From 9c51026509d7fd11d84e0035008e1a9768960f2b Mon Sep 17 00:00:00 2001 From: Syed Asifful Dayyan Date: Fri, 6 Mar 2015 18:40:42 +0100 Subject: brcmfmac: Add support for BCM4345 SDIO chipset. These changes add support for BCM4345 SDIO chipset. Reviewed-by: Pieter-Paul Giesberts Reviewed-by: Arend Van Spriel Reviewed-by: Hante Meuleman Reviewed-by: Daniel (Deognyoun) Kim Signed-off-by: Syed Asifful Dayyan Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo --- include/linux/mmc/sdio_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h index 996807963716..91397858dc95 100644 --- a/include/linux/mmc/sdio_ids.h +++ b/include/linux/mmc/sdio_ids.h @@ -33,6 +33,7 @@ #define SDIO_DEVICE_ID_BROADCOM_43341 0xa94d #define SDIO_DEVICE_ID_BROADCOM_4335_4339 0x4335 #define SDIO_DEVICE_ID_BROADCOM_43362 0xa962 +#define SDIO_DEVICE_ID_BROADCOM_4345 0x4345 #define SDIO_DEVICE_ID_BROADCOM_4354 0x4354 #define SDIO_VENDOR_ID_INTEL 0x0089 -- cgit v1.2.3 From 702131e2a393b45174be326f1dbe20b658b4f157 Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Thu, 5 Mar 2015 18:25:10 +0100 Subject: bcma: move PCI IRQ control function to host specific code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This function isn't really related to any bus core. It touches PCI device config registers only, so move it to the (PCI) host file. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo --- drivers/bcma/driver_pci.c | 33 ------------------------- drivers/bcma/host_pci.c | 34 ++++++++++++++++++++++++++ drivers/net/wireless/b43/main.c | 2 +- drivers/net/wireless/brcm80211/brcmsmac/main.c | 2 +- include/linux/bcma/bcma.h | 9 +++++++ include/linux/bcma/bcma_driver_pci.h | 2 -- 6 files changed, 45 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/drivers/bcma/driver_pci.c b/drivers/bcma/driver_pci.c index cfd35bc1c5a3..f499a469e66d 100644 --- a/drivers/bcma/driver_pci.c +++ b/drivers/bcma/driver_pci.c @@ -282,39 +282,6 @@ void bcma_core_pci_power_save(struct bcma_bus *bus, bool up) } EXPORT_SYMBOL_GPL(bcma_core_pci_power_save); -int bcma_core_pci_irq_ctl(struct bcma_bus *bus, struct bcma_device *core, - bool enable) -{ - struct pci_dev *pdev; - u32 coremask, tmp; - int err = 0; - - if (bus->hosttype != BCMA_HOSTTYPE_PCI) { - /* This bcma device is not on a PCI host-bus. So the IRQs are - * not routed through the PCI core. - * So we must not enable routing through the PCI core. */ - goto out; - } - - pdev = bus->host_pci; - - err = pci_read_config_dword(pdev, BCMA_PCI_IRQMASK, &tmp); - if (err) - goto out; - - coremask = BIT(core->core_index) << 8; - if (enable) - tmp |= coremask; - else - tmp &= ~coremask; - - err = pci_write_config_dword(pdev, BCMA_PCI_IRQMASK, tmp); - -out: - return err; -} -EXPORT_SYMBOL_GPL(bcma_core_pci_irq_ctl); - static void bcma_core_pci_extend_L1timer(struct bcma_drv_pci *pc, bool extend) { u32 w; diff --git a/drivers/bcma/host_pci.c b/drivers/bcma/host_pci.c index a62a2f9091f5..0856189c065f 100644 --- a/drivers/bcma/host_pci.c +++ b/drivers/bcma/host_pci.c @@ -351,3 +351,37 @@ void bcma_host_pci_down(struct bcma_bus *bus) bcma_core_pci_down(&bus->drv_pci[0]); } EXPORT_SYMBOL_GPL(bcma_host_pci_down); + +/* See also si_pci_setup */ +int bcma_host_pci_irq_ctl(struct bcma_bus *bus, struct bcma_device *core, + bool enable) +{ + struct pci_dev *pdev; + u32 coremask, tmp; + int err = 0; + + if (bus->hosttype != BCMA_HOSTTYPE_PCI) { + /* This bcma device is not on a PCI host-bus. So the IRQs are + * not routed through the PCI core. + * So we must not enable routing through the PCI core. */ + goto out; + } + + pdev = bus->host_pci; + + err = pci_read_config_dword(pdev, BCMA_PCI_IRQMASK, &tmp); + if (err) + goto out; + + coremask = BIT(core->core_index) << 8; + if (enable) + tmp |= coremask; + else + tmp &= ~coremask; + + err = pci_write_config_dword(pdev, BCMA_PCI_IRQMASK, tmp); + +out: + return err; +} +EXPORT_SYMBOL_GPL(bcma_host_pci_irq_ctl); diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c index ac99798570e8..998490a7b167 100644 --- a/drivers/net/wireless/b43/main.c +++ b/drivers/net/wireless/b43/main.c @@ -4866,7 +4866,7 @@ static int b43_wireless_core_init(struct b43_wldev *dev) switch (dev->dev->bus_type) { #ifdef CONFIG_B43_BCMA case B43_BUS_BCMA: - bcma_core_pci_irq_ctl(dev->dev->bdev->bus, + bcma_host_pci_irq_ctl(dev->dev->bdev->bus, dev->dev->bdev, true); bcma_host_pci_up(dev->dev->bdev->bus); break; diff --git a/drivers/net/wireless/brcm80211/brcmsmac/main.c b/drivers/net/wireless/brcm80211/brcmsmac/main.c index c84af1dfc88f..369527e27689 100644 --- a/drivers/net/wireless/brcm80211/brcmsmac/main.c +++ b/drivers/net/wireless/brcm80211/brcmsmac/main.c @@ -4959,7 +4959,7 @@ static int brcms_b_up_prep(struct brcms_hardware *wlc_hw) * Configure pci/pcmcia here instead of in brcms_c_attach() * to allow mfg hotswap: down, hotswap (chip power cycle), up. */ - bcma_core_pci_irq_ctl(wlc_hw->d11core->bus, wlc_hw->d11core, + bcma_host_pci_irq_ctl(wlc_hw->d11core->bus, wlc_hw->d11core, true); /* diff --git a/include/linux/bcma/bcma.h b/include/linux/bcma/bcma.h index 44057b45ed32..e34f906647d3 100644 --- a/include/linux/bcma/bcma.h +++ b/include/linux/bcma/bcma.h @@ -437,6 +437,8 @@ static inline struct bcma_device *bcma_find_core(struct bcma_bus *bus, #ifdef CONFIG_BCMA_HOST_PCI extern void bcma_host_pci_up(struct bcma_bus *bus); extern void bcma_host_pci_down(struct bcma_bus *bus); +extern int bcma_host_pci_irq_ctl(struct bcma_bus *bus, + struct bcma_device *core, bool enable); #else static inline void bcma_host_pci_up(struct bcma_bus *bus) { @@ -444,6 +446,13 @@ static inline void bcma_host_pci_up(struct bcma_bus *bus) static inline void bcma_host_pci_down(struct bcma_bus *bus) { } +static inline int bcma_host_pci_irq_ctl(struct bcma_bus *bus, + struct bcma_device *core, bool enable) +{ + if (bus->hosttype == BCMA_HOSTTYPE_PCI) + return -ENOTSUPP; + return 0; +} #endif extern bool bcma_core_is_enabled(struct bcma_device *core); diff --git a/include/linux/bcma/bcma_driver_pci.h b/include/linux/bcma/bcma_driver_pci.h index 8e90004fdfd7..3a468687c170 100644 --- a/include/linux/bcma/bcma_driver_pci.h +++ b/include/linux/bcma/bcma_driver_pci.h @@ -238,8 +238,6 @@ struct bcma_drv_pci { #define pcicore_write16(pc, offset, val) bcma_write16((pc)->core, offset, val) #define pcicore_write32(pc, offset, val) bcma_write32((pc)->core, offset, val) -extern int bcma_core_pci_irq_ctl(struct bcma_bus *bus, - struct bcma_device *core, bool enable); extern void bcma_core_pci_power_save(struct bcma_bus *bus, bool up); extern int bcma_core_pci_pcibios_map_irq(const struct pci_dev *dev); -- cgit v1.2.3 From 982a40f5c0bb03368989a6b1ae833b474854e931 Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Thu, 5 Mar 2015 18:25:11 +0100 Subject: bcma: allow disabling (not building) PCI driver MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It isn't required for bcma bus on SoCs, so provide some empty functions and allow disabling it. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo --- drivers/bcma/Kconfig | 4 ++-- drivers/bcma/bcma_private.h | 20 ++++++++++++++++++++ include/linux/bcma/bcma_driver_pci.h | 6 ++++++ 3 files changed, 28 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/bcma/Kconfig b/drivers/bcma/Kconfig index 9be17d3431bb..1500b7120fc7 100644 --- a/drivers/bcma/Kconfig +++ b/drivers/bcma/Kconfig @@ -45,9 +45,9 @@ config BCMA_HOST_SOC If unsure, say N -# TODO: make it depend on PCI when ready config BCMA_DRIVER_PCI - bool + bool "BCMA Broadcom PCI core driver" + depends on BCMA && PCI default y help BCMA bus may have many versions of PCIe core. This driver diff --git a/drivers/bcma/bcma_private.h b/drivers/bcma/bcma_private.h index 5a1d22489afc..15f2b2e242ea 100644 --- a/drivers/bcma/bcma_private.h +++ b/drivers/bcma/bcma_private.h @@ -106,15 +106,35 @@ static inline void __exit bcma_host_soc_unregister_driver(void) #endif /* CONFIG_BCMA_HOST_SOC && CONFIG_OF */ /* driver_pci.c */ +#ifdef CONFIG_BCMA_DRIVER_PCI u32 bcma_pcie_read(struct bcma_drv_pci *pc, u32 address); void bcma_core_pci_early_init(struct bcma_drv_pci *pc); void bcma_core_pci_init(struct bcma_drv_pci *pc); void bcma_core_pci_up(struct bcma_drv_pci *pc); void bcma_core_pci_down(struct bcma_drv_pci *pc); +#else +static inline void bcma_core_pci_early_init(struct bcma_drv_pci *pc) +{ + WARN_ON(pc->core->bus->hosttype == BCMA_HOSTTYPE_PCI); +} +static inline void bcma_core_pci_init(struct bcma_drv_pci *pc) +{ + /* Initialization is required for PCI hosted bus */ + WARN_ON(pc->core->bus->hosttype == BCMA_HOSTTYPE_PCI); +} +#endif /* driver_pcie2.c */ +#ifdef CONFIG_BCMA_DRIVER_PCI void bcma_core_pcie2_init(struct bcma_drv_pcie2 *pcie2); void bcma_core_pcie2_up(struct bcma_drv_pcie2 *pcie2); +#else +static inline void bcma_core_pcie2_init(struct bcma_drv_pcie2 *pcie2) +{ + /* Initialization is required for PCI hosted bus */ + WARN_ON(pcie2->core->bus->hosttype == BCMA_HOSTTYPE_PCI); +} +#endif extern int bcma_chipco_watchdog_register(struct bcma_drv_cc *cc); diff --git a/include/linux/bcma/bcma_driver_pci.h b/include/linux/bcma/bcma_driver_pci.h index 3a468687c170..5ba6918ca20b 100644 --- a/include/linux/bcma/bcma_driver_pci.h +++ b/include/linux/bcma/bcma_driver_pci.h @@ -238,7 +238,13 @@ struct bcma_drv_pci { #define pcicore_write16(pc, offset, val) bcma_write16((pc)->core, offset, val) #define pcicore_write32(pc, offset, val) bcma_write32((pc)->core, offset, val) +#ifdef CONFIG_BCMA_DRIVER_PCI extern void bcma_core_pci_power_save(struct bcma_bus *bus, bool up); +#else +static inline void bcma_core_pci_power_save(struct bcma_bus *bus, bool up) +{ +} +#endif extern int bcma_core_pci_pcibios_map_irq(const struct pci_dev *dev); extern int bcma_core_pci_plat_dev_init(struct pci_dev *dev); -- cgit v1.2.3 From e03826d5045e81a66a4fad7be9a8ecdaeb7911cf Mon Sep 17 00:00:00 2001 From: Keerthy Date: Tue, 17 Mar 2015 15:56:04 +0530 Subject: regulator: palmas: Correct TPS659038 register definition for REGEN2 The register offset for REGEN2_CTRL in different for TPS659038 chip as when compared with other Palmas family PMICs. In the case of TPS659038 the wrong offset pointed to PLLEN_CTRL and was causing a hang. Correcting the same. Signed-off-by: Keerthy Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- drivers/regulator/palmas-regulator.c | 4 ++++ include/linux/mfd/palmas.h | 3 +++ 2 files changed, 7 insertions(+) (limited to 'include/linux') diff --git a/drivers/regulator/palmas-regulator.c b/drivers/regulator/palmas-regulator.c index 9205f433573c..18198316b6cf 100644 --- a/drivers/regulator/palmas-regulator.c +++ b/drivers/regulator/palmas-regulator.c @@ -1572,6 +1572,10 @@ static int palmas_regulators_probe(struct platform_device *pdev) if (!pmic) return -ENOMEM; + if (of_device_is_compatible(node, "ti,tps659038-pmic")) + palmas_generic_regs_info[PALMAS_REG_REGEN2].ctrl_addr = + TPS659038_REGEN2_CTRL; + pmic->dev = &pdev->dev; pmic->palmas = palmas; palmas->pmic = pmic; diff --git a/include/linux/mfd/palmas.h b/include/linux/mfd/palmas.h index fb0390a1a498..ee7b1ce7a6f8 100644 --- a/include/linux/mfd/palmas.h +++ b/include/linux/mfd/palmas.h @@ -2999,6 +2999,9 @@ enum usb_irq_events { #define PALMAS_GPADC_TRIM15 0x0E #define PALMAS_GPADC_TRIM16 0x0F +/* TPS659038 regen2_ctrl offset iss different from palmas */ +#define TPS659038_REGEN2_CTRL 0x12 + /* TPS65917 Interrupt registers */ /* Registers for function INTERRUPT */ -- cgit v1.2.3 From a2f4870697a5bcf4a87073ec6b32dd2928c1211d Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 17 Mar 2015 12:23:19 -0400 Subject: fs: make sure the timestamps for lazytime inodes eventually get written Jan Kara pointed out that if there is an inode which is constantly getting dirtied with I_DIRTY_PAGES, an inode with an updated timestamp will never be written since inode->dirtied_when is constantly getting updated. We fix this by adding an extra field to the inode, dirtied_time_when, so inodes with a stale dirtytime can get detected and handled. In addition, if we have a dirtytime inode caused by an atime update, and there is no write activity on the file system, we need to have a secondary system to make sure these inodes get written out. We do this by setting up a second delayed work structure which wakes up the CPU much more rarely compared to writeback_expire_centisecs. Signed-off-by: Theodore Ts'o Reviewed-by: Jan Kara --- fs/fs-writeback.c | 82 +++++++++++++++++++++++++++++++++++++++++++++++------- include/linux/fs.h | 1 + 2 files changed, 73 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index e907052eeadb..2cfcd74faf87 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -53,6 +53,18 @@ struct wb_writeback_work { struct completion *done; /* set if the caller waits */ }; +/* + * If an inode is constantly having its pages dirtied, but then the + * updates stop dirtytime_expire_interval seconds in the past, it's + * possible for the worst case time between when an inode has its + * timestamps updated and when they finally get written out to be two + * dirtytime_expire_intervals. We set the default to 12 hours (in + * seconds), which means most of the time inodes will have their + * timestamps written to disk after 12 hours, but in the worst case a + * few inodes might not their timestamps updated for 24 hours. + */ +unsigned int dirtytime_expire_interval = 12 * 60 * 60; + /** * writeback_in_progress - determine whether there is writeback in progress * @bdi: the device's backing_dev_info structure. @@ -275,8 +287,8 @@ static int move_expired_inodes(struct list_head *delaying_queue, if ((flags & EXPIRE_DIRTY_ATIME) == 0) older_than_this = work->older_than_this; - else if ((work->reason == WB_REASON_SYNC) == 0) { - expire_time = jiffies - (HZ * 86400); + else if (!work->for_sync) { + expire_time = jiffies - (dirtytime_expire_interval * HZ); older_than_this = &expire_time; } while (!list_empty(delaying_queue)) { @@ -458,6 +470,7 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, */ redirty_tail(inode, wb); } else if (inode->i_state & I_DIRTY_TIME) { + inode->dirtied_when = jiffies; list_move(&inode->i_wb_list, &wb->b_dirty_time); } else { /* The inode is clean. Remove from writeback lists. */ @@ -505,12 +518,17 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) spin_lock(&inode->i_lock); dirty = inode->i_state & I_DIRTY; - if (((dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) && - (inode->i_state & I_DIRTY_TIME)) || - (inode->i_state & I_DIRTY_TIME_EXPIRED)) { - dirty |= I_DIRTY_TIME | I_DIRTY_TIME_EXPIRED; - trace_writeback_lazytime(inode); - } + if (inode->i_state & I_DIRTY_TIME) { + if ((dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) || + unlikely(inode->i_state & I_DIRTY_TIME_EXPIRED) || + unlikely(time_after(jiffies, + (inode->dirtied_time_when + + dirtytime_expire_interval * HZ)))) { + dirty |= I_DIRTY_TIME | I_DIRTY_TIME_EXPIRED; + trace_writeback_lazytime(inode); + } + } else + inode->i_state &= ~I_DIRTY_TIME_EXPIRED; inode->i_state &= ~dirty; /* @@ -1131,6 +1149,45 @@ void wakeup_flusher_threads(long nr_pages, enum wb_reason reason) rcu_read_unlock(); } +/* + * Wake up bdi's periodically to make sure dirtytime inodes gets + * written back periodically. We deliberately do *not* check the + * b_dirtytime list in wb_has_dirty_io(), since this would cause the + * kernel to be constantly waking up once there are any dirtytime + * inodes on the system. So instead we define a separate delayed work + * function which gets called much more rarely. (By default, only + * once every 12 hours.) + * + * If there is any other write activity going on in the file system, + * this function won't be necessary. But if the only thing that has + * happened on the file system is a dirtytime inode caused by an atime + * update, we need this infrastructure below to make sure that inode + * eventually gets pushed out to disk. + */ +static void wakeup_dirtytime_writeback(struct work_struct *w); +static DECLARE_DELAYED_WORK(dirtytime_work, wakeup_dirtytime_writeback); + +static void wakeup_dirtytime_writeback(struct work_struct *w) +{ + struct backing_dev_info *bdi; + + rcu_read_lock(); + list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) { + if (list_empty(&bdi->wb.b_dirty_time)) + continue; + bdi_wakeup_thread(bdi); + } + rcu_read_unlock(); + schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); +} + +static int __init start_dirtytime_writeback(void) +{ + schedule_delayed_work(&dirtytime_work, dirtytime_expire_interval * HZ); + return 0; +} +__initcall(start_dirtytime_writeback); + static noinline void block_dump___mark_inode_dirty(struct inode *inode) { if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) { @@ -1269,8 +1326,13 @@ void __mark_inode_dirty(struct inode *inode, int flags) } inode->dirtied_when = jiffies; - list_move(&inode->i_wb_list, dirtytime ? - &bdi->wb.b_dirty_time : &bdi->wb.b_dirty); + if (dirtytime) + inode->dirtied_time_when = jiffies; + if (inode->i_state & (I_DIRTY_INODE | I_DIRTY_PAGES)) + list_move(&inode->i_wb_list, &bdi->wb.b_dirty); + else + list_move(&inode->i_wb_list, + &bdi->wb.b_dirty_time); spin_unlock(&bdi->wb.list_lock); trace_writeback_dirty_inode_enqueue(inode); diff --git a/include/linux/fs.h b/include/linux/fs.h index b4d71b5e1ff2..f4131e8ead74 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -604,6 +604,7 @@ struct inode { struct mutex i_mutex; unsigned long dirtied_when; /* jiffies of first dirtying */ + unsigned long dirtied_time_when; struct hlist_node i_hash; struct list_head i_wb_list; /* backing dev IO list */ -- cgit v1.2.3 From 1efff914afac8a965ad63817ecf8861a927c2ace Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 17 Mar 2015 12:23:32 -0400 Subject: fs: add dirtytime_expire_seconds sysctl Add a tuning knob so we can adjust the dirtytime expiration timeout, which is very useful for testing lazytime. Signed-off-by: Theodore Ts'o Reviewed-by: Jan Kara --- fs/fs-writeback.c | 11 +++++++++++ include/linux/writeback.h | 3 +++ kernel/sysctl.c | 8 ++++++++ 3 files changed, 22 insertions(+) (limited to 'include/linux') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 2cfcd74faf87..32a8bbd7a9ad 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1188,6 +1188,17 @@ static int __init start_dirtytime_writeback(void) } __initcall(start_dirtytime_writeback); +int dirtytime_interval_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int ret; + + ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + if (ret == 0 && write) + mod_delayed_work(system_wq, &dirtytime_work, 0); + return ret; +} + static noinline void block_dump___mark_inode_dirty(struct inode *inode) { if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) { diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 00048339c23e..b2dd371ec0ca 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -130,6 +130,7 @@ extern int vm_dirty_ratio; extern unsigned long vm_dirty_bytes; extern unsigned int dirty_writeback_interval; extern unsigned int dirty_expire_interval; +extern unsigned int dirtytime_expire_interval; extern int vm_highmem_is_dirtyable; extern int block_dump; extern int laptop_mode; @@ -146,6 +147,8 @@ extern int dirty_ratio_handler(struct ctl_table *table, int write, extern int dirty_bytes_handler(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); +int dirtytime_interval_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); struct ctl_table; int dirty_writeback_centisecs_handler(struct ctl_table *, int, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 88ea2d6e0031..ce410bb9f2e1 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1227,6 +1227,14 @@ static struct ctl_table vm_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &zero, }, + { + .procname = "dirtytime_expire_seconds", + .data = &dirtytime_expire_interval, + .maxlen = sizeof(dirty_expire_interval), + .mode = 0644, + .proc_handler = dirtytime_interval_handler, + .extra1 = &zero, + }, { .procname = "nr_pdflush_threads", .mode = 0444 /* read-only */, -- cgit v1.2.3 From cf39284d41f67964cf42b21bb386c012cf5b7f65 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 18 Mar 2015 08:57:41 +0800 Subject: regulator: Fix documentation for regmap in the config dev_get_regulator() does not exist, fix the typo. Signed-off-by: Axel Lin Signed-off-by: Mark Brown --- include/linux/regulator/driver.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/regulator/driver.h b/include/linux/regulator/driver.h index d4ad5b5a02bb..045f709cb89b 100644 --- a/include/linux/regulator/driver.h +++ b/include/linux/regulator/driver.h @@ -316,7 +316,7 @@ struct regulator_desc { * @driver_data: private regulator data * @of_node: OpenFirmware node to parse for device tree bindings (may be * NULL). - * @regmap: regmap to use for core regmap helpers if dev_get_regulator() is + * @regmap: regmap to use for core regmap helpers if dev_get_regmap() is * insufficient. * @ena_gpio_initialized: GPIO controlling regulator enable was properly * initialized, meaning that >= 0 is a valid gpio -- cgit v1.2.3 From 5067c0469c643512f24786990e315f9c15cc7d24 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 12 Mar 2015 10:32:18 -0700 Subject: ata: Add a new flag to destinguish sas controller SAS controller has its own tag allocation, which doesn't directly match to ATA tag, so SAS and SATA have different code path for ata tags. Originally we use port->scsi_host (98bd4be1) to destinguish SAS controller, but libsas set ->scsi_host too, so we can't use it for the destinguish, we add a new flag for this purpose. Without this patch, the following oops can happen because scsi-mq uses a host-wide tag map shared among all devices with some integer tag values >= ATA_MAX_QUEUE. These unexpectedly high tag values cause __ata_qc_from_tag() to return NULL, which is then dereferenced in ata_qc_new_init(). BUG: unable to handle kernel NULL pointer dereference at 0000000000000058 IP: [] ata_qc_new_init+0x3e/0x120 PGD 32adf0067 PUD 32adf1067 PMD 0 Oops: 0002 [#1] SMP DEBUG_PAGEALLOC Modules linked in: iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi igb i2c_algo_bit ptp pps_core pm80xx libsas scsi_transport_sas sg coretemp eeprom w83795 i2c_i801 CPU: 4 PID: 1450 Comm: cydiskbench Not tainted 4.0.0-rc3 #1 Hardware name: Supermicro X8DTH-i/6/iF/6F/X8DTH, BIOS 2.1b 05/04/12 task: ffff8800ba86d500 ti: ffff88032a064000 task.ti: ffff88032a064000 RIP: 0010:[] [] ata_qc_new_init+0x3e/0x120 RSP: 0018:ffff88032a067858 EFLAGS: 00010046 RAX: 0000000000000000 RBX: ffff8800ba0d2230 RCX: 000000000000002a RDX: ffffffff80505ae0 RSI: 0000000000000020 RDI: ffff8800ba0d2230 RBP: ffff88032a067868 R08: 0000000000000201 R09: 0000000000000001 R10: 0000000000000000 R11: 0000000000000000 R12: ffff8800ba0d0000 R13: ffff8800ba0d2230 R14: ffffffff80505ae0 R15: ffff8800ba0d0000 FS: 0000000041223950(0063) GS:ffff88033e480000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000058 CR3: 000000032a0a3000 CR4: 00000000000006e0 Stack: ffff880329eee758 ffff880329eee758 ffff88032a0678a8 ffffffff80502dad ffff8800ba167978 ffff880329eee758 ffff88032bf9c520 ffff8800ba167978 ffff88032bf9c520 ffff88032bf9a290 ffff88032a0678b8 ffffffff80506909 Call Trace: [] ata_scsi_translate+0x3d/0x1b0 [] ata_sas_queuecmd+0x149/0x2a0 [] sas_queuecommand+0xa0/0x1f0 [libsas] [] scsi_dispatch_cmd+0xd4/0x1a0 [] scsi_queue_rq+0x66f/0x7f0 [] __blk_mq_run_hw_queue+0x208/0x3f0 [] blk_mq_run_hw_queue+0x88/0xc0 [] blk_mq_insert_request+0xc4/0x130 [] blk_execute_rq_nowait+0x73/0x160 [] sg_common_write+0x3da/0x720 [sg] [] sg_new_write+0x250/0x360 [sg] [] sg_write+0x13b/0x450 [sg] [] vfs_write+0xd1/0x1b0 [] SyS_write+0x54/0xc0 [] system_call_fastpath+0x12/0x17 tj: updated description. Fixes: 12cb5ce101ab ("libata: use blk taging") Reported-and-tested-by: Tony Battersby Signed-off-by: Shaohua Li Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 4 ++-- drivers/scsi/ipr.c | 3 ++- drivers/scsi/libsas/sas_ata.c | 3 ++- include/linux/libata.h | 1 + 4 files changed, 7 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 4c35f0822d06..ef150ebb4c30 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4737,7 +4737,7 @@ struct ata_queued_cmd *ata_qc_new_init(struct ata_device *dev, int tag) return NULL; /* libsas case */ - if (!ap->scsi_host) { + if (ap->flags & ATA_FLAG_SAS_HOST) { tag = ata_sas_allocate_tag(ap); if (tag < 0) return NULL; @@ -4776,7 +4776,7 @@ void ata_qc_free(struct ata_queued_cmd *qc) tag = qc->tag; if (likely(ata_tag_valid(tag))) { qc->tag = ATA_TAG_POISON; - if (!ap->scsi_host) + if (ap->flags & ATA_FLAG_SAS_HOST) ata_sas_free_tag(tag, ap); } } diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 9219953ee949..d9afc51af7d3 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -6815,7 +6815,8 @@ static struct ata_port_operations ipr_sata_ops = { }; static struct ata_port_info sata_port_info = { - .flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA, + .flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA | + ATA_FLAG_SAS_HOST, .pio_mask = ATA_PIO4_ONLY, .mwdma_mask = ATA_MWDMA2, .udma_mask = ATA_UDMA6, diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index 932d9cc98d2f..9c706d8c1441 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -547,7 +547,8 @@ static struct ata_port_operations sas_sata_ops = { }; static struct ata_port_info sata_port_info = { - .flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA | ATA_FLAG_NCQ, + .flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA | ATA_FLAG_NCQ | + ATA_FLAG_SAS_HOST, .pio_mask = ATA_PIO4, .mwdma_mask = ATA_MWDMA2, .udma_mask = ATA_UDMA6, diff --git a/include/linux/libata.h b/include/linux/libata.h index fc03efa64ffe..6b08cc106c21 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -232,6 +232,7 @@ enum { * led */ ATA_FLAG_NO_DIPM = (1 << 23), /* host not happy with DIPM */ ATA_FLAG_LOWTAG = (1 << 24), /* host wants lowest available tag */ + ATA_FLAG_SAS_HOST = (1 << 25), /* SAS host */ /* bits 24:31 of ap->flags are reserved for LLD specific flags */ -- cgit v1.2.3 From 25911556283e5093fca235d7ba03faae7ed5dbf2 Mon Sep 17 00:00:00 2001 From: Arend van Spriel Date: Wed, 18 Mar 2015 13:25:26 +0100 Subject: brcmfmac: add support for BCM43430 SDIO chipset This patch added support for the BCM43430 802.11n SDIO chipset. Reviewed-by: Hante Meuleman Reviewed-by: Daniel (Deognyoun) Kim Reviewed-by: Franky (Zhenhui) Lin Reviewed-by: Pieter-Paul Giesberts Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo --- drivers/net/wireless/brcm80211/brcmfmac/bcmsdh.c | 1 + drivers/net/wireless/brcm80211/brcmfmac/chip.c | 18 ++++++++++++++++++ drivers/net/wireless/brcm80211/brcmfmac/sdio.c | 5 +++++ drivers/net/wireless/brcm80211/include/brcm_hw_ids.h | 1 + drivers/net/wireless/brcm80211/include/chipcommon.h | 9 ++++++++- include/linux/mmc/sdio_ids.h | 1 + 6 files changed, 34 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/net/wireless/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/brcm80211/brcmfmac/bcmsdh.c index 9667b6aabc00..9b508bd3b839 100644 --- a/drivers/net/wireless/brcm80211/brcmfmac/bcmsdh.c +++ b/drivers/net/wireless/brcm80211/brcmfmac/bcmsdh.c @@ -1098,6 +1098,7 @@ static const struct sdio_device_id brcmf_sdmmc_ids[] = { BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43341), BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43362), BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4335_4339), + BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_43430), BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4345), BRCMF_SDIO_DEVICE(SDIO_DEVICE_ID_BROADCOM_4354), { /* end: all zeroes */ } diff --git a/drivers/net/wireless/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/brcm80211/brcmfmac/chip.c index 734172570054..ab2fac8b2760 100644 --- a/drivers/net/wireless/brcm80211/brcmfmac/chip.c +++ b/drivers/net/wireless/brcm80211/brcmfmac/chip.c @@ -600,6 +600,12 @@ static void brcmf_chip_socram_ramsize(struct brcmf_core_priv *sr, u32 *ramsize, if (sr->chip->pub.chiprev < 2) *srsize = (32 * 1024); break; + case BRCM_CC_43430_CHIP_ID: + /* assume sr for now as we can not check + * firmware sr capability at this point. + */ + *srsize = (64 * 1024); + break; default: break; } @@ -1072,6 +1078,7 @@ static void brcmf_chip_cm3_set_passive(struct brcmf_chip_priv *chip) { struct brcmf_core *core; + struct brcmf_core_priv *sr; brcmf_chip_disable_arm(chip, BCMA_CORE_ARM_CM3); core = brcmf_chip_get_core(&chip->pub, BCMA_CORE_80211); @@ -1081,6 +1088,13 @@ brcmf_chip_cm3_set_passive(struct brcmf_chip_priv *chip) D11_BCMA_IOCTL_PHYCLOCKEN); core = brcmf_chip_get_core(&chip->pub, BCMA_CORE_INTERNAL_MEM); brcmf_chip_resetcore(core, 0, 0, 0); + + /* disable bank #3 remap for this device */ + if (chip->pub.chip == BRCM_CC_43430_CHIP_ID) { + sr = container_of(core, struct brcmf_core_priv, pub); + brcmf_chip_core_write32(sr, SOCRAMREGOFFS(bankidx), 3); + brcmf_chip_core_write32(sr, SOCRAMREGOFFS(bankpda), 0); + } } static bool brcmf_chip_cm3_set_active(struct brcmf_chip_priv *chip) @@ -1188,6 +1202,10 @@ bool brcmf_chip_sr_capable(struct brcmf_chip *pub) addr = CORE_CC_REG(base, chipcontrol_data); reg = chip->ops->read32(chip->ctx, addr); return (reg & pmu_cc3_mask) != 0; + case BRCM_CC_43430_CHIP_ID: + addr = CORE_CC_REG(base, sr_control1); + reg = chip->ops->read32(chip->ctx, addr); + return reg != 0; default: addr = CORE_CC_REG(base, pmucapabilities_ext); reg = chip->ops->read32(chip->ctx, addr); diff --git a/drivers/net/wireless/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/brcm80211/brcmfmac/sdio.c index 38fa0e8f859d..e162860b265c 100644 --- a/drivers/net/wireless/brcm80211/brcmfmac/sdio.c +++ b/drivers/net/wireless/brcm80211/brcmfmac/sdio.c @@ -615,6 +615,8 @@ static const struct sdiod_drive_str sdiod_drvstr_tab2_3v3[] = { #define BCM43362_NVRAM_NAME "brcm/brcmfmac43362-sdio.txt" #define BCM4339_FIRMWARE_NAME "brcm/brcmfmac4339-sdio.bin" #define BCM4339_NVRAM_NAME "brcm/brcmfmac4339-sdio.txt" +#define BCM43430_FIRMWARE_NAME "brcm/brcmfmac43430-sdio.bin" +#define BCM43430_NVRAM_NAME "brcm/brcmfmac43430-sdio.txt" #define BCM4345_FIRMWARE_NAME "brcm/brcmfmac4345-sdio.bin" #define BCM4345_NVRAM_NAME "brcm/brcmfmac4345-sdio.txt" #define BCM4354_FIRMWARE_NAME "brcm/brcmfmac4354-sdio.bin" @@ -640,6 +642,8 @@ MODULE_FIRMWARE(BCM43362_FIRMWARE_NAME); MODULE_FIRMWARE(BCM43362_NVRAM_NAME); MODULE_FIRMWARE(BCM4339_FIRMWARE_NAME); MODULE_FIRMWARE(BCM4339_NVRAM_NAME); +MODULE_FIRMWARE(BCM43430_FIRMWARE_NAME); +MODULE_FIRMWARE(BCM43430_NVRAM_NAME); MODULE_FIRMWARE(BCM4345_FIRMWARE_NAME); MODULE_FIRMWARE(BCM4345_NVRAM_NAME); MODULE_FIRMWARE(BCM4354_FIRMWARE_NAME); @@ -671,6 +675,7 @@ static const struct brcmf_firmware_names brcmf_fwname_data[] = { { BRCM_CC_4335_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM4335) }, { BRCM_CC_43362_CHIP_ID, 0xFFFFFFFE, BRCMF_FIRMWARE_NVRAM(BCM43362) }, { BRCM_CC_4339_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM4339) }, + { BRCM_CC_43430_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM43430) }, { BRCM_CC_4345_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM4345) }, { BRCM_CC_4354_CHIP_ID, 0xFFFFFFFF, BRCMF_FIRMWARE_NVRAM(BCM4354) } }; diff --git a/drivers/net/wireless/brcm80211/include/brcm_hw_ids.h b/drivers/net/wireless/brcm80211/include/brcm_hw_ids.h index b599e7e41148..4efdd51af9c8 100644 --- a/drivers/net/wireless/brcm80211/include/brcm_hw_ids.h +++ b/drivers/net/wireless/brcm80211/include/brcm_hw_ids.h @@ -37,6 +37,7 @@ #define BRCM_CC_43362_CHIP_ID 43362 #define BRCM_CC_4335_CHIP_ID 0x4335 #define BRCM_CC_4339_CHIP_ID 0x4339 +#define BRCM_CC_43430_CHIP_ID 43430 #define BRCM_CC_4345_CHIP_ID 0x4345 #define BRCM_CC_4354_CHIP_ID 0x4354 #define BRCM_CC_4356_CHIP_ID 0x4356 diff --git a/drivers/net/wireless/brcm80211/include/chipcommon.h b/drivers/net/wireless/brcm80211/include/chipcommon.h index d242333b7559..e1fd499930a0 100644 --- a/drivers/net/wireless/brcm80211/include/chipcommon.h +++ b/drivers/net/wireless/brcm80211/include/chipcommon.h @@ -183,7 +183,14 @@ struct chipcregs { u8 uart1lsr; u8 uart1msr; u8 uart1scratch; - u32 PAD[126]; + u32 PAD[62]; + + /* save/restore, corerev >= 48 */ + u32 sr_capability; /* 0x500 */ + u32 sr_control0; /* 0x504 */ + u32 sr_control1; /* 0x508 */ + u32 gpio_control; /* 0x50C */ + u32 PAD[60]; /* PMU registers (corerev >= 20) */ u32 pmucontrol; /* 0x600 */ diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h index 91397858dc95..83430f2ea757 100644 --- a/include/linux/mmc/sdio_ids.h +++ b/include/linux/mmc/sdio_ids.h @@ -33,6 +33,7 @@ #define SDIO_DEVICE_ID_BROADCOM_43341 0xa94d #define SDIO_DEVICE_ID_BROADCOM_4335_4339 0x4335 #define SDIO_DEVICE_ID_BROADCOM_43362 0xa962 +#define SDIO_DEVICE_ID_BROADCOM_43430 0xa9a6 #define SDIO_DEVICE_ID_BROADCOM_4345 0x4345 #define SDIO_DEVICE_ID_BROADCOM_4354 0x4354 -- cgit v1.2.3 From 074c238177a75f5e79af3b2cb6a84e54823ef950 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 25 Mar 2015 15:55:42 -0700 Subject: mm: numa: slow PTE scan rate if migration failures occur Dave Chinner reported the following on https://lkml.org/lkml/2015/3/1/226 Across the board the 4.0-rc1 numbers are much slower, and the degradation is far worse when using the large memory footprint configs. Perf points straight at the cause - this is from 4.0-rc1 on the "-o bhash=101073" config: - 56.07% 56.07% [kernel] [k] default_send_IPI_mask_sequence_phys - default_send_IPI_mask_sequence_phys - 99.99% physflat_send_IPI_mask - 99.37% native_send_call_func_ipi smp_call_function_many - native_flush_tlb_others - 99.85% flush_tlb_page ptep_clear_flush try_to_unmap_one rmap_walk try_to_unmap migrate_pages migrate_misplaced_page - handle_mm_fault - 99.73% __do_page_fault trace_do_page_fault do_async_page_fault + async_page_fault 0.63% native_send_call_func_single_ipi generic_exec_single smp_call_function_single This is showing excessive migration activity even though excessive migrations are meant to get throttled. Normally, the scan rate is tuned on a per-task basis depending on the locality of faults. However, if migrations fail for any reason then the PTE scanner may scan faster if the faults continue to be remote. This means there is higher system CPU overhead and fault trapping at exactly the time we know that migrations cannot happen. This patch tracks when migration failures occur and slows the PTE scanner. Signed-off-by: Mel Gorman Reported-by: Dave Chinner Tested-by: Dave Chinner Cc: Ingo Molnar Cc: Aneesh Kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 9 +++++---- kernel/sched/fair.c | 8 ++++++-- mm/huge_memory.c | 3 ++- mm/memory.c | 3 ++- 4 files changed, 15 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6d77432e14ff..a419b65770d6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1625,11 +1625,11 @@ struct task_struct { /* * numa_faults_locality tracks if faults recorded during the last - * scan window were remote/local. The task scan period is adapted - * based on the locality of the faults with different weights - * depending on whether they were shared or private faults + * scan window were remote/local or failed to migrate. The task scan + * period is adapted based on the locality of the faults with different + * weights depending on whether they were shared or private faults */ - unsigned long numa_faults_locality[2]; + unsigned long numa_faults_locality[3]; unsigned long numa_pages_migrated; #endif /* CONFIG_NUMA_BALANCING */ @@ -1719,6 +1719,7 @@ struct task_struct { #define TNF_NO_GROUP 0x02 #define TNF_SHARED 0x04 #define TNF_FAULT_LOCAL 0x08 +#define TNF_MIGRATE_FAIL 0x10 #ifdef CONFIG_NUMA_BALANCING extern void task_numa_fault(int last_node, int node, int pages, int flags); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7ce18f3c097a..bcfe32088b37 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -1609,9 +1609,11 @@ static void update_task_scan_period(struct task_struct *p, /* * If there were no record hinting faults then either the task is * completely idle or all activity is areas that are not of interest - * to automatic numa balancing. Scan slower + * to automatic numa balancing. Related to that, if there were failed + * migration then it implies we are migrating too quickly or the local + * node is overloaded. In either case, scan slower */ - if (local + shared == 0) { + if (local + shared == 0 || p->numa_faults_locality[2]) { p->numa_scan_period = min(p->numa_scan_period_max, p->numa_scan_period << 1); @@ -2080,6 +2082,8 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags) if (migrated) p->numa_pages_migrated += pages; + if (flags & TNF_MIGRATE_FAIL) + p->numa_faults_locality[2] += pages; p->numa_faults[task_faults_idx(NUMA_MEMBUF, mem_node, priv)] += pages; p->numa_faults[task_faults_idx(NUMA_CPUBUF, cpu_node, priv)] += pages; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 0a42d1521aa4..51b3e7c64622 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1350,7 +1350,8 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, if (migrated) { flags |= TNF_MIGRATED; page_nid = target_nid; - } + } else + flags |= TNF_MIGRATE_FAIL; goto out; clear_pmdnuma: diff --git a/mm/memory.c b/mm/memory.c index d20e12da3a3c..97839f5c8c30 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3103,7 +3103,8 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, if (migrated) { page_nid = target_nid; flags |= TNF_MIGRATED; - } + } else + flags |= TNF_MIGRATE_FAIL; out: if (page_nid != -1) -- cgit v1.2.3 From 1e9e39f4a29857a396ac7b669d109f697f66695e Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 26 Feb 2015 19:34:37 +0000 Subject: usbnet: Fix tx_packets stat for FLAG_MULTI_FRAME drivers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently the usbnet core does not update the tx_packets statistic for drivers with FLAG_MULTI_PACKET and there is no hook in the TX completion path where they could do this. cdc_ncm and dependent drivers are bumping tx_packets stat on the transmit path while asix and sr9800 aren't updating it at all. Add a packet count in struct skb_data so these drivers can fill it in, initialise it to 1 for other drivers, and add the packet count to the tx_packets statistic on completion. Signed-off-by: Ben Hutchings Tested-by: Bjørn Mork Signed-off-by: David S. Miller --- drivers/net/usb/asix_common.c | 2 ++ drivers/net/usb/cdc_ncm.c | 3 ++- drivers/net/usb/sr9800.c | 1 + drivers/net/usb/usbnet.c | 5 +++-- include/linux/usb/usbnet.h | 12 ++++++++++++ 5 files changed, 20 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/usb/asix_common.c b/drivers/net/usb/asix_common.c index 5c55f11572ba..724a9b50df7a 100644 --- a/drivers/net/usb/asix_common.c +++ b/drivers/net/usb/asix_common.c @@ -188,6 +188,8 @@ struct sk_buff *asix_tx_fixup(struct usbnet *dev, struct sk_buff *skb, memcpy(skb_tail_pointer(skb), &padbytes, sizeof(padbytes)); skb_put(skb, sizeof(padbytes)); } + + usbnet_set_skb_tx_stats(skb, 1); return skb; } diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index 80a844e0ae03..70cbea551139 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -1172,7 +1172,6 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) /* return skb */ ctx->tx_curr_skb = NULL; - dev->net->stats.tx_packets += ctx->tx_curr_frame_num; /* keep private stats: framing overhead and number of NTBs */ ctx->tx_overhead += skb_out->len - ctx->tx_curr_frame_payload; @@ -1184,6 +1183,8 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) */ dev->net->stats.tx_bytes -= skb_out->len - ctx->tx_curr_frame_payload; + usbnet_set_skb_tx_stats(skb_out, n); + return skb_out; exit_no_skb: diff --git a/drivers/net/usb/sr9800.c b/drivers/net/usb/sr9800.c index b94a0fbb8b3b..7650cdc8fe6b 100644 --- a/drivers/net/usb/sr9800.c +++ b/drivers/net/usb/sr9800.c @@ -144,6 +144,7 @@ static struct sk_buff *sr_tx_fixup(struct usbnet *dev, struct sk_buff *skb, skb_put(skb, sizeof(padbytes)); } + usbnet_set_skb_tx_stats(skb, 1); return skb; } diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 449835f4331e..0f3ff285f6a1 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -1188,8 +1188,7 @@ static void tx_complete (struct urb *urb) struct usbnet *dev = entry->dev; if (urb->status == 0) { - if (!