| Message ID | 20201218174011.340514-2-s.miroshnichenko@yadro.com (mailing list archive) |
|---|---|
| State | New, archived |
| Delegated to: | Bjorn Helgaas |
| Headers | show |
| Series | PCI: Allow BAR movement during boot and hotplug | expand |
Hi, The kbuild test robot has reported a build error on some configs, pointing out that another #ifdef should be used. I've checked, the DEBUG_LOCK_ALLOC seems to suit better here: On Fri, 2020-12-18 at 20:39 +0300, Sergei Miroshnichenko wrote: > This is a yet another approach to fix an old [1-2] concurrency issue, > when: > - two or more devices are being hot-added into a bridge which was > initially empty; > - a bridge with two or more devices is being hot-added; > - the BIOS/bootloader/firmware doesn't pre-enable bridges during > boot; > > ... > > +#ifdef CONFIG_PROVE_LOCKING +#ifdef CONFIG_DEBUG_LOCK_ALLOC > +static int pci_bridge_depth(struct pci_dev *dev) > +{ > + struct pci_dev *bridge = pci_upstream_bridge(dev); > + > + if (!bridge) > + return 0; > + > + return 1 + pci_bridge_depth(bridge); > +} > +#endif /* CONFIG_PROVE_LOCKING */ +#endif /* CONFIG_DEBUG_LOCK_ALLOC */ > + > static void pci_enable_bridge(struct pci_dev *dev) > { > struct pci_dev *bridge; > int retval; > > + mutex_lock_nested(&dev->enable_mutex, pci_bridge_depth(dev)); > + > bridge = pci_upstream_bridge(dev); > if (bridge) > pci_enable_bridge(bridge); Is there a proper way to send a "hotfix" for a single patch of the series of 26, without resending them all? Best regards, Serge
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index b9fecc25d213..076b908127fe 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1844,11 +1844,25 @@ int pci_reenable_device(struct pci_dev *dev) } EXPORT_SYMBOL(pci_reenable_device); +#ifdef CONFIG_PROVE_LOCKING +static int pci_bridge_depth(struct pci_dev *dev) +{ + struct pci_dev *bridge = pci_upstream_bridge(dev); + + if (!bridge) + return 0; + + return 1 + pci_bridge_depth(bridge); +} +#endif /* CONFIG_PROVE_LOCKING */ + static void pci_enable_bridge(struct pci_dev *dev) { struct pci_dev *bridge; int retval; + mutex_lock_nested(&dev->enable_mutex, pci_bridge_depth(dev)); + bridge = pci_upstream_bridge(dev); if (bridge) pci_enable_bridge(bridge); @@ -1856,6 +1870,7 @@ static void pci_enable_bridge(struct pci_dev *dev) if (pci_is_enabled(dev)) { if (!dev->is_busmaster) pci_set_master(dev); + mutex_unlock(&dev->enable_mutex); return; } @@ -1864,6 +1879,7 @@ static void pci_enable_bridge(struct pci_dev *dev) pci_err(dev, "Error enabling bridge (%d), continuing\n", retval); pci_set_master(dev); + mutex_unlock(&dev->enable_mutex); } static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 953f15abc850..2f9631287719 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -2240,6 +2240,7 @@ struct pci_dev *pci_alloc_dev(struct pci_bus *bus) INIT_LIST_HEAD(&dev->bus_list); dev->dev.type = &pci_dev_type; dev->bus = pci_bus_get(bus); + mutex_init(&dev->enable_mutex); return dev; } diff --git a/include/linux/pci.h b/include/linux/pci.h index b32126d26997..81d54889bd51 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -455,6 +455,7 @@ struct pci_dev { unsigned int no_command_memory:1; /* No PCI_COMMAND_MEMORY */ pci_dev_flags_t dev_flags; atomic_t enable_cnt; /* pci_enable_device has been called */ + struct mutex enable_mutex; u32 saved_config_space[16]; /* Config space saved at suspend time */ struct hlist_head saved_cap_space;
This is a yet another approach to fix an old [1-2] concurrency issue, when: - two or more devices are being hot-added into a bridge which was initially empty; - a bridge with two or more devices is being hot-added; - the BIOS/bootloader/firmware doesn't pre-enable bridges during boot; The problem is that a bridge is reported as enabled before the MEM/IO bits are actually written to the PCI_COMMAND register, so another driver thread starts memory requests through the not-yet-enabled bridge: CPU0 CPU1 pci_enable_device_mem() pci_enable_device_mem() pci_enable_bridge() pci_enable_bridge() pci_is_enabled() return false; atomic_inc_return(enable_cnt) Start actual enabling the bridge ... pci_is_enabled() ... return true; ... Start memory requests <-- FAIL ... Set the PCI_COMMAND_MEMORY bit <-- Must wait for this Protect the pci_enable_bridge(), similarly to the previous solution from commit 40f11adc7cd9 ("PCI: Avoid race while enabling upstream bridges"), but adding per-device mutexes. To prevent false positives from the lockdep, use a lock_nested() with a "depth" of a bridge within the PCI topology. CC: Srinath Mannam <srinath.mannam@broadcom.com> CC: Marta Rybczynska <mrybczyn@kalray.eu> Signed-off-by: Sergei Miroshnichenko <s.miroshnichenko@yadro.com> [1] https://lore.kernel.org/linux-pci/1501858648-22228-1-git-send-email-srinath.mannam@broadcom.com/T/#u [RFC PATCH v3] pci: Concurrency issue during pci enable bridge [2] https://lore.kernel.org/linux-pci/744877924.5841545.1521630049567.JavaMail.zimbra@kalray.eu/T/#u [RFC PATCH] nvme: avoid race-conditions when enabling devices --- drivers/pci/pci.c | 16 ++++++++++++++++ drivers/pci/probe.c | 1 + include/linux/pci.h | 1 + 3 files changed, 18 insertions(+)