42e4d8 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -302,6 +302,10 @@ extern bool osc_sb_apei_support_acked; OSC_PCI_EXPRESS_PME_CONTROL | \ OSC_PCI_EXPRESS_AER_CONTROL | \ OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL) + +#define OSC_PCI_NATIVE_HOTPLUG (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL | \ + OSC_SHPC_NATIVE_HP_CONTROL) + extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 *mask, u32 req); extern void acpi_early_init(void); From a424948dde8421089826e2f782d0efe9e565707e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 6 Nov 2011 22:21:46 +0100 Subject: [PATCH 07/10] PCI/ACPI/PM: Avoid resuming devices that don't signal PME Modify pci_acpi_wake_dev() to avoid resuming PME-capable devices whose PME Status bits are not set, which may happen currently if several devices are associated with the same wakeup GPE and all of them are notified whenever at least one of them signals PME. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/pci-acpi.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index c8e75851a314..060fd22a1103 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -45,16 +45,20 @@ static void pci_acpi_wake_dev(acpi_handle handle, u32 event, void *context) { struct pci_dev *pci_dev = context; - if (event == ACPI_NOTIFY_DEVICE_WAKE && pci_dev) { + if (event != ACPI_NOTIFY_DEVICE_WAKE || !pci_dev) + return; + + if (!pci_dev->pm_cap || !pci_dev->pme_support + || pci_check_pme_status(pci_dev)) { if (pci_dev->pme_poll) pci_dev->pme_poll = false; pci_wakeup_event(pci_dev); - pci_check_pme_status(pci_dev); pm_runtime_resume(&pci_dev->dev); - if (pci_dev->subordinate) - pci_pme_wakeup_bus(pci_dev->subordinate); } + + if (pci_dev->subordinate) + pci_pme_wakeup_bus(pci_dev->subordinate); } /** From a776c491ca5e38c26d9f66923ff574d041e747f4 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 17 Oct 2011 11:46:06 -0700 Subject: [PATCH 08/10] PCI: msi: Disable msi interrupts when we initialize a pci device I traced a nasty kexec on panic boot failure to the fact that we had screaming msi interrupts and we were not disabling the msi messages at kernel startup. The booting kernel had not enabled those interupts so was not prepared to handle them. I can see no reason why we would ever want to leave the msi interrupts enabled at boot if something else has enabled those interrupts. The pci spec specifies that msi interrupts should be off by default. Drivers are expected to enable the msi interrupts if they want to use them. Our interrupt handling code reprograms the interrupt handlers at boot and will not be be able to do anything useful with an unexpected interrupt. This patch applies cleanly all of the way back to 2.6.32 where I noticed the problem. Cc: stable@kernel.org Signed-off-by: Eric W. Biederman Signed-off-by: Jesse Barnes --- drivers/pci/msi.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index e6b6b9c67023..337e16ab4a92 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -981,5 +981,15 @@ EXPORT_SYMBOL(pci_msi_enabled); void pci_msi_init_pci_dev(struct pci_dev *dev) { + int pos; INIT_LIST_HEAD(&dev->msi_list); + + /* Disable the msi hardware to avoid screaming interrupts + * during boot. This is the power on reset default so + * usually this should be a noop. + */ + pos = pci_find_capability(dev, PCI_CAP_ID_MSI); + if (pos) + msi_set_enable(dev, pos, 0); + msix_set_enable(dev, 0); } From 60fe823837d10673500ff685c01eb2f896fe5849 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 11 Nov 2011 10:06:56 -0700 Subject: [PATCH 09/10] PCI: Enable is not exposed as a PASID capability The PASID ECN indicates bit 0 is reserved in the capability register. Switch pci_enable_pasid() to error if PASID is already enabled and don't expose enable as a feature in pci_pasid_features(). Reviewed-by: Joerg Roedel Tested-by: Joerg Roedel Signed-off-by: Alex Williamson Signed-off-by: Jesse Barnes --- drivers/pci/ats.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 831e1920386c..8e95a123d37a 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -348,7 +348,7 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) pci_read_config_word(pdev, pos + PCI_PASID_CONTROL_OFF, &control); pci_read_config_word(pdev, pos + PCI_PASID_CAP_OFF, &supported); - if (!