(dev->driver_info->flags & FLAG_MULTI_PACKET)) - dev->net->stats.tx_packets++; + dev->net->stats.tx_packets += entry->packets; dev->net->stats.tx_bytes += entry->length; } else { dev->net->stats.tx_errors++; @@ -1348,6 +1347,8 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, urb->transfer_flags |= URB_ZERO_PACKET; } entry->length = urb->transfer_buffer_length = length; + if (!(info->flags & FLAG_MULTI_PACKET)) + usbnet_set_skb_tx_stats(skb, 1); spin_lock_irqsave(&dev->txq.lock, flags); retval = usb_autopm_get_interface_async(dev->intf); diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h index d9a4905e01d0..ff3fb2bd0e90 100644 --- a/include/linux/usb/usbnet.h +++ b/include/linux/usb/usbnet.h @@ -228,8 +228,20 @@ struct skb_data { /* skb->cb is one of these */ struct usbnet *dev; enum skb_state state; size_t length; + unsigned long packets; }; +/* Drivers that set FLAG_MULTI_PACKET must call this in their + * tx_fixup method before returning an skb. + */ +static inline void +usbnet_set_skb_tx_stats(struct sk_buff *skb, unsigned long packets) +{ + struct skb_data *entry = (struct skb_data *) skb->cb; + + entry->packets = packets; +} + extern int usbnet_open(struct net_device *net); extern int usbnet_stop(struct net_device *net); extern netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, -- cgit v1.2.3 From 7a1e890e2168e33fb62d84528e996b8b4b478fea Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Wed, 25 Mar 2015 21:41:33 +0100 Subject: usbnet: Fix tx_bytes statistic running backward in cdc_ncm MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cdc_ncm disagrees with usbnet about how much framing overhead should be counted in the tx_bytes statistics, and tries 'fix' this by decrementing tx_bytes on the transmit path. But statistics must never be decremented except due to roll-over; this will thoroughly confuse user-space. Also, tx_bytes is only incremented by usbnet in the completion path. Fix this by requiring drivers that set FLAG_MULTI_FRAME to set a tx_bytes delta along with the tx_packets count. Fixes: beeecd42c3b4 ("net: cdc_ncm/cdc_mbim: adding NCM protocol statistics") Signed-off-by: Ben Hutchings Signed-off-by: Bjørn Mork --- drivers/net/usb/asix_common.c | 2 +- drivers/net/usb/cdc_ncm.c | 7 +++---- drivers/net/usb/sr9800.c | 2 +- drivers/net/usb/usbnet.c | 16 +++++++++++++--- include/linux/usb/usbnet.h | 6 ++++-- 5 files changed, 22 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/usb/asix_common.c b/drivers/net/usb/asix_common.c index 724a9b50df7a..75d6f26729a3 100644 --- a/drivers/net/usb/asix_common.c +++ b/drivers/net/usb/asix_common.c @@ -189,7 +189,7 @@ struct sk_buff *asix_tx_fixup(struct usbnet *dev, struct sk_buff *skb, skb_put(skb, sizeof(padbytes)); } - usbnet_set_skb_tx_stats(skb, 1); + usbnet_set_skb_tx_stats(skb, 1, 0); return skb; } diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index 70cbea551139..c3e4da9e79ca 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -1177,13 +1177,12 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) ctx->tx_overhead += skb_out->len - ctx->tx_curr_frame_payload; ctx->tx_ntbs++; - /* usbnet has already counted all the framing overhead. + /* usbnet will count all the framing overhead by default. * Adjust the stats so that the tx_bytes counter show real * payload data instead. */ - dev->net->stats.tx_bytes -= skb_out->len - ctx->tx_curr_frame_payload; - - usbnet_set_skb_tx_stats(skb_out, n); + usbnet_set_skb_tx_stats(skb_out, n, + ctx->tx_curr_frame_payload - skb_out->len); return skb_out; diff --git a/drivers/net/usb/sr9800.c b/drivers/net/usb/sr9800.c index 7650cdc8fe6b..953de13267df 100644 --- a/drivers/net/usb/sr9800.c +++ b/drivers/net/usb/sr9800.c @@ -144,7 +144,7 @@ static struct sk_buff *sr_tx_fixup(struct usbnet *dev, struct sk_buff *skb, skb_put(skb, sizeof(padbytes)); } - usbnet_set_skb_tx_stats(skb, 1); + usbnet_set_skb_tx_stats(skb, 1, 0); return skb; } diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 0f3ff285f6a1..777757ae1973 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -1346,9 +1346,19 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, } else urb->transfer_flags |= URB_ZERO_PACKET; } - entry->length = urb->transfer_buffer_length = length; - if (!(info->flags & FLAG_MULTI_PACKET)) - usbnet_set_skb_tx_stats(skb, 1); + urb->transfer_buffer_length = length; + + if (info->flags & FLAG_MULTI_PACKET) { + /* Driver has set number of packets and a length delta. + * Calculate the complete length and ensure that it's + * positive. + */ + entry->length += length; + if (WARN_ON_ONCE(entry->length <= 0)) + entry->length = length; + } else { + usbnet_set_skb_tx_stats(skb, 1, length); + } spin_lock_irqsave(&dev->txq.lock, flags); retval = usb_autopm_get_interface_async(dev->intf); diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h index ff3fb2bd0e90..6e0ce8c7b8cb 100644 --- a/include/linux/usb/usbnet.h +++ b/include/linux/usb/usbnet.h @@ -227,7 +227,7 @@ struct skb_data { /* skb->cb is one of these */ struct urb *urb; struct usbnet *dev; enum skb_state state; - size_t length; + long length; unsigned long packets; }; @@ -235,11 +235,13 @@ struct skb_data { /* skb->cb is one of these */ * tx_fixup method before returning an skb. */ static inline void -usbnet_set_skb_tx_stats(struct sk_buff *skb, unsigned long packets) +usbnet_set_skb_tx_stats(struct sk_buff *skb, + unsigned long packets, long bytes_delta) { struct skb_data *entry = (struct skb_data *) skb->cb; entry->packets = packets; + entry->length = bytes_delta; } extern int usbnet_open(struct net_device *net); -- cgit v1.2.3 From 4ad3e3634a6cbe916722c7113c5b488d52c7a3dc Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 27 Mar 2015 14:15:04 +0000 Subject: irqchip: gicv3-its: Fix PROP/PEND and BASE/CBASE confusion The ITS driver sometime mixes up the use of GICR_PROPBASE bitfields for the GICR_PENDBASE register, and GITS_BASER for GICR_CBASE. This does not lead to any observable bug because similar bits are at the same location, but this just make the code even harder to understand... This patch provides the required #defines and fixes the mixup. Signed-off-by: Marc Zyngier Link: https://lkml.kernel.org/r/1427465705-17126-4-git-send-email-marc.zyngier@arm.com Signed-off-by: Jason Cooper --- drivers/irqchip/irq-gic-v3-its.c | 6 +++--- include/linux/irqchip/arm-gic-v3.h | 13 +++++++++++++ 2 files changed, 16 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index fa0c43660c8b..56353f6b5952 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -986,8 +986,8 @@ static void its_cpu_init_lpis(void) /* set PENDBASE */ val = (page_to_phys(pend_page) | - GICR_PROPBASER_InnerShareable | - GICR_PROPBASER_WaWb); + GICR_PENDBASER_InnerShareable | + GICR_PENDBASER_WaWb); writeq_relaxed(val, rbase + GICR_PENDBASER); @@ -1425,7 +1425,7 @@ static int its_probe(struct device_node *node, struct irq_domain *parent) writeq_relaxed(0, its->base + GITS_CWRITER); writel_relaxed(GITS_CTLR_ENABLE, its->base + GITS_CTLR); - if ((tmp ^ baser) & GITS_BASER_SHAREABILITY_MASK) { + if ((tmp ^ baser) & GITS_CBASER_SHAREABILITY_MASK) { pr_info("ITS: using cache flushing for cmd queue\n"); its->flags |= ITS_FLAGS_CMDQ_NEEDS_FLUSHING; } diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h index 781974afff9f..826a4bd63d4a 100644 --- a/include/linux/irqchip/arm-gic-v3.h +++ b/include/linux/irqchip/arm-gic-v3.h @@ -128,6 +128,19 @@ #define GICR_PROPBASER_RaWaWb (7U << 7) #define GICR_PROPBASER_IDBITS_MASK (0x1f) +#define GICR_PENDBASER_NonShareable (0U << 10) +#define GICR_PENDBASER_InnerShareable (1U << 10) +#define GICR_PENDBASER_OuterShareable (2U << 10) +#define GICR_PENDBASER_SHAREABILITY_MASK (3UL << 10) +#define GICR_PENDBASER_nCnB (0U << 7) +#define GICR_PENDBASER_nC (1U << 7) +#define GICR_PENDBASER_RaWt (2U << 7) +#define GICR_PENDBASER_RaWb (3U << 7) +#define GICR_PENDBASER_WaWt (4U << 7) +#define GICR_PENDBASER_WaWb (5U << 7) +#define GICR_PENDBASER_RaWaWt (6U << 7) +#define GICR_PENDBASER_RaWaWb (7U << 7) + /* * Re-Distributor registers, offsets from SGI_base */ -- cgit v1.2.3 From 241a386c7dbb8b0db400a1f92f2ebe3b10eb661d Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 27 Mar 2015 14:15:05 +0000 Subject: irqchip: gicv3-its: Use non-cacheable accesses when no shareability If the ITS or the redistributors report their shareability as zero, then it is important to make sure they will no generate any cacheable traffic, as this is unlikely to produce the expected result. Signed-off-by: Marc Zyngier Link: https://lkml.kernel.org/r/1427465705-17126-5-git-send-email-marc.zyngier@arm.com Signed-off-by: Jason Cooper --- drivers/irqchip/irq-gic-v3-its.c | 47 ++++++++++++++++++++++++++++++++++---- include/linux/irqchip/arm-gic-v3.h | 4 ++++ 2 files changed, 47 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 56353f6b5952..9687f8afebff 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -802,6 +802,7 @@ static int its_alloc_tables(struct its_node *its) int i; int psz = SZ_64K; u64 shr = GITS_BASER_InnerShareable; + u64 cache = GITS_BASER_WaWb; for (i = 0; i < GITS_BASER_NR_REGS; i++) { u64 val = readq_relaxed(its->base + GITS_BASER + i * 8); @@ -848,7 +849,7 @@ retry_baser: val = (virt_to_phys(base) | (type << GITS_BASER_TYPE_SHIFT) | ((entry_size - 1) << GITS_BASER_ENTRY_SIZE_SHIFT) | - GITS_BASER_WaWb | + cache | shr | GITS_BASER_VALID); @@ -874,9 +875,12 @@ retry_baser: * Shareability didn't stick. Just use * whatever the read reported, which is likely * to be the only thing this redistributor - * supports. + * supports. If that's zero, make it + * non-cacheable as well. */ shr = tmp & GITS_BASER_SHAREABILITY_MASK; + if (!shr) + cache = GITS_BASER_nC; goto retry_baser; } @@ -980,6 +984,17 @@ static void its_cpu_init_lpis(void) tmp = readq_relaxed(rbase + GICR_PROPBASER); if ((tmp ^ val) & GICR_PROPBASER_SHAREABILITY_MASK) { + if (!(tmp & GICR_PROPBASER_SHAREABILITY_MASK)) { + /* + * The HW reports non-shareable, we must + * remove the cacheability attributes as + * well. + */ + val &= ~(GICR_PROPBASER_SHAREABILITY_MASK | + GICR_PROPBASER_CACHEABILITY_MASK); + val |= GICR_PROPBASER_nC; + writeq_relaxed(val, rbase + GICR_PROPBASER); + } pr_info_once("GIC: using cache flushing for LPI property table\n"); gic_rdists->flags |= RDIST_FLAGS_PROPBASE_NEEDS_FLUSHING; } @@ -990,6 +1005,18 @@ static void its_cpu_init_lpis(void) GICR_PENDBASER_WaWb); writeq_relaxed(val, rbase + GICR_PENDBASER); + tmp = readq_relaxed(rbase + GICR_PENDBASER); + + if (!(tmp & GICR_PENDBASER_SHAREABILITY_MASK)) { + /* + * The HW reports non-shareable, we must remove the + * cacheability attributes as well. + */ + val &= ~(GICR_PENDBASER_SHAREABILITY_MASK | + GICR_PENDBASER_CACHEABILITY_MASK); + val |= GICR_PENDBASER_nC; + writeq_relaxed(val, rbase + GICR_PENDBASER); + } /* Enable LPIs */ val = readl_relaxed(rbase + GICR_CTLR); @@ -1422,14 +1449,26 @@ static int its_probe(struct device_node *node, struct irq_domain *parent) writeq_relaxed(baser, its->base + GITS_CBASER); tmp = readq_relaxed(its->base + GITS_CBASER); - writeq_relaxed(0, its->base + GITS_CWRITER); - writel_relaxed(GITS_CTLR_ENABLE, its->base + GITS_CTLR); if ((tmp ^ baser) & GITS_CBASER_SHAREABILITY_MASK) { + if (!(tmp & GITS_CBASER_SHAREABILITY_MASK)) { + /* + * The HW reports non-shareable, we must + * remove the cacheability attributes as + * well. + */ + baser &= ~(GITS_CBASER_SHAREABILITY_MASK | + GITS_CBASER_CACHEABILITY_MASK); + baser |= GITS_CBASER_nC; + writeq_relaxed(baser, its->base + GITS_CBASER); + } pr_info("ITS: using cache flushing for cmd queue\n"); its->flags |= ITS_FLAGS_CMDQ_NEEDS_FLUSHING; } + writeq_relaxed(0, its->base + GITS_CWRITER); + writel_relaxed(GITS_CTLR_ENABLE, its->base + GITS_CTLR); + if (of_property_read_bool(its->msi_chip.of_node, "msi-controller")) { its->domain = irq_domain_add_tree(NULL, &its_domain_ops, its); if (!its->domain) { diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h index 826a4bd63d4a..ffbc034c8810 100644 --- a/include/linux/irqchip/arm-gic-v3.h +++ b/include/linux/irqchip/arm-gic-v3.h @@ -126,6 +126,7 @@ #define GICR_PROPBASER_WaWb (5U << 7) #define GICR_PROPBASER_RaWaWt (6U << 7) #define GICR_PROPBASER_RaWaWb (7U << 7) +#define GICR_PROPBASER_CACHEABILITY_MASK (7U << 7) #define GICR_PROPBASER_IDBITS_MASK (0x1f) #define GICR_PENDBASER_NonShareable (0U << 10) @@ -140,6 +141,7 @@ #define GICR_PENDBASER_WaWb (5U << 7) #define GICR_PENDBASER_RaWaWt (6U << 7) #define GICR_PENDBASER_RaWaWb (7U << 7) +#define GICR_PENDBASER_CACHEABILITY_MASK (7U << 7) /* * Re-Distributor registers, offsets from SGI_base @@ -195,6 +197,7 @@ #define GITS_CBASER_WaWb (5UL << 59) #define GITS_CBASER_RaWaWt (6UL << 59) #define GITS_CBASER_RaWaWb (7UL << 59) +#define GITS_CBASER_CACHEABILITY_MASK (7UL << 59) #define GITS_CBASER_NonShareable (0UL << 10) #define GITS_CBASER_InnerShareable (1UL << 10) #define GITS_CBASER_OuterShareable (2UL << 10) @@ -211,6 +214,7 @@ #define GITS_BASER_WaWb (5UL << 59) #define GITS_BASER_RaWaWt (6UL << 59) #define GITS_BASER_RaWaWb (7UL << 59) +#define GITS_BASER_CACHEABILITY_MASK (7UL << 59) #define GITS_BASER_TYPE_SHIFT (56) #define GITS_BASER_TYPE(r) (((r) >> GITS_BASER_TYPE_SHIFT) & 7) #define GITS_BASER_ENTRY_SIZE_SHIFT (48) -- cgit v1.2.3 From e9637415a92cf25ad800b7fdeddcd30cce7b44ab Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Mon, 30 Mar 2015 13:39:09 -0400 Subject: block: fix blk_stack_limits() regression due to lcm() change Linux 3.19 commit 69c953c ("lib/lcm.c: lcm(n,0)=lcm(0,n) is 0, not n") caused blk_stack_limits() to not properly stack queue_limits for stacked devices (e.g. DM). Fix this regression by establishing lcm_not_zero() and switching blk_stack_limits() over to using it. DM uses blk_set_stacking_limits() to establish the initial top-level queue_limits that are then built up based on underlying devices' limits using blk_stack_limits(). In the case of optimal_io_size (io_opt) blk_set_stacking_limits() establishes a default value of 0. With commit 69c953c, lcm(0, n) is no longer n, which compromises proper stacking of the underlying devices' io_opt. Test: $ modprobe scsi_debug dev_size_mb=10 num_tgts=1 opt_blks=1536 $ cat /sys/block/sde/queue/optimal_io_size 786432 $ dmsetup create node --table "0 100 linear /dev/sde 0" Before this fix: $ cat /sys/block/dm-5/queue/optimal_io_size 0 After this fix: $ cat /sys/block/dm-5/queue/optimal_io_size 786432 Signed-off-by: Mike Snitzer Cc: stable@vger.kernel.org # 3.19+ Acked-by: Martin K. Petersen Signed-off-by: Jens Axboe --- block/blk-settings.c | 6 +++--- include/linux/lcm.h | 1 + lib/lcm.c | 11 +++++++++++ 3 files changed, 15 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/block/blk-settings.c b/block/blk-settings.c index 6ed2cbe5e8c9..12600bfffca9 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -585,7 +585,7 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, b->physical_block_size); t->io_min = max(t->io_min, b->io_min); - t->io_opt = lcm(t->io_opt, b->io_opt); + t->io_opt = lcm_not_zero(t->io_opt, b->io_opt); t->cluster &= b->cluster; t->discard_zeroes_data &= b->discard_zeroes_data; @@ -616,7 +616,7 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, b->raid_partial_stripes_expensive); /* Find lowest common alignment_offset */ - t->alignment_offset = lcm(t->alignment_offset, alignment) + t->alignment_offset = lcm_not_zero(t->alignment_offset, alignment) % max(t->physical_block_size, t->io_min); /* Verify that new alignment_offset is on a logical block boundary */ @@ -643,7 +643,7 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, b->max_discard_sectors); t->discard_granularity = max(t->discard_granularity, b->discard_granularity); - t->discard_alignment = lcm(t->discard_alignment, alignment) % + t->discard_alignment = lcm_not_zero(t->discard_alignment, alignment) % t->discard_granularity; } diff --git a/include/linux/lcm.h b/include/linux/lcm.h index 7bf01d779b45..1ce79a7f1daa 100644 --- a/include/linux/lcm.h +++ b/include/linux/lcm.h @@ -4,5 +4,6 @@ #include unsigned long lcm(unsigned long a, unsigned long b) __attribute_const__; +unsigned long lcm_not_zero(unsigned long a, unsigned long b) __attribute_const__; #endif /* _LCM_H */ diff --git a/lib/lcm.c b/lib/lcm.c index e97dbd51e756..03d7fcb420b5 100644 --- a/lib/lcm.c +++ b/lib/lcm.c @@ -12,3 +12,14 @@ unsigned long lcm(unsigned long a, unsigned long b) return 0; } EXPORT_SYMBOL_GPL(lcm); + +unsigned long lcm_not_zero(unsigned long a, unsigned long b) +{ + unsigned long l = lcm(a, b); + + if (l) + return l; + + return (b ? : a); +} +EXPORT_SYMBOL_GPL(lcm_not_zero); -- cgit v1.2.3 From f9c72d10d6fbf949558cd088389a42213ed7b12d Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 31 Mar 2015 12:03:28 -0400 Subject: sunrpc: make debugfs file creation failure non-fatal We currently have a problem that SELinux policy is being enforced when creating debugfs files. If a debugfs file is created as a side effect of doing some syscall, then that creation can fail if the SELinux policy for that process prevents it. This seems wrong. We don't do that for files under /proc, for instance, so Bruce has proposed a patch to fix that. While discussing that patch however, Greg K.H. stated: "No kernel code should care / fail if a debugfs function fails, so please fix up the sunrpc code first." This patch converts all of the sunrpc debugfs setup code to be void return functins, and the callers to not look for errors from those functions. This should allow rpc_clnt and rpc_xprt creation to work, even if the kernel fails to create debugfs files for some reason. Symptoms were failing krb5 mounts on systems using gss-proxy and selinux. Fixes: 388f0c776781 "sunrpc: add a debugfs rpc_xprt directory..." Cc: stable@vger.kernel.org Signed-off-by: Jeff Layton Acked-by: Greg Kroah-Hartman Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/debug.h | 18 +++++++-------- net/sunrpc/clnt.c | 4 +--- net/sunrpc/debugfs.c | 52 ++++++++++++++++++++++++-------------------- net/sunrpc/sunrpc_syms.c | 7 +----- net/sunrpc/xprt.c | 7 +----- 5 files changed, 41 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h index c57d8ea0716c..59a7889e15db 100644 --- a/include/linux/sunrpc/debug.h +++ b/include/linux/sunrpc/debug.h @@ -60,17 +60,17 @@ struct rpc_xprt; #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) void rpc_register_sysctl(void); void rpc_unregister_sysctl(void); -int sunrpc_debugfs_init(void); +void sunrpc_debugfs_init(void); void sunrpc_debugfs_exit(void); -int rpc_clnt_debugfs_register(struct rpc_clnt *); +void rpc_clnt_debugfs_register(struct rpc_clnt *); void rpc_clnt_debugfs_unregister(struct rpc_clnt *); -int rpc_xprt_debugfs_register(struct rpc_xprt *); +void rpc_xprt_debugfs_register(struct rpc_xprt *); void rpc_xprt_debugfs_unregister(struct rpc_xprt *); #else -static inline int +static inline void sunrpc_debugfs_init(void) { - return 0; + return; } static inline void @@ -79,10 +79,10 @@ sunrpc_debugfs_exit(void) return; } -static inline int +static inline void rpc_clnt_debugfs_register(struct rpc_clnt *clnt) { - return 0; + return; } static inline void @@ -91,10 +91,10 @@ rpc_clnt_debugfs_unregister(struct rpc_clnt *clnt) return; } -static inline int +static inline void rpc_xprt_debugfs_register(struct rpc_xprt *xprt) { - return 0; + return; } static inline void diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 612aa73bbc60..e6ce1517367f 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -303,9 +303,7 @@ static int rpc_client_register(struct rpc_clnt *clnt, struct super_block *pipefs_sb; int err; - err = rpc_clnt_debugfs_register(clnt); - if (err) - return err; + rpc_clnt_debugfs_register(clnt); pipefs_sb = rpc_get_sb_net(net); if (pipefs_sb) { diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c index e811f390f9f6..82962f7e6e88 100644 --- a/net/sunrpc/debugfs.c +++ b/net/sunrpc/debugfs.c @@ -129,48 +129,52 @@ static const struct file_operations tasks_fops = { .release = tasks_release, }; -int +void rpc_clnt_debugfs_register(struct rpc_clnt *clnt) { - int len, err; + int len; char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */ + struct rpc_xprt *xprt; /* Already registered? */ - if (clnt->cl_debugfs) - return 0; + if (clnt->cl_debugfs || !rpc_clnt_dir) + return; len = snprintf(name, sizeof(name), "%x", clnt->cl_clid); if (len >= sizeof(name)) - return -EINVAL; + return; /* make the per-client dir */ clnt->cl_debugfs = debugfs_create_dir(name, rpc_clnt_dir); if (!clnt->cl_debugfs) - return -ENOMEM; + return; /* make tasks file */ - err = -ENOMEM; if (!debugfs_create_file("tasks", S_IFREG | S_IRUSR, clnt->cl_debugfs, clnt, &tasks_fops)) goto out_err; - err = -EINVAL; rcu_read_lock(); + xprt = rcu_dereference(clnt->cl_xprt); + /* no "debugfs" dentry? Don't bother with the symlink. */ + if (!xprt->debugfs) { + rcu_read_unlock(); + return; + } len = snprintf(name, sizeof(name), "../../rpc_xprt/%s", - rcu_dereference(clnt->cl_xprt)->debugfs->d_name.name); + xprt->debugfs->d_name.name); rcu_read_unlock(); + if (len >= sizeof(name)) goto out_err; - err = -ENOMEM; if (!debugfs_create_symlink("xprt", clnt->cl_debugfs, name)) goto out_err; - return 0; + return; out_err: debugfs_remove_recursive(clnt->cl_debugfs); clnt->cl_debugfs = NULL; - return err; } void @@ -226,33 +230,33 @@ static const struct file_operations xprt_info_fops = { .release = xprt_info_release, }; -int +void rpc_xprt_debugfs_register(struct rpc_xprt *xprt) { int len, id; static atomic_t cur_id; char name[9]; /* 8 hex digits + NULL term */ + if (!rpc_xprt_dir) + return; + id = (unsigned int)atomic_inc_return(&cur_id); len = snprintf(name, sizeof(name), "%x", id); if (len >= sizeof(name)) - return -EINVAL; + return; /* make the per-client dir */ xprt->debugfs = debugfs_create_dir(name, rpc_xprt_dir); if (!xprt->debugfs) - return -ENOMEM; + return; /* make tasks file */ if (!debugfs_create_file("info", S_IFREG | S_IRUSR, xprt->debugfs, xprt, &xprt_info_fops)) { debugfs_remove_recursive(xprt->debugfs); xprt->debugfs = NULL; - return -ENOMEM; } - - return 0; } void @@ -266,14 +270,17 @@ void __exit sunrpc_debugfs_exit(void) { debugfs_remove_recursive(topdir); + topdir = NULL; + rpc_clnt_dir = NULL; + rpc_xprt_dir = NULL; } -int __init +void __init sunrpc_debugfs_init(void) { topdir = debugfs_create_dir("sunrpc", NULL); if (!topdir) - goto out; + return; rpc_clnt_dir = debugfs_create_dir("rpc_clnt", topdir); if (!rpc_clnt_dir) @@ -283,10 +290,9 @@ sunrpc_debugfs_init(void) if (!rpc_xprt_dir) goto out_remove; - return 0; + return; out_remove: debugfs_remove_recursive(topdir); topdir = NULL; -out: - return -ENOMEM; + rpc_clnt_dir = NULL; } diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index e37fbed87956..ee5d3d253102 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -98,10 +98,7 @@ init_sunrpc(void) if (err) goto out4; - err = sunrpc_debugfs_init(); - if (err) - goto out5; - + sunrpc_debugfs_init(); #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) rpc_register_sysctl(); #endif @@ -109,8 +106,6 @@ init_sunrpc(void) init_socket_xprt(); /* clnt sock transport */ return 0; -out5: - unregister_rpc_pipefs(); out4: unregister_pernet_subsys(&sunrpc_net_ops); out3: diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index e3015aede0d9..9949722d99ce 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -1331,7 +1331,6 @@ static void xprt_init(struct rpc_xprt *xprt, struct net *net) */ struct rpc_xprt *xprt_create_transport(struct xprt_create *args) { - int err; struct rpc_xprt *xprt; struct xprt_class *t; @@ -1372,11 +1371,7 @@ found: return ERR_PTR(-ENOMEM); } - err = rpc_xprt_debugfs_register(xprt); - if (err) { - xprt_destroy(xprt); - return ERR_PTR(err); - } + rpc_xprt_debugfs_register(xprt); dprintk("RPC: created transport %p with %u slots\n", xprt, xprt->max_reqs); -- cgit v1.2.3 From a54acb3a6f853e8394c4cb7b6a4d93c88f13eefd Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Thu, 2 Apr 2015 17:07:00 +0200 Subject: dev: introduce dev_get_iflink() The goal of this patch is to prepare the removal of the iflink field. It introduces a new ndo function, which will be implemented by virtual interfaces. There is no functional change into this patch. All readers of iflink field now call dev_get_iflink(). Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_core.c | 2 +- include/linux/netdevice.h | 4 ++++ net/batman-adv/hard-interface.c | 5 +++-- net/bridge/br_netlink.c | 4 ++-- net/core/dev.c | 21 +++++++++++++++++++-- net/core/link_watch.c | 4 ++-- net/core/net-sysfs.c | 10 +++++++++- net/core/rtnetlink.c | 8 ++++---- net/ipv4/ipmr.c | 2 +- net/ipv6/addrconf.c | 4 ++-- net/ipv6/ip6mr.c | 2 +- 11 files changed, 48 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 2a175006028b..131bde98188d 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -330,7 +330,7 @@ static int ipvlan_process_v4_outbound(struct sk_buff *skb) struct rtable *rt; int err, ret = NET_XMIT_DROP; struct flowi4 fl4 = { - .flowi4_oif = dev->iflink, + .flowi4_oif = dev_get_iflink(dev), .flowi4_tos = RT_TOS(ip4h->tos), .flowi4_flags = FLOWI_FLAG_ANYSRC, .daddr = ip4h->daddr, diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 967bb4c8caf1..788eb7a622ad 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1030,6 +1030,8 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * int queue_index, u32 maxrate); * Called when a user wants to set a max-rate limitation of specific * TX queue. + * int (*ndo_get_iflink)(const struct net_device *dev); + * Called to get the iflink value of this device. */ struct net_device_ops { int (*ndo_init)(struct net_device *dev); @@ -1191,6 +1193,7 @@ struct net_device_ops { int (*ndo_set_tx_maxrate)(struct net_device *dev, int queue_index, u32 maxrate); + int (*ndo_get_iflink)(const struct net_device *dev); }; /** @@ -2149,6 +2152,7 @@ void __dev_remove_pack(struct packet_type *pt); void dev_add_offload(struct packet_offload *po); void dev_remove_offload(struct packet_offload *po); +int dev_get_iflink(const struct net_device *dev); struct net_device *__dev_get_by_flags(struct net *net, unsigned short flags, unsigned short mask); struct net_device *dev_get_by_name(struct net *net, const char *name); diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index fbda6b54baff..baf1f9843f2c 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -83,11 +83,12 @@ static bool batadv_is_on_batman_iface(const struct net_device *net_dev) return true; /* no more parents..stop recursion */ - if (net_dev->iflink == 0 || net_dev->iflink == net_dev->ifindex) + if (dev_get_iflink(net_dev) == 0 || + dev_get_iflink(net_dev) == net_dev->ifindex) return false; /* recurse over the parent device */ - parent_dev = __dev_get_by_index(&init_net, net_dev->iflink); + parent_dev = __dev_get_by_index(&init_net, dev_get_iflink(net_dev)); /* if we got a NULL parent_dev there is something broken.. */ if (WARN(!parent_dev, "Cannot find parent device")) return false; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index e1115a224a95..0e4ddb81610d 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -305,8 +305,8 @@ static int br_fill_ifinfo(struct sk_buff *skb, nla_put_u8(skb, IFLA_OPERSTATE, operstate) || (dev->addr_len && nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) || - (dev->ifindex != dev->iflink && - nla_put_u32(skb, IFLA_LINK, dev->iflink))) + (dev->ifindex != dev_get_iflink(dev) && + nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev)))) goto nla_put_failure; if (event == RTM_NEWLINK && port) { diff --git a/net/core/dev.c b/net/core/dev.c index 65492b0354c0..77172d085760 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -659,6 +659,23 @@ __setup("netdev=", netdev_boot_setup); *******************************************************************************/ +/** + * dev_get_iflink - get 'iflink' value of a interface + * @dev: targeted interface + * + * Indicates the ifindex the interface is linked to. + * Physical interfaces have the same 'ifindex' and 'iflink' values. + */ + +int dev_get_iflink(const struct net_device *dev) +{ + if (dev->netdev_ops && dev->netdev_ops->ndo_get_iflink) + return dev->netdev_ops->ndo_get_iflink(dev); + + return dev->iflink; +} +EXPORT_SYMBOL(dev_get_iflink); + /** * __dev_get_by_name - find a device by its name * @net: the applicable net namespace @@ -6345,7 +6362,7 @@ int register_netdevice(struct net_device *dev) else if (__dev_get_by_index(net, dev->ifindex)) goto err_uninit; - if (dev->iflink == -1) + if (dev_get_iflink(dev) == -1) dev->iflink = dev->ifindex; /* Transfer changeable features to wanted_features and enable @@ -7061,7 +7078,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* If there is an ifindex conflict assign a new one */ if (__dev_get_by_index(net, dev->ifindex)) { - int iflink = (dev->iflink == dev->ifindex); + int iflink = (dev_get_iflink(dev) == dev->ifindex); dev->ifindex = dev_new_index(net); if (iflink) dev->iflink = dev->ifindex; diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 49a9e3e06c08..982861607f88 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -40,7 +40,7 @@ static DEFINE_SPINLOCK(lweventlist_lock); static unsigned char default_operstate(const struct net_device *dev) { if (!netif_carrier_ok(dev)) - return (dev->ifindex != dev->iflink ? + return (dev->ifindex != dev_get_iflink(dev) ? IF_OPER_LOWERLAYERDOWN : IF_OPER_DOWN); if (netif_dormant(dev)) @@ -89,7 +89,7 @@ static bool linkwatch_urgent_event(struct net_device *dev) if (!netif_running(dev)) return false; - if (dev->ifindex != dev->iflink) + if (dev->ifindex != dev_get_iflink(dev)) return true; if (dev->priv_flags & IFF_TEAM_PORT) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index cc5cf689809c..4238d6da5c60 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -109,11 +109,19 @@ NETDEVICE_SHOW_RO(dev_id, fmt_hex); NETDEVICE_SHOW_RO(dev_port, fmt_dec); NETDEVICE_SHOW_RO(addr_assign_type, fmt_dec); NETDEVICE_SHOW_RO(addr_len, fmt_dec); -NETDEVICE_SHOW_RO(iflink, fmt_dec); NETDEVICE_SHOW_RO(ifindex, fmt_dec); NETDEVICE_SHOW_RO(type, fmt_dec); NETDEVICE_SHOW_RO(link_mode, fmt_dec); +static ssize_t iflink_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct net_device *ndev = to_net_dev(dev); + + return sprintf(buf, fmt_dec, dev_get_iflink(ndev)); +} +static DEVICE_ATTR_RO(iflink); + static ssize_t format_name_assign_type(const struct net_device *dev, char *buf) { return sprintf(buf, fmt_dec, dev->name_assign_type); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b96ac2109c82..ee0186cdd5cf 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1055,8 +1055,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, #ifdef CONFIG_RPS nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) || #endif - (dev->ifindex != dev->iflink && - nla_put_u32(skb, IFLA_LINK, dev->iflink)) || + (dev->ifindex != dev_get_iflink(dev) && + nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev))) || (upper_dev && nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) || nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) || @@ -2863,8 +2863,8 @@ int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) || (dev->addr_len && nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) || - (dev->ifindex != dev->iflink && - nla_put_u32(skb, IFLA_LINK, dev->iflink))) + (dev->ifindex != dev_get_iflink(dev) && + nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev)))) goto nla_put_failure; br_afspec = nla_nest_start(skb, IFLA_AF_SPEC); diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index b4a545d24adb..eec68b0c3bc8 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -801,7 +801,7 @@ static int vif_add(struct net *net, struct mr_table *mrt, v->pkt_out = 0; v->link = dev->ifindex; if (v->flags & (VIFF_TUNNEL | VIFF_REGISTER)) - v->link = dev->iflink; + v->link = dev_get_iflink(dev); /* And finish update writing critical data */ write_lock_bh(&mrt_lock); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5c9e94cb1b2c..37b70e82bff8 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4858,8 +4858,8 @@ static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, (dev->addr_len && nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) || nla_put_u32(skb, IFLA_MTU, dev->mtu) || - (dev->ifindex != dev->iflink && - nla_put_u32(skb, IFLA_LINK, dev->iflink))) + (dev->ifindex != dev_get_iflink(dev) && + nla_put_u32(skb, IFLA_LINK, dev_get_iflink(dev)))) goto nla_put_failure; protoinfo = nla_nest_start(skb, IFLA_PROTINFO); if (!protoinfo) diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index caf6b99374e6..18a5ab286420 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -992,7 +992,7 @@ static int mif6_add(struct net *net, struct mr6_table *mrt, v->pkt_out = 0; v->link = dev->ifindex; if (v->flags & MIFF_REGISTER) - v->link = dev->iflink; + v->link = dev_get_iflink(dev); /* And finish update writing critical data */ write_lock_bh(&mrt_lock); -- cgit v1.2.3 From 7a66bbc96ce9ad8261fa5f7f6ae65370eb6866ee Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Thu, 2 Apr 2015 17:07:09 +0200 Subject: net: remove iflink field from struct net_device Now that all users of iflink have the ndo_get_iflink handler available, it's possible to remove this field. By default, dev_get_iflink() returns the ifindex of the interface. Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +-- net/core/dev.c | 13 ++----------- 2 files changed, 3 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 788eb7a622ad..846a1f5bc9db 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1538,7 +1538,7 @@ struct net_device { netdev_features_t mpls_features; int ifindex; - int iflink; + int group; struct net_device_stats stats; @@ -1741,7 +1741,6 @@ struct net_device { #endif struct phy_device *phydev; struct lock_class_key *qdisc_tx_busylock; - int group; struct pm_qos_request pm_qos_req; }; #define to_net_dev(d) container_of(d, struct net_device, dev) diff --git a/net/core/dev.c b/net/core/dev.c index 77172d085760..3be107e0bc93 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -672,7 +672,7 @@ int dev_get_iflink(const struct net_device *dev) if (dev->netdev_ops && dev->netdev_ops->ndo_get_iflink) return dev->netdev_ops->ndo_get_iflink(dev); - return dev->iflink; + return dev->ifindex; } EXPORT_SYMBOL(dev_get_iflink); @@ -6331,8 +6331,6 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->addr_list_lock); netdev_set_addr_lockdep_class(dev); - dev->iflink = -1; - ret = dev_get_valid_name(net, dev, dev->name); if (ret < 0) goto out; @@ -6362,9 +6360,6 @@ int register_netdevice(struct net_device *dev) else if (__dev_get_by_index(net, dev->ifindex)) goto err_uninit; - if (dev_get_iflink(dev) == -1) - dev->iflink = dev->ifindex; - /* Transfer changeable features to wanted_features and enable * software offloads (GSO and GRO). */ @@ -7077,12 +7072,8 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char dev_net_set(dev, net); /* If there is an ifindex conflict assign a new one */ - if (__dev_get_by_index(net, dev->ifindex)) { - int iflink = (dev_get_iflink(dev) == dev->ifindex); + if (__dev_get_by_index(net, dev->ifindex)) dev->ifindex = dev_new_index(net); - if (iflink) - dev->iflink = dev->ifindex; - } /* Send a netdev-add uevent to the new namespace */ kobject_uevent(&dev->dev.kobj, KOBJ_ADD); -- cgit v1.2.3 From 802f42a8d99254b8602bf65cc50e8333bf479f13 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:06 +0300 Subject: net/mlx4: Add RSS support for fragmented IP datagrams Enable RSS support for fragmented IP packets, when device supports it. Until now, fragmented IP packets were directed only to the default_qpn. Since IP fragments (datagram) have no upper protocols (L3 IP packets), hash is performed on 3-tuple - dst MAC, source IP and dest IP. The HW makes sure that this holds for the 1st fragment too, so all fragments go to the same QP. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/fw.c | 12 ++++++++++++ drivers/net/ethernet/mellanox/mlx4/fw.h | 1 + drivers/net/ethernet/mellanox/mlx4/main.c | 2 ++ include/linux/mlx4/device.h | 1 + 4 files changed, 16 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index 209a6171e59b..412e019e8b2e 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -105,6 +105,7 @@ static void dump_dev_cap_flags(struct mlx4_dev *dev, u64 flags) [41] = "Unicast VEP steering support", [42] = "Multicast VEP steering support", [48] = "Counters support", + [52] = "RSS IP fragments support", [53] = "Port ETS Scheduler support", [55] = "Port link type sensing support", [59] = "Port management change event support", @@ -1138,6 +1139,9 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave, } for (; slave_port < dev->caps.num_ports; ++slave_port) flags &= ~(MLX4_DEV_CAP_FLAG_WOL_PORT1 << slave_port); + + /* Not exposing RSS IP fragments to guests */ + flags &= ~MLX4_DEV_CAP_FLAG_RSS_IP_FRAG; MLX4_PUT(outbox->buf, flags, QUERY_DEV_CAP_EXT_FLAGS_OFFSET); MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_VL_PORT_OFFSET); @@ -1701,6 +1705,10 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param) if (dev->caps.flags & MLX4_DEV_CAP_FLAG_COUNTERS) *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 4); + /* Enable RSS spread to fragmented IP packets when supported */ + if (dev->caps.flags & MLX4_DEV_CAP_FLAG_RSS_IP_FRAG) + *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 13); + /* CX3 is capable of extending CQEs/EQEs from 32 to 64 bytes */ if (dev->caps.flags & MLX4_DEV_CAP_FLAG_64B_EQE) { *(inbox + INIT_HCA_EQE_CQE_OFFSETS / 4) |= cpu_to_be32(1 << 29); @@ -1889,6 +1897,10 @@ int mlx4_QUERY_HCA(struct mlx4_dev *dev, else param->steering_mode = MLX4_STEERING_MODE_A0; } + + if (dword_field & (1 << 13)) + param->rss_ip_frags = 1; + /* steering attributes */ if (param->steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED) { MLX4_GET(param->mc_base, outbox, INIT_HCA_FS_BASE_OFFSET); diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h index 863655bd3947..07cb7c2461ad 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw.h @@ -203,6 +203,7 @@ struct mlx4_init_hca_param { u64 dev_cap_enabled; u16 cqe_size; /* For use only when CQE stride feature enabled */ u16 eqe_size; /* For use only when EQE stride feature enabled */ + u8 rss_ip_frags; }; struct mlx4_init_ib_param { diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 43aa76775b5f..ee0a67069b57 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -885,6 +885,8 @@ static int mlx4_slave_cap(struct mlx4_dev *dev) mlx4_warn(dev, "Timestamping is not supported in slave mode\n"); slave_adjust_steering_mode(dev, &dev_cap, &hca_param); + mlx4_dbg(dev, "RSS support for IP fragments is %s\n", + hca_param.rss_ip_frags ? "on" : "off"); if (func_cap.extra_flags & MLX4_QUERY_FUNC_FLAGS_BF_RES_QP && dev->caps.bf_reg_size) diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index ab7ebec943b8..41eaafdfb1cd 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -174,6 +174,7 @@ enum { MLX4_DEV_CAP_FLAG_VEP_UC_STEER = 1LL << 41, MLX4_DEV_CAP_FLAG_VEP_MC_STEER = 1LL << 42, MLX4_DEV_CAP_FLAG_COUNTERS = 1LL << 48, + MLX4_DEV_CAP_FLAG_RSS_IP_FRAG = 1LL << 52, MLX4_DEV_CAP_FLAG_SET_ETH_SCHED = 1LL << 53, MLX4_DEV_CAP_FLAG_SENSE_SUPPORT = 1LL << 55, MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV = 1LL << 59, -- cgit v1.2.3 From fccea6436a5bd16aed6f722efce13ec2c90427d7 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:08 +0300 Subject: net/mlx4: Make mlx4_is_eth visible inline funcion Currently implemented as static function in resource_tracker.c -- this change will allow other files in mlx4_core to use it as well. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/resource_tracker.c | 5 ----- include/linux/mlx4/device.h | 5 +++++ 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index 11bcd27e218f..fd2520aa8a8c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -221,11 +221,6 @@ struct res_fs_rule { int qpn; }; -static int mlx4_is_eth(struct mlx4_dev *dev, int port) -{ - return dev->caps.port_mask[port] == MLX4_PORT_TYPE_IB ? 0 : 1; -} - static void *res_tracker_lookup(struct rb_root *root, u64 res_id) { struct rb_node *node = root->rb_node; diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 41eaafdfb1cd..0f7f13d6a03e 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -1002,6 +1002,11 @@ static inline int mlx4_is_slave(struct mlx4_dev *dev) return dev->flags & MLX4_FLAG_SLAVE; } +static inline int mlx4_is_eth(struct mlx4_dev *dev, int port) +{ + return dev->caps.port_type[port] == MLX4_PORT_TYPE_IB ? 0 : 1; +} + int mlx4_buf_alloc(struct mlx4_dev *dev, int size, int max_direct, struct mlx4_buf *buf, gfp_t gfp); void mlx4_buf_free(struct mlx4_dev *dev, int size, struct mlx4_buf *buf); -- cgit v1.2.3 From 12a889c057504fbf307dd237aedb87263ef2848a Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:10 +0300 Subject: net/mlx4: New file for QoS related firmware commands Create two new files fw_qos.h and fw_qos.c in mlx4_core module. It gathers all relevant QoS firmware related commands etc, thus improving encapsulation of the mlx4_core module. For now it contains the QoS existing commands: mlx4_SET_PORT_SCHEDULER and mlx4_SET_PORT_PRIO2TC. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/Makefile | 5 +- drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c | 1 + drivers/net/ethernet/mellanox/mlx4/fw_qos.c | 125 +++++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx4/fw_qos.h | 75 +++++++++++++++ drivers/net/ethernet/mellanox/mlx4/mlx4.h | 15 --- drivers/net/ethernet/mellanox/mlx4/port.c | 71 -------------- include/linux/mlx4/device.h | 5 - 7 files changed, 204 insertions(+), 93 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx4/fw_qos.c create mode 100644 drivers/net/ethernet/mellanox/mlx4/fw_qos.h (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/Makefile b/drivers/net/ethernet/mellanox/mlx4/Makefile index 3e9c70f15b42..c82217e0d22d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/Makefile +++ b/drivers/net/ethernet/mellanox/mlx4/Makefile @@ -1,7 +1,8 @@ obj-$(CONFIG_MLX4_CORE) += mlx4_core.o -mlx4_core-y := alloc.o catas.o cmd.o cq.o eq.o fw.o icm.o intf.o main.o mcg.o \ - mr.o pd.o port.o profile.o qp.o reset.o sense.o srq.o resource_tracker.o +mlx4_core-y := alloc.o catas.o cmd.o cq.o eq.o fw.o fw_qos.o icm.o intf.o \ + main.o mcg.o mr.o pd.o port.o profile.o qp.o reset.o sense.o \ + srq.o resource_tracker.o obj-$(CONFIG_MLX4_EN) += mlx4_en.o diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index 8e3260c0eaa5..f01918c63f28 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -35,6 +35,7 @@ #include #include "mlx4_en.h" +#include "fw_qos.h" /* Definitions for QCN */ diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.c b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c new file mode 100644 index 000000000000..0f5af7cf4547 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c @@ -0,0 +1,125 @@ +/* + * Copyright (c) 2004, 2005 Topspin Communications. All rights reserved. + * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. + * All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include "fw_qos.h" + +struct mlx4_set_port_prio2tc_context { + u8 prio2tc[4]; +}; + +struct mlx4_port_scheduler_tc_cfg_be { + __be16 pg; + __be16 bw_precentage; + __be16 max_bw_units; /* 3-100Mbps, 4-1Gbps, other values - reserved */ + __be16 max_bw_value; +}; + +struct mlx4_set_port_scheduler_context { + struct mlx4_port_scheduler_tc_cfg_be tc[MLX4_NUM_TC]; +}; + +int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc) +{ + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_set_port_prio2tc_context *context; + int err; + u32 in_mod; + int i; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + context = mailbox->buf; + + for (i = 0; i < MLX4_NUM_UP; i += 2) + context->prio2tc[i >> 1] = prio2tc[i] << 4 | prio2tc[i + 1]; + + in_mod = MLX4_SET_PORT_PRIO2TC << 8 | port; + err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_SET_PORT_PRIO2TC); + +int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, + u8 *pg, u16 *ratelimit) +{ + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_set_port_scheduler_context *context; + int err; + u32 in_mod; + int i; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + context = mailbox->buf; + + for (i = 0; i < MLX4_NUM_TC; i++) { + struct mlx4_port_scheduler_tc_cfg_be *tc = &context->tc[i]; + u16 r; + + if (ratelimit && ratelimit[i]) { + if (ratelimit[i] <= MLX4_MAX_100M_UNITS_VAL) { + r = ratelimit[i]; + tc->max_bw_units = + htons(MLX4_RATELIMIT_100M_UNITS); + } else { + r = ratelimit[i] / 10; + tc->max_bw_units = + htons(MLX4_RATELIMIT_1G_UNITS); + } + tc->max_bw_value = htons(r); + } else { + tc->max_bw_value = htons(MLX4_RATELIMIT_DEFAULT); + tc->max_bw_units = htons(MLX4_RATELIMIT_1G_UNITS); + } + + tc->pg = htons(pg[i]); + tc->bw_precentage = htons(tc_tx_bw[i]); + } + + in_mod = MLX4_SET_PORT_SCHEDULER << 8 | port; + err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_SET_PORT_SCHEDULER); diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h new file mode 100644 index 000000000000..ab9be0fad9c8 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h @@ -0,0 +1,75 @@ +/* + * Copyright (c) 2004, 2005 Topspin Communications. All rights reserved. + * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. + * All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef MLX4_FW_QOS_H +#define MLX4_FW_QOS_H + +#include +#include + +#define MLX4_NUM_UP 8 +#define MLX4_NUM_TC 8 + +/** + * mlx4_SET_PORT_PRIO2TC - This routine maps user priorities to traffic + * classes of a given port and device. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @prio2tc: Array of TC associated with each priorities. + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc); + +/** + * mlx4_SET_PORT_SCHEDULER - This routine configures the arbitration between + * traffic classes (ETS) and configured rate limit for traffic classes. + * tc_tx_bw, pg and ratelimit are arrays where each index represents a TC. + * The description for those parameters below refers to a single TC. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @tc_tx_bw: The percentage of the bandwidth allocated for traffic class + * within a TC group. The sum of the bw_percentage of all the traffic + * classes within a TC group must equal 100% for correct operation. + * @pg: The TC group the traffic class is associated with. + * @ratelimit: The maximal bandwidth allowed for the use by this traffic class. + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, + u8 *pg, u16 *ratelimit); + +#endif /* MLX4_FW_QOS_H */ diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 0b16db015745..9e5236b1b5f8 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -64,21 +64,6 @@ #define INIT_HCA_TPT_MW_ENABLE (1 << 7) -struct mlx4_set_port_prio2tc_context { - u8 prio2tc[4]; -}; - -struct mlx4_port_scheduler_tc_cfg_be { - __be16 pg; - __be16 bw_precentage; - __be16 max_bw_units; /* 3-100Mbps, 4-1Gbps, other values - reserved */ - __be16 max_bw_value; -}; - -struct mlx4_set_port_scheduler_context { - struct mlx4_port_scheduler_tc_cfg_be tc[MLX4_NUM_TC]; -}; - enum { MLX4_HCR_BASE = 0x80680, MLX4_HCR_SIZE = 0x0001c, diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index b97f173ab062..6a53d42db52f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -1016,77 +1016,6 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, } EXPORT_SYMBOL(mlx4_SET_PORT_qpn_calc); -int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc) -{ - struct mlx4_cmd_mailbox *mailbox; - struct mlx4_set_port_prio2tc_context *context; - int err; - u32 in_mod; - int i; - - mailbox = mlx4_alloc_cmd_mailbox(dev); - if (IS_ERR(mailbox)) - return PTR_ERR(mailbox); - context = mailbox->buf; - for (i = 0; i < MLX4_NUM_UP; i += 2) - context->prio2tc[i >> 1] = prio2tc[i] << 4 | prio2tc[i + 1]; - - in_mod = MLX4_SET_PORT_PRIO2TC << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); - - mlx4_free_cmd_mailbox(dev, mailbox); - return err; -} -EXPORT_SYMBOL(mlx4_SET_PORT_PRIO2TC); - -int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, - u8 *pg, u16 *ratelimit) -{ - struct mlx4_cmd_mailbox *mailbox; - struct mlx4_set_port_scheduler_context *context; - int err; - u32 in_mod; - int i; - - mailbox = mlx4_alloc_cmd_mailbox(dev); - if (IS_ERR(mailbox)) - return PTR_ERR(mailbox); - context = mailbox->buf; - - for (i = 0; i < MLX4_NUM_TC; i++) { - struct mlx4_port_scheduler_tc_cfg_be *tc = &context->tc[i]; - u16 r; - - if (ratelimit && ratelimit[i]) { - if (ratelimit[i] <= MLX4_MAX_100M_UNITS_VAL) { - r = ratelimit[i]; - tc->max_bw_units = - htons(MLX4_RATELIMIT_100M_UNITS); - } else { - r = ratelimit[i]/10; - tc->max_bw_units = - htons(MLX4_RATELIMIT_1G_UNITS); - } - tc->max_bw_value = htons(r); - } else { - tc->max_bw_value = htons(MLX4_RATELIMIT_DEFAULT); - tc->max_bw_units = htons(MLX4_RATELIMIT_1G_UNITS); - } - - tc->pg = htons(pg[i]); - tc->bw_precentage = htons(tc_tx_bw[i]); - } - - in_mod = MLX4_SET_PORT_SCHEDULER << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); - - mlx4_free_cmd_mailbox(dev, mailbox); - return err; -} -EXPORT_SYMBOL(mlx4_SET_PORT_SCHEDULER); - enum { VXLAN_ENABLE_MODIFY = 1 << 7, VXLAN_STEERING_MODIFY = 1 << 6, diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 0f7f13d6a03e..b676d0c0111a 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -49,8 +49,6 @@ #define MSIX_LEGACY_SZ 4 #define MIN_MSIX_P_PORT 5 -#define MLX4_NUM_UP 8 -#define MLX4_NUM_TC 8 #define MLX4_MAX_100M_UNITS_VAL 255 /* * work around: can't set values * greater then this value when @@ -1311,9 +1309,6 @@ int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu, u8 pptx, u8 pfctx, u8 pprx, u8 pfcrx); int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, u8 promisc); -int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc); -int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, - u8 *pg, u16 *ratelimit); int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable); int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx); int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx); -- cgit v1.2.3 From 7e95bb99a8f969d4534867452793e44cc4731ca9 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:11 +0300 Subject: net/mlx4: Add mlx4_ALLOCATE_VPP implementation Implements device ALLOCATE_VPP command, to be used for granular QoS configuration of VFs by the PF device. Defines and queries the amount of VPPs assigned to each port, and the amount of VPPs assigned to each priority of each port. Once the total VPPs are split between the priorities of a port, they may be assigned with a share of the BW or a rate limit. Split into two functions (get/set) whoch are supplied with mlx4_alloc_vpp_context and physical port number. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/cmd.c | 9 ++++ drivers/net/ethernet/mellanox/mlx4/fw_qos.c | 75 +++++++++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx4/fw_qos.h | 28 +++++++++++ include/linux/mlx4/cmd.h | 1 + 4 files changed, 113 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 179ca2889434..79cc5ec57dca 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -1454,6 +1454,15 @@ static struct mlx4_cmd_info cmd_info[] = { .verify = NULL, .wrapper = mlx4_CMD_EPERM_wrapper, }, + { + .opcode = MLX4_CMD_ALLOCATE_VPP, + .has_inbox = false, + .has_outbox = true, + .out_is_imm = false, + .encode_slave_id = false, + .verify = NULL, + .wrapper = mlx4_CMD_EPERM_wrapper, + }, { .opcode = MLX4_CMD_CONF_SPECIAL_QP, .has_inbox = false, diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.c b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c index 0f5af7cf4547..5ce3440b99c7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw_qos.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c @@ -34,6 +34,13 @@ #include #include "fw_qos.h" +#include "fw.h" + +enum { + /* allocate vpp opcode modifiers */ + MLX4_ALLOCATE_VPP_ALLOCATE = 0x0, + MLX4_ALLOCATE_VPP_QUERY = 0x1 +}; struct mlx4_set_port_prio2tc_context { u8 prio2tc[4]; @@ -50,6 +57,12 @@ struct mlx4_set_port_scheduler_context { struct mlx4_port_scheduler_tc_cfg_be tc[MLX4_NUM_TC]; }; +/* Granular Qos (per VF) section */ +struct mlx4_alloc_vpp_param { + __be32 availible_vpp; + __be32 vpp_p_up[MLX4_NUM_UP]; +}; + int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc) { struct mlx4_cmd_mailbox *mailbox; @@ -123,3 +136,65 @@ int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, return err; } EXPORT_SYMBOL(mlx4_SET_PORT_SCHEDULER); + +int mlx4_ALLOCATE_VPP_get(struct mlx4_dev *dev, u8 port, + u16 *availible_vpp, u8 *vpp_p_up) +{ + int i; + int err; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_alloc_vpp_param *out_param; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + out_param = mailbox->buf; + + err = mlx4_cmd_box(dev, 0, mailbox->dma, port, + MLX4_ALLOCATE_VPP_QUERY, + MLX4_CMD_ALLOCATE_VPP, + MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_NATIVE); + if (err) + goto out; + + /* Total number of supported VPPs */ + *availible_vpp = (u16)be32_to_cpu(out_param->availible_vpp); + + for (i = 0; i < MLX4_NUM_UP; i++) + vpp_p_up[i] = (u8)be32_to_cpu(out_param->vpp_p_up[i]); + +out: + mlx4_free_cmd_mailbox(dev, mailbox); + + return err; +} +EXPORT_SYMBOL(mlx4_ALLOCATE_VPP_get); + +int mlx4_ALLOCATE_VPP_set(struct mlx4_dev *dev, u8 port, u8 *vpp_p_up) +{ + int i; + int err; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_alloc_vpp_param *in_param; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + in_param = mailbox->buf; + + for (i = 0; i < MLX4_NUM_UP; i++) + in_param->vpp_p_up[i] = cpu_to_be32(vpp_p_up[i]); + + err = mlx4_cmd(dev, mailbox->dma, port, + MLX4_ALLOCATE_VPP_ALLOCATE, + MLX4_CMD_ALLOCATE_VPP, + MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_ALLOCATE_VPP_set); diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h index ab9be0fad9c8..be79951c44ef 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h @@ -71,5 +71,33 @@ int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc); **/ int mlx4_SET_PORT_SCHEDULER(struct mlx4_dev *dev, u8 port, u8 *tc_tx_bw, u8 *pg, u16 *ratelimit); +/** + * mlx4_ALLOCATE_VPP_get - Query port VPP availible resources and allocation. + * Before distribution of VPPs to priorities, only availible_vpp is returned. + * After initialization it returns the distribution of VPPs among priorities. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @availible_vpp: Pointer to variable where number of availible VPPs is stored + * @vpp_p_up: Distribution of VPPs to priorities is stored in this array + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_ALLOCATE_VPP_get(struct mlx4_dev *dev, u8 port, + u16 *availible_vpp, u8 *vpp_p_up); +/** + * mlx4_ALLOCATE_VPP_set - Distribution of VPPs among differnt priorities. + * The total number of VPPs assigned to all for a port must not exceed + * the value reported by availible_vpp in mlx4_ALLOCATE_VPP_get. + * VPP allocation is allowed only after the port type has been set, + * and while no QPs are open for this port. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @vpp_p_up: Allocation of VPPs to different priorities. + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_ALLOCATE_VPP_set(struct mlx4_dev *dev, u8 port, u8 *vpp_p_up); #endif /* MLX4_FW_QOS_H */ diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h index 7299e9548906..e0f88a098fb8 100644 --- a/include/linux/mlx4/cmd.h +++ b/include/linux/mlx4/cmd.h @@ -68,6 +68,7 @@ enum { MLX4_CMD_UNMAP_ICM_AUX = 0xffb, MLX4_CMD_SET_ICM_SIZE = 0xffd, MLX4_CMD_ACCESS_REG = 0x3b, + MLX4_CMD_ALLOCATE_VPP = 0x80, /*master notify fw on finish for slave's flr*/ MLX4_CMD_INFORM_FLR_DONE = 0x5b, -- cgit v1.2.3 From 1c29146d3821be984030b49cf9509a269edc3b8b Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:12 +0300 Subject: net/mlx4: Add mlx4_SET_VPORT_QOS implementation Add the SET_VPORT_QOS device command, which is ntended for virtual granular QoS configuration per VF in SRIOV mode. The SET_VPORT_QOS command sets and queries QoS parameters of a VPort. Each priority allowed for a VPort is assigned with a share of the BW, and a BW limitation. QoS parameters can be modified at any time, but must be initialized before any QP is associated with the VPort. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/cmd.c | 9 +++ drivers/net/ethernet/mellanox/mlx4/fw_qos.c | 89 +++++++++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx4/fw_qos.h | 36 ++++++++++++ include/linux/mlx4/cmd.h | 1 + 4 files changed, 135 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 79cc5ec57dca..af95231ba3e2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -1463,6 +1463,15 @@ static struct mlx4_cmd_info cmd_info[] = { .verify = NULL, .wrapper = mlx4_CMD_EPERM_wrapper, }, + { + .opcode = MLX4_CMD_SET_VPORT_QOS, + .has_inbox = false, + .has_outbox = true, + .out_is_imm = false, + .encode_slave_id = false, + .verify = NULL, + .wrapper = mlx4_CMD_EPERM_wrapper, + }, { .opcode = MLX4_CMD_CONF_SPECIAL_QP, .has_inbox = false, diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.c b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c index 5ce3440b99c7..8f2fde0487c4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw_qos.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.c @@ -42,6 +42,12 @@ enum { MLX4_ALLOCATE_VPP_QUERY = 0x1 }; +enum { + /* set vport qos opcode modifiers */ + MLX4_SET_VPORT_QOS_SET = 0x0, + MLX4_SET_VPORT_QOS_QUERY = 0x1 +}; + struct mlx4_set_port_prio2tc_context { u8 prio2tc[4]; }; @@ -63,6 +69,19 @@ struct mlx4_alloc_vpp_param { __be32 vpp_p_up[MLX4_NUM_UP]; }; +struct mlx4_prio_qos_param { + __be32 bw_share; + __be32 max_avg_bw; + __be32 reserved; + __be32 enable; + __be32 reserved1[4]; +}; + +struct mlx4_set_vport_context { + __be32 reserved[8]; + struct mlx4_prio_qos_param qos_p_up[MLX4_NUM_UP]; +}; + int mlx4_SET_PORT_PRIO2TC(struct mlx4_dev *dev, u8 port, u8 *prio2tc) { struct mlx4_cmd_mailbox *mailbox; @@ -198,3 +217,73 @@ int mlx4_ALLOCATE_VPP_set(struct mlx4_dev *dev, u8 port, u8 *vpp_p_up) return err; } EXPORT_SYMBOL(mlx4_ALLOCATE_VPP_set); + +int mlx4_SET_VPORT_QOS_get(struct mlx4_dev *dev, u8 port, u8 vport, + struct mlx4_vport_qos_param *out_param) +{ + int i; + int err; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_set_vport_context *ctx; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + ctx = mailbox->buf; + + err = mlx4_cmd_box(dev, 0, mailbox->dma, (vport << 8) | port, + MLX4_SET_VPORT_QOS_QUERY, + MLX4_CMD_SET_VPORT_QOS, + MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_NATIVE); + if (err) + goto out; + + for (i = 0; i < MLX4_NUM_UP; i++) { + out_param[i].bw_share = be32_to_cpu(ctx->qos_p_up[i].bw_share); + out_param[i].max_avg_bw = + be32_to_cpu(ctx->qos_p_up[i].max_avg_bw); + out_param[i].enable = + !!(be32_to_cpu(ctx->qos_p_up[i].enable) & 31); + } + +out: + mlx4_free_cmd_mailbox(dev, mailbox); + + return err; +} +EXPORT_SYMBOL(mlx4_SET_VPORT_QOS_get); + +int mlx4_SET_VPORT_QOS_set(struct mlx4_dev *dev, u8 port, u8 vport, + struct mlx4_vport_qos_param *in_param) +{ + int i; + int err; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_set_vport_context *ctx; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + ctx = mailbox->buf; + + for (i = 0; i < MLX4_NUM_UP; i++) { + ctx->qos_p_up[i].bw_share = cpu_to_be32(in_param[i].bw_share); + ctx->qos_p_up[i].max_avg_bw = + cpu_to_be32(in_param[i].max_avg_bw); + ctx->qos_p_up[i].enable = + cpu_to_be32(in_param[i].enable << 31); + } + + err = mlx4_cmd(dev, mailbox->dma, (vport << 8) | port, + MLX4_SET_VPORT_QOS_SET, + MLX4_CMD_SET_VPORT_QOS, + MLX4_CMD_TIME_CLASS_A, + MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_SET_VPORT_QOS_set); diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h index be79951c44ef..b3fffafff062 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h @@ -41,6 +41,12 @@ #define MLX4_NUM_UP 8 #define MLX4_NUM_TC 8 +struct mlx4_vport_qos_param { + u32 bw_share; + u32 max_avg_bw; + u8 enable; +}; + /** * mlx4_SET_PORT_PRIO2TC - This routine maps user priorities to traffic * classes of a given port and device. @@ -100,4 +106,34 @@ int mlx4_ALLOCATE_VPP_get(struct mlx4_dev *dev, u8 port, **/ int mlx4_ALLOCATE_VPP_set(struct mlx4_dev *dev, u8 port, u8 *vpp_p_up); +/** + * mlx4_SET_VPORT_QOS_get - Query QoS proporties of a Vport. + * Each priority allowed for the Vport is assigned with a share of the BW, + * and a BW limitation. This commands query the current QoS values. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @vport: Vport id. + * @out_param: Array of mlx4_vport_qos_param that will contain the values. + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_SET_VPORT_QOS_get(struct mlx4_dev *dev, u8 port, u8 vport, + struct mlx4_vport_qos_param *out_param); + +/** + * mlx4_SET_VPORT_QOS_set - Set QoS proporties of a Vport. + * QoS parameters can be modified at any time, but must be initialized + * before any QP is associated with the VPort. + * + * @dev: mlx4_dev. + * @port: Physical port number. + * @vport: Vport id. + * @out_param: Array of mlx4_vport_qos_param which holds the requested values. + * + * Returns 0 on success or a negative mlx4_core errno code. + **/ +int mlx4_SET_VPORT_QOS_set(struct mlx4_dev *dev, u8 port, u8 vport, + struct mlx4_vport_qos_param *in_param); + #endif /* MLX4_FW_QOS_H */ diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h index e0f88a098fb8..88326ed0033a 100644 --- a/include/linux/mlx4/cmd.h +++ b/include/linux/mlx4/cmd.h @@ -69,6 +69,7 @@ enum { MLX4_CMD_SET_ICM_SIZE = 0xffd, MLX4_CMD_ACCESS_REG = 0x3b, MLX4_CMD_ALLOCATE_VPP = 0x80, + MLX4_CMD_SET_VPORT_QOS = 0x81, /*master notify fw on finish for slave's flr*/ MLX4_CMD_INFORM_FLR_DONE = 0x5b, -- cgit v1.2.3 From d019fcb2244519d453694bfce868f3e717bfcebb Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:13 +0300 Subject: net/mlx4: Query device for QoS per VF support Checks in QUERY_DEV_CAP if the granular QoS per VF feature is supported by the device. Disabled for guests. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/fw.c | 10 +++++++++- include/linux/mlx4/device.h | 3 ++- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index 412e019e8b2e..94e857e39063 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -147,7 +147,8 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags) [21] = "Port Remap support", [22] = "QCN support", [23] = "QP rate limiting support", - [24] = "Ethernet Flow control statistics support" + [24] = "Ethernet Flow control statistics support", + [25] = "Granular QoS per VF support", }; int i; @@ -871,6 +872,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) MLX4_GET(size, outbox, QUERY_DEV_CAP_MAX_DESC_SZ_RQ_OFFSET); dev_cap->max_rq_desc_sz = size; MLX4_GET(field, outbox, QUERY_DEV_CAP_CQ_EQ_CACHE_LINE_STRIDE); + if (field & (1 << 4)) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_QOS_VPP; if (field & (1 << 5)) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_ETH_PROT_CTRL; if (field & (1 << 6)) @@ -1199,6 +1202,11 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave, field16 = 0; MLX4_PUT(outbox->buf, field16, QUERY_DEV_CAP_QP_RATE_LIMIT_NUM_OFFSET); + /* turn off QoS per VF support for guests */ + MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_CQ_EQ_CACHE_LINE_STRIDE); + field &= 0xef; + MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_CQ_EQ_CACHE_LINE_STRIDE); + return 0; } diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index b676d0c0111a..c37208f7869f 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -205,7 +205,8 @@ enum { MLX4_DEV_CAP_FLAG2_PORT_REMAP = 1LL << 21, MLX4_DEV_CAP_FLAG2_QCN = 1LL << 22, MLX4_DEV_CAP_FLAG2_QP_RATE_LIMIT = 1LL << 23, - MLX4_DEV_CAP_FLAG2_FLOWSTATS_EN = 1LL << 24 + MLX4_DEV_CAP_FLAG2_FLOWSTATS_EN = 1LL << 24, + MLX4_DEV_CAP_FLAG2_QOS_VPP = 1LL << 25, }; enum { -- cgit v1.2.3 From 08068cd5683f11e4505aa9c8cc6ed5942f8ad299 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:15 +0300 Subject: net/mlx4: Added qos_vport QP configuration in VST mode Granular QoS per VF feature introduce a new QP field, qos_vport. PF administrator can connect VF QPs to a certain QoS Vport, to inherit its proporties. Connecting QPs to the default QoS Vport (defined as 0) is always allowed, even when there are no allocated VPPs. At this point, only the default vport is connected to QPs. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/cmd.c | 9 ++++++++- drivers/net/ethernet/mellanox/mlx4/fw_qos.h | 3 +++ drivers/net/ethernet/mellanox/mlx4/mlx4.h | 2 ++ drivers/net/ethernet/mellanox/mlx4/qp.c | 5 +++++ drivers/net/ethernet/mellanox/mlx4/resource_tracker.c | 6 ++++++ include/linux/mlx4/qp.h | 8 ++++++-- 6 files changed, 30 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 7d3c64275b75..4c0c6c7e63e7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -48,6 +48,7 @@ #include "mlx4.h" #include "fw.h" +#include "fw_qos.h" #define CMD_POLL_TOKEN 0xffff #define INBOX_MASK 0xffffffffffffff00ULL @@ -1808,7 +1809,8 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv, if (vp_oper->state.default_vlan == vp_admin->default_vlan && vp_oper->state.default_qos == vp_admin->default_qos && - vp_oper->state.link_state == vp_admin->link_state) + vp_oper->state.link_state == vp_admin->link_state && + vp_oper->state.qos_vport == vp_admin->qos_vport) return 0; if (!(priv->mfunc.master.slave_state[slave].active && @@ -1866,6 +1868,7 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv, vp_oper->state.default_vlan = vp_admin->default_vlan; vp_oper->state.default_qos = vp_admin->default_qos; vp_oper->state.link_state = vp_admin->link_state; + vp_oper->state.qos_vport = vp_admin->qos_vport; if (vp_admin->link_state == IFLA_VF_LINK_STATE_DISABLE) work->flags |= MLX4_VF_IMMED_VLAN_FLAG_LINK_DISABLE; @@ -1874,6 +1877,7 @@ static int mlx4_master_immediate_activate_vlan_qos(struct mlx4_priv *priv, work->port = port; work->slave = slave; work->qos = vp_oper->state.default_qos; + work->qos_vport = vp_oper->state.qos_vport; work->vlan_id = vp_oper->state.default_vlan; work->vlan_ix = vp_oper->vlan_idx; work->priv = priv; @@ -2339,6 +2343,9 @@ int mlx4_multi_func_init(struct mlx4_dev *dev) INIT_LIST_HEAD(&s_state->mcast_filters[port]); admin_vport->default_vlan = MLX4_VGT; oper_vport->default_vlan = MLX4_VGT; + admin_vport->qos_vport = + MLX4_VPP_DEFAULT_VPORT; + oper_vport->qos_vport = MLX4_VPP_DEFAULT_VPORT; vf_oper->vport[port].vlan_idx = NO_INDX; vf_oper->vport[port].mac_idx = NO_INDX; } diff --git a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h index 4a5320f79094..ac1f331878e6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw_qos.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw_qos.h @@ -44,6 +44,9 @@ /* Default supported priorities for VPP allocation */ #define MLX4_DEFAULT_QOS_PRIO (0) +/* Derived from FW feature definition, 0 is the default vport fo all QPs */ +#define MLX4_VPP_DEFAULT_VPORT (0) + struct mlx4_vport_qos_param { u32 bw_share; u32 max_avg_bw; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 1add698dba5b..32469c64d580 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -498,6 +498,7 @@ struct mlx4_vport_state { u32 tx_rate; bool spoofchk; u32 link_state; + u8 qos_vport; }; struct mlx4_vf_admin_state { @@ -636,6 +637,7 @@ struct mlx4_vf_immed_vlan_work { int orig_vlan_ix; u8 port; u8 qos; + u8 qos_vport; u16 vlan_id; u16 orig_vlan_id; }; diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index 69e4462e4ee4..b75214a80d0e 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -447,6 +447,11 @@ int mlx4_update_qp(struct mlx4_dev *dev, u32 qpn, cmd->qp_context.rate_limit_params = cpu_to_be16((params->rate_unit << 14) | params->rate_val); } + if (attr & MLX4_UPDATE_QP_QOS_VPORT) { + qp_mask |= 1ULL << MLX4_UPD_QP_MASK_QOS_VPP; + cmd->qp_context.qos_vport = params->qos_vport; + } + cmd->primary_addr_path_mask = cpu_to_be64(pri_addr_path_mask); cmd->qp_mask = cpu_to_be64(qp_mask); diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index fd2520aa8a8c..c7f28bf4b8e2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -765,6 +765,7 @@ static int update_vport_qp_param(struct mlx4_dev *dev, qpc->pri_path.feup |= MLX4_FEUP_FORCE_ETH_UP | MLX4_FVL_FORCE_ETH_VLAN; qpc->pri_path.sched_queue &= 0xC7; qpc->pri_path.sched_queue |= (vp_oper->state.default_qos) << 3; + qpc->qos_vport = vp_oper->state.qos_vport; } if (vp_oper->state.spoofchk) { qpc->pri_path.feup |= MLX4_FSM_FORCE_ETH_SRC_MAC; @@ -4917,6 +4918,11 @@ void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work) qp->sched_queue & 0xC7; upd_context->qp_context.pri_path.sched_queue |= ((work->qos & 0x7) << 3); + upd_context->qp_mask |= + cpu_to_be64(1ULL << + MLX4_UPD_QP_MASK_QOS_VPP); + upd_context->qp_context.qos_vport = + work->qos_vport; } err = mlx4_cmd(dev, mailbox->dma, diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h index 1023ebe035b7..6fed539e5456 100644 --- a/include/linux/mlx4/qp.h +++ b/include/linux/mlx4/qp.h @@ -209,7 +209,8 @@ struct mlx4_qp_context { __be16 sq_wqe_counter; u32 reserved3; __be16 rate_limit_params; - __be16 reserved4; + u8 reserved4; + u8 qos_vport; __be32 param3; __be32 nummmcpeers_basemkey; u8 log_page_size; @@ -231,6 +232,7 @@ struct mlx4_update_qp_context { enum { MLX4_UPD_QP_MASK_PM_STATE = 32, MLX4_UPD_QP_MASK_VSD = 33, + MLX4_UPD_QP_MASK_QOS_VPP = 34, MLX4_UPD_QP_MASK_RATE_LIMIT = 35, }; @@ -432,7 +434,8 @@ enum mlx4_update_qp_attr { MLX4_UPDATE_QP_SMAC = 1 << 0, MLX4_UPDATE_QP_VSD = 1 << 1, MLX4_UPDATE_QP_RATE_LIMIT = 1 << 2, - MLX4_UPDATE_QP_SUPPORTED_ATTRS = (1 << 3) - 1 + MLX4_UPDATE_QP_QOS_VPORT = 1 << 3, + MLX4_UPDATE_QP_SUPPORTED_ATTRS = (1 << 4) - 1 }; enum mlx4_update_qp_params_flags { @@ -441,6 +444,7 @@ enum mlx4_update_qp_params_flags { struct mlx4_update_qp_params { u8 smac_index; + u8 qos_vport; u32 flags; u16 rate_unit; u16 rate_val; -- cgit v1.2.3 From cda373f4849d5dd6fedceb4aeba35682a0e1a833 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:16 +0300 Subject: net/mlx4_en: Enable TX rate limit per VF Support granular QoS per VF, by implementing the ndo_set_vf_rate. Enforce a rate limit per VF when called, and enabled only for VFs in VST mode with user priority supported by the device. We don't enforce VFs to be in VST mode at the moment of configuration, but rather save the given rate limit and enforce it when the VF is moved to VST with user priority which is supported (currently 0). VST<->VGT or VST qos value state changes are disallowed when a rate limit is configured. Minimum BW share is not supported yet. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/cmd.c | 177 ++++++++++++++++++++++++- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 11 ++ include/linux/mlx4/cmd.h | 2 + 3 files changed, 189 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 4c0c6c7e63e7..bc6e4c9b6bb5 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -2781,6 +2781,103 @@ static int mlx4_slaves_closest_port(struct mlx4_dev *dev, int slave, int port) return port; } +static int mlx4_set_vport_qos(struct mlx4_priv *priv, int slave, int port, + int max_tx_rate) +{ + int i; + int err; + struct mlx4_qos_manager *port_qos; + struct mlx4_dev *dev = &priv->dev; + struct mlx4_vport_qos_param vpp_qos[MLX4_NUM_UP]; + + port_qos = &priv->mfunc.master.qos_ctl[port]; + memset(vpp_qos, 0, sizeof(struct mlx4_vport_qos_param) * MLX4_NUM_UP); + + if (slave > port_qos->num_of_qos_vfs) { + mlx4_info(dev, "No availible VPP resources for this VF\n"); + return -EINVAL; + } + + /* Query for default QoS values from Vport 0 is needed */ + err = mlx4_SET_VPORT_QOS_get(dev, port, 0, vpp_qos); + if (err) { + mlx4_info(dev, "Failed to query Vport 0 QoS values\n"); + return err; + } + + for (i = 0; i < MLX4_NUM_UP; i++) { + if (test_bit(i, port_qos->priority_bm) && max_tx_rate) { + vpp_qos[i].max_avg_bw = max_tx_rate; + vpp_qos[i].enable = 1; + } else { + /* if user supplied tx_rate == 0, meaning no rate limit + * configuration is required. so we are leaving the + * value of max_avg_bw as queried from Vport 0. + */ + vpp_qos[i].enable = 0; + } + } + + err = mlx4_SET_VPORT_QOS_set(dev, port, slave, vpp_qos); + if (err) { + mlx4_info(dev, "Failed to set Vport %d QoS values\n", slave); + return err; + } + + return 0; +} + +static bool mlx4_is_vf_vst_and_prio_qos(struct mlx4_dev *dev, int port, + struct mlx4_vport_state *vf_admin) +{ + struct mlx4_qos_manager *info; + struct mlx4_priv *priv = mlx4_priv(dev); + + if (!mlx4_is_master(dev) || + !(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_QOS_VPP)) + return false; + + info = &priv->mfunc.master.qos_ctl[port]; + + if (vf_admin->default_vlan != MLX4_VGT && + test_bit(vf_admin->default_qos, info->priority_bm)) + return true; + + return false; +} + +static bool mlx4_valid_vf_state_change(struct mlx4_dev *dev, int port, + struct mlx4_vport_state *vf_admin, + int vlan, int qos) +{ + struct mlx4_vport_state dummy_admin = {0}; + + if (!mlx4_is_vf_vst_and_prio_qos(dev, port, vf_admin) || + !vf_admin->tx_rate) + return true; + + dummy_admin.default_qos = qos; + dummy_admin.default_vlan = vlan; + + /* VF wants to move to other VST state which is valid with current + * rate limit. Either differnt default vlan in VST or other + * supported QoS priority. Otherwise we don't allow this change when + * the TX rate is still configured. + */ + if (mlx4_is_vf_vst_and_prio_qos(dev, port, &dummy_admin)) + return true; + + mlx4_info(dev, "Cannot change VF state to %s while rate is set\n", + (vlan == MLX4_VGT) ? "VGT" : "VST"); + + if (vlan != MLX4_VGT) + mlx4_info(dev, "VST priority %d not supported for QoS\n", qos); + + mlx4_info(dev, "Please set rate to 0 prior to this VF state change\n"); + + return false; +} + int mlx4_set_vf_mac(struct mlx4_dev *dev, int port, int vf, u64 mac) { struct mlx4_priv *priv = mlx4_priv(dev); @@ -2824,12 +2921,22 @@ int mlx4_set_vf_vlan(struct mlx4_dev *dev, int port, int vf, u16 vlan, u8 qos) port = mlx4_slaves_closest_port(dev, slave, port); vf_admin = &priv->mfunc.master.vf_admin[slave].vport[port]; + if (!mlx4_valid_vf_state_change(dev, port, vf_admin, vlan, qos)) + return -EPERM; + if ((0 == vlan) && (0 == qos)) vf_admin->default_vlan = MLX4_VGT; else vf_admin->default_vlan = vlan; vf_admin->default_qos = qos; + /* If rate was configured prior to VST, we saved the configured rate + * in vf_admin->rate and now, if priority supported we enforce the QoS + */ + if (mlx4_is_vf_vst_and_prio_qos(dev, port, vf_admin) && + vf_admin->tx_rate) + vf_admin->qos_vport = slave; + if (mlx4_master_immediate_activate_vlan_qos(priv, slave, port)) mlx4_info(dev, "updating vf %d port %d config will take effect on next VF restart\n", @@ -2838,6 +2945,69 @@ int mlx4_set_vf_vlan(struct mlx4_dev *dev, int port, int vf, u16 vlan, u8 qos) } EXPORT_SYMBOL_GPL(mlx4_set_vf_vlan); +int mlx4_set_vf_rate(struct mlx4_dev *dev, int port, int vf, int min_tx_rate, + int max_tx_rate) +{ + int err; + int slave; + struct mlx4_vport_state *vf_admin; + struct mlx4_priv *priv = mlx4_priv(dev); + + if (!mlx4_is_master(dev) || + !