In some specific scenario it's possible that the pci_create_resource_files() gets called multiple times and the created entry actually gets wrongly deleted with extreme case of having a NULL pointer dereference when the PCI is removed.
This mainly happen due to bad timing where the PCI bus is adding PCI devices and at the same time the sysfs code is adding the entry causing double execution of the pci_create_resource_files function and kernel WARNING.
To be more precise there is a race between the late_initcall of pci-sysfs with pci_sysfs_init and PCI bus.c pci_bus_add_device that also call pci_create_sysfs_dev_files.
With correct amount of ""luck"" (or better say bad luck) pci_create_sysfs_dev_files in bus.c might be called with pci_sysfs_init is executing the loop.
This has been reported multiple times and on multiple system, like imx6 system, ipq806x systems...
To address this, imlement multiple improvement to the implementation: 1. Add a bool to pci_dev to flag when sysfs entry are created (sysfs_init) 2. Implement a simple completion to wait pci_sysfs_init execution. 3. Permit additional call of pci_create_sysfs_dev_files only after pci_sysfs_init has finished.
With such logic in place, we address al kind of timing problem with minimal change to any driver.
A notice worth to mention is that the remove function are not affected by this as the pci_remove_resource_files have enough check in place to always work and it's always called by pci_stop_dev.
Cc: stable@vger.kernel.org Reported-by: Krzysztof Hałasa khalasa@piap.pl Closes: https://bugzilla.kernel.org/show_bug.cgi?id=215515 Signed-off-by: Christian Marangi ansuelsmth@gmail.com --- drivers/pci/pci-sysfs.c | 34 +++++++++++++++++++++++++++++----- include/linux/pci.h | 1 + 2 files changed, 30 insertions(+), 5 deletions(-)
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 71a36f57ef57..cab3aa27f947 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -14,6 +14,7 @@ */
#include <linux/bitfield.h> +#include <linux/completion.h> #include <linux/kernel.h> #include <linux/sched.h> #include <linux/pci.h> @@ -37,6 +38,7 @@ #endif
static int sysfs_initialized; /* = 0 */ +static DECLARE_COMPLETION(sysfs_init_completion);
/* show configuration fields */ #define pci_config_attr(field, format_string) \ @@ -1652,12 +1654,32 @@ static const struct attribute_group pci_dev_resource_resize_group = { .is_visible = resource_resize_is_visible, };
+static int __pci_create_sysfs_dev_files(struct pci_dev *pdev) +{ + int ret; + + ret = pci_create_resource_files(pdev); + if (ret) + return ret; + + /* on success set sysfs correctly created */ + pdev->sysfs_init = true; + return 0; +} + int __must_check pci_create_sysfs_dev_files(struct pci_dev *pdev) { if (!sysfs_initialized) return -EACCES;
- return pci_create_resource_files(pdev); + /* sysfs entry already created */ + if (pdev->sysfs_init) + return 0; + + /* wait for pci_sysfs_init */ + wait_for_completion(&sysfs_init_completion); + + return __pci_create_sysfs_dev_files(pdev); }
/** @@ -1678,21 +1700,23 @@ static int __init pci_sysfs_init(void) { struct pci_dev *pdev = NULL; struct pci_bus *pbus = NULL; - int retval; + int retval = 0;
sysfs_initialized = 1; for_each_pci_dev(pdev) { - retval = pci_create_sysfs_dev_files(pdev); + retval = __pci_create_sysfs_dev_files(pdev); if (retval) { pci_dev_put(pdev); - return retval; + goto exit; } }
while ((pbus = pci_find_next_bus(pbus))) pci_create_legacy_files(pbus);
- return 0; +exit: + complete_all(&sysfs_init_completion); + return retval; } late_initcall(pci_sysfs_init);
diff --git a/include/linux/pci.h b/include/linux/pci.h index f3f6d6dee3ae..f417a0528f01 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -480,6 +480,7 @@ struct pci_dev { unsigned int non_mappable_bars:1; /* BARs can't be mapped to user-space */ pci_dev_flags_t dev_flags; atomic_t enable_cnt; /* pci_enable_device has been called */ + bool sysfs_init; /* sysfs entry has been created */
spinlock_t pcie_cap_lock; /* Protects RMW ops in capability accessors */ u32 saved_config_space[16]; /* Config space saved at suspend time */