From a9149d243f259ad8f02b1e23dfe8ba06128f15e1 Mon Sep 17 00:00:00 2001 From: Dan Moulding Date: Tue, 28 Jan 2020 02:31:07 -0700 Subject: iwlwifi: mvm: Do not require PHY_SKU NVM section for 3168 devices The logic for checking required NVM sections was recently fixed in commit b3f20e098293 ("iwlwifi: mvm: fix NVM check for 3168 devices"). However, with that fixed the else is now taken for 3168 devices and within the else clause there is a mandatory check for the PHY_SKU section. This causes the parsing to fail for 3168 devices. The PHY_SKU section is really only mandatory for the IWL_NVM_EXT layout (the phy_sku parameter of iwl_parse_nvm_data is only used when the NVM type is IWL_NVM_EXT). So this changes the PHY_SKU section check so that it's only mandatory for IWL_NVM_EXT. Fixes: b3f20e098293 ("iwlwifi: mvm: fix NVM check for 3168 devices") Signed-off-by: Dan Moulding Signed-off-by: Kalle Valo --- drivers/net/wireless/intel/iwlwifi/mvm/nvm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c index 70b29bf16bb9..60296a754af2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c @@ -308,7 +308,8 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm) } /* PHY_SKU section is mandatory in B0 */ - if (!mvm->nvm_sections[NVM_SECTION_TYPE_PHY_SKU].data) { + if (mvm->trans->cfg->nvm_type == IWL_NVM_EXT && + !mvm->nvm_sections[NVM_SECTION_TYPE_PHY_SKU].data) { IWL_ERR(mvm, "Can't parse phy_sku in B0, empty sections\n"); return NULL; -- cgit From 8e8ce08198de193e3d21d42e96945216e3d9ac7f Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 16 Feb 2020 13:02:06 +0100 Subject: batman-adv: Don't schedule OGM for disabled interface A transmission scheduling for an interface which is currently dropped by batadv_iv_ogm_iface_disable could still be in progress. The B.A.T.M.A.N. V is simply cancelling the workqueue item in an synchronous way but this is not possible with B.A.T.M.A.N. IV because the OGM submissions are intertwined. Instead it has to stop submitting the OGM when it detect that the buffer pointer is set to NULL. Reported-by: syzbot+a98f2016f40b9cd3818a@syzkaller.appspotmail.com Reported-by: syzbot+ac36b6a33c28a491e929@syzkaller.appspotmail.com Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Signed-off-by: Sven Eckelmann Cc: Hillf Danton Signed-off-by: Simon Wunderlich --- net/batman-adv/bat_iv_ogm.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index f0209505e41a..a7c8dd7ae513 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -789,6 +789,10 @@ static void batadv_iv_ogm_schedule_buff(struct batadv_hard_iface *hard_iface) lockdep_assert_held(&hard_iface->bat_iv.ogm_buff_mutex); + /* interface already disabled by batadv_iv_ogm_iface_disable */ + if (!*ogm_buff) + return; + /* the interface gets activated here to avoid race conditions between * the moment of activating the interface in * hardif_activate_interface() where the originator mac is set and -- cgit From d64c7a08034b32c285e576208ae44fc3ba3fa7df Mon Sep 17 00:00:00 2001 From: You-Sheng Yang Date: Wed, 26 Feb 2020 23:37:10 +0800 Subject: r8152: check disconnect status after long sleep Dell USB Type C docking WD19/WD19DC attaches additional peripherals as: /: Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M |__ Port 1: Dev 11, If 0, Class=Hub, Driver=hub/4p, 5000M |__ Port 3: Dev 12, If 0, Class=Hub, Driver=hub/4p, 5000M |__ Port 4: Dev 13, If 0, Class=Vendor Specific Class, Driver=r8152, 5000M where usb 2-1-3 is a hub connecting all USB Type-A/C ports on the dock. When hotplugging such dock with additional usb devices already attached on it, the probing process may reset usb 2.1 port, therefore r8152 ethernet device is also reset. However, during r8152 device init there are several for-loops that, when it's unable to retrieve hardware registers due to being disconnected from USB, may take up to 14 seconds each in practice, and that has to be completed before USB may re-enumerate devices on the bus. As a result, devices attached to the