(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_QOS_VPP)) + return -EPROTONOSUPPORT; + + if (min_tx_rate) { + mlx4_info(dev, "Minimum BW share not supported\n"); + return -EPROTONOSUPPORT; + } + + slave = mlx4_get_slave_indx(dev, vf); + if (slave < 0) + return -EINVAL; + + port = mlx4_slaves_closest_port(dev, slave, port); + vf_admin = &priv->mfunc.master.vf_admin[slave].vport[port]; + + err = mlx4_set_vport_qos(priv, slave, port, max_tx_rate); + if (err) { + mlx4_info(dev, "vf %d failed to set rate %d\n", vf, + max_tx_rate); + return err; + } + + vf_admin->tx_rate = max_tx_rate; + /* if VF is not in supported mode (VST with supported prio), + * we do not change vport configuration for its QPs, but save + * the rate, so it will be enforced when it moves to supported + * mode next time. + */ + if (!mlx4_is_vf_vst_and_prio_qos(dev, port, vf_admin)) { + mlx4_info(dev, + "rate set for VF %d when not in valid state\n", vf); + + if (vf_admin->default_vlan != MLX4_VGT) + mlx4_info(dev, "VST priority not supported by QoS\n"); + else + mlx4_info(dev, "VF in VGT mode (needed VST)\n"); + + mlx4_info(dev, + "rate %d take affect when VF moves to valid state\n", + max_tx_rate); + return 0; + } + + /* If user sets rate 0 assigning default vport for its QPs */ + vf_admin->qos_vport = max_tx_rate ? slave : MLX4_VPP_DEFAULT_VPORT; + + if (priv->mfunc.master.slave_state[slave].active && + dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_UPDATE_QP) + mlx4_master_immediate_activate_vlan_qos(priv, slave, port); + + return 0; +} +EXPORT_SYMBOL_GPL(mlx4_set_vf_rate); + /* mlx4_get_slave_default_vlan - * return true if VST ( default vlan) * if VST, will return vlan & qos (if not NULL) @@ -2911,7 +3081,12 @@ int mlx4_get_vf_config(struct mlx4_dev *dev, int port, int vf, struct ifla_vf_in ivf->vlan = s_info->default_vlan; ivf->qos = s_info->default_qos; - ivf->max_tx_rate = s_info->tx_rate; + + if (mlx4_is_vf_vst_and_prio_qos(dev, port, s_info)) + ivf->max_tx_rate = s_info->tx_rate; + else + ivf->max_tx_rate = 0; + ivf->min_tx_rate = 0; ivf->spoofchk = s_info->spoofchk; ivf->linkstate = s_info->link_state; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index d73ab1ba5a78..2a5df0aa2228 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -2240,6 +2240,16 @@ static int mlx4_en_set_vf_vlan(struct net_device *dev, int vf, u16 vlan, u8 qos) return mlx4_set_vf_vlan(mdev->dev, en_priv->port, vf, vlan, qos); } +static int mlx4_en_set_vf_rate(struct net_device *dev, int vf, int min_tx_rate, + int max_tx_rate) +{ + struct mlx4_en_priv *en_priv = netdev_priv(dev); + struct mlx4_en_dev *mdev = en_priv->mdev; + + return mlx4_set_vf_rate(mdev->dev, en_priv->port, vf, min_tx_rate, + max_tx_rate); +} + static int mlx4_en_set_vf_spoofchk(struct net_device *dev, int vf, bool setting) { struct mlx4_en_priv *en_priv = netdev_priv(dev); @@ -2458,6 +2468,7 @@ static const struct net_device_ops mlx4_netdev_ops_master = { .ndo_vlan_rx_kill_vid = mlx4_en_vlan_rx_kill_vid, .ndo_set_vf_mac = mlx4_en_set_vf_mac, .ndo_set_vf_vlan = mlx4_en_set_vf_vlan, + .ndo_set_vf_rate = mlx4_en_set_vf_rate, .ndo_set_vf_spoofchk = mlx4_en_set_vf_spoofchk, .ndo_set_vf_link_state = mlx4_en_set_vf_link_state, .ndo_get_vf_config = mlx4_en_get_vf_config, diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h index 88326ed0033a..a788839f54bb 100644 --- a/include/linux/mlx4/cmd.h +++ b/include/linux/mlx4/cmd.h @@ -296,6 +296,8 @@ void mlx4_free_cmd_mailbox(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbo u32 mlx4_comm_get_version(void); int mlx4_set_vf_mac(struct mlx4_dev *dev, int port, int vf, u64 mac); int mlx4_set_vf_vlan(struct mlx4_dev *dev, int port, int vf, u16 vlan, u8 qos); +int mlx4_set_vf_rate(struct mlx4_dev *dev, int port, int vf, int min_tx_rate, + int max_tx_rate); int mlx4_set_vf_spoofchk(struct mlx4_dev *dev, int port, int vf, bool setting); int mlx4_get_vf_config(struct mlx4_dev *dev, int port, int vf, struct ifla_vf_info *ivf); int mlx4_set_vf_link_state(struct mlx4_dev *dev, int port, int vf, int link_state); -- cgit v1.2.3 From 3742cc65512cd4897a63dce94104f9a6e74997a0 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:17 +0300 Subject: net/mlx4: Warn users of depracated QoS Firmware A new capability bit was introduced in the past to to differ devices using the QoS ETS feature. The old was deprecated since then. If driver sees device which set only the old capabilty, it will print warning to user suggesting to upgrade the FW. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 +- drivers/net/ethernet/mellanox/mlx4/fw.c | 7 +++++-- drivers/net/ethernet/mellanox/mlx4/main.c | 8 ++++++++ include/linux/mlx4/device.h | 1 + 4 files changed, 15 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 2a5df0aa2228..be9e07a72193 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -2814,7 +2814,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, priv->msg_enable = MLX4_EN_MSG_LEVEL; #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { - if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_SET_ETH_SCHED) { + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) { dev->dcbnl_ops = &mlx4_en_dcbnl_ops; } else { en_info(priv, "enabling only PFC DCB ops\n"); diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index 94e857e39063..16bbd6cdff6c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -149,6 +149,7 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags) [23] = "QP rate limiting support", [24] = "Ethernet Flow control statistics support", [25] = "Granular QoS per VF support", + [26] = "Port ETS Scheduler support", }; int i; @@ -900,6 +901,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) MLX4_GET(field, outbox, QUERY_DEV_CAP_VXLAN); if (field & 1<<3) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS; + if (field & (1 << 5)) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_ETS_CFG; MLX4_GET(dev_cap->max_icm_sz, outbox, QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET); if (dev_cap->flags & MLX4_DEV_CAP_FLAG_COUNTERS) @@ -1157,9 +1160,9 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave, field &= 0x7f; MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_CQ_TS_SUPPORT_OFFSET); - /* For guests, disable vxlan tunneling */ + /* For guests, disable vxlan tunneling and QoS support */ MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_VXLAN); - field &= 0xf7; + field &= 0xd7; MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_VXLAN); /* For guests, report Blueflame disabled */ diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index ee0a67069b57..3becad37aec8 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -528,6 +528,14 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) dev->caps.alloc_res_qp_mask = (dev->caps.bf_reg_size ? MLX4_RESERVE_ETH_BF_QP : 0) | MLX4_RESERVE_A0_QP; + + if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) && + dev->caps.flags & MLX4_DEV_CAP_FLAG_SET_ETH_SCHED) { + mlx4_warn(dev, "Old device ETS support detected\n"); + mlx4_warn(dev, "Consider upgrading device FW.\n"); + dev->caps.flags2 |= MLX4_DEV_CAP_FLAG2_ETS_CFG; + } + } else { dev->caps.alloc_res_qp_mask = 0; } diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index c37208f7869f..15f46767342e 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -207,6 +207,7 @@ enum { MLX4_DEV_CAP_FLAG2_QP_RATE_LIMIT = 1LL << 23, MLX4_DEV_CAP_FLAG2_FLOWSTATS_EN = 1LL << 24, MLX4_DEV_CAP_FLAG2_QOS_VPP = 1LL << 25, + MLX4_DEV_CAP_FLAG2_ETS_CFG = 1LL << 26, }; enum { -- cgit v1.2.3 From a130b59057320192b4b00ed0ba4bc8a38f66f289 Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:19 +0300 Subject: net/mlx4: Add SET_PORT opcode modifiers enumeration The calls to SET_PORT used hard-code numbers, when supplying command's opcode modifiers, fix that to use well defined constants. Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx4/main.c | 11 +++++----- drivers/net/ethernet/mellanox/mlx4/cmd.c | 3 ++- drivers/net/ethernet/mellanox/mlx4/port.c | 36 ++++++++++++++++++------------- include/linux/mlx4/cmd.h | 8 ++++++- 4 files changed, 36 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index b972c0b41799..976bea794b5f 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -587,8 +587,9 @@ static int mlx4_ib_SET_PORT(struct mlx4_ib_dev *dev, u8 port, int reset_qkey_vio ((__be32 *) mailbox->buf)[1] = cpu_to_be32(cap_mask); } - err = mlx4_cmd(dev->dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); + err = mlx4_cmd(dev->dev, mailbox->dma, port, MLX4_SET_PORT_IB_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_WRAPPED); mlx4_free_cmd_mailbox(dev->dev, mailbox); return err; @@ -1525,8 +1526,8 @@ static void update_gids_task(struct work_struct *work) memcpy(gids, gw->gids, sizeof gw->gids); err = mlx4_cmd(dev, mailbox->dma, MLX4_SET_PORT_GID_TABLE << 8 | gw->port, - 1, MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, - MLX4_CMD_WRAPPED); + MLX4_SET_PORT_ETH_OPCODE, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); if (err) pr_warn("set port command failed\n"); else @@ -1564,7 +1565,7 @@ static void reset_gids_task(struct work_struct *work) IB_LINK_LAYER_ETHERNET) { err = mlx4_cmd(dev, mailbox->dma, MLX4_SET_PORT_GID_TABLE << 8 | gw->port, - 1, MLX4_CMD_SET_PORT, + MLX4_SET_PORT_ETH_OPCODE, MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); if (err) diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index bc6e4c9b6bb5..06993ea9e6ba 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -726,7 +726,8 @@ static int mlx4_cmd_wait(struct mlx4_dev *dev, u64 in_param, u64 *out_param, * specific command/input_mod/opcode_mod/fw-status to be debug. */ if (op == MLX4_CMD_SET_PORT && in_modifier == 1 && - op_modifier == 0 && context->fw_status == CMD_STAT_BAD_SIZE) + op_modifier == MLX4_SET_PORT_IB_OPCODE && + context->fw_status == CMD_STAT_BAD_SIZE) mlx4_dbg(dev, "command 0x%x failed: fw status = 0x%x\n", op, context->fw_status); else diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 6a53d42db52f..ca4488ed561c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -123,8 +123,9 @@ static int mlx4_set_port_mac_table(struct mlx4_dev *dev, u8 port, in_mod = MLX4_SET_PORT_MAC_TABLE << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_NATIVE); mlx4_free_cmd_mailbox(dev, mailbox); return err; @@ -337,8 +338,9 @@ static int mlx4_set_port_vlan_table(struct mlx4_dev *dev, u8 port, memcpy(mailbox->buf, entries, MLX4_VLAN_TABLE_SIZE); in_mod = MLX4_SET_PORT_VLAN_TABLE << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_NATIVE); mlx4_free_cmd_mailbox(dev, mailbox); @@ -625,9 +627,9 @@ static int mlx4_reset_roce_port_gids(struct mlx4_dev *dev, int slave, MLX4_ROCE_GID_ENTRY_SIZE); err = mlx4_cmd(dev, mailbox->dma, - ((u32)port) | (MLX4_SET_PORT_GID_TABLE << 8), 1, - MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, - MLX4_CMD_NATIVE); + ((u32)port) | (MLX4_SET_PORT_GID_TABLE << 8), + MLX4_SET_PORT_ETH_OPCODE, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); mutex_unlock(&(priv->port[port].gid_table.mutex)); return err; } @@ -941,8 +943,9 @@ int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port, int pkey_tbl_sz) (pkey_tbl_flag << MLX4_CHANGE_PORT_PKEY_TBL_SZ) | (dev->caps.port_ib_mtu[port] << MLX4_SET_PORT_MTU_CAP) | (vl_cap << MLX4_SET_PORT_VL_CAP)); - err = mlx4_cmd(dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); + err = mlx4_cmd(dev, mailbox->dma, port, + MLX4_SET_PORT_IB_OPCODE, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); if (err != -ENOMEM) break; } @@ -971,8 +974,9 @@ int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu, context->pfcrx = pfcrx; in_mod = MLX4_SET_PORT_GENERAL << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); + err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_WRAPPED); mlx4_free_cmd_mailbox(dev, mailbox); return err; @@ -1008,8 +1012,9 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, context->vlan_miss = MLX4_VLAN_MISS_IDX; in_mod = MLX4_SET_PORT_RQP_CALC << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED); + err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_WRAPPED); mlx4_free_cmd_mailbox(dev, mailbox); return err; @@ -1050,8 +1055,9 @@ int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable) context->steering = steering; in_mod = MLX4_SET_PORT_VXLAN << 8 | port; - err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, - MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + err = mlx4_cmd(dev, mailbox->dma, in_mod, MLX4_SET_PORT_ETH_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_NATIVE); mlx4_free_cmd_mailbox(dev, mailbox); return err; diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h index a788839f54bb..62c4d4def474 100644 --- a/include/linux/mlx4/cmd.h +++ b/include/linux/mlx4/cmd.h @@ -188,7 +188,13 @@ enum { }; enum { - /* set port opcode modifiers */ + /* Set port opcode modifiers */ + MLX4_SET_PORT_IB_OPCODE = 0x0, + MLX4_SET_PORT_ETH_OPCODE = 0x1, +}; + +enum { + /* Set port Ethernet input modifiers */ MLX4_SET_PORT_GENERAL = 0x0, MLX4_SET_PORT_RQP_CALC = 0x1, MLX4_SET_PORT_MAC_TABLE = 0x2, -- cgit v1.2.3 From 51af33cfed248dc8f36fa82df06b85e10038a01e Mon Sep 17 00:00:00 2001 From: Ido Shamay Date: Thu, 2 Apr 2015 16:31:20 +0300 Subject: net/mlx4_en: Add interface identify support Add support for the interface ethtool identify feature. Make the physical port LED to blink with green and yellow colors. The device handles the LED blink by itself (synchrous use of set_phys_id), by returning 0 to ETHTOOL_ID_ACTIVE command. Signed-off-by: Eyal Grossman Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 27 +++++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx4/fw.c | 10 +++++++++ drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 + drivers/net/ethernet/mellanox/mlx4/port.c | 26 ++++++++++++++++++++++++ include/linux/mlx4/cmd.h | 1 + include/linux/mlx4/device.h | 2 ++ 6 files changed, 67 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index eba969b08dd1..3f44e2bbb982 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1939,6 +1939,32 @@ static int mlx4_en_get_module_eeprom(struct net_device *dev, return 0; } +static int mlx4_en_set_phys_id(struct net_device *dev, + enum ethtool_phys_id_state state) +{ + int err; + u16 beacon_duration; + struct mlx4_en_priv *priv = netdev_priv(dev); + struct mlx4_en_dev *mdev = priv->mdev; + + if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PORT_BEACON)) + return -EOPNOTSUPP; + + switch (state) { + case ETHTOOL_ID_ACTIVE: + beacon_duration = PORT_BEACON_MAX_LIMIT; + break; + case ETHTOOL_ID_INACTIVE: + beacon_duration = 0; + break; + default: + return -EOPNOTSUPP; + } + + err = mlx4_SET_PORT_BEACON(mdev->dev, priv->port, beacon_duration); + return err; +} + const struct ethtool_ops mlx4_en_ethtool_ops = { .get_drvinfo = mlx4_en_get_drvinfo, .get_settings = mlx4_en_get_settings, @@ -1948,6 +1974,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = { .get_sset_count = mlx4_en_get_sset_count, .get_ethtool_stats = mlx4_en_get_ethtool_stats, .self_test = mlx4_en_self_test, + .set_phys_id = mlx4_en_set_phys_id, .get_wol = mlx4_en_get_wol, .set_wol = mlx4_en_set_wol, .get_msglevel = mlx4_en_get_msglevel, diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index b93e23bef6c9..6317844b291f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -150,6 +150,7 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags) [24] = "Ethernet Flow control statistics support", [25] = "Granular QoS per VF support", [26] = "Port ETS Scheduler support", + [27] = "Port beacon support", }; int i; @@ -647,6 +648,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) #define QUERY_DEV_CAP_RSS_OFFSET 0x2e #define QUERY_DEV_CAP_MAX_RDMA_OFFSET 0x2f #define QUERY_DEV_CAP_RSZ_SRQ_OFFSET 0x33 +#define QUERY_DEV_CAP_PORT_BEACON_OFFSET 0x34 #define QUERY_DEV_CAP_ACK_DELAY_OFFSET 0x35 #define QUERY_DEV_CAP_MTU_WIDTH_OFFSET 0x36 #define QUERY_DEV_CAP_VL_PORT_OFFSET 0x37 @@ -786,6 +788,9 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) if (field & 0x80) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_FS_EN; dev_cap->fs_log_max_ucast_qp_range_size = field & 0x1f; + MLX4_GET(field, outbox, QUERY_DEV_CAP_PORT_BEACON_OFFSET); + if (field & 0x80) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_PORT_BEACON; MLX4_GET(field, outbox, QUERY_DEV_CAP_FLOW_STEERING_IPOIB_OFFSET); if (field & 0x80) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_DMFS_IPOIB; @@ -1165,6 +1170,11 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave, field &= 0xd7; MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_VXLAN); + /* For guests, disable port BEACON */ + MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_PORT_BEACON_OFFSET); + field &= 0x7f; + MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_PORT_BEACON_OFFSET); + /* For guests, report Blueflame disabled */ MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_BF_OFFSET); field &= 0x7f; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 67eeea244eff..9de30216b146 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -482,6 +482,7 @@ enum { MLX4_EN_FLAG_RX_CSUM_NON_TCP_UDP = (1 << 5), }; +#define PORT_BEACON_MAX_LIMIT (65535) #define MLX4_EN_MAC_HASH_SIZE (1 << BITS_PER_BYTE) #define MLX4_EN_MAC_HASH_IDX 5 diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index ca4488ed561c..e8d716dfd5cb 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -835,6 +835,12 @@ static int mlx4_common_set_port(struct mlx4_dev *dev, int slave, u32 in_mod, MLX4_CMD_NATIVE); } + /* Slaves are not allowed to SET_PORT beacon (LED) blink */ + if (op_mod == MLX4_SET_PORT_BEACON_OPCODE) { + mlx4_warn(dev, "denying SET_PORT Beacon slave:%d\n", slave); + return -EPERM; + } + /* For IB, we only consider: * - The capability mask, which is set to the aggregate of all * slave function capabilities @@ -1064,6 +1070,26 @@ int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable) } EXPORT_SYMBOL(mlx4_SET_PORT_VXLAN); +int mlx4_SET_PORT_BEACON(struct mlx4_dev *dev, u8 port, u16 time) +{ + int err; + struct mlx4_cmd_mailbox *mailbox; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + + *((__be32 *)mailbox->buf) = cpu_to_be32(time); + + err = mlx4_cmd(dev, mailbox->dma, port, MLX4_SET_PORT_BEACON_OPCODE, + MLX4_CMD_SET_PORT, MLX4_CMD_TIME_CLASS_B, + MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_SET_PORT_BEACON); + int mlx4_SET_MCAST_FLTR_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr, struct mlx4_cmd_mailbox *inbox, diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h index 62c4d4def474..f62e7cf227c6 100644 --- a/include/linux/mlx4/cmd.h +++ b/include/linux/mlx4/cmd.h @@ -191,6 +191,7 @@ enum { /* Set port opcode modifiers */ MLX4_SET_PORT_IB_OPCODE = 0x0, MLX4_SET_PORT_ETH_OPCODE = 0x1, + MLX4_SET_PORT_BEACON_OPCODE = 0x4, }; enum { diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 15f46767342e..b761be7c88c8 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -208,6 +208,7 @@ enum { MLX4_DEV_CAP_FLAG2_FLOWSTATS_EN = 1LL << 24, MLX4_DEV_CAP_FLAG2_QOS_VPP = 1LL << 25, MLX4_DEV_CAP_FLAG2_ETS_CFG = 1LL << 26, + MLX4_DEV_CAP_FLAG2_PORT_BEACON = 1LL << 27, }; enum { @@ -1311,6 +1312,7 @@ int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu, u8 pptx, u8 pfctx, u8 pprx, u8 pfcrx); int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, u8 promisc); +int mlx4_SET_PORT_BEACON(struct mlx4_dev *dev, u8 port, u16 time); int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable); int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx); int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx); -- cgit v1.2.3 From 78500b8c03236a18d454c9cc8a24cccca506b200 Mon Sep 17 00:00:00 2001 From: Muhammad Mahajna Date: Thu, 2 Apr 2015 16:31:22 +0300 Subject: net/mlx4_en: Add RX-ALL support Enabled when the device supports KEEP FCS and IGNORE FCS. When the flag is set, pass all received frames up the stack, even ones with invalid FCS, controlled by ethtool. Signed-off-by: Muhammad Mahajna Signed-off-by: Ido Shamay Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 14 +++++++++++ drivers/net/ethernet/mellanox/mlx4/fw.c | 8 ++++++ drivers/net/ethernet/mellanox/mlx4/main.c | 21 ++++++++++++++++ drivers/net/ethernet/mellanox/mlx4/mlx4.h | 6 +++-- drivers/net/ethernet/mellanox/mlx4/port.c | 34 ++++++++++++++++++++++++++ include/linux/mlx4/device.h | 3 +++ 6 files changed, 84 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 333af3405a80..0f1afc085d58 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -2204,6 +2204,17 @@ static int mlx4_en_set_features(struct net_device *netdev, reset = true; } + if (DEV_FEATURE_CHANGED(netdev, features, NETIF_F_RXALL)) { + u8 ignore_fcs_value = (features & NETIF_F_RXALL) ? 1 : 0; + + en_info(priv, "Turn %s RX-ALL\n", + ignore_fcs_value ? "ON" : "OFF"); + ret = mlx4_SET_PORT_fcs_check(priv->mdev->dev, + priv->port, ignore_fcs_value); + if (ret) + return ret; + } + if (DEV_FEATURE_CHANGED(netdev, features, NETIF_F_HW_VLAN_CTAG_RX)) { en_info(priv, "Turn %s RX vlan strip offload\n", (features & NETIF_F_HW_VLAN_CTAG_RX) ? "ON" : "OFF"); @@ -2914,6 +2925,9 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP) dev->hw_features |= NETIF_F_RXFCS; + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_IGNORE_FCS) + dev->hw_features |= NETIF_F_RXALL; + if (mdev->dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED && mdev->dev->caps.dmfs_high_steer_mode != MLX4_STEERING_DMFS_A0_STATIC) diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index 6317844b291f..b9881fc1252f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -151,6 +151,7 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags) [25] = "Granular QoS per VF support", [26] = "Port ETS Scheduler support", [27] = "Port beacon support", + [28] = "RX-ALL support", }; int i; @@ -893,6 +894,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) MLX4_GET(field, outbox, QUERY_DEV_CAP_CONFIG_DEV_OFFSET); if (field & 0x20) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_CONFIG_DEV; + if (field & (1 << 2)) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_IGNORE_FCS; MLX4_GET(dev_cap->reserved_lkey, outbox, QUERY_DEV_CAP_RSVD_LKEY_OFFSET); MLX4_GET(field32, outbox, QUERY_DEV_CAP_ETH_BACKPL_OFFSET); @@ -1220,6 +1223,11 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave, field &= 0xef; MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_CQ_EQ_CACHE_LINE_STRIDE); + /* turn off ignore FCS feature for guests */ + MLX4_GET(field, outbox->buf, QUERY_DEV_CAP_CONFIG_DEV_OFFSET); + field &= 0xfb; + MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_CONFIG_DEV_OFFSET); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 3becad37aec8..acceb75e8c44 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -297,6 +297,25 @@ static int mlx4_dev_port(struct mlx4_dev *dev, int port, return err; } +static inline void mlx4_enable_ignore_fcs(struct mlx4_dev *dev) +{ + if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_IGNORE_FCS)) + return; + + if (mlx4_is_mfunc(dev)) { + mlx4_dbg(dev, "SRIOV mode - Disabling Ignore FCS"); + dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_IGNORE_FCS; + return; + } + + if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP)) { + mlx4_dbg(dev, + "Keep FCS is not supported - Disabling Ignore FCS"); + dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_IGNORE_FCS; + return; + } +} + #define MLX4_A0_STEERING_TABLE_SIZE 256 static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) { @@ -540,6 +559,8 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) dev->caps.alloc_res_qp_mask = 0; } + mlx4_enable_ignore_fcs(dev); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 32469c64d580..f30eeb730a86 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -763,9 +763,11 @@ enum { struct mlx4_set_port_general_context { - u8 reserved[3]; + u16 reserved1; + u8 v_ignore_fcs; u8 flags; - u16 reserved2; + u8 ignore_fcs; + u8 reserved2; __be16 mtu; u8 pptx; u8 pfctx; diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index e8d716dfd5cb..c2b21313dba7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -45,6 +45,14 @@ #define MLX4_VLAN_VALID (1u << 31) #define MLX4_VLAN_MASK 0xfff +#define MLX4_STATS_TRAFFIC_COUNTERS_MASK 0xfULL +#define MLX4_STATS_TRAFFIC_DROPS_MASK 0xc0ULL +#define MLX4_STATS_ERROR_COUNTERS_MASK 0x1ffc30ULL +#define MLX4_STATS_PORT_COUNTERS_MASK 0x1fe00000ULL + +#define MLX4_FLAG_V_IGNORE_FCS_MASK 0x2 +#define MLX4_IGNORE_FCS_MASK 0x1 + void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) { int i; @@ -1027,6 +1035,32 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, } EXPORT_SYMBOL(mlx4_SET_PORT_qpn_calc); +int mlx4_SET_PORT_fcs_check(struct mlx4_dev *dev, u8 port, u8 ignore_fcs_value) +{ + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_set_port_general_context *context; + u32 in_mod; + int err; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + context = mailbox->buf; + context->v_ignore_fcs |= MLX4_FLAG_V_IGNORE_FCS_MASK; + if (ignore_fcs_value) + context->ignore_fcs |= MLX4_IGNORE_FCS_MASK; + else + context->ignore_fcs &= ~MLX4_IGNORE_FCS_MASK; + + in_mod = MLX4_SET_PORT_GENERAL << 8 | port; + err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT, + MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} +EXPORT_SYMBOL(mlx4_SET_PORT_fcs_check); + enum { VXLAN_ENABLE_MODIFY = 1 << 7, VXLAN_STEERING_MODIFY = 1 << 6, diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index b761be7c88c8..f9ce34bec45b 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -209,6 +209,7 @@ enum { MLX4_DEV_CAP_FLAG2_QOS_VPP = 1LL << 25, MLX4_DEV_CAP_FLAG2_ETS_CFG = 1LL << 26, MLX4_DEV_CAP_FLAG2_PORT_BEACON = 1LL << 27, + MLX4_DEV_CAP_FLAG2_IGNORE_FCS = 1LL << 28, }; enum { @@ -1313,6 +1314,8 @@ int mlx4_SET_PORT_general(struct mlx4_dev *dev, u8 port, int mtu, int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, u8 promisc); int mlx4_SET_PORT_BEACON(struct mlx4_dev *dev, u8 port, u16 time); +int mlx4_SET_PORT_fcs_check(struct mlx4_dev *dev, u8 port, + u8 ignore_fcs_value); int mlx4_SET_PORT_VXLAN(struct mlx4_dev *dev, u8 port, u8 steering, int enable); int mlx4_find_cached_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *idx); int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx); -- cgit v1.2.3 From 64599cca51de08cef94bc13a0f98351e5bb01f41 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 2 Apr 2015 17:07:25 +0300 Subject: net/mlx5_core: Use coherent memory for command interface page Use coherent memory for the commands descriptor page. Take measures to make sure the page is aligned to MLX5_ADAPTER_PAGE_SIZE as required by the hardware. Reported-by: Yevgeny Kliteynik Signed-off-by: Eli Cohen Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 72 ++++++++++++++++++--------- include/linux/mlx5/driver.h | 3 ++ 2 files changed, 52 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index a2853057c779..43a73d36b01d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1319,6 +1319,45 @@ ex_err: return err; } +static int alloc_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) +{ + struct device *ddev = &dev->pdev->dev; + + cmd->cmd_alloc_buf = dma_zalloc_coherent(ddev, MLX5_ADAPTER_PAGE_SIZE, + &cmd->alloc_dma, GFP_KERNEL); + if (!cmd->cmd_alloc_buf) + return -ENOMEM; + + /* make sure it is aligned to 4K */ + if (!((uintptr_t)cmd->cmd_alloc_buf & (MLX5_ADAPTER_PAGE_SIZE - 1))) { + cmd->cmd_buf = cmd->cmd_alloc_buf; + cmd->dma = cmd->alloc_dma; + cmd->alloc_size = MLX5_ADAPTER_PAGE_SIZE; + return 0; + } + + dma_free_coherent(ddev, MLX5_ADAPTER_PAGE_SIZE, cmd->cmd_alloc_buf, + cmd->alloc_dma); + cmd->cmd_alloc_buf = dma_zalloc_coherent(ddev, + 2 * MLX5_ADAPTER_PAGE_SIZE - 1, + &cmd->alloc_dma, GFP_KERNEL); + if (!cmd->cmd_alloc_buf) + return -ENOMEM; + + cmd->cmd_buf = PTR_ALIGN(cmd->cmd_alloc_buf, MLX5_ADAPTER_PAGE_SIZE); + cmd->dma = ALIGN(cmd->alloc_dma, MLX5_ADAPTER_PAGE_SIZE); + cmd->alloc_size = 2 * MLX5_ADAPTER_PAGE_SIZE - 1; + return 0; +} + +static void free_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) +{ + struct device *ddev = &dev->pdev->dev; + + dma_free_coherent(ddev, cmd->alloc_size, cmd->cmd_alloc_buf, + cmd->alloc_dma); +} + int mlx5_cmd_init(struct mlx5_core_dev *dev) { int size = sizeof(struct mlx5_cmd_prot_block); @@ -1341,17 +1380,9 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) if (!cmd->pool) return -ENOMEM; - cmd->cmd_buf = (void *)__get_free_pages(GFP_ATOMIC, 0); - if (!cmd->cmd_buf) { - err = -ENOMEM; + err = alloc_cmd_page(dev, cmd); + if (err) goto err_free_pool; - } - cmd->dma = dma_map_single(&dev->pdev->dev, cmd->cmd_buf, PAGE_SIZE, - DMA_BIDIRECTIONAL); - if (dma_mapping_error(&dev->pdev->dev, cmd->dma)) { - err = -ENOMEM; - goto err_free; - } cmd_l = ioread32be(&dev->iseg->cmdq_addr_l_sz) & 0xff; cmd->log_sz = cmd_l >> 4 & 0xf; @@ -1360,13 +1391,13 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) dev_err(&dev->pdev->dev, "firmware reports too many outstanding commands %d\n", 1 << cmd->log_sz); err = -EINVAL; - goto err_map; + goto err_free_page; } if (cmd->log_sz + cmd->log_stride > MLX5_ADAPTER_PAGE_SHIFT) { dev_err(&dev->pdev->dev, "command queue size overflow\n"); err = -EINVAL; - goto err_map; + goto err_free_page; } cmd->checksum_disabled = 1; @@ -1378,7 +1409,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) dev_err(&dev->pdev->dev, "driver does not support command interface version. driver %d, firmware %d\n", CMD_IF_REV, cmd->cmdif_rev); err = -ENOTSUPP; - goto err_map; + goto err_free_page; } spin_lock_init(&cmd->alloc_lock); @@ -1394,7 +1425,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) if (cmd_l & 0xfff) { dev_err(&dev->pdev->dev, "invalid command queue address\n"); err = -ENOMEM; - goto err_map; + goto err_free_page; } iowrite32be(cmd_h, &dev->iseg->cmdq_addr_h); @@ -1410,7 +1441,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) err = create_msg_cache(dev); if (err) { dev_err(&dev->pdev->dev, "failed to create command cache\n"); - goto err_map; + goto err_free_page; } set_wqname(dev); @@ -1435,11 +1466,8 @@ err_wq: err_cache: destroy_msg_cache(dev); -err_map: - dma_unmap_single(&dev->pdev->dev, cmd->dma, PAGE_SIZE, - DMA_BIDIRECTIONAL); -err_free: - free_pages((unsigned long)cmd->cmd_buf, 0); +err_free_page: + free_cmd_page(dev, cmd); err_free_pool: pci_pool_destroy(cmd->pool); @@ -1455,9 +1483,7 @@ void mlx5_cmd_cleanup(struct mlx5_core_dev *dev) clean_debug_files(dev); destroy_workqueue(cmd->wq); destroy_msg_cache(dev); - dma_unmap_single(&dev->pdev->dev, cmd->dma, PAGE_SIZE, - DMA_BIDIRECTIONAL); - free_pages((unsigned long)cmd->cmd_buf, 0); + free_cmd_page(dev, cmd); pci_pool_destroy(cmd->pool); } EXPORT_SYMBOL(mlx5_cmd_cleanup); diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 166d9315fe4b..8d8ca6d9b03b 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -232,6 +232,9 @@ struct mlx5_cmd_stats { }; struct mlx5_cmd { + void *cmd_alloc_buf; + dma_addr_t alloc_dma; + int alloc_size; void *cmd_buf; dma_addr_t dma; u16 cmdif_rev; -- cgit v1.2.3 From 302bdf68fc56a6330bc6b10ce435b4d466417537 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 2 Apr 2015 17:07:29 +0300 Subject: net/mlx5_core: Fix Mellanox copyright note Signed-off-by: Achiad Shochat Signed-off-by: Saeed Mahameed Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/alloc.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/cq.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/debugfs.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/eq.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/fw.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/health.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/mad.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/mcg.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h | 2 +- drivers/net/ethernet/mellanox/mlx5/core/mr.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/pd.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/port.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/qp.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/srq.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/uar.c | 2 +- include/linux/mlx5/cmd.h | 2 +- include/linux/mlx5/cq.h | 2 +- include/linux/mlx5/device.h | 2 +- include/linux/mlx5/doorbell.h | 2 +- include/linux/mlx5/driver.h | 2 +- include/linux/mlx5/mlx5_ifc.h | 2 +- include/linux/mlx5/qp.h | 2 +- include/linux/mlx5/srq.h | 2 +- 26 files changed, 26 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c index 63dd75485204..ac0f7bf4be95 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 697d538163f2..e3273faf4568 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c index 43c5f4809526..eb0cf81f5f45 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c index 4878025e231c..5210d92e6bc7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index da82991239a8..dbf190d9b9ad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 06f9036acd83..4b4cda3bcc5f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index 3e6670c4a7cd..292d76f2a904 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mad.c b/drivers/net/ethernet/mellanox/mlx5/core/mad.c index fd80ecfa7195..ee1b0b965f34 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mad.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mad.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index f29fd5592305..78dfdab1ead4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mcg.c b/drivers/net/ethernet/mellanox/mlx5/core/mcg.c index 44837640bd7c..d79fd85d1dd5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mcg.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index f0c9f9a7a361..a051b906afdf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c index 1a9b75139c2d..1adb300dd850 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index cf8591a0c48d..df2238372ea7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pd.c b/drivers/net/ethernet/mellanox/mlx5/core/pd.c index 790da5c4ca4f..f2d3aee909e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pd.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 72c2d002c3b8..49e90f2612d8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c index 575d853dbe05..dc7dbf7e9d98 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/srq.c b/drivers/net/ethernet/mellanox/mlx5/core/srq.c index 38bce93f8314..f9d25dcd03c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/srq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/srq.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/drivers/net/ethernet/mellanox/mlx5/core/uar.c b/drivers/net/ethernet/mellanox/mlx5/core/uar.c index 06801d6f595e..5a89bb1d678a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/uar.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/uar.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/cmd.h b/include/linux/mlx5/cmd.h index 2826a4b6071e..68cd08f02c2f 100644 --- a/include/linux/mlx5/cmd.h +++ b/include/linux/mlx5/cmd.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/cq.h b/include/linux/mlx5/cq.h index f6b17ac601bd..72ee0d732a26 100644 --- a/include/linux/mlx5/cq.h +++ b/include/linux/mlx5/cq.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 4e5bd813bb9a..abf65c790421 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/doorbell.h b/include/linux/mlx5/doorbell.h index 163a818411e7..afc78a3f4462 100644 --- a/include/linux/mlx5/doorbell.h +++ b/include/linux/mlx5/doorbell.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 8d8ca6d9b03b..8fedd39a9a60 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 5f48b8f592c5..cb3ad17edd1f 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/qp.h b/include/linux/mlx5/qp.h index 61f7a342d1bf..310b5f7fd6ae 100644 --- a/include/linux/mlx5/qp.h +++ b/include/linux/mlx5/qp.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU diff --git a/include/linux/mlx5/srq.h b/include/linux/mlx5/srq.h index e1a363a33663..f43ed054a3e0 100644 --- a/include/linux/mlx5/srq.h +++ b/include/linux/mlx5/srq.