(supported & PCI_PASID_ENABLE)) + if (control & PCI_PASID_ENABLE) return -EINVAL; supported &= PCI_PASID_EXEC | PCI_PASID_PRIV; @@ -390,7 +390,6 @@ EXPORT_SYMBOL_GPL(pci_disable_pasid); * Returns a negative value when no PASI capability is present. * Otherwise is returns a bitmask with supported features. Current * features reported are: - * PCI_PASID_ENABLE - PASID capability can be enabled * PCI_PASID_EXEC - Execute permission supported * PCI_PASID_PRIV - Priviledged mode supported */ @@ -405,7 +404,7 @@ int pci_pasid_features(struct pci_dev *pdev) pci_read_config_word(pdev, pos + PCI_PASID_CAP_OFF, &supported); - supported &= PCI_PASID_ENABLE | PCI_PASID_EXEC | PCI_PASID_PRIV; + supported &= PCI_PASID_EXEC | PCI_PASID_PRIV; return supported; } From 91f57d5e1be3db1e079c8696f1eab214f1c7922d Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 11 Nov 2011 10:07:36 -0700 Subject: [PATCH 10/10] PCI: More PRI/PASID cleanup More consistency cleanups. Drop the _OFF, separate and indent CTRL/CAP/STATUS bit definitions. This helped find the previous mis-use of bit 0 in the PASID capability register. Reviewed-by: Joerg Roedel Tested-by: Joerg Roedel Signed-off-by: Alex Williamson Signed-off-by: Jesse Barnes --- drivers/pci/ats.c | 69 ++++++++++++++++++++-------------------- include/linux/pci_regs.h | 30 +++++++++-------- 2 files changed, 51 insertions(+), 48 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 8e95a123d37a..2df49af6cc90 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -178,17 +178,18 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); - pci_read_config_word(pdev, pos + PCI_PRI_STATUS_OFF, &status); - if ((control & PCI_PRI_ENABLE) || !(status & PCI_PRI_STATUS_STOPPED)) + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); + if ((control & PCI_PRI_CTRL_ENABLE) || + !(status & PCI_PRI_STATUS_STOPPED)) return -EBUSY; - pci_read_config_dword(pdev, pos + PCI_PRI_MAX_REQ_OFF, &max_requests); + pci_read_config_dword(pdev, pos + PCI_PRI_MAX_REQ, &max_requests); reqs = min(max_requests, reqs); - pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ_OFF, reqs); + pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ, reqs); - control |= PCI_PRI_ENABLE; - pci_write_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, control); + control |= PCI_PRI_CTRL_ENABLE; + pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); return 0; } @@ -209,9 +210,9 @@ void pci_disable_pri(struct pci_dev *pdev) if (!pos) return; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); - control &= ~PCI_PRI_ENABLE; - pci_write_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, control); + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + control &= ~PCI_PRI_CTRL_ENABLE; + pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); } EXPORT_SYMBOL_GPL(pci_disable_pri); @@ -230,9 +231,9 @@ bool pci_pri_enabled(struct pci_dev *pdev) if (!pos) return false; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); - return (control & PCI_PRI_ENABLE) ? true : false; + return (control & PCI_PRI_CTRL_ENABLE) ? true : false; } EXPORT_SYMBOL_GPL(pci_pri_enabled); @@ -252,13 +253,13 @@ int pci_reset_pri(struct pci_dev *pdev) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); - if (control & PCI_PRI_ENABLE) + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + if (control & PCI_PRI_CTRL_ENABLE) return -EBUSY; - control |= PCI_PRI_RESET; + control |= PCI_PRI_CTRL_RESET; - pci_write_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, control); + pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); return 0; } @@ -285,10 +286,10 @@ bool pci_pri_stopped(struct pci_dev *pdev) if (!pos) return true; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); - pci_read_config_word(pdev, pos + PCI_PRI_STATUS_OFF, &status); + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); - if (control & PCI_PRI_ENABLE) + if (control & PCI_PRI_CTRL_ENABLE) return false; return (status & PCI_PRI_STATUS_STOPPED) ? true : false; @@ -314,11 +315,11 @@ int pci_pri_status(struct pci_dev *pdev) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PRI_CONTROL_OFF, &control); - pci_read_config_word(pdev, pos + PCI_PRI_STATUS_OFF, &status); + pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); /* Stopped bit is undefined when enable == 1, so clear it */ - if (control & PCI_PRI_ENABLE) + if (control & PCI_PRI_CTRL_ENABLE) status &= ~PCI_PRI_STATUS_STOPPED; return status; @@ -345,21 +346,21 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CONTROL_OFF, &control); - pci_read_config_word(pdev, pos + PCI_PASID_CAP_OFF, &supported); + pci_read_config_word(pdev, pos + PCI_PASID_CTRL, &control); + pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); - if (control & PCI_PASID_ENABLE) + if (control & PCI_PASID_CTRL_ENABLE) return -EINVAL; - supported &= PCI_PASID_EXEC | PCI_PASID_PRIV; + supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV; /* User wants to enable anything unsupported? */ if ((supported & features) != features) return -EINVAL; - control = PCI_PASID_ENABLE | features; + control = PCI_PASID_CTRL_ENABLE | features; - pci_write_config_word(pdev, pos + PCI_PASID_CONTROL_OFF, control); + pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control); return 0; } @@ -379,7 +380,7 @@ void pci_disable_pasid(struct pci_dev *pdev) if (!pos) return; - pci_write_config_word(pdev, pos + PCI_PASID_CONTROL_OFF, control); + pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control); } EXPORT_SYMBOL_GPL(pci_disable_pasid); @@ -390,8 +391,8 @@ EXPORT_SYMBOL_GPL(pci_disable_pasid); * Returns a negative value when no PASI capability is present. * Otherwise is returns a bitmask with supported features. Current * features reported are: - * PCI_PASID_EXEC - Execute permission supported - * PCI_PASID_PRIV - Priviledged mode supported + * PCI_PASID_CAP_EXEC - Execute permission supported + * PCI_PASID_CAP_PRIV - Priviledged mode supported */ int pci_pasid_features(struct pci_dev *pdev) { @@ -402,9 +403,9 @@ int pci_pasid_features(struct pci_dev *pdev) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CAP_OFF, &supported); + pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); - supported &= PCI_PASID_EXEC | PCI_PASID_PRIV; + supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV; return supported; } @@ -428,7 +429,7 @@ int pci_max_pasids(struct pci_dev *pdev) if (!pos) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CAP_OFF, &supported); + pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); supported = (supported & PASID_NUMBER_MASK) >> PASID_NUMBER_SHIFT; diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index 090d3a9f5b26..28fe380cb19d 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -666,22 +666,24 @@ #define PCI_ATS_MIN_STU 12 /* shift of minimum STU block */ /* Page Request Interface */ -#define PCI_PRI_CONTROL_OFF 0x04 /* Offset of control register */ -#define PCI_PRI_STATUS_OFF 0x06 /* Offset of status register */ -#define PCI_PRI_ENABLE 0x0001 /* Enable mask */ -#define PCI_PRI_RESET 0x0002 /* Reset bit mask */ -#define PCI_PRI_STATUS_RF 0x0001 /* Request Failure */ -#define PCI_PRI_STATUS_UPRGI 0x0002 /* Unexpected PRG index */ -#define PCI_PRI_STATUS_STOPPED 0x0100 /* PRI Stopped */ -#define PCI_PRI_MAX_REQ_OFF 0x08 /* Cap offset for max reqs supported */ -#define PCI_PRI_ALLOC_REQ_OFF 0x0c /* Cap offset for max reqs allowed */ +#define PCI_PRI_CTRL 0x04 /* PRI control register */ +#define PCI_PRI_CTRL_ENABLE 0x01 /* Enable */ +#define PCI_PRI_CTRL_RESET 0x02 /* Reset */ +#define PCI_PRI_STATUS 0x06 /* PRI status register */ +#define PCI_PRI_STATUS_RF 0x001 /* Response Failure */ +#define PCI_PRI_STATUS_UPRGI 0x002 /* Unexpected PRG index */ +#define PCI_PRI_STATUS_STOPPED 0x100 /* PRI Stopped */ +#define PCI_PRI_MAX_REQ 0x08 /* PRI max reqs supported */ +#define PCI_PRI_ALLOC_REQ 0x0c /* PRI max reqs allowed */ /* PASID capability */ -#define PCI_PASID_CAP_OFF 0x04 /* PASID feature register */ -#define PCI_PASID_CONTROL_OFF 0x06 /* PASID control register */ -#define PCI_PASID_ENABLE 0x01 /* Enable/Supported bit */ -#define PCI_PASID_EXEC 0x02 /* Exec permissions Enable/Supported */ -#define PCI_PASID_PRIV 0x04 /* Priviledge Mode Enable/Support */ +#define PCI_PASID_CAP 0x04 /* PASID feature register */ +#define PCI_PASID_CAP_EXEC 0x02 /* Exec permissions Supported */ +#define PCI_PASID_CAP_PRIV 0x04 /* Priviledge Mode Supported */ +#define PCI_PASID_CTRL 0x06 /* PASID control register */ +#define PCI_PASID_CTRL_ENABLE 0x01 /* Enable bit */ +#define PCI_PASID_CTRL_EXEC 0x02 /* Exec permissions Enable */ +#define PCI_PASID_CTRL_PRIV 0x04 /* Priviledge Mode Enable */ /* Single Root I/O Virtualization */ #define PCI_SRIOV_CAP 0x04 /* SR-IOV Capabilities */