dock will only be available after nearly 1 minute after the dock was plugged in: [ 216.388290] [250] r8152 2-1.4:1.0: usb_probe_interface [ 216.388292] [250] r8152 2-1.4:1.0: usb_probe_interface - got id [ 258.830410] r8152 2-1.4:1.0 (unnamed net_device) (uninitialized): PHY not ready [ 258.830460] r8152 2-1.4:1.0 (unnamed net_device) (uninitialized): Invalid header when reading pass-thru MAC addr [ 258.830464] r8152 2-1.4:1.0 (unnamed net_device) (uninitialized): Get ether addr fail This happens in, for example, r8153_init: static int generic_ocp_read(struct r8152 *tp, u16 index, u16 size, void *data, u16 type) { if (test_bit(RTL8152_UNPLUG, &tp->flags)) return -ENODEV; ... } static u16 ocp_read_word(struct r8152 *tp, u16 type, u16 index) { u32 data; ... generic_ocp_read(tp, index, sizeof(tmp), &tmp, type | byen); data = __le32_to_cpu(tmp); ... return (u16)data; } static void r8153_init(struct r8152 *tp) { ... if (test_bit(RTL8152_UNPLUG, &tp->flags)) return; for (i = 0; i < 500; i++) { if (ocp_read_word(tp, MCU_TYPE_PLA, PLA_BOOT_CTRL) & AUTOLOAD_DONE) break; msleep(20); } ... } Since ocp_read_word() doesn't check the return status of generic_ocp_read(), and the only exit condition for the loop is to have a match in the returned value, such loops will only ends after exceeding its maximum runs when the device has been marked as disconnected, which takes 500 * 20ms = 10 seconds in theory, 14 in practice. To solve this long latency another test to RTL8152_UNPLUG flag should be added after those 20ms sleep to skip unnecessary loops, so that the device probe can complete early and proceed to parent port reset/reprobe process. This can be reproduced on all kernel versions up to latest v5.6-rc2, but after v5.5-rc7 the reproduce rate is dramatically lowered to 1/30 or less while it was around 1/2. Signed-off-by: You-Sheng Yang Signed-off-by: David S. Miller --- drivers/net/usb/r8152.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 78ddbaf6401b..95b19ce96513 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3221,6 +3221,8 @@ static u16 r8153_phy_status(struct r8152 *tp, u16 desired) } msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } return data; @@ -5402,7 +5404,10 @@ static void r8153_init(struct r8152 *tp) if (ocp_read_word(tp, MCU_TYPE_PLA, PLA_BOOT_CTRL) & AUTOLOAD_DONE) break; + msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } data = r8153_phy_status(tp, 0); @@ -5539,7 +5544,10 @@ static void r8153b_init(struct r8152 *tp) if (ocp_read_word(tp, MCU_TYPE_PLA, PLA_BOOT_CTRL) & AUTOLOAD_DONE) break; + msleep(20); + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + break; } data = r8153_phy_status(tp, 0); -- cgit From 84b3268027641401bb8ad4427a90a3cce2eb86f5 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 26 Feb 2020 19:47:34 +0100 Subject: netlink: Use netlink header as base to calculate bad attribute offset Userspace might send a batch that is composed of several netlink messages. The netlink_ack() function must use the pointer to the netlink header as base to calculate the bad attribute offset. Fixes: 2d4bc93368f5 ("netlink: extended ACK reporting") Signed-off-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index edf3e285e242..5313f1cec170 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2434,7 +2434,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err, in_skb->len)) WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_OFFS, (u8 *)extack->bad_attr - - in_skb->data)); + (u8 *)nlh)); } else { if (extack->cookie_len) WARN_ON(nla_put(skb, NLMSGERR_ATTR_COOKIE, -- cgit From 8a171c5cc9de7fb0a485341b790aba736978ed73 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 27 Feb 2020 18:09:44 -0800 Subject: atm: nicstar: fix if-statement empty body warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When debugging via PRINTK() is not enabled, make the PRINTK() macro be an empty do-while block. Thix fixes a gcc warning when -Wextra is set: ../drivers/atm/nicstar.c:1819:23: warning: suggest braces around empty body in an ‘else’ statement [-Wempty-body] I have verified that there is no object code change (with gcc 7.5.0). Signed-off-by: Randy Dunlap Cc: Chas Williams <3chas3@gmail.com> Cc: linux-atm-general@lists.sourceforge.net Cc: netdev@vger.kernel.org Cc: David S. Miller Signed-off-by: David S. Miller --- drivers/atm/nicstar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/atm/nicstar.c b/drivers/atm/nicstar.c index 8db8c0fb5e2d..7af74fb450a0 100644 --- a/drivers/atm/nicstar.c +++ b/drivers/atm/nicstar.c @@ -91,7 +91,7 @@ #ifdef GENERAL_DEBUG #define PRINTK(args...) printk(args) #else -#define PRINTK(args...) +#define PRINTK(args...) do {} while (0) #endif /* GENERAL_DEBUG */ #ifdef EXTRA_DEBUG -- cgit From 0395823b8d9a4d87bd1bf74359123461c2ae801b Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 28 Feb 2020 19:39:41 +0000 Subject: net: dsa: mv88e6xxx: fix lockup on warm boot If the switch is not hardware reset on a warm boot, interrupts can be left enabled, and possibly pending. This will cause us to enter an infinite loop trying to service an interrupt we are unable to handle, thereby preventing the kernel from booting. Ensure that the global 2 interrupt sources are disabled before we claim the parent interrupt. Observed on the ZII development revision B and C platforms with reworked serdes support, and using reboot -f to reboot the platform. Fixes: dc30c35be720 ("net: dsa: mv88e6xxx: Implement interrupt support.") Signed-off-by: Russell King Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx/global2.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c index 01503014b1ee..8fd483020c5b 100644 --- a/drivers/net/dsa/mv88e6xxx/global2.c +++ b/drivers/net/dsa/mv88e6xxx/global2.c @@ -1099,6 +1099,13 @@ int mv88e6xxx_g2_irq_setup(struct mv88e6xxx_chip *chip) { int err, irq, virq; + chip->g2_irq.masked = ~0; + mv88e6xxx_reg_lock(chip); + err = mv88e6xxx_g2_int_mask(chip, ~chip->g2_irq.masked); + mv88e6xxx_reg_unlock(chip); + if (err) + return err; + chip->g2_irq.domain = irq_domain_add_simple( chip->dev->of_node, 16, 0, &mv88e6xxx_g2_irq_domain_ops, chip); if (!chip->g2_irq.domain) @@ -1108,7 +1115,6 @@ int mv88e6xxx_g2_irq_setup(struct mv88e6xxx_chip *chip) irq_create_mapping(chip->g2_irq.domain, irq); chip->g2_irq.chip = mv88e6xxx_g2_irq_chip; - chip->g2_irq.masked = ~0; chip->device_irq = irq_find_mapping(chip->g1_irq.domain, MV88E6XXX_G1_STS_IRQ_DEVICE); -- cgit From 07758eb9ff52794fba15d03aa88d92dbd1b7d125 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Sat, 29 Feb 2020 17:27:13 +0800 Subject: net/ipv6: use configured metric when add peer route When we add peer address with metric configured, IPv4 could set the dest metric correctly, but IPv6 do not. e.g. ]# ip addr add 192.0.2.1 peer 192.0.2.2/32 dev eth1 metric 20 ]# ip route show dev eth1 192.0.2.2 proto kernel scope link src 192.0.2.1 metric 20 ]# ip addr add 2001:db8::1 peer 2001:db8::2/128 dev eth1 metric 20 ]# ip -6 route show dev eth1 2001:db8::1 proto kernel metric 20 pref medium 2001:db8::2 proto kernel metric 256 pref medium Fix this by using configured metric instead of default one. Reported-by: Jianlin Shi Fixes: 8308f3ff1753 ("net/ipv6: Add support for specifying metric of connected routes") Reviewed-by: David Ahern Signed-off-by: Hangbin Liu Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index cb493e15959c..164c71c54b5c 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5983,9 +5983,9 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp) if (ifp->idev->cnf.forwarding) addrconf_join_anycast(ifp); if (!ipv6_addr_any(&ifp->peer_addr)) - addrconf_prefix_route(&ifp->peer_addr, 128, 0, - ifp->idev->dev, 0, 0, - GFP_ATOMIC); + addrconf_prefix_route(&ifp->peer_addr, 128, + ifp->rt_priority, ifp->idev->dev, + 0, 0, GFP_ATOMIC); break; case RTM_DELADDR: if (ifp->idev->cnf.forwarding) -- cgit From 52c0d4e306ca0261a896f5e2ad823112195b7f4b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 