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013, Mellanox Technologies inc. All rights reserved. + * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU -- cgit v1.2.3 From 233d05d28ad942929b6b4fbc48aa8dd083c16484 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 2 Apr 2015 17:07:32 +0300 Subject: net/mlx5_core: Move completion eqs from mlx5_ib to mlx5_core Preparation for ethernet driver. These functions will be used in drivers other than mlx5_ib. Signed-off-by: Achiad Shochat Signed-off-by: Saeed Mahameed Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx5/cq.c | 2 +- drivers/infiniband/hw/mlx5/main.c | 104 +------------------------ drivers/infiniband/hw/mlx5/mlx5_ib.h | 3 - drivers/net/ethernet/mellanox/mlx5/core/main.c | 91 ++++++++++++++++++++++ include/linux/mlx5/driver.h | 3 +- 5 files changed, 98 insertions(+), 105 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index 6efd9f73ff60..c94e2651820d 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -780,7 +780,7 @@ struct ib_cq *mlx5_ib_create_cq(struct ib_device *ibdev, int entries, cq->cqe_size = cqe_size; cqb->ctx.cqe_sz_flags = cqe_sz_to_mlx_sz(cqe_size) << 5; cqb->ctx.log_sz_usr_page = cpu_to_be32((ilog2(entries) << 24) | index); - err = mlx5_vector2eqn(dev, vector, &eqn, &irqn); + err = mlx5_vector2eqn(dev->mdev, vector, &eqn, &irqn); if (err) goto err_cqb; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 45816b10af5e..e59e2a1f8726 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -62,95 +62,6 @@ static char mlx5_version[] = DRIVER_NAME ": Mellanox Connect-IB Infiniband driver v" DRIVER_VERSION " (" DRIVER_RELDATE ")\n"; -int mlx5_vector2eqn(struct mlx5_ib_dev *dev, int vector, int *eqn, int *irqn) -{ - struct mlx5_eq_table *table = &dev->mdev->priv.eq_table; - struct mlx5_eq *eq, *n; - int err = -ENOENT; - - spin_lock(&table->lock); - list_for_each_entry_safe(eq, n, &dev->eqs_list, list) { - if (eq->index == vector) { - *eqn = eq->eqn; - *irqn = eq->irqn; - err = 0; - break; - } - } - spin_unlock(&table->lock); - - return err; -} - -static int alloc_comp_eqs(struct mlx5_ib_dev *dev) -{ - struct mlx5_eq_table *table = &dev->mdev->priv.eq_table; - char name[MLX5_MAX_EQ_NAME]; - struct mlx5_eq *eq, *n; - int ncomp_vec; - int nent; - int err; - int i; - - INIT_LIST_HEAD(&dev->eqs_list); - ncomp_vec = table->num_comp_vectors; - nent = MLX5_COMP_EQ_SIZE; - for (i = 0; i < ncomp_vec; i++) { - eq = kzalloc(sizeof(*eq), GFP_KERNEL); - if (!eq) { - err = -ENOMEM; - goto clean; - } - - snprintf(name, MLX5_MAX_EQ_NAME, "mlx5_comp%d", i); - err = mlx5_create_map_eq(dev->mdev, eq, - i + MLX5_EQ_VEC_COMP_BASE, nent, 0, - name, &dev->mdev->priv.uuari.uars[0]); - if (err) { - kfree(eq); - goto clean; - } - mlx5_ib_dbg(dev, "allocated completion EQN %d\n", eq->eqn); - eq->index = i; - spin_lock(&table->lock); - list_add_tail(&eq->list, &dev->eqs_list); - spin_unlock(&table->lock); - } - - dev->num_comp_vectors = ncomp_vec; - return 0; - -clean: - spin_lock(&table->lock); - list_for_each_entry_safe(eq, n, &dev->eqs_list, list) { - list_del(&eq->list); - spin_unlock(&table->lock); - if (mlx5_destroy_unmap_eq(dev->mdev, eq)) - mlx5_ib_warn(dev, "failed to destroy EQ 0x%x\n", eq->eqn); - kfree(eq); - spin_lock(&table->lock); - } - spin_unlock(&table->lock); - return err; -} - -static void free_comp_eqs(struct mlx5_ib_dev *dev) -{ - struct mlx5_eq_table *table = &dev->mdev->priv.eq_table; - struct mlx5_eq *eq, *n; - - spin_lock(&table->lock); - list_for_each_entry_safe(eq, n, &dev->eqs_list, list) { - list_del(&eq->list); - spin_unlock(&table->lock); - if (mlx5_destroy_unmap_eq(dev->mdev, eq)) - mlx5_ib_warn(dev, "failed to destroy EQ 0x%x\n", eq->eqn); - kfree(eq); - spin_lock(&table->lock); - } - spin_unlock(&table->lock); -} - static int mlx5_ib_query_device(struct ib_device *ibdev, struct ib_device_attr *props) { @@ -1291,10 +1202,6 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) get_ext_port_caps(dev); - err = alloc_comp_eqs(dev); - if (err) - goto err_dealloc; - MLX5_INIT_DOORBELL_LOCK(&dev->uar_lock); strlcpy(dev->ib_dev.name, "mlx5_%d", IB_DEVICE_NAME_MAX); @@ -1303,7 +1210,8 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) dev->ib_dev.local_dma_lkey = mdev->caps.gen.reserved_lkey; dev->num_ports = mdev->caps.gen.num_ports; dev->ib_dev.phys_port_cnt = dev->num_ports; - dev->ib_dev.num_comp_vectors = dev->num_comp_vectors; + dev->ib_dev.num_comp_vectors = + dev->mdev->priv.eq_table.num_comp_vectors; dev->ib_dev.dma_device = &mdev->pdev->dev; dev->ib_dev.uverbs_abi_ver = MLX5_IB_UVERBS_ABI_VERSION; @@ -1390,13 +1298,13 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) err = init_node_data(dev); if (err) - goto err_eqs; + goto err_dealloc; mutex_init(&dev->cap_mask_mutex); err = create_dev_resources(&dev->devr); if (err) - goto err_eqs; + goto err_dealloc; err = mlx5_ib_odp_init_one(dev); if (err) @@ -1433,9 +1341,6 @@ err_odp: err_rsrc: destroy_dev_resources(&dev->devr); -err_eqs: - free_comp_eqs(dev); - err_dealloc: ib_dealloc_device((struct ib_device *)dev); @@ -1450,7 +1355,6 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context) destroy_umrc_res(dev); mlx5_ib_odp_remove_one(dev); destroy_dev_resources(&dev->devr); - free_comp_eqs(dev); ib_dealloc_device(&dev->ib_dev); } diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index 8f94f69fbea5..dff1cfcdf476 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -421,9 +421,7 @@ struct mlx5_ib_dev { struct ib_device ib_dev; struct mlx5_core_dev *mdev; MLX5_DECLARE_DOORBELL_LOCK(uar_lock); - struct list_head eqs_list; int num_ports; - int num_comp_vectors; /* serialize update of capability mask */ struct mutex cap_mask_mutex; @@ -594,7 +592,6 @@ struct ib_xrcd *mlx5_ib_alloc_xrcd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata); int mlx5_ib_dealloc_xrcd(struct ib_xrcd *xrcd); -int mlx5_vector2eqn(struct mlx5_ib_dev *dev, int vector, int *eqn, int *irqn); int mlx5_ib_get_buf_offset(u64 addr, int page_shift, u32 *offset); int mlx5_query_ext_port_caps(struct mlx5_ib_dev *dev, u8 port); int mlx5_ib_query_port(struct ib_device *ibdev, u8 port, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index fe41a7c2c629..cfc29027f07a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -507,6 +507,87 @@ static int mlx5_core_disable_hca(struct mlx5_core_dev *dev) return 0; } +int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn, int *irqn) +{ + struct mlx5_eq_table *table = &dev->priv.eq_table; + struct mlx5_eq *eq, *n; + int err = -ENOENT; + + spin_lock(&table->lock); + list_for_each_entry_safe(eq, n, &table->comp_eqs_list, list) { + if (eq->index == vector) { + *eqn = eq->eqn; + *irqn = eq->irqn; + err = 0; + break; + } + } + spin_unlock(&table->lock); + + return err; +} +EXPORT_SYMBOL(mlx5_vector2eqn); + +static void free_comp_eqs(struct mlx5_core_dev *dev) +{ + struct mlx5_eq_table *table = &dev->priv.eq_table; + struct mlx5_eq *eq, *n; + + spin_lock(&table->lock); + list_for_each_entry_safe(eq, n, &table->comp_eqs_list, list) { + list_del(&eq->list); + spin_unlock(&table->lock); + if (mlx5_destroy_unmap_eq(dev, eq)) + mlx5_core_warn(dev, "failed to destroy EQ 0x%x\n", + eq->eqn); + kfree(eq); + spin_lock(&table->lock); + } + spin_unlock(&table->lock); +} + +static int alloc_comp_eqs(struct mlx5_core_dev *dev) +{ + struct mlx5_eq_table *table = &dev->priv.eq_table; + char name[MLX5_MAX_EQ_NAME]; + struct mlx5_eq *eq; + int ncomp_vec; + int nent; + int err; + int i; + + INIT_LIST_HEAD(&table->comp_eqs_list); + ncomp_vec = table->num_comp_vectors; + nent = MLX5_COMP_EQ_SIZE; + for (i = 0; i < ncomp_vec; i++) { + eq = kzalloc(sizeof(*eq), GFP_KERNEL); + if (!eq) { + err = -ENOMEM; + goto clean; + } + + snprintf(name, MLX5_MAX_EQ_NAME, "mlx5_comp%d", i); + err = mlx5_create_map_eq(dev, eq, + i + MLX5_EQ_VEC_COMP_BASE, nent, 0, + name, &dev->priv.uuari.uars[0]); + if (err) { + kfree(eq); + goto clean; + } + mlx5_core_dbg(dev, "allocated completion EQN %d\n", eq->eqn); + eq->index = i; + spin_lock(&table->lock); + list_add_tail(&eq->list, &table->comp_eqs_list); + spin_unlock(&table->lock); + } + + return 0; + +clean: + free_comp_eqs(dev); + return err; +} + static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) { struct mlx5_priv *priv = &dev->priv; @@ -643,6 +724,12 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) goto err_free_uar; } + err = alloc_comp_eqs(dev); + if (err) { + dev_err(&pdev->dev, "Failed to alloc completion EQs\n"); + goto err_stop_eqs; + } + MLX5_INIT_DOORBELL_LOCK(&priv->cq_uar_lock); mlx5_init_cq_table(dev); @@ -652,6 +739,9 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) return 0; +err_stop_eqs: + mlx5_stop_eqs(dev); + err_free_uar: mlx5_free_uuars(dev, &priv->uuari); @@ -703,6 +793,7 @@ static void mlx5_dev_cleanup(struct mlx5_core_dev *dev) mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); mlx5_cleanup_cq_table(dev); + free_comp_eqs(dev); mlx5_stop_eqs(dev); mlx5_free_uuars(dev, &priv->uuari); mlx5_eq_cleanup(dev); diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 8fedd39a9a60..f250f6580dad 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -410,7 +410,7 @@ struct mlx5_core_srq { struct mlx5_eq_table { void __iomem *update_ci; void __iomem *update_arm_ci; - struct list_head *comp_eq_head; + struct list_head comp_eqs_list; struct mlx5_eq pages_eq; struct mlx5_eq async_eq; struct mlx5_eq cmd_eq; @@ -725,6 +725,7 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq); int mlx5_start_eqs(struct mlx5_core_dev *dev); int mlx5_stop_eqs(struct mlx5_core_dev *dev); +int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn, int *irqn); int mlx5_core_attach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid, u32 qpn); int mlx5_core_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid, u32 qpn); -- cgit v1.2.3 From ce0f75093282c5dca1e79ae3e3e893deaea86166 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 2 Apr 2015 17:07:33 +0300 Subject: net/mlx5_core: Modify arm CQ in preparation for upcoming Ethernet driver Pass consumer index as a parameter to arm CQ Signed-off-by: Saeed Mahameed Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx5/cq.c | 8 ++++++-- include/linux/mlx5/cq.h | 5 +++-- 2 files changed, 9 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index c94e2651820d..2ee6b1051975 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -572,11 +572,15 @@ int mlx5_ib_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) int mlx5_ib_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) { + struct mlx5_core_dev *mdev = to_mdev(ibcq->device)->mdev; + void __iomem *uar_page = mdev->priv.uuari.uars[0].map; + mlx5_cq_arm(&to_mcq(ibcq)->mcq, (flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED ? MLX5_CQ_DB_REQ_NOT_SOL : MLX5_CQ_DB_REQ_NOT, - to_mdev(ibcq->device)->mdev->priv.uuari.uars[0].map, - MLX5_GET_DOORBELL_LOCK(&to_mdev(ibcq->device)->mdev->priv.cq_uar_lock)); + uar_page, + MLX5_GET_DOORBELL_LOCK(&mdev->priv.cq_uar_lock), + to_mcq(ibcq)->mcq.cons_index); return 0; } diff --git a/include/linux/mlx5/cq.h b/include/linux/mlx5/cq.h index 72ee0d732a26..2695ced222df 100644 --- a/include/linux/mlx5/cq.h +++ b/include/linux/mlx5/cq.h @@ -137,14 +137,15 @@ enum { static inline void mlx5_cq_arm(struct mlx5_core_cq *cq, u32 cmd, void __iomem *uar_page, - spinlock_t *doorbell_lock) + spinlock_t *doorbell_lock, + u32 cons_index) { __be32 doorbell[2]; u32 sn; u32 ci; sn = cq->arm_sn & 3; - ci = cq->cons_index & 0xffffff; + ci = cons_index & 0xffffff; *cq->arm_db = cpu_to_be32(sn << 28 | cmd | ci); -- cgit v1.2.3 From 64613d9499c4887485d4350387919ea507330d90 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 2 Apr 2015 17:07:34 +0300 Subject: net/mlx5_core: Extend struct mlx5_interface to support multiple protocols Preparation for ethernet driver. Signed-off-by: Achiad Shochat Signed-off-by: Saeed Mahameed Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx5/main.c | 1 + drivers/net/ethernet/mellanox/mlx5/core/main.c | 22 ++++++++++++++++++++++ include/linux/mlx5/driver.h | 8 ++++++++ 3 files changed, 31 insertions(+) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index e59e2a1f8726..57c9809e8b87 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1362,6 +1362,7 @@ static struct mlx5_interface mlx5_ib_interface = { .add = mlx5_ib_add, .remove = mlx5_ib_remove, .event = mlx5_ib_event, + .protocol = MLX5_INTERFACE_PROTOCOL_IB, }; static int __init mlx5_ib_init(void) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index cfc29027f07a..28425e5ea91f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -908,6 +908,28 @@ void mlx5_unregister_interface(struct mlx5_interface *intf) } EXPORT_SYMBOL(mlx5_unregister_interface); +void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol) +{ + struct mlx5_priv *priv = &mdev->priv; + struct mlx5_device_context *dev_ctx; + unsigned long flags; + void *result = NULL; + + spin_lock_irqsave(&priv->ctx_lock, flags); + + list_for_each_entry(dev_ctx, &mdev->priv.ctx_list, list) + if ((dev_ctx->intf->protocol == protocol) && + dev_ctx->intf->get_dev) { + result = dev_ctx->intf->get_dev(dev_ctx->context); + break; + } + + spin_unlock_irqrestore(&priv->ctx_lock, flags); + + return result; +} +EXPORT_SYMBOL(mlx5_get_protocol_dev); + static void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event, unsigned long param) { diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index f250f6580dad..9a90e7523dc2 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -781,14 +781,22 @@ enum { MAX_MR_CACHE_ENTRIES = 16, }; +enum { + MLX5_INTERFACE_PROTOCOL_IB = 0, + MLX5_INTERFACE_PROTOCOL_ETH = 1, +}; + struct mlx5_interface { void * (*add)(struct mlx5_core_dev *dev); void (*remove)(struct mlx5_core_dev *dev, void *context); void (*event)(struct mlx5_core_dev *dev, void *context, enum mlx5_dev_event event, unsigned long param); + void * (*get_dev)(void *context); + int protocol; struct list_head list; }; +void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol); int mlx5_register_interface(struct mlx5_interface *intf); void mlx5_unregister_interface(struct mlx5_interface *intf); -- cgit v1.2.3 From e79d8429aac95a5cbe4c235795c7cd554c91f924 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Fri, 3 Apr 2015 22:17:17 +1030 Subject: netdevice: document NETDEV_TX_BUSY deprecation. This paraphrases DaveM (and steals some of his words) explaining why a device shouldn't return NETDEV_TX_BUSY, even though it looks so inviting to driver authors. See http://www.spinics.net/lists/netdev/msg322350.html Inspired-by: David Miller Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 846a1f5bc9db..a710d22b174f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -795,7 +795,10 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb, * struct net_device *dev); * Called when a packet needs to be transmitted. - * Must return NETDEV_TX_OK , NETDEV_TX_BUSY. + * Returns NETDEV_TX_OK. Can return NETDEV_TX_BUSY, but you should stop + * the queue before that can happen; it's for obsolete devices and weird + * corner cases, but the stack really does a non-trivial amount + * of useless work if you return NETDEV_TX_BUSY. * (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX) * Required can not be NULL. * -- cgit v1.2.3 From 2e7056c433216f406b90a003aa0ba42e19d3bdcf Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 31 Mar 2015 14:19:10 -0700 Subject: jhash: Update jhash_[321]words functions to use correct initval Looking over the implementation for jhash2 and comparing it to jhash_3words I realized that the two hashes were in fact very different. Doing a bit of digging led me to "The new jhash implementation" in which lookup2 was supposed to have been replaced with lookup3. In reviewing the patch I noticed that jhash2 had originally initialized a and b to JHASH_GOLDENRATIO and c to initval, but after the patch a, b, and c were initialized to initval + (length << 2) + JHASH_INITVAL. However the changes in jhash_3words simply replaced the initialization of a and b with JHASH_INITVAL. This change corrects what I believe was an oversight so that a, b, and c in jhash_3words all have the same value added consisting of initval + (length << 2) + JHASH_INITVAL so that jhash2 and jhash_3words will now produce the same hash result given the same inputs. Fixes: 60d509c823cca ("The new jhash implementation") Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller --- include/linux/jhash.h | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jhash.h b/include/linux/jhash.h index 47cb09edec1a..348c6f47e4cc 100644 --- a/include/linux/jhash.h +++ b/include/linux/jhash.h @@ -145,11 +145,11 @@ static inline u32 jhash2(const u32 *k, u32 length, u32 initval) } -/* jhash_3words - hash exactly 3, 2 or 1 word(s) */ -static inline u32 jhash_3words(u32 a, u32 b, u32 c, u32 initval) +/* __jhash_nwords - hash exactly 3, 2 or 1 word(s) */ +static inline u32 __jhash_nwords(u32 a, u32 b, u32 c, u32 initval) { - a += JHASH_INITVAL; - b += JHASH_INITVAL; + a += initval; + b += initval; c += initval; __jhash_final(a, b, c); @@ -157,14 +157,19 @@ static inline u32 jhash_3words(u32 a, u32 b, u32 c, u32 initval) return c; } +static inline u32 jhash_3words(u32 a, u32 b, u32 c, u32 initval) +{ + return __jhash_nwords(a, b, c, initval + JHASH_INITVAL + (3 << 2)); +} + static inline u32 jhash_2words(u32 a, u32 b, u32 initval) { - return jhash_3words(a, b, 0, initval); + return __jhash_nwords(a, b, 0, initval + JHASH_INITVAL + (2 << 2)); } static inline u32 jhash_1word(u32 a, u32 initval) { - return jhash_3words(a, 0, 0, initval); + return __jhash_nwords(a, 0, 0, initval + JHASH_INITVAL + (1 << 2)); } #endif /* _LINUX_JHASH_H */ -- cgit v1.2.3 From a3bebdce4135a44d09e96ba66c40797c8f9fa902 Mon Sep 17 00:00:00 2001 From: Stas Sergeev Date: Wed, 1 Apr 2015 20:30:31 +0300 Subject: add fixed_phy_update_state() - update state of fixed_phy Currently fixed_phy uses a callback to periodically poll the link state. This patch adds the fixed_phy_update_state() API. It solves the following problems: - On link state interrupt, MAC driver can't update status. Instead it needs to provide the callback to periodically query the HW about the link state. It is more efficient to update status after interrupt. - The callback needs to be unregistered before phy_disconnect(), or otherwise it will be called with net_dev==NULL. phy_disconnect() does not have enough info to unregister the callback automatically. - The callback needs to be registered before of_phy_connect() to avoid running with outdated state, but of_phy_connect() returns the phy_device pointer, which is needed to register the callback. Registering it before of_phy_connect() will therefore require a hack to get the pointer earlier. Overall, this addition makes the subsequent patch that implements SGMII link status for mvneta, much cleaner. CC: Florian Fainelli CC: netdev@vger.kernel.org CC: linux-kernel@vger.kernel.org Signed-off-by: Stas Sergeev Signed-off-by: David S. Miller --- drivers/net/phy/fixed_phy.c | 29 +++++++++++++++++++++++++++++ include/linux/phy_fixed.h | 9 +++++++++ 2 files changed, 38 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index a08a3c78ba97..1960b46add65 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -183,6 +183,35 @@ int fixed_phy_set_link_update(struct phy_device *phydev, } EXPORT_SYMBOL_GPL(fixed_phy_set_link_update); +int fixed_phy_update_state(struct phy_device *phydev, + const struct fixed_phy_status *status, + const struct fixed_phy_status *changed) +{ + struct fixed_mdio_bus *fmb = &platform_fmb; + struct fixed_phy *fp; + + if (!phydev || !phydev->bus) + return -EINVAL; + + list_for_each_entry(fp, &fmb->phys, node) { + if (fp->addr == phydev->addr) { +#define _UPD(x) if (changed->x) \ + fp->status.x = status->x + _UPD(link); + _UPD(speed); + _UPD(duplex); + _UPD(pause); + _UPD(asym_pause); +#undef _UPD + fixed_phy_update_regs(fp); + return 0; + } + } + + return -ENOENT; +} +EXPORT_SYMBOL(fixed_phy_update_state); + int fixed_phy_add(unsigned int irq, int phy_addr, struct fixed_phy_status *status) { diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h index 7e75bfe37cc7..fe5732d53eda 100644 --- a/include/linux/phy_fixed.h +++ b/include/linux/phy_fixed.h @@ -21,6 +21,9 @@ extern void fixed_phy_del(int phy_addr); extern int fixed_phy_set_link_update(struct phy_device *phydev, int (*link_update)(struct net_device *, struct fixed_phy_status *)); +extern int fixed_phy_update_state(struct phy_device *phydev, + const struct fixed_phy_status *status, + const struct fixed_phy_status *changed); #else static inline int fixed_phy_add(unsigned int irq, int phy_id, struct fixed_phy_status *status) @@ -43,6 +46,12 @@ static inline int fixed_phy_set_link_update(struct phy_device *phydev, { return -ENODEV; } +static inline int fixed_phy_update_state(struct phy_device *phydev, + const struct fixed_phy_status *status, + const struct fixed_phy_status *changed) +{ + return -ENODEV; +} #endif /* CONFIG_FIXED_PHY */ #endif /* __PHY_FIXED_H */ -- cgit v1.2.3 From cfdfab314647b1755afedc33ab66f3f247e161ae Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Apr 2015 16:23:58 -0400 Subject: netfilter: Create and use nf_hook_state. Instead of passing a large number of arguments down into the nf_hook() entry points, create a structure which carries this state down through the hook processing layers. This makes is so that if we want to change the types or signatures of any of these pieces of state, there are less places that need to be changed. Signed-off-by: David S. Miller --- include/linux/netfilter.h | 28 +++++++++++++++++++++++----- net/netfilter/core.c | 32 +++++++++++++------------------- net/netfilter/nf_internals.h | 11 +++-------- net/netfilter/nf_queue.c | 38 ++++++++++++++++++++------------------ 4 files changed, 59 insertions(+), 50 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 2517ece98820..aee7ef1e23ed 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -44,6 +44,16 @@ int netfilter_init(void); struct sk_buff; struct nf_hook_ops; + +struct nf_hook_state { + unsigned int hook; + int thresh; + u_int8_t pf; + struct net_device *in; + struct net_device *out; + int (*okfn)(struct sk_buff *); +}; + typedef unsigned int nf_hookfn(const struct nf_hook_ops *ops, struct sk_buff *skb, const struct net_device *in, @@ -118,9 +128,7 @@ static inline bool nf_hooks_active(u_int8_t pf, unsigned int hook) } #endif -int nf_hook_slow(u_int8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh); +int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state); /** * nf_hook_thresh - call a netfilter hook @@ -135,8 +143,18 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, struct net_device *outdev, int (*okfn)(struct sk_buff *), int thresh) { - if (nf_hooks_active(pf, hook)) - return nf_hook_slow(pf, hook, skb, indev, outdev, okfn, thresh); + if (nf_hooks_active(pf, hook)) { + struct nf_hook_state state = { + .hook = hook, + .thresh = thresh, + .pf = pf, + .in = indev, + .out = outdev, + .okfn = okfn + }; + + return nf_hook_slow(skb, &state); + } return 1; } diff --git a/net/netfilter/core.c b/net/netfilter/core.c index fea9ef566427..11d04ebfc5e3 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -120,12 +120,8 @@ EXPORT_SYMBOL(nf_unregister_hooks); unsigned int nf_iterate(struct list_head *head, struct sk_buff *skb, - unsigned int hook, - const struct net_device *indev, - const struct net_device *outdev, - struct nf_hook_ops **elemp, - int (*okfn)(struct sk_buff *), - int hook_thresh) + struct nf_hook_state *state, + struct nf_hook_ops **elemp) { unsigned int verdict; @@ -134,19 +130,20 @@ unsigned int nf_iterate(struct list_head *head, * function because of risk of continuing from deleted element. */ list_for_each_entry_continue_rcu((*elemp), head, list) { - if (hook_thresh > (*elemp)->priority) + if (state->thresh > (*elemp)->priority) continue; /* Optimization: we don't need to hold module reference here, since function can't sleep. --RR */ repeat: - verdict = (*elemp)->hook(*elemp, skb, indev, outdev, okfn); + verdict = (*elemp)->hook(*elemp, skb, state->in, state->out, + state->okfn); if (verdict != NF_ACCEPT) { #ifdef CONFIG_NETFILTER_DEBUG if (unlikely((verdict & NF_VERDICT_MASK) > NF_MAX_VERDICT)) { NFDEBUG("Evil return from %p(%u).\n", - (*elemp)->hook, hook); + (*elemp)->hook, state->hook); continue; } #endif @@ -161,11 +158,7 @@ repeat: /* Returns 1 if okfn() needs to be executed by the caller, * -EPERM for NF_DROP, 0 otherwise. */ -int nf_hook_slow(u_int8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *indev, - struct net_device *outdev, - int (*okfn)(struct sk_buff *), - int hook_thresh) +int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state) { struct nf_hook_ops *elem; unsigned int verdict; @@ -174,10 +167,11 @@ int nf_hook_slow(u_int8_t pf, unsigned int hook, struct sk_buff *skb, /* We may already have this, but read-locks nest anyway */ rcu_read_lock(); - elem = list_entry_rcu(&nf_hooks[pf][hook], struct nf_hook_ops, list); + elem = list_entry_rcu(&nf_hooks[state->pf][state->hook], + struct nf_hook_ops, list); next_hook: - verdict = nf_iterate(&nf_hooks[pf][hook], skb, hook, indev, - outdev, &elem, okfn, hook_thresh); + verdict = nf_iterate(&nf_hooks[state->pf][state->hook], skb, state, + &elem); if (verdict == NF_ACCEPT || verdict == NF_STOP) { ret = 1; } else if ((verdict & NF_VERDICT_MASK) == NF_DROP) { @@ -186,8 +180,8 @@ next_hook: if (ret == 0) ret = -EPERM; } else if ((verdict & NF_VERDICT_MASK) == NF_QUEUE) { - int err = nf_queue(skb, elem, pf, hook, indev, outdev, okfn, - verdict >> NF_VERDICT_QBITS); + int err = nf_queue(skb, elem, state, + verdict >> NF_VERDICT_QBITS); if (err < 0) { if (err == -ECANCELED) goto next_hook; diff --git a/net/netfilter/nf_internals.h b/net/netfilter/nf_internals.h index 61a3c927e63c..ea7f36784b3d 100644 --- a/net/netfilter/nf_internals.h +++ b/net/netfilter/nf_internals.h @@ -14,16 +14,11 @@ /* core.c */ unsigned int nf_iterate(struct list_head *head, struct sk_buff *skb, - unsigned int hook, const struct net_device *indev, - const struct net_device *outdev, - struct nf_hook_ops **elemp, - int (*okfn)(struct sk_buff *), int hook_thresh); + struct nf_hook_state *state, struct nf_hook_ops **elemp); /* nf_queue.c */ -int nf_queue(struct sk_buff *skb, struct nf_hook_ops *elem, u_int8_t pf, - unsigned int hook, struct net_device *indev, - struct net_device *outdev, int (*okfn)(struct sk_buff *), - unsigned int queuenum); +int nf_queue(struct sk_buff *skb, struct nf_hook_ops *elem, + struct nf_hook_state *state, unsigned int queuenum); int __init netfilter_queue_init(void); /* nf_log.c */ diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 4c8b68e5fa16..6f8e9485cc83 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -100,12 +100,9 @@ EXPORT_SYMBOL_GPL(nf_queue_entry_get_refs); * through nf_reinject(). */ int nf_queue(struct sk_buff *skb, - struct nf_hook_ops *elem, - u_int8_t pf, unsigned int hook, - struct net_device *indev, - struct net_device *outdev, - int (*okfn)(struct sk_buff *), - unsigned int queuenum) + struct nf_hook_ops *elem, + struct nf_hook_state *state, + unsigned int queuenum) { int status = -ENOENT; struct nf_queue_entry *entry = NULL; @@ -121,7 +118,7 @@ int nf_queue(struct sk_buff *skb, goto err_unlock; } - afinfo = nf_get_afinfo(pf); + afinfo = nf_get_afinfo(state->pf); if (!afinfo) goto err_unlock; @@ -134,11 +131,11 @@ int nf_queue(struct sk_buff *skb, *entry = (struct nf_queue_entry) { .skb = skb, .elem = elem, - .pf = pf, - .hook = hook, - .indev = indev, - .outdev = outdev, - .okfn = okfn, + .pf = state->pf, + .hook = state->hook, + .indev = state->in, + .outdev = state->out, + .okfn = state->okfn, .size = sizeof(*entry) + afinfo->route_key_size, }; @@ -171,6 +168,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) struct sk_buff *skb = entry->skb; struct nf_hook_ops *elem = entry->elem; const struct nf_afinfo *afinfo; + struct nf_hook_state state; int err; rcu_read_lock(); @@ -189,12 +187,17 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) verdict = NF_DROP; } + state.hook = entry->hook; + state.thresh = INT_MIN; + state.pf = entry->pf; + state.in = entry->indev; + state.out = entry->outdev; + state.okfn = entry->okfn; + if (verdict == NF_ACCEPT) { next_hook: verdict = nf_iterate(&nf_hooks[entry->pf][entry->hook], - skb, entry->hook, - entry->indev, entry->outdev, &elem, - entry->okfn, INT_MIN); + skb, &state, &elem); } switch (verdict & NF_VERDICT_MASK) { @@ -205,9 +208,8 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) local_bh_enable(); break; case NF_QUEUE: - err = nf_queue(skb, elem, entry->pf, entry->hook, - entry->indev, entry->outdev, entry->okfn, - verdict >> NF_VERDICT_QBITS); + err = nf_queue(skb, elem, &state, + verdict >> NF_VERDICT_QBITS); if (err < 0) { if (err == -ECANCELED) goto next_hook; -- cgit v1.2.3 From 238e54c9cb9385a1ba99e92801f3615a2fb398b6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Apr 2015 20:32:56 -0400 Subject: netfilter: Make nf_hookfn use nf_hook_state. Pass the nf_hook_state all the way down into the hook functions themselves. Signed-off-by: David S. Miller --- include/linux/netfilter.h | 4 +-- net/bridge/br_netfilter.c | 46 +++++++++----------------- net/bridge/netfilter/ebtable_filter.c | 14 ++++---- net/bridge/netfilter/ebtable_nat.c | 14 ++++---- net/bridge/netfilter/nf_tables_bridge.c | 10 +++--- net/decnet/netfilter/dn_rtmsg.c | 4 +-- net/ipv4/netfilter/arptable_filter.c | 7 ++-- net/ipv4/netfilter/ipt_CLUSTERIP.c | 10 +++--- net/ipv4/netfilter/ipt_SYNPROXY.c | 6 ++-- net/ipv4/netfilter/iptable_filter.c | 7 ++-- net/ipv4/netfilter/iptable_mangle.c | 14 ++++---- net/ipv4/netfilter/iptable_nat.c | 28 +++++++--------- net/ipv4/netfilter/iptable_raw.c | 8 ++--- net/ipv4/netfilter/iptable_security.c | 8 ++--- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 20 ++++------- net/ipv4/netfilter/nf_defrag_ipv4.c | 4 +-- net/ipv4/netfilter/nf_tables_arp.c | 6 ++-- net/ipv4/netfilter/nf_tables_ipv4.c | 12 +++---- net/ipv4/netfilter/nft_chain_nat_ipv4.c | 25 +++++--------- net/ipv4/netfilter/nft_chain_route_ipv4.c | 6 ++-- net/ipv6/netfilter/ip6t_SYNPROXY.c | 6 ++-- net/ipv6/netfilter/ip6table_filter.c | 7 ++-- net/ipv6/netfilter/ip6table_mangle.c | 13 ++++---- net/ipv6/netfilter/ip6table_nat.c | 28 +++++++--------- net/ipv6/netfilter/ip6table_raw.c | 7 ++-- net/ipv6/netfilter/ip6table_security.c | 8 ++--- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 20 ++++------- net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 8 ++--- net/ipv6/netfilter/nf_tables_ipv6.c | 12 +++---- net/ipv6/netfilter/nft_chain_nat_ipv6.c | 24 +++++--------- net/ipv6/netfilter/nft_chain_route_ipv6.c | 6 ++-- net/netfilter/core.c | 3 +- net/netfilter/ipvs/ip_vs_core.c | 32 ++++++------------ security/selinux/hooks.c | 28 +++++----------- security/smack/smack_netfilter.c | 8 ++--- 35 files changed, 169 insertions(+), 294 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index aee7ef1e23ed..c480c43ad8f7 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -56,9 +56,7 @@ struct nf_hook_state { typedef unsigned int nf_hookfn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)); + const struct nf_hook_state *state); struct nf_hook_ops { struct list_head list; diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index f3884a1b942f..7527e94dd5dc 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -562,9 +562,7 @@ bad: * to ip6tables, which doesn't support NAT, so things are fairly simple. */ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { const struct ipv6hdr *hdr; u32 pkt_len; @@ -612,9 +610,7 @@ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops, * address to be able to detect DNAT afterwards. */ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct net_bridge_port *p; struct net_bridge *br; @@ -623,7 +619,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, if (unlikely(!pskb_may_pull(skb, len))) return NF_DROP; - p = br_port_get_rcu(in); + p = br_port_get_rcu(state->in); if (p == NULL) return NF_DROP; br = p->br; @@ -633,7 +629,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, return NF_ACCEPT; nf_bridge_pull_encap_header_rcsum(skb); - return br_nf_pre_routing_ipv6(ops, skb, in, out, okfn); + return br_nf_pre_routing_ipv6(ops, skb, state); } if (!brnf_call_iptables && !br->nf_call_iptables) @@ -671,9 +667,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, * prevent this from happening. */ static unsigned int br_nf_local_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { br_drop_fake_rtable(skb); return NF_ACCEPT; @@ -710,9 +704,7 @@ static int br_nf_forward_finish(struct sk_buff *skb) * bridge ports. */ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_bridge_info *nf_bridge; struct net_device *parent; @@ -726,7 +718,7 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops, if (!nf_bridge_unshare(skb)) return NF_DROP; - parent = bridge_parent(out); + parent = bridge_parent(state->out); if (!parent) return NF_DROP; @@ -754,23 +746,21 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops, else skb->protocol = htons(ETH_P_IPV6); - NF_HOOK(pf, NF_INET_FORWARD, skb, brnf_get_logical_dev(skb, in), parent, - br_nf_forward_finish); + NF_HOOK(pf, NF_INET_FORWARD, skb, brnf_get_logical_dev(skb, state->in), + parent, br_nf_forward_finish); return NF_STOLEN; } static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct net_bridge_port *p; struct net_bridge *br; struct net_device **d = (struct net_device **)(skb->cb); - p = br_port_get_rcu(out); + p = br_port_get_rcu(state->out); if (p == NULL) return NF_ACCEPT; br = p->br; @@ -789,9 +779,9 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops, nf_bridge_push_encap_header(skb); return NF_ACCEPT; } - *d = (struct net_device *)in; - NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, skb, (struct net_device *)in, - (struct net_device *)out, br_nf_forward_finish); + *d = state->in; + NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, skb, state->in, + state->out, br_nf_forward_finish); return NF_STOLEN; } @@ -859,9 +849,7 @@ static int br_nf_dev_queue_xmit(struct sk_buff *skb) /* PF_BRIDGE/POST_ROUTING ********************************************/ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; struct net_device *realoutdev = bridge_parent(skb->dev); @@ -910,9 +898,7 @@ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops, * for the second time. */ static unsigned int ip_sabotage_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { if (skb->nf_bridge && !(skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING)) { diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index ce205aabf9c5..8a3f63b2e807 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -58,20 +58,18 @@ static const struct ebt_table frame_filter = { static unsigned int ebt_in_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return ebt_do_table(ops->hooknum, skb, in, out, - dev_net(in)->xt.frame_filter); + return ebt_do_table(ops->hooknum, skb, state->in, state->out, + dev_net(state->in)->xt.frame_filter); } static unsigned int ebt_out_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return ebt_do_table(ops->hooknum, skb, in, out, - dev_net(out)->xt.frame_filter); + return ebt_do_table(ops->hooknum, skb, state->in, state->out, + dev_net(state->out)->xt.frame_filter); } static struct nf_hook_ops ebt_ops_filter[] __read_mostly = { diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index a0ac2984fb6c..c5ef5b1ab678 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -58,20 +58,18 @@ static struct ebt_table frame_nat = { static unsigned int ebt_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return ebt_do_table(ops->hooknum, skb, in, out, - dev_net(in)->xt.frame_nat); + return ebt_do_table(ops->hooknum, skb, state->in, state->out, + dev_net(state->in)->xt.frame_nat); } static unsigned int ebt_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return ebt_do_table(ops->hooknum, skb, in, out, - dev_net(out)->xt.frame_nat); + return ebt_do_table(ops->hooknum, skb, state->in, state->out, + dev_net(state->out)->xt.frame_nat); } static struct nf_hook_ops ebt_ops_nat[] __read_mostly = { diff --git a/net/bridge/netfilter/nf_tables_bridge.c b/net/bridge/netfilter/nf_tables_bridge.c index 19473a9371b8..2c46a47160a8 100644 --- a/net/bridge/netfilter/nf_tables_bridge.c +++ b/net/bridge/netfilter/nf_tables_bridge.c @@ -93,21 +93,19 @@ static inline void nft_bridge_set_pktinfo_ipv6(struct nft_pktinfo *pkt, static unsigned int nft_do_chain_bridge(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nft_pktinfo pkt; switch (eth_hdr(skb)->h_proto) { case htons(ETH_P_IP): - nft_bridge_set_pktinfo_ipv4(&pkt, ops, skb, in, out); + nft_bridge_set_pktinfo_ipv4(&pkt, ops, skb, state->in, state->out); break; case htons(ETH_P_IPV6): - nft_bridge_set_pktinfo_ipv6(&pkt, ops, skb, in, out); + nft_bridge_set_pktinfo_ipv6(&pkt, ops, skb, state->in, state->out); break; default: - nft_set_pktinfo(&pkt, ops, skb, in, out); + nft_set_pktinfo(&pkt, ops, skb, state->in, state->out); break; } diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index e4d9560a910b..af34fc9bdf69 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -89,9 +89,7 @@ static void dnrmg_send_peer(struct sk_buff *skb) static unsigned int dnrmg_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { dnrmg_send_peer(skb); return NF_ACCEPT; diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 802ddecb30b8..6a641cb41062 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -28,12 +28,11 @@ static const struct xt_table packet_filter = { /* The work comes in here from netfilter.c */ static unsigned int arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - const struct net *net = dev_net((in != NULL) ? in : out); + const struct net *net = dev_net(state->in ? state->in : state->out); - return arpt_do_table(skb, ops->hooknum, in, out, + return arpt_do_table(skb, ops->hooknum, state->in, state->out, net->ipv4.arptable_filter); } diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index f75e9df5e017..771ab3d01ad3 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -504,14 +504,12 @@ static void arp_print(struct arp_payload *payload) static unsigned int arp_mangle(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct arphdr *arp = arp_hdr(skb); struct arp_payload *payload; struct clusterip_config *c; - struct net *net = dev_net(in ? in : out); + struct net *net = dev_net(state->in ? state->in : state->out); /* we don't care about non-ethernet and non-ipv4 ARP */ if (arp->ar_hrd != htons(ARPHRD_ETHER) || @@ -536,10 +534,10 @@ arp_mangle(const struct nf_hook_ops *ops, * addresses on different interfacs. However, in the CLUSTERIP case * this wouldn't work, since we didn't subscribe the mcast group on * other interfaces */ - if (c->dev != out) { + if (c->dev != state->out) { pr_debug("not mangling arp reply on different " "interface: cip'%s'-skb'%s'\n", - c->dev->name, out->name); + c->dev->name, state->out->name); clusterip_config_put(c); return NF_ACCEPT; } diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c index a313c3fbeb46..e9e67793055f 100644 --- a/net/ipv4/netfilter/ipt_SYNPROXY.c +++ b/net/ipv4/netfilter/ipt_SYNPROXY.c @@ -300,11 +300,9 @@ synproxy_tg4(struct sk_buff *skb, const struct xt_action_param *par) static unsigned int ipv4_synproxy_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *nhs) { - struct synproxy_net *snet = synproxy_pernet(dev_net(in ? : out)); + struct synproxy_net *snet = synproxy_pernet(dev_net(nhs->in ? : nhs->out)); enum ip_conntrack_info ctinfo; struct nf_conn *ct; struct nf_conn_synproxy *synproxy; diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index e08a74a243a8..1df0d42bfd39 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -34,8 +34,7 @@ static const struct xt_table packet_filter = { static unsigned int iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { const struct net *net; @@ -45,8 +44,8 @@ iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, /* root is playing with raw sockets. */ return NF_ACCEPT; - net = dev_net((in != NULL) ? in : out); - return ipt_do_table(skb, ops->hooknum, in, out, + net = dev_net(state->in ? state->in : state->out); + return ipt_do_table(skb, ops->hooknum, state->in, state->out, net->ipv4.iptable_filter); } diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 6a5079c34bb3..7a825e740045 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -81,18 +81,16 @@ ipt_mangle_out(struct sk_buff *skb, const struct net_device *out) static unsigned int iptable_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { if (ops->hooknum == NF_INET_LOCAL_OUT) - return ipt_mangle_out(skb, out); + return ipt_mangle_out(skb, state->out); if (ops->hooknum == NF_INET_POST_ROUTING) - return ipt_do_table(skb, ops->hooknum, in, out, - dev_net(out)->ipv4.iptable_mangle); + return ipt_do_table(skb, ops->hooknum, state->in, state->out, + dev_net(state->out)->ipv4.iptable_mangle); /* PREROUTING/INPUT/FORWARD: */ - return ipt_do_table(skb, ops->hooknum, in, out, - dev_net(in)->ipv4.iptable_mangle); + return ipt_do_table(skb, ops->hooknum, state->in, state->out, + dev_net(state->in)->ipv4.iptable_mangle); } static struct nf_hook_ops *mangle_ops __read_mostly; diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 6b67d7e9a75d..7a7fea4711e5 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -41,38 +41,34 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops, static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_fn(ops, skb, in, out, iptable_nat_do_chain); + return nf_nat_ipv4_fn(ops, skb, state->in, state->out, + iptable_nat_do_chain); } static unsigned int