summaryrefslogtreecommitdiff
path: root/arch/s390/pci/pci.c
diff options
context:
space:
mode:
authorGerd Bayer <gbayer@linux.ibm.com>2023-11-10 18:27:06 +0300
committerHeiko Carstens <hca@linux.ibm.com>2024-02-20 16:37:32 +0300
commitbcb5d6c769039c8358a2359e7c3ea5d97ce93108 (patch)
tree502f45fa792da2dbe28ae4d3a8b4364e7fa59b7a /arch/s390/pci/pci.c
parent0d48566d4b58946c8e1b0baac0347616060a81c9 (diff)
downloadlinux-bcb5d6c769039c8358a2359e7c3ea5d97ce93108.tar.xz
s390/pci: introduce lock to synchronize state of zpci_dev's
There's a number of tasks that need the state of a zpci device to be stable. Other tasks need to be synchronized as they change the state. State changes could be generated by the system as availability or error events, or be requested by the user through manipulations in sysfs. Some other actions accessible through sysfs - like device resets - need the state to be stable. Unsynchronized state handling could lead to unusable devices. This has been observed in cases of concurrent state changes through systemd udev rules and DPM boot control. Some breakage can be provoked by artificial tests, e.g. through repetitively injecting "recover" on a PCI function through sysfs while running a "hotplug remove/add" in a loop through a PCI slot's "power" attribute in sysfs. After a few iterations this could result in a kernel oops. So introduce a new mutex "state_lock" to guard the state property of the struct zpci_dev. Acquire this lock in all task that modify the state: - hotplug add and remove, through the PCI hotplug slot entry, - avaiability events, as reported by the platform, - error events, as reported by the platform, - during device resets, explicit through sysfs requests or implict through the common PCI layer. Break out an inner _do_recover() routine out of recover_store() to separte the necessary synchronizations from the actual manipulations of the zpci_dev required for the reset. With the following changes I was able to run the inject loops for hours without hitting an error. Signed-off-by: Gerd Bayer <gbayer@linux.ibm.com> Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com> Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Diffstat (limited to 'arch/s390/pci/pci.c')
-rw-r--r--arch/s390/pci/pci.c11
1 files changed, 9 insertions, 2 deletions
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index dff609e8a2a0..17267f659d22 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -28,6 +28,7 @@
#include <linux/jump_label.h>
#include <linux/pci.h>
#include <linux/printk.h>
+#include <linux/lockdep.h>
#include <asm/isc.h>
#include <asm/airq.h>
@@ -730,12 +731,12 @@ EXPORT_SYMBOL_GPL(zpci_disable_device);
* equivalent to its state during boot when first probing a driver.
* Consequently after reset the PCI function requires re-initialization via the
* common PCI code including re-enabling IRQs via pci_alloc_irq_vectors()
- * and enabling the function via e.g.pci_enablde_device_flags().The caller
+ * and enabling the function via e.g. pci_enable_device_flags(). The caller
* must guard against concurrent reset attempts.
*
* In most cases this function should not be called directly but through
* pci_reset_function() or pci_reset_bus() which handle the save/restore and
- * locking.
+ * locking - asserted by lockdep.
*
* Return: 0 on success and an error value otherwise
*/
@@ -744,6 +745,7 @@ int zpci_hot_reset_device(struct zpci_dev *zdev)
u8 status;
int rc;
+ lockdep_assert_held(&zdev->state_lock);
zpci_dbg(3, "rst fid:%x, fh:%x\n", zdev->fid, zdev->fh);
if (zdev_enabled(zdev)) {
/* Disables device access, DMAs and IRQs (reset state) */
@@ -806,6 +808,7 @@ struct zpci_dev *zpci_create_device(u32 fid, u32 fh, enum zpci_state state)
zdev->state = state;
kref_init(&zdev->kref);
+ mutex_init(&zdev->state_lock);
mutex_init(&zdev->fmb_lock);
mutex_init(&zdev->kzdev_lock);
@@ -870,6 +873,10 @@ int zpci_deconfigure_device(struct zpci_dev *zdev)
{
int rc;
+ lockdep_assert_held(&zdev->state_lock);
+ if (zdev->state != ZPCI_FN_STATE_CONFIGURED)
+ return 0;
+
if (zdev->zbus->bus)
zpci_bus_remove_device(zdev, false);