iptable_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_in(ops, skb, in, out, iptable_nat_do_chain); + return nf_nat_ipv4_in(ops, skb, state->in, state->out, + iptable_nat_do_chain); } static unsigned int iptable_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_out(ops, skb, in, out, iptable_nat_do_chain); + return nf_nat_ipv4_out(ops, skb, state->in, state->out, + iptable_nat_do_chain); } static unsigned int iptable_nat_ipv4_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_local_fn(ops, skb, in, out, iptable_nat_do_chain); + return nf_nat_ipv4_local_fn(ops, skb, state->in, state->out, + iptable_nat_do_chain); } static struct nf_hook_ops nf_nat_ipv4_ops[] __read_mostly = { diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index b2f7e8f98316..fac8f607c70b 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -21,8 +21,7 @@ static const struct xt_table packet_raw = { /* The work comes in here from netfilter.c. */ static unsigned int iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { const struct net *net; @@ -32,8 +31,9 @@ iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, /* root is playing with raw sockets. */ return NF_ACCEPT; - net = dev_net((in != NULL) ? in : out); - return ipt_do_table(skb, ops->hooknum, in, out, net->ipv4.iptable_raw); + net = dev_net(state->in ? state->in : state->out); + return ipt_do_table(skb, ops->hooknum, state->in, state->out, + net->ipv4.iptable_raw); } static struct nf_hook_ops *rawtable_ops __read_mostly; diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index c86647ed2078..d9ad60a57413 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -38,9 +38,7 @@ static const struct xt_table security_table = { static unsigned int iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { const struct net *net; @@ -50,8 +48,8 @@ iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, /* Somebody is playing with raw sockets. */ return NF_ACCEPT; - net = dev_net((in != NULL) ? in : out); - return ipt_do_table(skb, ops->hooknum, in, out, + net = dev_net(state->in ? state->in : state->out); + return ipt_do_table(skb, ops->hooknum, state->in, state->out, net->ipv4.iptable_security); } diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 8c8d6642cbb0..30ad9554b5e9 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -94,9 +94,7 @@ static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, static unsigned int ipv4_helper(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_conn *ct; enum ip_conntrack_info ctinfo; @@ -123,9 +121,7 @@ static unsigned int ipv4_helper(const struct nf_hook_ops *ops, static unsigned int ipv4_confirm(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_conn *ct; enum ip_conntrack_info ctinfo; @@ -149,24 +145,20 @@ out: static unsigned int ipv4_conntrack_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_conntrack_in(dev_net(in), PF_INET, ops->hooknum, skb); + return nf_conntrack_in(dev_net(state->in), PF_INET, ops->hooknum, skb); } static unsigned int ipv4_conntrack_local(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { /* root is playing with raw sockets. */ if (skb->len < sizeof(struct iphdr) || ip_hdrlen(skb) < sizeof(struct iphdr)) return NF_ACCEPT; - return nf_conntrack_in(dev_net(out), PF_INET, ops->hooknum, skb); + return nf_conntrack_in(dev_net(state->out), PF_INET, ops->hooknum, skb); } /* Connection tracking may drop packets, but never alters them, so diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index 7e5ca6f2d0cd..c88b7d434718 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -63,9 +63,7 @@ static enum ip_defrag_users nf_ct_defrag_user(unsigned int hooknum, static unsigned int ipv4_conntrack_defrag(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(skb->sk); diff --git a/net/ipv4/netfilter/nf_tables_arp.c b/net/ipv4/netfilter/nf_tables_arp.c index 19412a4063fb..fceb50e1e87d 100644 --- a/net/ipv4/netfilter/nf_tables_arp.c +++ b/net/ipv4/netfilter/nf_tables_arp.c @@ -17,13 +17,11 @@ static unsigned int nft_do_chain_arp(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nft_pktinfo pkt; - nft_set_pktinfo(&pkt, ops, skb, in, out); + nft_set_pktinfo(&pkt, ops, skb, state->in, state->out); return nft_do_chain(&pkt, ops); } diff --git a/net/ipv4/netfilter/nf_tables_ipv4.c b/net/ipv4/netfilter/nf_tables_ipv4.c index 6820c8c40842..708e388e3dbe 100644 --- a/net/ipv4/netfilter/nf_tables_ipv4.c +++ b/net/ipv4/netfilter/nf_tables_ipv4.c @@ -20,22 +20,18 @@ static unsigned int nft_do_chain_ipv4(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nft_pktinfo pkt; - nft_set_pktinfo_ipv4(&pkt, ops, skb, in, out); + nft_set_pktinfo_ipv4(&pkt, ops, skb, state->in, state->out); return nft_do_chain(&pkt, ops); } static unsigned int nft_ipv4_output(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { if (unlikely(skb->len < sizeof(struct iphdr) || ip_hdr(skb)->ihl < sizeof(struct iphdr) / 4)) { @@ -45,7 +41,7 @@ static unsigned int nft_ipv4_output(const struct nf_hook_ops *ops, return NF_ACCEPT; } - return nft_do_chain_ipv4(ops, skb, in, out, okfn); + return nft_do_chain_ipv4(ops, skb, state); } struct nft_af_info nft_af_ipv4 __read_mostly = { diff --git a/net/ipv4/netfilter/nft_chain_nat_ipv4.c b/net/ipv4/netfilter/nft_chain_nat_ipv4.c index df547bf50078..d08db6b0fcc3 100644 --- a/net/ipv4/netfilter/nft_chain_nat_ipv4.c +++ b/net/ipv4/netfilter/nft_chain_nat_ipv4.c @@ -41,38 +41,31 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops, static unsigned int nft_nat_ipv4_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_fn(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv4_fn(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv4_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_in(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv4_in(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv4_out(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_out(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv4_out(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv4_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv4_local_fn(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv4_local_fn(ops, skb, state->in, state->out, + nft_nat_do_chain); } static const struct nf_chain_type nft_chain_nat_ipv4 = { diff --git a/net/ipv4/netfilter/nft_chain_route_ipv4.c b/net/ipv4/netfilter/nft_chain_route_ipv4.c index 125b66766c0a..073d0776ae7f 100644 --- a/net/ipv4/netfilter/nft_chain_route_ipv4.c +++ b/net/ipv4/netfilter/nft_chain_route_ipv4.c @@ -23,9 +23,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { unsigned int ret; struct nft_pktinfo pkt; @@ -39,7 +37,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops, ip_hdrlen(skb) < sizeof(struct iphdr)) return NF_ACCEPT; - nft_set_pktinfo_ipv4(&pkt, ops, skb, in, out); + nft_set_pktinfo_ipv4(&pkt, ops, skb, state->in, state->out); mark = skb->mark; iph = ip_hdr(skb); diff --git a/net/ipv6/netfilter/ip6t_SYNPROXY.c b/net/ipv6/netfilter/ip6t_SYNPROXY.c index a0d17270117c..6edb7b106de7 100644 --- a/net/ipv6/netfilter/ip6t_SYNPROXY.c +++ b/net/ipv6/netfilter/ip6t_SYNPROXY.c @@ -315,11 +315,9 @@ synproxy_tg6(struct sk_buff *skb, const struct xt_action_param *par) static unsigned int ipv6_synproxy_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *nhs) { - struct synproxy_net *snet = synproxy_pernet(dev_net(in ? : out)); + struct synproxy_net *snet = synproxy_pernet(dev_net(nhs->in ? : nhs->out)); enum ip_conntrack_info ctinfo; struct nf_conn *ct; struct nf_conn_synproxy *synproxy; diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index ca7f6c128086..eb9ef093454f 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -33,12 +33,11 @@ static const struct xt_table packet_filter = { /* The work comes in here from netfilter.c. */ static unsigned int ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - const struct net *net = dev_net((in != NULL) ? in : out); + const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, in, out, + return ip6t_do_table(skb, ops->hooknum, state->in, state->out, net->ipv6.ip6table_filter); } diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 307bbb782d14..e713b8d3dbbc 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -77,17 +77,16 @@ ip6t_mangle_out(struct sk_buff *skb, const struct net_device *out) /* The work comes in here from netfilter.c. */ static unsigned int ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { if (ops->hooknum == NF_INET_LOCAL_OUT) - return ip6t_mangle_out(skb, out); + return ip6t_mangle_out(skb, state->out); if (ops->hooknum == NF_INET_POST_ROUTING) - return ip6t_do_table(skb, ops->hooknum, in, out, - dev_net(out)->ipv6.ip6table_mangle); + return ip6t_do_table(skb, ops->hooknum, state->in, state->out, + dev_net(state->out)->ipv6.ip6table_mangle); /* INPUT/FORWARD */ - return ip6t_do_table(skb, ops->hooknum, in, out, - dev_net(in)->ipv6.ip6table_mangle); + return ip6t_do_table(skb, ops->hooknum, state->in, state->out, + dev_net(state->in)->ipv6.ip6table_mangle); } static struct nf_hook_ops *mangle_ops __read_mostly; diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index b0634ac996b7..e32b0d0315e6 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -43,38 +43,34 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops, static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_fn(ops, skb, in, out, ip6table_nat_do_chain); + return nf_nat_ipv6_fn(ops, skb, state->in, state->out, + ip6table_nat_do_chain); } static unsigned int ip6table_nat_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_in(ops, skb, in, out, ip6table_nat_do_chain); + return nf_nat_ipv6_in(ops, skb, state->in, state->out, + ip6table_nat_do_chain); } static unsigned int ip6table_nat_out(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_out(ops, skb, in, out, ip6table_nat_do_chain); + return nf_nat_ipv6_out(ops, skb, state->in, state->out, + ip6table_nat_do_chain); } static unsigned int ip6table_nat_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_local_fn(ops, skb, in, out, ip6table_nat_do_chain); + return nf_nat_ipv6_local_fn(ops, skb, state->in, state->out, + ip6table_nat_do_chain); } static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = { diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 5274740acecc..937908e25862 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -20,12 +20,11 @@ static const struct xt_table packet_raw = { /* The work comes in here from netfilter.c. */ static unsigned int ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - const struct net *net = dev_net((in != NULL) ? in : out); + const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, in, out, + return ip6t_do_table(skb, ops->hooknum, state->in, state->out, net->ipv6.ip6table_raw); } diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index ab3b0219ecfa..f33b41e8e294 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -37,13 +37,11 @@ static const struct xt_table security_table = { static unsigned int ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - const struct net *net = dev_net((in != NULL) ? in : out); + const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, in, out, + return ip6t_do_table(skb, ops->hooknum, state->in, state->out, net->ipv6.ip6table_security); } diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index fba91c6fc7ca..4ba0c34c627b 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -97,9 +97,7 @@ static int ipv6_get_l4proto(const struct sk_buff *skb, unsigned int nhoff, static unsigned int ipv6_helper(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_conn *ct; const struct nf_conn_help *help; @@ -135,9 +133,7 @@ static unsigned int ipv6_helper(const struct nf_hook_ops *ops, static unsigned int ipv6_confirm(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nf_conn *ct; enum ip_conntrack_info ctinfo; @@ -171,25 +167,21 @@ out: static unsigned int ipv6_conntrack_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_conntrack_in(dev_net(in), PF_INET6, ops->hooknum, skb); + return nf_conntrack_in(dev_net(state->in), PF_INET6, ops->hooknum, skb); } static unsigned int ipv6_conntrack_local(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { /* root is playing with raw sockets. */ if (skb->len < sizeof(struct ipv6hdr)) { net_notice_ratelimited("ipv6_conntrack_local: packet too short\n"); return NF_ACCEPT; } - return nf_conntrack_in(dev_net(out), PF_INET6, ops->hooknum, skb); + return nf_conntrack_in(dev_net(state->out), PF_INET6, ops->hooknum, skb); } static struct nf_hook_ops ipv6_conntrack_ops[] __read_mostly = { diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index e70382e4dfb5..e2b882056751 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -54,9 +54,7 @@ static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, static unsigned int ipv6_defrag(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct sk_buff *reasm; @@ -78,8 +76,8 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops, nf_ct_frag6_consume_orig(reasm); NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, reasm, - (struct net_device *) in, (struct net_device *) out, - okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1); + state->in, state->out, + state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1); return NF_STOLEN; } diff --git a/net/ipv6/netfilter/nf_tables_ipv6.c b/net/ipv6/netfilter/nf_tables_ipv6.c index 0d812b31277d..224bc8971a0b 100644 --- a/net/ipv6/netfilter/nf_tables_ipv6.c +++ b/net/ipv6/netfilter/nf_tables_ipv6.c @@ -18,14 +18,12 @@ static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct nft_pktinfo pkt; /* malformed packet, drop it */ - if (nft_set_pktinfo_ipv6(&pkt, ops, skb, in, out) < 0) + if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state->in, state->out) < 0) return NF_DROP; return nft_do_chain(&pkt, ops); @@ -33,9 +31,7 @@ static unsigned int nft_do_chain_ipv6(const struct nf_hook_ops *ops, static unsigned int nft_ipv6_output(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { if (unlikely(skb->len < sizeof(struct ipv6hdr))) { if (net_ratelimit()) @@ -44,7 +40,7 @@ static unsigned int nft_ipv6_output(const struct nf_hook_ops *ops, return NF_ACCEPT; } - return nft_do_chain_ipv6(ops, skb, in, out, okfn); + return nft_do_chain_ipv6(ops, skb, state); } struct nft_af_info nft_af_ipv6 __read_mostly = { diff --git a/net/ipv6/netfilter/nft_chain_nat_ipv6.c b/net/ipv6/netfilter/nft_chain_nat_ipv6.c index 1c4b75dd425b..f73f4ae25bc2 100644 --- a/net/ipv6/netfilter/nft_chain_nat_ipv6.c +++ b/net/ipv6/netfilter/nft_chain_nat_ipv6.c @@ -39,38 +39,30 @@ static unsigned int nft_nat_do_chain(const struct nf_hook_ops *ops, static unsigned int nft_nat_ipv6_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_fn(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv6_fn(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv6_in(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_in(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv6_in(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv6_out(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_out(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv6_out(ops, skb, state->in, state->out, nft_nat_do_chain); } static unsigned int nft_nat_ipv6_local_fn(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return nf_nat_ipv6_local_fn(ops, skb, in, out, nft_nat_do_chain); + return nf_nat_ipv6_local_fn(ops, skb, state->in, state->out, nft_nat_do_chain); } static const struct nf_chain_type nft_chain_nat_ipv6 = { diff --git a/net/ipv6/netfilter/nft_chain_route_ipv6.c b/net/ipv6/netfilter/nft_chain_route_ipv6.c index 42031299585e..c826c3c854b2 100644 --- a/net/ipv6/netfilter/nft_chain_route_ipv6.c +++ b/net/ipv6/netfilter/nft_chain_route_ipv6.c @@ -24,9 +24,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { unsigned int ret; struct nft_pktinfo pkt; @@ -35,7 +33,7 @@ static unsigned int nf_route_table_hook(const struct nf_hook_ops *ops, u32 mark, flowlabel; /* malformed packet, drop it */ - if (nft_set_pktinfo_ipv6(&pkt, ops, skb, in, out) < 0) + if (nft_set_pktinfo_ipv6(&pkt, ops, skb, state->in, state->out) < 0) return NF_DROP; /* save source/dest address, mark, hoplimit, flowlabel, priority */ diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 11d04ebfc5e3..e6163017c42d 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -136,8 +136,7 @@ unsigned int nf_iterate(struct list_head *head, /* Optimization: we don't need to hold module reference here, since function can't sleep. --RR */ repeat: - verdict = (*elemp)->hook(*elemp, skb, state->in, state->out, - state->okfn); + verdict = (*elemp)->hook(*elemp, skb, state); if (verdict != NF_ACCEPT) { #ifdef CONFIG_NETFILTER_DEBUG if (unlikely((verdict & NF_VERDICT_MASK) diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 04dbd9c7213f..5d2b806a862e 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1272,8 +1272,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) */ static unsigned int ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_out(ops->hooknum, skb, AF_INET); } @@ -1284,8 +1283,7 @@ ip_vs_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_out(ops->hooknum, skb, AF_INET); } @@ -1299,8 +1297,7 @@ ip_vs_local_reply4(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_out(ops->hooknum, skb, AF_INET6); } @@ -1311,8 +1308,7 @@ ip_vs_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_local_reply6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_out(ops->hooknum, skb, AF_INET6); } @@ -1769,9 +1765,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) */ static unsigned int ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_in(ops->hooknum, skb, AF_INET); } @@ -1782,8 +1776,7 @@ ip_vs_remote_request4(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_in(ops->hooknum, skb, AF_INET); } @@ -1796,9 +1789,7 @@ ip_vs_local_request4(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_in(ops->hooknum, skb, AF_INET6); } @@ -1809,8 +1800,7 @@ ip_vs_remote_request6(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return ip_vs_in(ops->hooknum, skb, AF_INET6); } @@ -1829,8 +1819,7 @@ ip_vs_local_request6(const struct nf_hook_ops *ops, struct sk_buff *skb, */ static unsigned int ip_vs_forward_icmp(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { int r; struct net *net; @@ -1851,8 +1840,7 @@ ip_vs_forward_icmp(const struct nf_hook_ops *ops, struct sk_buff *skb, #ifdef CONFIG_IP_VS_IPV6 static unsigned int ip_vs_forward_icmp_v6(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { int r; struct net *net; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index edc66de39f2e..7e392edaab97 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -4852,21 +4852,17 @@ static unsigned int selinux_ip_forward(struct sk_buff *skb, static unsigned int selinux_ipv4_forward(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return selinux_ip_forward(skb, in, PF_INET); + return selinux_ip_forward(skb, state->in, PF_INET); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) static unsigned int selinux_ipv6_forward(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return selinux_ip_forward(skb, in, PF_INET6); + return selinux_ip_forward(skb, state->in, PF_INET6); } #endif /* IPV6 */ @@ -4914,9 +4910,7 @@ static unsigned int selinux_ip_output(struct sk_buff *skb, static unsigned int selinux_ipv4_output(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { return selinux_ip_output(skb, PF_INET); } @@ -5091,21 +5085,17 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb, static unsigned int selinux_ipv4_postroute(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return selinux_ip_postroute(skb, out, PF_INET); + return selinux_ip_postroute(skb, state->out, PF_INET); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) static unsigned int selinux_ipv6_postroute(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { - return selinux_ip_postroute(skb, out, PF_INET6); + return selinux_ip_postroute(skb, state->out, PF_INET6); } #endif /* IPV6 */ diff --git a/security/smack/smack_netfilter.c b/security/smack/smack_netfilter.c index c952632afb0d..a455cfc9ec1f 100644 --- a/security/smack/smack_netfilter.c +++ b/security/smack/smack_netfilter.c @@ -23,9 +23,7 @@ static unsigned int smack_ipv6_output(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct socket_smack *ssp; struct smack_known *skp; @@ -42,9 +40,7 @@ static unsigned int smack_ipv6_output(const struct nf_hook_ops *ops, static unsigned int smack_ipv4_output(const struct nf_hook_ops *ops, struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) + const struct nf_hook_state *state) { struct socket_smack *ssp; struct smack_known *skp; -- cgit v1.2.3 From 1c491ba2592f621f21a693d43fab06302527fc0f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Apr 2015 20:56:08 -0400 Subject: netfilter: Pass nf_hook_state through ipt_do_table(). Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_tables.h | 3 +-- net/ipv4/netfilter/ip_tables.c | 13 ++++++------- net/ipv4/netfilter/iptable_filter.c | 3 +-- net/ipv4/netfilter/iptable_mangle.c | 11 ++++++----- net/ipv4/netfilter/iptable_nat.c | 3 +-- net/ipv4/netfilter/iptable_raw.c | 3 +-- net/ipv4/netfilter/iptable_security.c | 2 +- 7 files changed, 17 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 901e84db847d..4073510da485 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -65,8 +65,7 @@ struct ipt_error { extern void *ipt_alloc_initial_table(const struct xt_table *); extern unsigned int ipt_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table); #ifdef CONFIG_COMPAT diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index cf5e82f39d3b..c69db7fa25ee 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -288,8 +288,7 @@ struct ipt_entry *ipt_next_entry(const struct ipt_entry *entry) unsigned int ipt_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table) { static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); @@ -306,8 +305,8 @@ ipt_do_table(struct sk_buff *skb, /* Initialization */ ip = ip_hdr(skb); - indev = in ? in->name : nulldevname; - outdev = out ? out->name : nulldevname; + indev = state->in ? state->in->name : nulldevname; + outdev = state->out ? state->out->name : nulldevname; /* We handle fragments by dealing with the first fragment as * if it was a normal packet. All other fragments are treated * normally, except that they will NEVER match rules that ask @@ -317,8 +316,8 @@ ipt_do_table(struct sk_buff *skb, acpar.fragoff = ntohs(ip->frag_off) & IP_OFFSET; acpar.thoff = ip_hdrlen(skb); acpar.hotdrop = false; - acpar.in = in; - acpar.out = out; + acpar.in = state->in; + acpar.out = state->out; acpar.family = NFPROTO_IPV4; acpar.hooknum = hook; @@ -370,7 +369,7 @@ ipt_do_table(struct sk_buff *skb, #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) /* The packet is traced: log it */ if (unlikely(skb->nf_trace)) - trace_packet(skb, hook, in, out, + trace_packet(skb, hook, state->in, state->out, table->name, private, e); #endif /* Standard target? */ diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 1df0d42bfd39..a0f3beca52d2 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -45,8 +45,7 @@ iptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, return NF_ACCEPT; net = dev_net(state->in ? state->in : state->out); - return ipt_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv4.iptable_filter); + return ipt_do_table(skb, ops->hooknum, state, net->ipv4.iptable_filter); } static struct nf_hook_ops *filter_ops __read_mostly; diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 7a825e740045..62cbb8c5f4a8 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -37,8 +37,9 @@ static const struct xt_table packet_mangler = { }; static unsigned int -ipt_mangle_out(struct sk_buff *skb, const struct net_device *out) +ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state) { + struct net_device *out = state->out; unsigned int ret; const struct iphdr *iph; u_int8_t tos; @@ -58,7 +59,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct net_device *out) daddr = iph->daddr; tos = iph->tos; - ret = ipt_do_table(skb, NF_INET_LOCAL_OUT, NULL, out, + ret = ipt_do_table(skb, NF_INET_LOCAL_OUT, state, dev_net(out)->ipv4.iptable_mangle); /* Reroute for ANY change. */ if (ret != NF_DROP && ret != NF_STOLEN) { @@ -84,12 +85,12 @@ iptable_mangle_hook(const struct nf_hook_ops *ops, const struct nf_hook_state *state) { if (ops->hooknum == NF_INET_LOCAL_OUT) - return ipt_mangle_out(skb, state->out); + return ipt_mangle_out(skb, state); if (ops->hooknum == NF_INET_POST_ROUTING) - return ipt_do_table(skb, ops->hooknum, state->in, state->out, + return ipt_do_table(skb, ops->hooknum, state, dev_net(state->out)->ipv4.iptable_mangle); /* PREROUTING/INPUT/FORWARD: */ - return ipt_do_table(skb, ops->hooknum, state->in, state->out, + return ipt_do_table(skb, ops->hooknum, state, dev_net(state->in)->ipv4.iptable_mangle); } diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 086e2311438e..0d4d9cdf98a4 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -35,8 +35,7 @@ static unsigned int iptable_nat_do_chain(const struct nf_hook_ops *ops, { struct net *net = nf_ct_net(ct); - return ipt_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv4.nat_table); + return ipt_do_table(skb, ops->hooknum, state, net->ipv4.nat_table); } static unsigned int iptable_nat_ipv4_fn(const struct nf_hook_ops *ops, diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index fac8f607c70b..0356e6da4bb7 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -32,8 +32,7 @@ iptable_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, return NF_ACCEPT; net = dev_net(state->in ? state->in : state->out); - return ipt_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv4.iptable_raw); + return ipt_do_table(skb, ops->hooknum, state, net->ipv4.iptable_raw); } static struct nf_hook_ops *rawtable_ops __read_mostly; diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index d9ad60a57413..4bce3980ccd9 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -49,7 +49,7 @@ iptable_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, return NF_ACCEPT; net = dev_net(state->in ? state->in : state->out); - return ipt_do_table(skb, ops->hooknum, state->in, state->out, + return ipt_do_table(skb, ops->hooknum, state, net->ipv4.iptable_security); } -- cgit v1.2.3 From 8f8a37152df49d541c43f010543f2b0176fcfb8e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Apr 2015 21:09:51 -0400 Subject: netfilter: Pass nf_hook_state through ip6t_do_table(). Signed-off-by: David S. Miller --- include/linux/netfilter_ipv6/ip6_tables.h | 3 +-- net/ipv6/netfilter/ip6_tables.c | 13 ++++++------- net/ipv6/netfilter/ip6table_filter.c | 3 +-- net/ipv6/netfilter/ip6table_mangle.c | 12 ++++++------ net/ipv6/netfilter/ip6table_nat.c | 3 +-- net/ipv6/netfilter/ip6table_raw.c | 3 +-- net/ipv6/netfilter/ip6table_security.c | 2 +- 7 files changed, 17 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index 610208b18c05..b40d2b635778 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -31,8 +31,7 @@ extern struct xt_table *ip6t_register_table(struct net *net, extern void ip6t_unregister_table(struct net *net, struct xt_table *table); extern unsigned int ip6t_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table); /* Check for an extension */ diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 83f59dc3cccc..1a732a1d3c8e 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -317,8 +317,7 @@ ip6t_next_entry(const struct ip6t_entry *entry) unsigned int ip6t_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table) { static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); @@ -333,8 +332,8 @@ ip6t_do_table(struct sk_buff *skb, unsigned int addend; /* Initialization */ - indev = in ? in->name : nulldevname; - outdev = out ? out->name : nulldevname; + indev = state->in ? state->in->name : nulldevname; + outdev = state->out ? state->out->name : nulldevname; /* We handle fragments by dealing with the first fragment as * if it was a normal packet. All other fragments are treated * normally, except that they will NEVER match rules that ask @@ -342,8 +341,8 @@ ip6t_do_table(struct sk_buff *skb, * rule is also a fragment-specific rule, non-fragments won't * match it. */ acpar.hotdrop = false; - acpar.in = in; - acpar.out = out; + acpar.in = state->in; + acpar.out = state->out; acpar.family = NFPROTO_IPV6; acpar.hooknum = hook; @@ -393,7 +392,7 @@ ip6t_do_table(struct sk_buff *skb, #if IS_ENABLED(CONFIG_NETFILTER_XT_TARGET_TRACE) /* The packet is traced: log it */ if (unlikely(skb->nf_trace)) - trace_packet(skb, hook, in, out, + trace_packet(skb, hook, state->in, state->out, table->name, private, e); #endif /* Standard target? */ diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index eb9ef093454f..5c33d8abc077 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -37,8 +37,7 @@ ip6table_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, { const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv6.ip6table_filter); + return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_filter); } static struct nf_hook_ops *filter_ops __read_mostly; diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index e713b8d3dbbc..b551f5b79fe2 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -32,7 +32,7 @@ static const struct xt_table packet_mangler = { }; static unsigned int -ip6t_mangle_out(struct sk_buff *skb, const struct net_device *out) +ip6t_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state) { unsigned int ret; struct in6_addr saddr, daddr; @@ -57,8 +57,8 @@ ip6t_mangle_out(struct sk_buff *skb, const struct net_device *out) /* flowlabel and prio (includes version, which shouldn't change either */ flowlabel = *((u_int32_t *)ipv6_hdr(skb)); - ret = ip6t_do_table(skb, NF_INET_LOCAL_OUT, NULL, out, - dev_net(out)->ipv6.ip6table_mangle); + ret = ip6t_do_table(skb, NF_INET_LOCAL_OUT, state, + dev_net(state->out)->ipv6.ip6table_mangle); if (ret != NF_DROP && ret != NF_STOLEN && (!ipv6_addr_equal(&ipv6_hdr(skb)->saddr, &saddr) || @@ -80,12 +80,12 @@ ip6table_mangle_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, const struct nf_hook_state *state) { if (ops->hooknum == NF_INET_LOCAL_OUT) - return ip6t_mangle_out(skb, state->out); + return ip6t_mangle_out(skb, state); if (ops->hooknum == NF_INET_POST_ROUTING) - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, + return ip6t_do_table(skb, ops->hooknum, state, dev_net(state->out)->ipv6.ip6table_mangle); /* INPUT/FORWARD */ - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, + return ip6t_do_table(skb, ops->hooknum, state, dev_net(state->in)->ipv6.ip6table_mangle); } diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index d78f69c7abce..c3a7f7af0ed4 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -37,8 +37,7 @@ static unsigned int ip6table_nat_do_chain(const struct nf_hook_ops *ops, { struct net *net = nf_ct_net(ct); - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv6.ip6table_nat); + return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_nat); } static unsigned int ip6table_nat_fn(const struct nf_hook_ops *ops, diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 937908e25862..0b33caad2b69 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -24,8 +24,7 @@ ip6table_raw_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, { const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, - net->ipv6.ip6table_raw); + return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_raw); } static struct nf_hook_ops *rawtable_ops __read_mostly; diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index f33b41e8e294..fcef83c25f7b 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -41,7 +41,7 @@ ip6table_security_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, { const struct net *net = dev_net(state->in ? state->in : state->out); - return ip6t_do_table(skb, ops->hooknum, state->in, state->out, + return ip6t_do_table(skb, ops->hooknum, state, net->ipv6.ip6table_security); } -- cgit v1.2.3 From b85c3dc9bd5347ad9540ec8d103b7c049c48b7cd Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Apr 2015 21:18:46 -0400 Subject: netfilter: Pass nf_hook_state through arpt_do_table(). Signed-off-by: David S. Miller --- include/linux/netfilter_arp/arp_tables.h | 3 +-- net/ipv4/netfilter/arp_tables.c | 11 +++++------ net/ipv4/netfilter/arptable_filter.c | 2 +- 3 files changed, 7 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index cfb7191e6efa..c22a7fb8d0df 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -54,8 +54,7 @@ extern struct xt_table *arpt_register_table(struct net *net, extern void arpt_unregister_table(struct xt_table *table); extern unsigned int arpt_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table); #ifdef CONFIG_COMPAT diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index f95b6f93814b..13bfe84bf3ca 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -248,8 +248,7 @@ struct arpt_entry *arpt_next_entry(const struct arpt_entry *entry) unsigned int arpt_do_table(struct sk_buff *skb, unsigned int hook, - const struct net_device *in, - const struct net_device *out, + const struct nf_hook_state *state, struct xt_table *table) { static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); @@ -265,8 +264,8 @@ unsigned int arpt_do_table(struct sk_buff *skb, if (!pskb_may_pull(skb, arp_hdr_len(skb->dev))) return NF_DROP; - indev = in ? in->name : nulldevname; - outdev = out ? out->name : nulldevname; + indev = state->in ? state->in->name : nulldevname; + outdev = state->out ? state->out->name : nulldevname; local_bh_disable(); addend = xt_write_recseq_begin(); @@ -281,8 +280,8 @@ unsigned int arpt_do_table(struct sk_buff *skb, e = get_entry(table_base, private->hook_entry[hook]); back = get_entry(table_base, private->underflow[hook]); - acpar.in = in; - acpar.out = out; + acpar.in = state->in; + acpar.out = state->out; acpar.hooknum = hook; acpar.family = NFPROTO_ARP; acpar.hotdrop = false; diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 6a641cb41062..93876d03120c 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -32,7 +32,7 @@ arptable_filter_hook(const struct nf_hook_ops *ops, struct sk_buff *skb, { const struct net *net = dev_net(state->in ? state->in : state->out); - return arpt_do_table(skb, ops->hooknum, state->in, state->out, + return arpt_do_table(skb, ops->hooknum, state, net->ipv4.arptable_filter); } -- cgit v1.2.3 From f60e5990d9c1424af9dbca60a23ba2a1c7c1ce90 Mon Sep 17 00:00:00 2001 From: "hannes@stressinduktion.org" Date: Wed, 1 Apr 2015 17:07:44 +0200 Subject: ipv6: protect skb->sk accesses from recursive dereference inside the stack We should not consult skb->sk for output decisions in xmit recursion levels > 0 in the stack. Otherwise local socket settings could influence the result of e.g. tunnel encapsulation process. ipv6 does not conform with this in three places: 1) ip6_fragment: we do consult ipv6_npinfo for frag_size 2) sk_mc_loop in ipv6 uses skb->sk and checks if we should loop the packet back to the local socket 3) ip6_skb_dst_mtu could query the settings from the user socket and force a wrong MTU Furthermore: In sk_mc_loop we could potentially land in WARN_ON(1) if we use a PF_PACKET socket ontop of an IPv6-backed vxlan device. Reuse xmit_recursion as we are currently only interested in protecting tunnel devices. Cc: Jiri Pirko Signed-off-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++++ include/net/ip.h | 16 ---------------- include/net/ip6_route.h | 3 ++- include/net/sock.h | 2 ++ net/core/dev.c | 4 +++- net/core/sock.c | 19 +++++++++++++++++++ net/ipv6/ip6_output.c | 3 ++- 7 files changed, 34 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index dcf6ec27739b..278738873703 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2185,6 +2185,12 @@ void netdev_freemem(struct net_device *dev); void synchronize_net(void); int init_dummy_netdev(struct net_device *dev); +DECLARE_PER_CPU(int, xmit_recursion); +static inline int dev_recursion_level(void) +{ + return this_cpu_read(xmit_recursion); +} + struct net_device *dev_get_by_index(struct net *net, int ifindex); struct net_device *__dev_get_by_index(struct net *net, int ifindex); struct net_device *dev_get_by_index_rcu(struct net *net, int ifindex); diff --git a/include/net/ip.h b/include/net/ip.h index 025c61c0dffb..6cc1eafb153a 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -453,22 +453,6 @@ static __inline__ void inet_reset_saddr(struct sock *sk) #endif -static inline int sk_mc_loop(struct sock *sk) -{ - if (!sk) - return 1; - switch (sk->sk_family) { - case AF_INET: - return inet_sk(sk)->mc_loop; -#if IS_ENABLED(CONFIG_IPV6) - case AF_INET6: - return inet6_sk(sk)->mc_loop; -#endif - } - WARN_ON(1); - return 1; -} - bool ip_call_ra_chain(struct sk_buff *skb); /* diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 1d09b46c1e48..eda131d179d9 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -174,7 +174,8 @@ int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); static inline int ip6_skb_dst_mtu(struct sk_buff *skb) { - struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; + struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? + inet6_sk(skb->sk) : NULL; return (np && np->pmtudisc >= IPV6_PMTUDISC_PROBE) ? skb_dst(skb)->dev->mtu : dst_mtu(skb_dst(skb)); diff --git a/include/net/sock.h b/include/net/sock.h index ab186b1d31ff..e4079c28e6b8 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1762,6 +1762,8 @@ struct dst_entry *__sk_dst_check(struct sock *sk, u32 cookie); struct dst_entry *sk_dst_check(struct sock *sk, u32 cookie); +bool sk_mc_loop(struct sock *sk); + static inline bool sk_can_gso(const struct sock *sk) { return net_gso_ok(sk->sk_route_caps, sk->sk_gso_type); diff --git a/net/core/dev.c b/net/core/dev.c index 962ee9d71964..45109b70664e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2848,7 +2848,9 @@ static void skb_update_prio(struct sk_buff *skb) #define skb_update_prio(skb) #endif -static DEFINE_PER_CPU(int, xmit_recursion); +DEFINE_PER_CPU(int, xmit_recursion); +EXPORT_SYMBOL(xmit_recursion); + #define RECURSION_LIMIT 10 /** diff --git a/net/core/sock.c b/net/core/sock.c index 78e89eb7eb70..71e3e5f1eaa0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -653,6 +653,25 @@ static inline void sock_valbool_flag(struct sock *sk, int bit, int valbool) sock_reset_flag(sk, bit); } +bool sk_mc_loop(struct sock *sk) +{ + if (dev_recursion_level()) + return false; + if (!sk) + return true; + switch (sk->sk_family) { + case AF_INET: + return inet_sk(sk)->mc_loop; +#if IS_ENABLED(CONFIG_IPV6) + case AF_INET6: + return inet6_sk(sk)->mc_loop; +#endif + } + WARN_ON(1); + return true; +} +EXPORT_SYMBOL(sk_mc_loop); + /* * This is meant for all protocols to use and covers goings on * at the socket level. Everything here is generic. diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 7e80b61b51ff..36cf0ab685a0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -542,7 +542,8 @@ int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) { struct sk_buff *frag; struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); - struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL; + struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? + inet6_sk(skb->sk) : NULL; struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; unsigned int mtu, hlen, left, len; -- cgit v1.2.3 From 107a9f4dc9211c1f91703d1739d7fd22ac58b332 Mon Sep 17 00:00:00 2001 From: David Miller Date: Sun, 5 Apr 2015 22:18:54 -0400 Subject: netfilter: Add nf_hook_state initializer function. This way we can consolidate where we setup new nf_hook_state objects, to make sure the entire thing is initialized. The only other place an nf_hook_object is instantiated is nf_queue, wherein a structure copy is used. Signed-off-by: David S. Miller --- include/linux/netfilter.h | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index c480c43ad8f7..b8c88f3c85ff 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -54,6 +54,21 @@ struct nf_hook_state { int (*okfn)(struct sk_buff *); }; +static inline void nf_hook_state_init(struct nf_hook_state *p, + unsigned int hook, + int thresh, u_int8_t pf, + struct net_device *indev, + struct net_device *outdev, + int (*okfn)(struct sk_buff *)) +{ + p->hook = hook; + p->thresh = thresh; + p->pf = pf; + p->in = indev; + p->out = outdev; + p->okfn = okfn; +} + typedef unsigned int nf_hookfn(const struct nf_hook_ops *ops, struct sk_buff *skb, const struct nf_hook_state *state); @@ -142,15 +157,10 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, int (*okfn)(struct sk_buff *), int thresh) { if (nf_hooks_active(pf, hook)) { - struct nf_hook_state state = { - .hook = hook, - .thresh = thresh, - .pf = pf, - .in = indev, - .out = outdev, - .okfn = okfn - }; + struct nf_hook_state state; + nf_hook_state_init(&state, hook, thresh, pf, + indev, outdev, okfn); return nf_hook_slow(skb, &state); } return 1; -- cgit v1.2.3 From 1c984f8a5df085bcf35364a8a870bd4db4da4ed3 Mon Sep 17 00:00:00 2001 From: David Miller Date: Sun, 5 Apr 2015 22:19:00 -0400 Subject: netfilter: Add socket pointer to nf_hook_state. It is currently always set to NULL, but nf_queue is adjusted to be prepared for it being set to a real socket by taking and releasing a reference to that socket when necessary. Signed-off-by: David S. Miller --- include/linux/netfilter.h | 7 ++++++- net/netfilter/nf_queue.c | 4 ++++ 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index b8c88f3c85ff..f8f58fab2402 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -45,12 +45,15 @@ struct sk_buff; struct nf_hook_ops; +struct sock; + struct nf_hook_state { unsigned int hook; int thresh; u_int8_t pf; struct net_device *in; struct net_device *out; + struct sock *sk; int (*okfn)(struct sk_buff *); }; @@ -59,6 +62,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p, int thresh, u_int8_t pf, struct net_device *indev, struct net_device *outdev, + struct sock *sk, int (*okfn)(struct sk_buff *)) { p->hook = hook; @@ -66,6 +70,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p, p->pf = pf; p->in = indev; p->out = outdev; + p->sk = sk; p->okfn = okfn; } @@ -160,7 +165,7 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, struct nf_hook_state state; nf_hook_state_init(&state, hook, thresh, pf, - indev, outdev, okfn); + indev, outdev, NULL, okfn); return nf_hook_slow(skb, &state); } return 1; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index d3cd37edca18..c4a706678f88 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -54,6 +54,8 @@ void nf_queue_entry_release_refs(struct nf_queue_entry *entry) dev_put(state->in); if (state->out) dev_put(state->out); + if (state->sk) + sock_put(state->sk); #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) if (entry->skb->nf_bridge) { struct nf_bridge_info *nf_bridge = entry->skb->nf_bridge; @@ -81,6 +83,8 @@ bool nf_queue_entry_get_refs(struct nf_queue_entry *entry) dev_hold(state->in); if (state->out) dev_hold(state->out); + if (state->sk) + sock_hold(state->sk); #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) if (entry->skb->nf_bridge) { struct nf_bridge_info *nf_bridge = entry->skb->nf_bridge; -- cgit v1.2.3 From 7026b1ddb6b8d4e6ee33dc2bd06c0ca8746fa7ab Mon Sep 17 00:00:00 2001 From: David Miller Date: Sun, 5 Apr 2015 22:19:04 -0400 Subject: netfilter: Pass socket pointer down through okfn(). On the output paths in particular, we have to sometimes deal with two socket contexts. First, and usually skb->sk, is the local socket that generated the frame. And second, is potentially the socket used to control a tunneling socket, such as one the encapsulates using UDP. We do not want to disassociate skb->sk when encapsulating in order to fix this, because that would break socket memory accounting. The most extreme case where this can cause huge problems is an AF_PACKET socket transmitting over a vxlan device. We hit code paths doing checks that assume they are dealing with an ipv4 socket, but are actually operating upon the AF_PACKET one. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 14 +++++-- include/linux/netfilter.h | 62 +++++++++++++++++-------------- include/linux/netfilter_bridge.h | 2 +- include/net/dn_neigh.h | 6 +-- include/net/ip.h | 3 +- include/net/ip6_route.h | 3 +- include/net/ipv6.h | 2 +- include/net/xfrm.h | 8 ++-- net/bridge/br_forward.c | 13 ++++--- net/bridge/br_input.c | 16 ++++---- net/bridge/br_multicast.c | 3 +- net/bridge/br_netfilter.c | 56 +++++++++++++++------------- net/bridge/br_private.h | 6 +-- net/bridge/br_stp_bpdu.c | 5 ++- net/core/dev.c | 10 ++--- net/decnet/dn_neigh.c | 35 +++++++++-------- net/decnet/dn_nsp_in.c | 5 ++- net/decnet/dn_route.c | 26 +++++++------ net/ipv4/arp.c | 10 +++-- net/ipv4/ip_forward.c | 8 ++-- net/ipv4/ip_input.c | 10 +++-- net/ipv4/ip_output.c | 45 ++++++++++++---------- net/ipv4/ipmr.c | 7 ++-- net/ipv4/raw.c | 4 +- net/ipv4/xfrm4_input.c | 5 ++- net/ipv4/xfrm4_output.c | 12 +++--- net/ipv6/ip6_input.c | 11 +++--- net/ipv6/ip6_output.c | 33 ++++++++-------- net/ipv6/ip6mr.c | 7 ++-- net/ipv6/mcast.c | 9 +++-- net/ipv6/ndisc.c | 5 ++- net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 2 +- net/ipv6/output_core.c | 4 +- net/ipv6/raw.c | 4 +- net/ipv6/xfrm6_input.c | 3 +- net/ipv6/xfrm6_output.c | 15 ++++---- net/netfilter/ipvs/ip_vs_xmit.c | 8 ++-- net/netfilter/nf_queue.c | 2 +- net/xfrm/xfrm_output.c | 16 ++++---- 39 files changed, 277 insertions(+), 218 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 41bf58a2b936..45823db2efb0 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2165,8 +2165,12 @@ int dev_open(struct net_device *dev); int dev_close(struct net_device *dev); int dev_close_many(struct list_head *head, bool unlink); void dev_disable_lro(struct net_device *dev); -int dev_loopback_xmit(struct sk_buff *newskb); -int dev_queue_xmit(struct sk_buff *skb); +int dev_loopback_xmit(struct sock *sk, struct sk_buff *newskb); +int dev_queue_xmit_sk(struct sock *sk, struct sk_buff *skb); +static inline int dev_queue_xmit(struct sk_buff *skb) +{ + return dev_queue_xmit_sk(skb->sk, skb); +} int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv); int register_netdevice(struct net_device *dev); void unregister_netdevice_queue(struct net_device *dev, struct list_head *head); @@ -2927,7 +2931,11 @@ static inline void dev_consume_skb_any(struct sk_buff *skb) int netif_rx(struct sk_buff *skb); int netif_rx_ni(struct sk_buff *skb); -int netif_receive_skb(struct sk_buff *skb); +int netif_receive_skb_sk(struct sock *sk, struct sk_buff *skb); +static inline int netif_receive_skb(struct sk_buff *skb) +{ + return netif_receive_skb_sk(skb->sk, skb); +} gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); void napi_gro_flush(struct napi_struct *napi, bool flush_old); struct sk_buff *napi_get_frags(struct napi_struct *napi); diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index f8f58fab2402..63560d0a8dfe 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -54,7 +54,7 @@ struct nf_hook_state { struct net_device *in; struct net_device *out; struct sock *sk; - int (*okfn)(struct sk_buff *); + int (*okfn)(struct sock *, struct sk_buff *); }; static inline void nf_hook_state_init(struct nf_hook_state *p, @@ -63,7 +63,7 @@ static inline void nf_hook_state_init(struct nf_hook_state *p, struct net_device *indev, struct net_device *outdev, struct sock *sk, - int (*okfn)(struct sk_buff *)) + int (*okfn)(struct sock *, struct sk_buff *)) { p->hook = hook; p->thresh = thresh; @@ -156,26 +156,29 @@ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state); * value indicates the packet has been consumed by the hook. */ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, + struct sock *sk, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh) + int (*okfn)(struct sock *, struct sk_buff *), + int thresh) { if (nf_hooks_active(pf, hook)) { struct nf_hook_state state; nf_hook_state_init(&state, hook, thresh, pf, - indev, outdev, NULL, okfn); + indev, outdev, sk, okfn); return nf_hook_slow(skb, &state); } return 1; } -static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *)) +static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk, + struct sk_buff *skb, struct net_device *indev, + struct net_device *outdev, + int (*okfn)(struct sock *, struct sk_buff *)) { - return nf_hook_thresh(pf, hook, skb, indev, outdev, okfn, INT_MIN); + return nf_hook_thresh(pf, hook, sk, skb, indev, outdev, okfn, INT_MIN); } /* Activate hook; either okfn or kfree_skb called, unless a hook @@ -196,35 +199,36 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, */ static inline int -NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *in, struct net_device *out, - int (*okfn)(struct sk_buff *), int thresh) +NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sock *sk, + struct sk_buff *skb, struct net_device *in, + struct net_device *out, + int (*okfn)(struct sock *, struct sk_buff *), int thresh) { - int ret = nf_hook_thresh(pf, hook, skb, in, out, okfn, thresh); + int ret = nf_hook_thresh(pf, hook, sk, skb, in, out, okfn, thresh); if (ret == 1) - ret = okfn(skb); + ret = okfn(sk, skb); return ret; } static inline int -NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *in, struct net_device *out, - int (*okfn)(struct sk_buff *), bool cond) +NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sock *sk, + struct sk_buff *skb, struct net_device *in, struct net_device *out, + int (*okfn)(struct sock *, struct sk_buff *), bool cond) { int ret; if (!cond || - ((ret = nf_hook_thresh(pf, hook, skb, in, out, okfn, INT_MIN)) == 1)) - ret = okfn(skb); + ((ret = nf_hook_thresh(pf, hook, sk, skb, in, out, okfn, INT_MIN)) == 1)) + ret = okfn(sk, skb); return ret; } static inline int -NF_HOOK(uint8_t pf, unsigned int hook, struct sk_buff *skb, +NF_HOOK(uint8_t pf, unsigned int hook, struct sock *sk, struct sk_buff *skb, struct net_device *in, struct net_device *out, - int (*okfn)(struct sk_buff *)) + int (*okfn)(struct sock *, struct sk_buff *)) { - return NF_HOOK_THRESH(pf, hook, skb, in, out, okfn, INT_MIN); + return NF_HOOK_THRESH(pf, hook, sk, skb, in, out, okfn, INT_MIN); } /* Call setsockopt() */ @@ -324,19 +328,21 @@ nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, u_int8_t family) } #else /* !CONFIG_NETFILTER */ -#define NF_HOOK(pf, hook, skb, indev, outdev, okfn) (okfn)(skb) -#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) (okfn)(skb) +#define NF_HOOK(pf, hook, sk, skb, indev, outdev, okfn) (okfn)(sk, skb) +#define NF_HOOK_COND(pf, hook, sk, skb, indev, outdev, okfn, cond) (okfn)(sk, skb) static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, + struct sock *sk, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh) + int (*okfn)(struct sock *sk, struct sk_buff *), int thresh) { - return okfn(skb); + return okfn(sk, skb); } -static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, - struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *)) +static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sock *sk, + struct sk_buff *skb, struct net_device *indev, + struct net_device *outdev, + int (*okfn)(struct sock *, struct sk_buff *)) { return 1; } diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index 2734977199ca..5fc0a0fe244b 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -30,7 +30,7 @@ static inline unsigned int nf_bridge_mtu_reduction(const struct sk_buff *skb) return 0; } -int br_handle_frame_finish(struct sk_buff *skb); +int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb); static inline void br_drop_fake_rtable(struct sk_buff *skb) { diff --git a/include/net/dn_neigh.h b/include/net/dn_neigh.h index 0f26aa707e62..d0424269313f 100644 --- a/include/net/dn_neigh.h +++ b/include/net/dn_neigh.h @@ -18,11 +18,11 @@ struct dn_neigh { void dn_neigh_init(void); void dn_neigh_cleanup(void); -int dn_neigh_router_hello(struct sk_buff *skb); -int dn_neigh_endnode_hello(struct sk_buff *skb); +int dn_neigh_router_hello(struct sock *sk, struct sk_buff *skb); +int dn_neigh_endnode_hello(struct sock *sk, struct sk_buff *skb); void dn_neigh_pointopoint_hello(struct sk_buff *skb); int dn_neigh_elist(struct net_device *dev, unsigned char *ptr, int n); -int dn_to_neigh_output(struct sk_buff *skb); +int dn_to_neigh_output(struct sock *sk, struct sk_buff *skb); extern struct neigh_table dn_neigh_table; diff --git a/include/net/ip.h b/include/net/ip.h index 69cd9cb8400c..d14af7edd197 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -108,7 +108,8 @@ int ip_local_deliver(struct sk_buff *skb); int ip_mr_input(struct sk_buff *skb); int ip_output(struct sock *sk, struct sk_buff *skb); int ip_mc_output(struct sock *sk, struct sk_buff *skb); -int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); +int ip_fragment(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)); int ip_do_nat(struct sk_buff *skb); void ip_send_check(struct iphdr *ip); int __ip_local_out(struct sk_buff *skb); diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index eda131d179d9..5e192068e6cb 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -170,7 +170,8 @@ static inline bool ipv6_anycast_destination(const struct sk_buff *skb) return rt->rt6i_flags & RTF_ANYCAST; } -int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); +int ip6_fragment(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)); static inline int ip6_skb_dst_mtu(struct sk_buff *skb) { diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 65142e6af440..b6ae959824ff 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -769,7 +769,7 @@ static inline u8 ip6_tclass(__be32 flowinfo) int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev); -int ip6_rcv_finish(struct sk_buff *skb); +int ip6_rcv_finish(struct sock *sk, struct sk_buff *skb); /* * upper-layer output functions diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 461f83539493..36ac102c97c7 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -332,7 +332,7 @@ struct xfrm_state_afinfo { int (*tmpl_sort)(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n); int (*state_sort)(struct xfrm_state **dst, struct xfrm_state **src, int n); int (*output)(struct sock *sk, struct sk_buff *skb); - int (*output_finish)(struct sk_buff *skb); + int (*output_finish)(struct sock *sk, struct sk_buff *skb); int (*extract_input)(struct xfrm_state *x, struct sk_buff *skb); int (*extract_output)(struct xfrm_state *x, @@ -1503,7 +1503,7 @@ int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb); int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type); int xfrm_input_resume(struct sk_buff *skb, int nexthdr); int xfrm_output_resume(struct sk_buff *skb, int err); -int xfrm_output(struct sk_buff *skb); +int xfrm_output(struct sock *sk, struct sk_buff *skb); int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb); void xfrm_local_error(struct sk_buff *skb, int mtu); int xfrm4_extract_header(struct sk_buff *skb); @@ -1524,7 +1524,7 @@ static inline int xfrm4_rcv_spi(struct sk_buff *skb, int nexthdr, __be32 spi) int xfrm4_extract_output(struct xfrm_state *x, struct sk_buff *skb); int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb); int xfrm4_output(struct sock *sk, struct sk_buff *skb); -int xfrm4_output_finish(struct sk_buff *skb); +int xfrm4_output_finish(struct sock *sk, struct sk_buff *skb); int xfrm4_rcv_cb(struct sk_buff *skb, u8 protocol, int err); int xfrm4_protocol_register(struct xfrm4_protocol *handler, unsigned char protocol); int xfrm4_protocol_deregister(struct xfrm4_protocol *handler, unsigned char protocol); @@ -1549,7 +1549,7 @@ __be32 xfrm6_tunnel_spi_lookup(struct net *net, const xfrm_address_t *saddr); int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb); int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb); int xfrm6_output(struct sock *sk, struct sk_buff *skb); -int xfrm6_output_finish(struct sk_buff *skb); +int xfrm6_output_finish(struct sock *sk, struct sk_buff *skb); int xfrm6_find_1stfragopt(struct xfrm_state *x, struct sk_buff *skb, u8 **prevhdr); diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 3304a5442331..e97572b5d2cc 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -35,7 +35,7 @@ static inline int should_deliver(const struct net_bridge_port *p, p->state == BR_STATE_FORWARDING; } -int br_dev_queue_push_xmit(struct sk_buff *skb) +int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb) { if (!is_skb_forwardable(skb->dev, skb)) { kfree_skb(skb); @@ -49,9 +49,10 @@ int br_dev_queue_push_xmit(struct sk_buff *skb) } EXPORT_SYMBOL_GPL(br_dev_queue_push_xmit); -int br_forward_finish(struct sk_buff *skb) +int br_forward_finish(struct sock *sk, struct sk_buff *skb) { - return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, skb, NULL, skb->dev, + return NF_HOOK(NFPROTO_BRIDGE, NF_BR_POST_ROUTING, sk, skb, + NULL, skb->dev, br_dev_queue_push_xmit); } @@ -75,7 +76,8 @@ static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) return; } - NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev, + NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb, + NULL, skb->dev, br_forward_finish); } @@ -96,7 +98,8 @@ static void __br_forward(const struct net_bridge_port *to, struct sk_buff *skb) skb->dev = to->dev; skb_forward_csum(skb); - NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, skb, indev, skb->dev, + NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, NULL, skb, + indev, skb->dev, br_forward_finish); } diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 052c5ebbc947..f921a5dce22d 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -55,8 +55,9 @@ static int br_pass_frame_up(struct sk_buff *skb) if (!skb) return NET_RX_DROP; - return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, skb, indev, NULL, - netif_receive_skb); + return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb, + indev, NULL, + netif_receive_skb_sk); } static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, @@ -119,7 +120,7 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, } /* note: already called with rcu_read_lock */ -int br_handle_frame_finish(struct sk_buff *skb) +int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb) { const unsigned char *dest = eth_hdr(skb)->h_dest; struct net_bridge_port *p = br_port_get_rcu(skb->dev); @@ -207,7 +208,7 @@ drop: EXPORT_SYMBOL_GPL(br_handle_frame_finish); /* note: already called with rcu_read_lock */ -static int br_handle_local_finish(struct sk_buff *skb) +static int br_handle_local_finish(struct sock *sk, struct sk_buff *skb) { struct net_bridge_port *p = br_port_get_rcu(skb->dev); u16 vid = 0; @@ -277,8 +278,8 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb) } /* Deliver packet to local host only */ - if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, skb, skb->dev, - NULL, br_handle_local_finish)) { + if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, NULL, skb, + skb->dev, NULL, br_handle_local_finish)) { return RX_HANDLER_CONSUMED; /* consumed by filter */ } else { *pskb = skb; @@ -302,7 +303,8 @@ forward: if (ether_addr_equal(p->br->dev->dev_addr, dest)) skb->pkt_type = PACKET_HOST; - NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, NULL, skb, + skb->dev, NULL, br_handle_frame_finish); break; default: diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index c465876c7861..4b6722f8f179 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -814,7 +814,8 @@ static void __br_multicast_send_query(struct net_bridge *br, if (port) { skb->dev = port->dev; - NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev, + NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb, + NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 7527e94dd5dc..acd31c9f2116 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -261,7 +261,7 @@ static void nf_bridge_update_protocol(struct sk_buff *skb) /* PF_BRIDGE/PRE_ROUTING *********************************************/ /* Undo the changes made for ip6tables PREROUTING and continue the * bridge PRE_ROUTING hook. */ -static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) +static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; struct rtable *rt; @@ -282,7 +282,8 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) skb->dev = nf_bridge->physindev; nf_bridge_update_protocol(skb); nf_bridge_push_encap_header(skb); - NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb, + skb->dev, NULL, br_handle_frame_finish, 1); return 0; @@ -293,7 +294,7 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) * don't, we use the neighbour framework to find out. In both cases, we make * sure that br_handle_frame_finish() is called afterwards. */ -static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb) +static int br_nf_pre_routing_finish_bridge(struct sock *sk, struct sk_buff *skb) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; struct neighbour *neigh; @@ -310,7 +311,7 @@ static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb) if (neigh->hh.hh_len) { neigh_hh_bridge(&neigh->hh, skb); skb->dev = nf_bridge->physindev; - ret = br_handle_frame_finish(skb); + ret = br_handle_frame_finish(sk, skb); } else { /* the neighbour function below overwrites the complete * MAC header, so we save the Ethernet source address and @@ -387,7 +388,7 @@ static bool dnat_took_place(const struct sk_buff *skb) * device, we proceed as if ip_route_input() succeeded. If it differs from the * logical bridge port or if ip_route_output_key() fails we drop the packet. */ -static int br_nf_pre_routing_finish(struct sk_buff *skb) +static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb) { struct net_device *dev = skb->dev; struct iphdr *iph = ip_hdr(skb); @@ -440,7 +441,7 @@ bridged_dnat: nf_bridge_push_encap_header(skb); NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, - skb, skb->dev, NULL, + sk, skb, skb->dev, NULL, br_nf_pre_routing_finish_bridge, 1); return 0; @@ -460,7 +461,8 @@ bridged_dnat: skb->dev = nf_bridge->physindev; nf_bridge_update_protocol(skb); nf_bridge_push_encap_header(skb); - NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb, + skb->dev, NULL, br_handle_frame_finish, 1); return 0; @@ -596,7 +598,8 @@ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops, return NF_DROP; skb->protocol = htons(ETH_P_IPV6); - NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, state->sk, skb, + skb->dev, NULL, br_nf_pre_routing_finish_ipv6); return NF_STOLEN; @@ -651,7 +654,8 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, skb->protocol = htons(ETH_P_IP); - NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, state->sk, skb, + skb->dev, NULL, br_nf_pre_routing_finish); return NF_STOLEN; @@ -674,7 +678,7 @@ static unsigned int br_nf_local_in(const struct nf_hook_ops *ops, } /* PF_BRIDGE/FORWARD *************************************************/ -static int br_nf_forward_finish(struct sk_buff *skb) +static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb) { struct nf_bridge_info *nf_bridge = skb->nf_bridge; struct net_device *in; @@ -691,8 +695,8 @@ static int br_nf_forward_finish(struct sk_buff *skb) } nf_bridge_push_encap_header(skb); - NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, skb, in, - skb->dev, br_forward_finish, 1); + NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_FORWARD, sk, skb, + in, skb->dev, br_forward_finish, 1); return 0; } @@ -746,7 +750,8 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops, else skb->protocol = htons(ETH_P_IPV6); - NF_HOOK(pf, NF_INET_FORWARD, skb, brnf_get_logical_dev(skb, state->in), + NF_HOOK(pf, NF_INET_FORWARD, NULL, skb, + brnf_get_logical_dev(skb, state->in), parent, br_nf_forward_finish); return NF_STOLEN; @@ -780,8 +785,8 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops, return NF_ACCEPT; } *d = state->in; - NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, skb, state->in, - state->out, br_nf_forward_finish); + NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, state->sk, skb, + state->in, state->out, br_nf_forward_finish); return NF_STOLEN; } @@ -804,24 +809,24 @@ static bool nf_bridge_copy_header(struct sk_buff *skb) return true; } -static int br_nf_push_frag_xmit(struct sk_buff *skb) +static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb) { if (!nf_bridge_copy_header(skb)) { kfree_skb(skb); return 0; } - return br_dev_queue_push_xmit(skb); + return br_dev_queue_push_xmit(sk, skb); } -static int br_nf_dev_queue_xmit(struct sk_buff *skb) +static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb) { int ret; int frag_max_size; unsigned int mtu_reserved; if (skb_is_gso(skb) || skb->protocol != htons(ETH_P_IP)) - return br_dev_queue_push_xmit(skb); + return br_dev_queue_push_xmit(sk, skb); mtu_reserved = nf_bridge_mtu_reduction(skb); /* This is wrong! We should preserve the original fragment @@ -833,16 +838,16 @@ static int br_nf_dev_queue_xmit(struct sk_buff *skb) /* Drop invalid packet */ return NF_DROP; IPCB(skb)->frag_max_size = frag_max_size; - ret = ip_fragment(skb, br_nf_push_frag_xmit); + ret = ip_fragment(sk, skb, br_nf_push_frag_xmit); } else - ret = br_dev_queue_push_xmit(skb); + ret = br_dev_queue_push_xmit(sk, skb); return ret; } #else -static int br_nf_dev_queue_xmit(struct sk_buff *skb) +static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb) { - return br_dev_queue_push_xmit(skb); + return br_dev_queue_push_xmit(sk, skb); } #endif @@ -887,7 +892,8 @@ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops, else skb->protocol = htons(ETH_P_IPV6); - NF_HOOK(pf, NF_INET_POST_ROUTING, skb, NULL, realoutdev, + NF_HOOK(pf, NF_INET_POST_ROUTING, state->sk, skb, + NULL, realoutdev, br_nf_dev_queue_xmit); return NF_STOLEN; @@ -927,7 +933,7 @@ static void br_nf_pre_routing_finish_bridge_slow(struct sk_buff *skb) skb_copy_to_linear_data_offset(skb, -(ETH_HLEN-ETH_ALEN), skb->nf_bridge->data, ETH_HLEN-ETH_ALEN); skb->dev = nf_bridge->physindev; - br_handle_frame_finish(skb); + br_handle_frame_finish(NULL, skb); } static int br_nf_dev_xmit(struct sk_buff *skb) diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index b46fa0c5b8ec..6ca0251cb478 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -410,10 +410,10 @@ int br_fdb_external_learn_del(struct net_bridge *br, struct net_bridge_port *p, /* br_forward.c */ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb); -int br_dev_queue_push_xmit(struct sk_buff *skb); +int br_dev_queue_push_xmit(struct sock *sk, struct sk_buff *skb); void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, struct sk_buff *skb0); -int br_forward_finish(struct sk_buff *skb); +int br_forward_finish(struct sock *sk, struct sk_buff *skb); void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast); void br_flood_forward(struct net_bridge *br, struct sk_buff *skb, struct sk_buff *skb2, bool unicast); @@ -431,7 +431,7 @@ void br_port_flags_change(struct net_bridge_port *port, unsigned long mask); void br_manage_promisc(struct net_bridge *br); /* br_input.c */ -int br_handle_frame_finish(struct sk_buff *skb); +int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb); rx_handler_result_t br_handle_frame(struct sk_buff **pskb); static inline bool br_rx_handler_check_rcu(const struct net_device *dev) diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c index bdb459d21ad8..534fc4cd263e 100644 --- a/net/bridge/br_stp_bpdu.c +++ b/net/bridge/br_stp_bpdu.c @@ -54,8 +54,9 @@ static void br_send_bpdu(struct net_bridge_port *p, skb_reset_mac_header(skb); - NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev, - dev_queue_xmit); + NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, NULL, skb, + NULL, skb->dev, + dev_queue_xmit_sk); } static inline void br_set_ticks(unsigned char *dest, int j) diff --git a/net/core/dev.c b/net/core/dev.c index 3b3965288f52..b2775f06c710 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2879,7 +2879,7 @@ EXPORT_SYMBOL(xmit_recursion); * dev_loopback_xmit - loop back @skb * @skb: buffer to transmit */ -int dev_loopback_xmit(struct sk_buff *skb) +int dev_loopback_xmit(struct sock *sk, struct sk_buff *skb) { skb_reset_mac_header(skb); __skb_pull(skb, skb_network_offset(skb)); @@ -3017,11 +3017,11 @@ out: return rc; } -int dev_queue_xmit(struct sk_buff *skb) +int dev_queue_xmit_sk(struct sock *sk, struct sk_buff *skb) { return __dev_queue_xmit(skb, NULL); } -EXPORT_SYMBOL(dev_queue_xmit); +EXPORT_SYMBOL(dev_queue_xmit_sk); int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv) { @@ -3853,13 +3853,13 @@ static int netif_receive_skb_internal(struct sk_buff *skb) * NET_RX_SUCCESS: no congestion * NET_RX_DROP: packet was dropped */ -int netif_receive_skb(struct sk_buff *skb) +int netif_receive_skb_sk(struct sock *sk, struct sk_buff *skb) { trace_netif_receive_skb_entry(skb); return netif_receive_skb_internal(skb); } -EXPORT_SYMBOL(netif_receive_skb); +EXPORT_SYMBOL(netif_receive_skb_sk); /* Network device is going away, flush any packets still pending * Called with irqs disabled. diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index be1f08cdad29..4507b188fc51 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -194,7 +194,7 @@ static int dn_neigh_output(struct neighbour *neigh, struct sk_buff *skb) return err; } -static int dn_neigh_output_packet(struct sk_buff *skb) +static int dn_neigh_output_packet(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct dn_route *rt = (struct dn_route *)dst; @@ -206,7 +206,8 @@ static int dn_neigh_output_packet(struct sk_buff *skb) /* * For talking to broadcast devices: Ethernet & PPP */ -static int dn_long_output(struct neighbour *neigh, struct sk_buff *skb) +static int dn_long_output(struct neighbour *neigh, struct sock *sk, + struct sk_buff *skb) { struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_long_packet) + 3; @@ -245,14 +246,15 @@ static int dn_long_output(struct neighbour *neigh, struct sk_buff *skb) skb_reset_network_header(skb); - return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, skb, NULL, - neigh->dev, dn_neigh_output_packet); + return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb, + NULL, neigh->dev, dn_neigh_output_packet); } /* * For talking to pointopoint and multidrop devices: DDCMP and X.25 */ -static int dn_short_output(struct neighbour *neigh, struct sk_buff *skb) +static int dn_short_output(struct neighbour *neigh, struct sock *sk, + struct sk_buff *skb) { struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_short_packet) + 2; @@ -284,8 +286,8 @@ static int dn_short_output(struct neighbour *neigh, struct sk_buff *skb) skb_reset_network_header(skb); - return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, skb, NULL, - neigh->dev, dn_neigh_output_packet); + return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb, + NULL, neigh->dev, dn_neigh_output_packet); } /* @@ -293,7 +295,8 @@ static int dn_short_output(struct neighbour *neigh, struct sk_buff *skb) * Phase 3 output is the same as short output, execpt that * it clears the area bits before transmission. */ -static int dn_phase3_output(struct neighbour *neigh, struct sk_buff *skb) +static int dn_phase3_output(struct neighbour *neigh, struct sock *sk, + struct sk_buff *skb) { struct net_device *dev = neigh->dev; int headroom = dev->hard_header_len + sizeof(struct dn_short_packet) + 2; @@ -324,11 +327,11 @@ static int dn_phase3_output(struct neighbour *neigh, struct sk_buff *skb) skb_reset_network_header(skb); - return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, skb, NULL, - neigh->dev, dn_neigh_output_packet); + return NF_HOOK(NFPROTO_DECNET, NF_DN_POST_ROUTING, sk, skb, + NULL, neigh->dev, dn_neigh_output_packet); } -int dn_to_neigh_output(struct sk_buff *skb) +int dn_to_neigh_output(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct dn_route *rt = (struct dn_route *) dst; @@ -347,11 +350,11 @@ int dn_to_neigh_output(struct sk_buff *skb) rcu_read_unlock(); if (dn->flags & DN_NDFLAG_P3) - return dn_phase3_output(neigh, skb); + return dn_phase3_output(neigh, sk, skb); if (use_long) - return dn_long_output(neigh, skb); + return dn_long_output(neigh, sk, skb); else - return dn_short_output(neigh, skb); + return dn_short_output(neigh, sk, skb); } /* @@ -372,7 +375,7 @@ void dn_neigh_pointopoint_hello(struct sk_buff *skb) /* * Ethernet router hello message received */ -int dn_neigh_router_hello(struct sk_buff *skb) +int dn_neigh_router_hello(struct sock *sk, struct sk_buff *skb) { struct rtnode_hello_message *msg = (struct rtnode_hello_message *)skb->data; @@ -434,7 +437,7 @@ int dn_neigh_router_hello(struct sk_buff *skb) /* * Endnode hello message received */ -int dn_neigh_endnode_hello(struct sk_buff *skb) +int dn_neigh_endnode_hello(struct sock *sk, struct sk_buff *skb) { struct endnode_hello_message *msg = (struct endnode_hello_message *)skb->data; struct neighbour *neigh; diff --git a/net/decnet/dn_nsp_in.c b/net/decnet/dn_nsp_in.c index fe5f01485d33..a321eac9fd0c 100644 --- a/net/decnet/dn_nsp_in.c +++ b/net/decnet/dn_nsp_in.c @@ -714,7 +714,7 @@ out: return ret; } -static int dn_nsp_rx_packet(struct sk_buff *skb) +static int dn_nsp_rx_packet(struct sock *sk2, struct sk_buff *skb) { struct dn_skb_cb *cb = DN_SKB_CB(skb); struct sock *sk = NULL; @@ -814,7 +814,8 @@ free_out: int dn_nsp_rx(struct sk_buff *skb) { - return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_IN, skb, skb->dev, NULL, + return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_IN, NULL, skb, + skb->dev, NULL, dn_nsp_rx_packet); } diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 9ab0c4ba297f..03227ffd19ce 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -512,7 +512,7 @@ static int dn_return_long(struct sk_buff *skb) * * Returns: result of input function if route is found, error code otherwise */ -static int dn_route_rx_packet(struct sk_buff *skb) +static int dn_route_rx_packet(struct sock *sk, struct sk_buff *skb) { struct dn_skb_cb *cb; int err; @@ -573,7 +573,8 @@ static int dn_route_rx_long(struct sk_buff *skb) ptr++; cb->hops = *ptr++; /* Visit Count */ - return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, skb, skb->dev, NULL, + return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, NULL, skb, + skb->dev, NULL, dn_route_rx_packet); drop_it: @@ -600,7 +601,8 @@ static int dn_route_rx_short(struct sk_buff *skb) ptr += 2; cb->hops = *ptr & 0x3f; - return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, skb, skb->dev, NULL, + return NF_HOOK(NFPROTO_DECNET, NF_DN_PRE_ROUTING, NULL, skb, + skb->dev, NULL, dn_route_rx_packet); drop_it: @@ -608,7 +610,7 @@ drop_it: return NET_RX_DROP; } -static int dn_route_discard(struct sk_buff *skb) +static int dn_route_discard(struct sock *sk, struct sk_buff *skb) { /* * I know we drop the packet here, but thats considered success in @@ -618,7 +620,7 @@ static int dn_route_discard(struct sk_buff *skb) return NET_RX_SUCCESS; } -static int dn_route_ptp_hello(struct sk_buff *skb) +static int dn_route_ptp_hello(struct sock *sk, struct sk_buff *skb) { dn_dev_hello(skb); dn_neigh_pointopoint_hello(skb); @@ -704,22 +706,22 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type switch (flags & DN_RT_CNTL_MSK) { case DN_RT_PKT_HELO: return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO, - skb, skb->dev, NULL, + NULL, skb, skb->dev, NULL, dn_route_ptp_hello); case DN_RT_PKT_L1RT: case DN_RT_PKT_L2RT: return NF_HOOK(NFPROTO_DECNET, NF_DN_ROUTE, - skb, skb->dev, NULL, + NULL, skb, skb->dev, NULL, dn_route_discard); case DN_RT_PKT_ERTH: return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO, - skb, skb->dev, NULL, + NULL, skb, skb->dev, NULL, dn_neigh_router_hello); case DN_RT_PKT_EEDH: return NF_HOOK(NFPROTO_DECNET, NF_DN_HELLO, - skb, skb->dev, NULL, + NULL, skb, skb->dev, NULL, dn_neigh_endnode_hello); } } else { @@ -768,7 +770,8 @@ static int dn_output(struct sock *sk, struct sk_buff *skb) cb->rt_flags |= DN_RT_F_IE; cb->hops = 0; - return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_OUT, skb, NULL, dev, + return NF_HOOK(NFPROTO_DECNET, NF_DN_LOCAL_OUT, sk, skb, + NULL, dev, dn_to_neigh_output); error: @@ -816,7 +819,8 @@ static int dn_forward(struct sk_buff *skb) if (rt->rt_flags & RTCF_DOREDIRECT) cb->rt_flags |= DN_RT_F_IE; - return NF_HOOK(NFPROTO_DECNET, NF_DN_FORWARD, skb, dev, skb->dev, + return NF_HOOK(NFPROTO_DECNET, NF_DN_FORWARD, NULL, skb, + dev, skb->dev, dn_to_neigh_output); drop: diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index c6e67aa46c32..933a92820d26 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -591,7 +591,8 @@ EXPORT_SYMBOL(arp_create); void arp_xmit(struct sk_buff *skb) { /* Send it off, maybe filter it using firewalling first. */ - NF_HOOK(NFPROTO_ARP, NF_ARP_OUT, skb, NULL, skb->dev, dev_queue_xmit); + NF_HOOK(NFPROTO_ARP, NF_ARP_OUT, NULL, skb, + NULL, skb->dev, dev_queue_xmit_sk); } EXPORT_SYMBOL(arp_xmit); @@ -625,7 +626,7 @@ EXPORT_SYMBOL(arp_send); * Process an arp request. */ -static int arp_process(struct sk_buff *skb) +static int arp_process(struct sock *sk, struct sk_buff *skb) { struct net_device *dev = skb->dev; struct in_device *in_dev = __in_dev_get_rcu(dev); @@ -846,7 +847,7 @@ out: static void parp_redo(struct sk_buff *skb) { - arp_process(skb); + arp_process(NULL, skb); } @@ -879,7 +880,8 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev, memset(NEIGH_CB(skb), 0, sizeof(struct neighbour_cb)); - return NF_HOOK(NFPROTO_ARP, NF_ARP_IN, skb, dev, NULL, arp_process); + return NF_HOOK(NFPROTO_ARP, NF_ARP_IN, NULL, skb, + dev, NULL, arp_process); consumeskb: consume_skb(skb); diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index d9bc28ac5d1b..939992c456f3 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -57,7 +57,7 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) } -static int ip_forward_finish(struct sk_buff *skb) +static int ip_forward_finish(struct sock *sk, struct sk_buff *skb) { struct ip_options *opt = &(IPCB(skb)->opt); @@ -68,7 +68,7 @@ static int ip_forward_finish(struct sk_buff *skb) ip_forward_options(skb); skb_sender_cpu_clear(skb); - return dst_output(skb); + return dst_output_sk(sk, skb); } int ip_forward(struct sk_buff *skb) @@ -136,8 +136,8 @@ int ip_forward(struct sk_buff *skb) skb->priority = rt_tos2priority(iph->tos); - return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, skb, skb->dev, - rt->dst.dev, ip_forward_finish); + return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb, + skb->dev, rt->dst.dev, ip_forward_finish); sr_failed: /* diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 2e0410ed8f16..2db4c8773c1b 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -187,7 +187,7 @@ bool ip_call_ra_chain(struct sk_buff *skb) return false; } -static int ip_local_deliver_finish(struct sk_buff *skb) +static int ip_local_deliver_finish(struct sock *sk, struct sk_buff *skb) { struct net *net = dev_net(skb->dev); @@ -253,7 +253,8 @@ int ip_local_deliver(struct sk_buff *skb) return 0; } - return NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_IN, skb, skb->dev, NULL, + return NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_IN, NULL, skb, + skb->dev, NULL, ip_local_deliver_finish); } @@ -309,7 +310,7 @@ drop: int sysctl_ip_early_demux __read_mostly = 1; EXPORT_SYMBOL(sysctl_ip_early_demux); -static int ip_rcv_finish(struct sk_buff *skb) +static int ip_rcv_finish(struct sock *sk, struct sk_buff *skb) { const struct iphdr *iph = ip_hdr(skb); struct rtable *rt; @@ -451,7 +452,8 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, /* Must drop socket now because of tproxy. */ skb_orphan(skb); - return NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, dev, NULL, + return NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, NULL, skb, + dev, NULL, ip_rcv_finish); csum_error: diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 26f6f7956168..5da4d15262fd 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -91,14 +91,19 @@ void ip_send_check(struct iphdr *iph) } EXPORT_SYMBOL(ip_send_check); -int __ip_local_out(struct sk_buff *skb) +int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb) { struct iphdr *iph = ip_hdr(skb); iph->tot_len = htons(skb->len); ip_send_check(iph); - return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, skb, NULL, - skb_dst(skb)->dev, dst_output); + return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb, NULL, + skb_dst(skb)->dev, dst_output_sk); +} + +int __ip_local_out(struct sk_buff *skb) +{ + return __ip_local_out_sk(skb->sk, skb); } int ip_local_out_sk(struct sock *sk, struct sk_buff *skb) @@ -163,7 +168,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, } EXPORT_SYMBOL_GPL(ip_build_and_send_pkt); -static inline int ip_finish_output2(struct sk_buff *skb) +static inline int ip_finish_output2(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct rtable *rt = (struct rtable *)dst; @@ -211,7 +216,7 @@ static inline int ip_finish_output2(struct sk_buff *skb) return -EINVAL; } -static int ip_finish_output_gso(struct sk_buff *skb) +static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb) { netdev_features_t features; struct sk_buff *segs; @@ -220,7 +225,7 @@ static int ip_finish_output_gso(struct sk_buff *skb) /* common case: locally created skb or seglen is <= mtu */ if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || skb_gso_network_seglen(skb) <= ip_skb_dst_mtu(skb)) - return ip_finish_output2(skb); + return ip_finish_output2(sk, skb); /* Slowpath - GSO segment length is exceeding the dst MTU. * @@ -243,7 +248,7 @@ static int ip_finish_output_gso(struct sk_buff *skb) int err; segs->next = NULL; - err = ip_fragment(segs, ip_finish_output2); + err = ip_fragment(sk, segs, ip_finish_output2); if (err && ret == 0) ret = err; @@ -253,22 +258,22 @@ static int ip_finish_output_gso(struct sk_buff *skb) return ret; } -static int ip_finish_output(struct sk_buff *skb) +static int ip_finish_output(struct sock *sk, struct sk_buff *skb) { #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) /* Policy lookup after SNAT yielded a new policy */ if (skb_dst(skb)->xfrm) { IPCB(skb)->flags |= IPSKB_REROUTED; - return dst_output(skb); + return dst_output_sk(sk, skb); } #endif if (skb_is_gso(skb)) - return ip_finish_output_gso(skb); + return ip_finish_output_gso(sk, skb); if (skb->len > ip_skb_dst_mtu(skb)) - return ip_fragment(skb, ip_finish_output2); + return ip_fragment(sk, skb, ip_finish_output2); - return ip_finish_output2(skb); + return ip_finish_output2(sk, skb); } int ip_mc_output(struct sock *sk, struct sk_buff *skb) @@ -307,7 +312,7 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb) struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); if (newskb) NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, - newskb, NULL, newskb->dev, + sk, newskb, NULL, newskb->dev, dev_loopback_xmit); } @@ -322,11 +327,11 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb) if (rt->rt_flags&RTCF_BROADCAST) { struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); if (newskb) - NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, newskb, + NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, newskb, NULL, newskb->dev, dev_loopback_xmit); } - return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, NULL, + return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb, NULL, skb->dev, ip_finish_output, !(IPCB(skb)->flags & IPSKB_REROUTED)); } @@ -340,7 +345,8 @@ int ip_output(struct sock *sk, struct sk_buff *skb) skb->dev = dev; skb->protocol = htons(ETH_P_IP); - return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, NULL, dev, + return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb, + NULL, dev, ip_finish_output, !(IPCB(skb)->flags & IPSKB_REROUTED)); } @@ -480,7 +486,8 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) * single device frame, and queue such a frame for sending. */ -int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) +int ip_fragment(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)) { struct iphdr *iph; int ptr; @@ -593,7 +600,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) ip_send_check(iph); } - err = output(skb); + err = output(sk, skb); if (!err) IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES); @@ -730,7 +737,7 @@ slow_path: ip_send_check(iph); - err = output(skb2); + err = output(sk, skb2); if (err) goto fail; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 5f17d0e78071..3a2c0162c3ba 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1679,7 +1679,7 @@ static void ip_encap(struct net *net, struct sk_buff *skb, nf_reset(skb); } -static inline int ipmr_forward_finish(struct sk_buff *skb) +static inline int ipmr_forward_finish(struct sock *sk, struct sk_buff *skb) { struct ip_options *opt = &(IPCB(skb)->opt); @@ -1689,7 +1689,7 @@ static inline int ipmr_forward_finish(struct sk_buff *skb) if (unlikely(opt->optlen)) ip_forward_options(skb); - return dst_output(skb); + return dst_output_sk(sk, skb); } /* @@ -1788,7 +1788,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, * not mrouter) cannot join to more than one interface - it will * result in receiving multiple packets. */ - NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, skb, skb->dev, dev, + NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb, + skb->dev, dev, ipmr_forward_finish); return; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 6d0fa8fb8af0..c0bb648fb2f9 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -412,8 +412,8 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, icmp_out_count(net, ((struct icmphdr *) skb_transport_header(skb))->type); - err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT, skb, NULL, - rt->dst.dev, dst_output); + err = NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb, + NULL, rt->dst.dev, dst_output_sk); if (err > 0) err = net_xmit_errno(err); if (err) diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c index cac7468db0a1..60b032f58ccc 100644 --- a/net/ipv4/xfrm4_input.c +++ b/net/ipv4/xfrm4_input.c @@ -22,7 +22,7 @@ int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb) return xfrm4_extract_header(skb); } -static inline int xfrm4_rcv_encap_finish(struct sk_buff *skb) +static inline int xfrm4_rcv_encap_finish(struct sock *sk, struct sk_buff *skb) { if (!skb_dst(skb)) { const struct iphdr *iph = ip_hdr(skb); @@ -52,7 +52,8 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async) iph->tot_len = htons(skb->len); ip_send_check(iph); - NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, NULL, skb, + skb->dev, NULL, xfrm4_rcv_encap_finish); return 0; } diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index dab73813cb92..2878dbfffeb7 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -69,7 +69,7 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb) } EXPORT_SYMBOL(xfrm4_prepare_output); -int xfrm4_output_finish(struct sk_buff *skb) +int xfrm4_output_finish(struct sock *sk, struct sk_buff *skb) { memset(IPCB(skb), 0, sizeof(*IPCB(skb))); @@ -77,26 +77,26 @@ int xfrm4_output_finish(struct sk_buff *skb) IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED; #endif - return xfrm_output(skb); + return xfrm_output(sk, skb); } -static int __xfrm4_output(struct sk_buff *skb) +static int __xfrm4_output(struct sock *sk, struct sk_buff *skb) { struct xfrm_state *x = skb_dst(skb)->xfrm; #ifdef CONFIG_NETFILTER if (!x) { IPCB(skb)->flags |= IPSKB_REROUTED; - return dst_output(skb); + return dst_output_sk(sk, skb); } #endif - return x->outer_mode->afinfo->output_finish(skb); + return x->outer_mode->afinfo->output_finish(sk, skb); } int xfrm4_output(struct sock *sk, struct sk_buff *skb) { - return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, + return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb, NULL, skb_dst(skb)->dev, __xfrm4_output, !(IPCB(skb)->flags & IPSKB_REROUTED)); } diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index fb97f7f8d4ed..f2e464eba5ef 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -46,8 +46,7 @@ #include #include - -int ip6_rcv_finish(struct sk_buff *skb) +int ip6_rcv_finish(struct sock *sk, struct sk_buff *skb) { if (sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) { const struct inet6_protocol *ipprot; @@ -183,7 +182,8 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt /* Must drop socket now because of tproxy. */ skb_orphan(skb); - return NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, skb, dev, NULL, + return NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, NULL, skb, + dev, NULL, ip6_rcv_finish); err: IP6_INC_STATS_BH(net, idev, IPSTATS_MIB_INHDRERRORS); @@ -198,7 +198,7 @@ drop: */ -static int ip6_input_finish(struct sk_buff *skb) +static int ip6_input_finish(struct sock *sk, struct sk_buff *skb) { struct net *net = dev_net(skb_dst(skb)->dev); const struct inet6_protocol *ipprot; @@ -277,7 +277,8 @@ discard: int ip6_input(struct sk_buff *skb) { - return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN, skb, skb->dev, NULL, + return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN, NULL, skb, + skb->dev, NULL, ip6_input_finish); } diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 654f245aa930..7fde1f265c90 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -56,7 +56,7 @@ #include #include -static int ip6_finish_output2(struct sk_buff *skb) +static int ip6_finish_output2(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct net_device *dev = dst->dev; @@ -70,7 +70,7 @@ static int ip6_finish_output2(struct sk_buff *skb) if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr)) { struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); - if (!(dev->flags & IFF_LOOPBACK) && sk_mc_loop(skb->sk) && + if (!(dev->flags & IFF_LOOPBACK) && sk_mc_loop(sk) && ((mroute6_socket(dev_net(dev), skb) && !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) || ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, @@ -82,7 +82,7 @@ static int ip6_finish_output2(struct sk_buff *skb) */ if (newskb) NF_HOOK(NFPROTO_IPV6, NF_INET_POST_ROUTING, - newskb, NULL, newskb->dev, + sk, newskb, NULL, newskb->dev, dev_loopback_xmit); if (ipv6_hdr(skb)->hop_limit == 0) { @@ -122,14 +122,14 @@ static int ip6_finish_output2(struct sk_buff *skb) return -EINVAL; } -static int ip6_finish_output(struct sk_buff *skb) +static int ip6_finish_output(struct sock *sk, struct sk_buff *skb) { if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) || dst_allfrag(skb_dst(skb)) || (IP6CB(skb)->frag_max_size && skb->len > IP6CB(skb)->frag_max_size)) - return ip6_fragment(skb, ip6_finish_output2); + return ip6_fragment(sk, skb, ip6_finish_output2); else - return ip6_finish_output2(skb); + return ip6_finish_output2(sk, skb); } int ip6_output(struct sock *sk, struct sk_buff *skb) @@ -143,7 +143,8 @@ int ip6_output(struct sock *sk, struct sk_buff *skb) return 0; } - return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb, NULL, dev, + return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, sk, skb, + NULL, dev, ip6_finish_output, !(IP6CB(skb)->flags & IP6SKB_REROUTED)); } @@ -223,8 +224,8 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, if ((skb->len <= mtu) || skb->ignore_df || skb_is_gso(skb)) { IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUT, skb->len); - return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, - dst->dev, dst_output); + return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb, + NULL, dst->dev, dst_output_sk); } skb->dev = dst->dev; @@ -316,10 +317,10 @@ static int ip6_forward_proxy_check(struct sk_buff *skb) return 0; } -static inline int ip6_forward_finish(struct sk_buff *skb) +static inline int ip6_forward_finish(struct sock *sk, struct sk_buff *skb) { skb_sender_cpu_clear(skb); - return dst_output(skb); + return dst_output_sk(sk, skb); } static unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) @@ -511,7 +512,8 @@ int ip6_forward(struct sk_buff *skb) IP6_INC_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS); IP6_ADD_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTOCTETS, skb->len); - return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, skb, skb->dev, dst->dev, + return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, NULL, skb, + skb->dev, dst->dev, ip6_forward_finish); error: @@ -538,7 +540,8 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) skb_copy_secmark(to, from); } -int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) +int ip6_fragment(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)) { struct sk_buff *frag; struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); @@ -667,7 +670,7 @@ int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) ip6_copy_metadata(frag, skb); } - err = output(skb); + err = output(sk, skb); if (!err) IP6_INC_STATS(net, ip6_dst_idev(&rt->dst), IPSTATS_MIB_FRAGCREATES); @@ -800,7 +803,7 @@ slow_path: /* * Put this fragment into the sending queue. */ - err = output(frag); + err = output(sk, frag); if (err) goto fail; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 8493a22e74eb..74ceb73c1c9a 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1986,13 +1986,13 @@ int ip6mr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg) } #endif -static inline int ip6mr_forward2_finish(struct sk_buff *skb) +static inline int ip6mr_forward2_finish(struct sock *sk, struct sk_buff *skb) { IP6_INC_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTFORWDATAGRAMS); IP6_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUTOCTETS, skb->len); - return dst_output(skb); + return dst_output_sk(sk, skb); } /* @@ -2064,7 +2064,8 @@ static int ip6mr_forward2(struct net *net, struct mr6_table *mrt, IP6CB(skb)->flags |= IP6SKB_FORWARDED; - return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, skb, skb->dev, dev, + return NF_HOOK(NFPROTO_IPV6, NF_INET_FORWARD, NULL, skb, + skb->dev, dev, ip6mr_forward2_finish); out_free: diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index fac1f27e428e..083b2927fc67 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1644,8 +1644,9 @@ static void mld_sendpack(struct sk_buff *skb) payload_len = skb->len; - err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev, - dst_output); + err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, + net->ipv6.igmp_sk, skb, NULL, skb->dev, + dst_output_sk); out: if (!err) { ICMP6MSGOUT_INC_STATS(net, idev, ICMPV6_MLD2_REPORT); @@ -2007,8 +2008,8 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) } skb_dst_set(skb, dst); - err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, skb->dev, - dst_output); + err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb, + NULL, skb->dev, dst_output_sk); out: if (!err) { ICMP6MSGOUT_INC_STATS(net, idev, type); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 71fde6cafb35..96f153c0846b 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -463,8 +463,9 @@ static void ndisc_send_skb(struct sk_buff *skb, idev = __in6_dev_get(dst->dev); IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); - err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, dst->dev, - dst_output); + err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb, + NULL, dst->dev, + dst_output_sk); if (!err) { ICMP6MSGOUT_INC_STATS(net, idev, type); ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTMSGS); diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index e2b882056751..a45db0b4785c 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -75,7 +75,7 @@ static unsigned int ipv6_defrag(const struct nf_hook_ops *ops, nf_ct_frag6_consume_orig(reasm); - NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, reasm, + NF_HOOK_THRESH(NFPROTO_IPV6, ops->hooknum, state->sk, reasm, state->in, state->out, state->okfn, NF_IP6_PRI_CONNTRACK_DEFRAG + 1); diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 4016a6ef9d61..7d1131dc29fe 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -146,8 +146,8 @@ int __ip6_local_out(struct sk_buff *skb) ipv6_hdr(skb)->payload_len = htons(len); IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr); - return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, - skb_dst(skb)->dev, dst_output); + return nf_hook(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb->sk, skb, + NULL, skb_dst(skb)->dev, dst_output_sk); } EXPORT_SYMBOL_GPL(__ip6_local_out); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 79ccdb4c1b33..8072bd4139b7 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -652,8 +652,8 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length, goto error_fault; IP6_UPD_PO_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len); - err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, - rt->dst.dev, dst_output); + err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, sk, skb, + NULL, rt->dst.dev, dst_output_sk); if (err > 0) err = net_xmit_errno(err); if (err) diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index f48fbe4d16f5..74bd17882a2f 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -42,7 +42,8 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async) ipv6_hdr(skb)->payload_len = htons(skb->len); __skb_push(skb, skb->data - skb_network_header(skb)); - NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, skb, skb->dev, NULL, + NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, NULL, skb, + skb->dev, NULL, ip6_rcv_finish); return -1; } diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 010f8bd2d577..09c76a7b474d 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -120,7 +120,7 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb) } EXPORT_SYMBOL(xfrm6_prepare_output); -int xfrm6_output_finish(struct sk_buff *skb) +int xfrm6_output_finish(struct sock *sk, struct sk_buff *skb) { memset(IP6CB(skb), 0, sizeof(*IP6CB(skb))); @@ -128,10 +128,10 @@ int xfrm6_output_finish(struct sk_buff *skb) IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED; #endif - return xfrm_output(skb); + return xfrm_output(sk, skb); } -static int __xfrm6_output(struct sk_buff *skb) +static int __xfrm6_output(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); struct xfrm_state *x = dst->xfrm; @@ -140,7 +140,7 @@ static int __xfrm6_output(struct sk_buff *skb) #ifdef CONFIG_NETFILTER if (!x) { IP6CB(skb)->flags |= IP6SKB_REROUTED; - return dst_output(skb); + return dst_output_sk(sk, skb); } #endif @@ -160,14 +160,15 @@ static int __xfrm6_output(struct sk_buff *skb) if (x->props.mode == XFRM_MODE_TUNNEL && ((skb->len > mtu && !skb_is_gso(skb)) || dst_allfrag(skb_dst(skb)))) { - return ip6_fragment(skb, x->outer_mode->afinfo->output_finish); + return ip6_fragment(sk, skb, + x->outer_mode->afinfo->output_finish); } - return x->outer_mode->afinfo->output_finish(skb); + return x->outer_mode->afinfo->output_finish(sk, skb); } int xfrm6_output(struct sock *sk, struct sk_buff *skb) { - return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb, + return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, sk, skb, NULL, skb_dst(skb)->dev, __xfrm6_output, !(IP6CB(skb)->flags & IP6SKB_REROUTED)); } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index bf02932b7188..19986ec5f21a 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -536,8 +536,8 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb, ip_vs_update_conntrack(skb, cp, 1); if (!local) { skb_forward_csum(skb); - NF_HOOK(pf, NF_INET_LOCAL_OUT, skb, NULL, skb_dst(skb)->dev, - dst_output); + NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, + NULL, skb_dst(skb)->dev, dst_output_sk); } else ret = NF_ACCEPT; return ret; @@ -554,8 +554,8 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb, ip_vs_notrack(skb); if (!local) { skb_forward_csum(skb); - NF_HOOK(pf, NF_INET_LOCAL_OUT, skb, NULL, skb_dst(skb)->dev, - dst_output); + NF_HOOK(pf, NF_INET_LOCAL_OUT, NULL, skb, + NULL, skb_dst(skb)->dev, dst_output_sk); } else ret = NF_ACCEPT; return ret; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index c4a706678f88..3f3ac57b2998 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -202,7 +202,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) case NF_ACCEPT: case NF_STOP: local_bh_disable(); - entry->state.okfn(skb); + entry->state.okfn(entry->state.sk, skb); local_bh_enable(); break; case NF_QUEUE: diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index 7c532856b398..fbcedbe33190 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -19,7 +19,7 @@ #include #include -static int xfrm_output2(struct sk_buff *skb); +static int xfrm_output2(struct sock *sk, struct sk_buff *skb); static int xfrm_skb_check_space(struct sk_buff *skb) { @@ -130,7 +130,7 @@ int xfrm_output_resume(struct sk_buff *skb, int err) return dst_output(skb); err = nf_hook(skb_dst(skb)->ops->family, - NF_INET_POST_ROUTING, skb, + NF_INET_POST_ROUTING, skb->sk, skb, NULL, skb_dst(skb)->dev, xfrm_output2); if (unlikely(err != 1)) goto out; @@ -144,12 +144,12 @@ out: } EXPORT_SYMBOL_GPL(xfrm_output_resume); -static int xfrm_output2(struct sk_buff *skb) +static int xfrm_output2(struct sock *sk, struct sk_buff *skb) { return xfrm_output_resume(skb, 1); } -static int xfrm_output_gso(struct sk_buff *skb) +static int xfrm_output_gso(struct sock *sk, struct sk_buff *skb) { struct sk_buff *segs; @@ -165,7 +165,7 @@ static int xfrm_output_gso(struct sk_buff *skb) int err; segs->next = NULL; - err = xfrm_output2(segs); + err = xfrm_output2(sk, segs); if (unlikely(err)) { kfree_skb_list(nskb); @@ -178,13 +178,13 @@ static int xfrm_output_gso(struct sk_buff *skb) return 0; } -int xfrm_output(struct sk_buff *skb) +int xfrm_output(struct sock *sk, struct sk_buff *skb) { struct net *net = dev_net(skb_dst(skb)->dev); int err; if (skb_is_gso(skb)) - return xfrm_output_gso(skb); + return xfrm_output_gso(sk, skb); if (skb->ip_summed == CHECKSUM_PARTIAL) { err = skb_checksum_help(skb); @@ -195,7 +195,7 @@ int xfrm_output(struct sk_buff *skb) } } - return xfrm_output2(skb); + return xfrm_output2(sk, skb); } EXPORT_SYMBOL_GPL(xfrm_output); -- cgit v1.2.3 From 388069d30240e9524ad69ce6cc692fff2a5a7829 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 7 Apr 2015 12:10:16 +0200 Subject: netdevice.h: remove iflink description Also move 'group' description to match the order of the net_device structure. Fixes: 7a66bbc96ce9 ("net: remove iflink field from struct net_device") Reported-by: Fengguang Wu Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 45823db2efb0..bf6d9df34d7b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1328,7 +1328,7 @@ enum netdev_priv_flags { * @mpls_features: Mask of features inheritable by MPLS * * @ifindex: interface index - * @iflink: unique device identifier + * @group: The group, that the device belongs to * * @stats: Statistics struct, which was left as a legacy, use * rtnl_link_stats64 instead @@ -1488,7 +1488,6 @@ enum netdev_priv_flags { * * @qdisc_tx_busylock: XXX: need comments on this one * - * @group: The group, that the device belongs to * @pm_qos_req: Power Management QoS object * * FIXME: cleanup struct net_device such that network protocol info -- cgit v1.2.3 From 7f9b838b71eb78a27de27a12ca5de8542fac3115 Mon Sep 17 00:00:00 2001 From: Daniel Lee Date: Mon, 6 Apr 2015 14:37:26 -0700 Subject: tcp: RFC7413 option support for Fast Open server Fast Open has been using the experimental option with a magic number (RFC6994) to request and grant Fast Open cookies. This patch enables the server to support the official IANA option 34 in RFC7413 in addition. The change has passed all existing Fast Open tests with both old and new options at Google. Signed-off-by: Daniel Lee Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 + include/net/tcp.h | 2 ++ net/ipv4/tcp_fastopen.c | 1 + net/ipv4/tcp_input.c | 42 ++++++++++++++++++++++++++++++------------ net/ipv4/tcp_output.c | 34 +++++++++++++++++++++++----------- 5 files changed, 57 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index f869ae8afbaf..a48d00318683 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -58,6 +58,7 @@ static inline unsigned int tcp_optlen(const struct sk_buff *skb) struct tcp_fastopen_cookie { s8 len; u8 val[TCP_FASTOPEN_COOKIE_MAX]; + bool exp; /* In RFC6994 experimental option format */ }; /* This defines a selective acknowledgement block. */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 963303fb96ae..7292c3c575bc 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -179,6 +179,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOPT_SACK 5 /* SACK Block */ #define TCPOPT_TIMESTAMP 8 /* Better RTT estimations/PAWS */ #define TCPOPT_MD5SIG 19 /* MD5 Signature (RFC2385) */ +#define TCPOPT_FASTOPEN 34 /* Fast open (RFC7413) */ #define TCPOPT_EXP 254 /* Experimental */ /* Magic number to be after the option value for sharing TCP * experimental options. See draft-ietf-tcpm-experimental-options-00.txt @@ -194,6 +195,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_SACK_PERM 2 #define TCPOLEN_TIMESTAMP 10 #define TCPOLEN_MD5SIG 18 +#define TCPOLEN_FASTOPEN_BASE 2 #define TCPOLEN_EXP_FASTOPEN_BASE 4 /* But this is what stacks really send out. */ diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 5da55e2b5cd2..e3d87aca6be8 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -303,6 +303,7 @@ fastopen: } else if (foc->len > 0) /* Client presents an invalid cookie */ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENPASSIVEFAIL); + valid_foc.exp = foc->exp; *foc = valid_foc; return false; } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c1ce304ba8d2..24f1630b2afb 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3603,6 +3603,23 @@ old_ack: return 0; } +static void tcp_parse_fastopen_option(int len, const unsigned char *cookie, + bool syn, struct tcp_fastopen_cookie *foc, + bool exp_opt) +{ + /* Valid only in SYN or SYN-ACK with an even length. */ + if (!foc || !syn || len < 0 || (len & 1)) + return; + + if (len >= TCP_FASTOPEN_COOKIE_MIN && + len <= TCP_FASTOPEN_COOKIE_MAX) + memcpy(foc->val, cookie, len); + else if (len != 0) + len = -1; + foc->len = len; + foc->exp = exp_opt; +} + /* Look for tcp options. Normally only called on SYN and SYNACK packets. * But, this can also be called on packets in the established flow when * the fast version below fails. @@ -3692,21 +3709,22 @@ void tcp_parse_options(const struct sk_buff *skb, */ break; #endif + case TCPOPT_FASTOPEN: + tcp_parse_fastopen_option( + opsize - TCPOLEN_FASTOPEN_BASE, + ptr, th->syn, foc, false); + break; + case TCPOPT_EXP: /* Fast Open option shares code 254 using a - * 16 bits magic number. It's valid only in - * SYN or SYN-ACK with an even size. + * 16 bits magic number. */ - if (opsize < TCPOLEN_EXP_FASTOPEN_BASE || - get_unaligned_be16(ptr) != TCPOPT_FASTOPEN_MAGIC || - !foc || !th->syn || (opsize & 1)) - break; - foc->len = opsize - TCPOLEN_EXP_FASTOPEN_BASE; - if (foc->len >= TCP_FASTOPEN_COOKIE_MIN && - foc->len <= TCP_FASTOPEN_COOKIE_MAX) - memcpy(foc->val, ptr + 2, foc->len); - else if (foc->len != 0) - foc->len = -1; + if (opsize >= TCPOLEN_EXP_FASTOPEN_BASE && + get_unaligned_be16(ptr) == + TCPOPT_FASTOPEN_MAGIC) + tcp_parse_fastopen_option(opsize - + TCPOLEN_EXP_FASTOPEN_BASE, + ptr + 2, th->syn, foc, true); break; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 7404e5238e00..464bd8c5de69 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -518,17 +518,26 @@ static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp, if (unlikely(OPTION_FAST_OPEN_COOKIE & options)) { struct tcp_fastopen_cookie *foc = opts->fastopen_cookie; + u8 *p = (u8 *)ptr; + u32 len; /* Fast Open option length */ + + if (foc->exp) { + len = TCPOLEN_EXP_FASTOPEN_BASE + foc->len; + *ptr = htonl((TCPOPT_EXP << 24) | (len << 16) | + TCPOPT_FASTOPEN_MAGIC); + p += TCPOLEN_EXP_FASTOPEN_BASE; + } else { + len = TCPOLEN_FASTOPEN_BASE + foc->len; + *p++ = TCPOPT_FASTOPEN; + *p++ = len; + } - *ptr++ = htonl((TCPOPT_EXP << 24) | - ((TCPOLEN_EXP_FASTOPEN_BASE + foc->len) << 16) | - TCPOPT_FASTOPEN_MAGIC); - - memcpy(ptr, foc->val, foc->len); - if ((foc->len & 3) == 2) { - u8 *align = ((u8 *)ptr) + foc->len; - align[0] = align[1] = TCPOPT_NOP; + memcpy(p, foc->val, foc->len); + if ((len & 3) == 2) { + p[foc->len] = TCPOPT_NOP; + p[foc->len + 1] = TCPOPT_NOP; } - ptr += (foc->len + 3) >> 2; + ptr += (len + 3) >> 2; } } @@ -641,8 +650,11 @@ static unsigned int tcp_synack_options(struct sock *sk, if (unlikely(!ireq->tstamp_ok)) remaining -= TCPOLEN_SACKPERM_ALIGNED; } - if (foc && foc->len >= 0) { - u32 need = TCPOLEN_EXP_FASTOPEN_BASE + foc->len; + if (foc != NULL && foc->len >= 0) { + u32 need = foc->len; + + need += foc->exp ? TCPOLEN_EXP_FASTOPEN_BASE : + TCPOLEN_FASTOPEN_BASE; need = (need + 3) & ~3U; /* Align to 32 bits */ if (remaining >= need) { opts->options |= OPTION_FAST_OPEN_COOKIE; -- cgit v1.2.3 From 2646c831c00c5d22aa72b79d24069c1b412cda7c Mon Sep 17 00:00:00 2001 From: Daniel Lee Date: Mon, 6 Apr 2015 14:37:27 -0700 Subject: tcp: RFC7413 option support for Fast Open client Fast Open has been using an experimental option with a magic number (RFC6994). This patch makes the client by default use the RFC7413 option (34) to get and send Fast Open cookies. This patch makes the client solicit cookies from a given server first with the RFC7413 option. If that fails to elicit a cookie, then it tries the RFC6994 experimental option. If that also fails, it uses the RFC7413 option on all subsequent connect attempts. If the server returns a Fast Open cookie then the client caches the form of the option that successfully elicited a cookie, and uses that form on later connects when it presents that cookie. The idea is to gradually obsolete the use of experimental options as the servers and clients upgrade, while keeping the interoperability meanwhile. Signed-off-by: Daniel Lee Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 + include/net/tcp.h | 3 ++- net/ipv4/tcp_input.c | 29 +++++++++++++++++++---------- net/ipv4/tcp_metrics.c | 13 +++++++++++-- net/ipv4/tcp_output.c | 6 +++++- net/ipv4/tcp_timer.c | 2 +- 6 files changed, 39 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index a48d00318683..0caa3a2d4106 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -189,6 +189,7 @@ struct tcp_sock { u8 do_early_retrans:1,/* Enable RFC5827 early-retransmit */ syn_data:1, /* SYN includes data */ syn_fastopen:1, /* SYN includes Fast Open option */ + syn_fastopen_exp:1,/* SYN includes Fast Open exp. option */ syn_data_acked:1,/* data in SYN is acked by SYN-ACK */ is_cwnd_limited:1;/* forward progress limited by snd_cwnd? */ u32 tlp_high_seq; /* snd_nxt at the time of TLP retransmit. */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 7292c3c575bc..9598871485ce 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1339,7 +1339,8 @@ void tcp_fastopen_cache_get(struct sock *sk, u16 *mss, struct tcp_fastopen_cookie *cookie, int *syn_loss, unsigned long *last_syn_loss); void tcp_fastopen_cache_set(struct sock *sk, u16 mss, - struct tcp_fastopen_cookie *cookie, bool syn_lost); + struct tcp_fastopen_cookie *cookie, bool syn_lost, + u16 try_exp); struct tcp_fastopen_request { /* Fast Open cookie. Size 0 means a cookie request */ struct tcp_fastopen_cookie cookie; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 24f1630b2afb..031cf72cd05c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5378,8 +5378,8 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *data = tp->syn_data ? tcp_write_queue_head(sk) : NULL; - u16 mss = tp->rx_opt.mss_clamp; - bool syn_drop; + u16 mss = tp->rx_opt.mss_clamp, try_exp = 0; + bool syn_drop = false; if (mss == tp->rx_opt.user_mss) { struct tcp_options_received opt; @@ -5391,16 +5391,25 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, mss = opt.mss_clamp; } - if (!tp->syn_fastopen) /* Ignore an unsolicited cookie */ + if (!tp->syn_fastopen) { + /* Ignore an unsolicited cookie */ cookie->len = -1; + } else if (tp->total_retrans) { + /* SYN timed out and the SYN-ACK neither has a cookie nor + * acknowledges data. Presumably the remote received only + * the retransmitted (regular) SYNs: either the original + * SYN-data or the corresponding SYN-ACK was dropped. + */ + syn_drop = (cookie->len < 0 && data); + } else if (cookie->len < 0 && !tp->syn_data) { + /* We requested a cookie but didn't get it. If we did not use + * the (old) exp opt format then try so next time (try_exp=1). + * Otherwise we go back to use the RFC7413 opt (try_exp=2). + */ + try_exp = tp->syn_fastopen_exp ? 2 : 1; + } - /* The SYN-ACK neither has cookie nor acknowledges the data. Presumably - * the remote receives only the retransmitted (regular) SYNs: either - * the original SYN-data or the corresponding SYN-ACK is lost. - */ - syn_drop = (cookie->len <= 0 && data && tp->total_retrans); - - tcp_fastopen_cache_set(sk, mss, cookie, syn_drop); + tcp_fastopen_cache_set(sk, mss, cookie, syn_drop, try_exp); if (data) { /* Retransmit unacked data in SYN */ tcp_for_write_queue_from(data, sk) { diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 78ecc4a01712..a51d63a43e33 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -28,7 +28,8 @@ static struct tcp_metrics_block *__tcp_get_metrics(const struct inetpeer_addr *s struct tcp_fastopen_metrics { u16 mss; - u16 syn_loss:10; /* Recurring Fast Open SYN losses */ + u16 syn_loss:10, /* Recurring Fast Open SYN losses */ + try_exp:2; /* Request w/ exp. option (once) */ unsigned long last_syn_loss; /* Last Fast Open SYN loss */ struct tcp_fastopen_cookie cookie; }; @@ -131,6 +132,8 @@ static void tcpm_suck_dst(struct tcp_metrics_block *tm, if (fastopen_clear) { tm->tcpm_fastopen.mss = 0; tm->tcpm_fastopen.syn_loss = 0; + tm->tcpm_fastopen.try_exp = 0; + tm->tcpm_fastopen.cookie.exp = false; tm->tcpm_fastopen.cookie.len = 0; } } @@ -713,6 +716,8 @@ void tcp_fastopen_cache_get(struct sock *sk, u16 *mss, if (tfom->mss) *mss = tfom->mss; *cookie = tfom->cookie; + if (cookie->len <= 0 && tfom->try_exp == 1) + cookie->exp = true; *syn_loss = tfom->syn_loss; *last_syn_loss = *syn_loss ? tfom->last_syn_loss : 0; } while (read_seqretry(&fastopen_seqlock, seq)); @@ -721,7 +726,8 @@ void tcp_fastopen_cache_get(struct sock *sk, u16 *mss, } void tcp_fastopen_cache_set(struct sock *sk, u16 mss, - struct tcp_fastopen_cookie *cookie, bool syn_lost) + struct tcp_fastopen_cookie *cookie, bool syn_lost, + u16 try_exp) { struct dst_entry *dst = __sk_dst_get(sk); struct tcp_metrics_block *tm; @@ -738,6 +744,9 @@ void tcp_fastopen_cache_set(struct sock *sk, u16 mss, tfom->mss = mss; if (cookie && cookie->len > 0) tfom->cookie = *cookie; + else if (try_exp > tfom->try_exp && + tfom->cookie.len <= 0 && !tfom->cookie.exp) + tfom->try_exp = try_exp; if (syn_lost) { ++tfom->syn_loss; tfom->last_syn_loss = jiffies; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 464bd8c5de69..e662d85d1635 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -592,13 +592,17 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb, } if (fastopen && fastopen->cookie.len >= 0) { - u32 need = TCPOLEN_EXP_FASTOPEN_BASE + fastopen->cookie.len; + u32 need = fastopen->cookie.len; + + need += fastopen->cookie.exp ? TCPOLEN_EXP_FASTOPEN_BASE : + TCPOLEN_FASTOPEN_BASE; need = (need + 3) & ~3U; /* Align to 32 bits */ if (remaining >= need) { opts->options |= OPTION_FAST_OPEN_COOKIE; opts->fastopen_cookie = &fastopen->cookie; remaining -= need; tp->syn_fastopen = 1; + tp->syn_fastopen_exp = fastopen->cookie.exp ? 1 : 0; } } diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 2568fd282873..8c65dc147d8b 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -167,7 +167,7 @@ static int tcp_write_timeout(struct sock *sk) if (icsk->icsk_retransmits) { dst_negative_advice(sk); if (tp->syn_fastopen || tp->syn_data) - tcp_fastopen_cache_set(sk, 0, NULL, true); + tcp_fastopen_cache_set(sk, 0, NULL, true, 0); if (tp->syn_data) NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENACTIVEFAIL); -- cgit v1.2.3