29 Feb 2020 22:30:07 +0200 Subject: net: dsa: sja1105: Don't destroy not-yet-created xmit_worker Fixes the following NULL pointer dereference on PHY connect error path teardown: [ 2.291010] sja1105 spi0.1: Probed switch chip: SJA1105T [ 2.310044] sja1105 spi0.1: Enabled switch tagging [ 2.314970] fsl-gianfar soc:ethernet@2d90000 eth2: error -19 setting up slave phy [ 2.322463] 8<--- cut here --- [ 2.325497] Unable to handle kernel NULL pointer dereference at virtual address 00000018 [ 2.333555] pgd = (ptrval) [ 2.336241] [00000018] *pgd=00000000 [ 2.339797] Internal error: Oops: 5 [#1] SMP ARM [ 2.344384] Modules linked in: [ 2.347420] CPU: 1 PID: 64 Comm: kworker/1:1 Not tainted 5.5.0-rc5 #1 [ 2.353820] Hardware name: Freescale LS1021A [ 2.358070] Workqueue: events deferred_probe_work_func [ 2.363182] PC is at kthread_destroy_worker+0x4/0x74 [ 2.368117] LR is at sja1105_teardown+0x70/0xb4 [ 2.372617] pc : [] lr : [] psr: 60000013 [ 2.378845] sp : eeac3d30 ip : eeab1900 fp : eef45480 [ 2.384036] r10: eef4549c r9 : 00000001 r8 : 00000000 [ 2.389227] r7 : eef527c0 r6 : 00000034 r5 : ed8ddd0c r4 : ed8ddc40 [ 2.395714] r3 : 00000000 r2 : 00000000 r1 : eef4549c r0 : 00000000 [ 2.402204] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none [ 2.409297] Control: 10c5387d Table: 8020406a DAC: 00000051 [ 2.415008] Process kworker/1:1 (pid: 64, stack limit = 0x(ptrval)) [ 2.421237] Stack: (0xeeac3d30 to 0xeeac4000) [ 2.612635] [] (kthread_destroy_worker) from [] (sja1105_teardown+0x70/0xb4) [ 2.621379] [] (sja1105_teardown) from [] (dsa_switch_teardown.part.1+0x48/0x74) [ 2.630467] [] (dsa_switch_teardown.part.1) from [] (dsa_register_switch+0x8b0/0xbf4) [ 2.639984] [] (dsa_register_switch) from [] (sja1105_probe+0x2ac/0x464) [ 2.648378] [] (sja1105_probe) from [] (spi_drv_probe+0x7c/0xa0) [ 2.656081] [] (spi_drv_probe) from [] (really_probe+0x208/0x480) [ 2.663871] [] (really_probe) from [] (driver_probe_device+0x78/0x1c4) [ 2.672093] [] (driver_probe_device) from [] (bus_for_each_drv+0x80/0xc4) [ 2.680574] [] (bus_for_each_drv) from [] (__device_attach+0xd0/0x168) [ 2.688794] [] (__device_attach) from [] (bus_probe_device+0x84/0x8c) [ 2.696927] [] (bus_probe_device) from [] (deferred_probe_work_func+0x84/0xc4) [ 2.705842] [] (deferred_probe_work_func) from [] (process_one_work+0x22c/0x560) [ 2.714926] [] (process_one_work) from [] (worker_thread+0x2a8/0x5d4) [ 2.723059] [] (worker_thread) from [] (kthread+0x150/0x154) [ 2.730416] [] (kthread) from [] (ret_from_fork+0x14/0x2c) Checking for NULL pointer is correct because the per-port xmit kernel threads are created in sja1105_probe immediately after calling dsa_register_switch. Fixes: a68578c20a96 ("net: dsa: Make deferred_xmit private to sja1105") Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- drivers/net/dsa/sja1105/sja1105_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c index 03ba6d25f7fe..7edea5741a5f 100644 --- a/drivers/net/dsa/sja1105/sja1105_main.c +++ b/drivers/net/dsa/sja1105/sja1105_main.c @@ -1741,7 +1741,8 @@ static void sja1105_teardown(struct dsa_switch *ds) if (!dsa_is_user_port(ds, port)) continue; - kthread_destroy_worker(sp->xmit_worker); + if (sp->xmit_worker) + kthread_destroy_worker(sp->xmit_worker); } sja1105_tas_teardown(ds); -- cgit From 249bc9744e165abe74ae326f43e9d70bad54c3b7 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sun, 1 Mar 2020 21:36:09 +0100 Subject: net: phy: avoid clearing PHY interrupts twice in irq handler On all PHY drivers that implement did_interrupt() reading the interrupt status bits clears them. This means we may loose an interrupt that is triggered between calling did_interrupt() and phy_clear_interrupt(). As part of the fix make it a requirement that did_interrupt() clears the interrupt. The Fixes tag refers to the first commit where the patch applies cleanly. Fixes: 49644e68f472 ("net: phy: add callback for custom interrupt handler to struct phy_driver") Reported-by: Michael Walle Signed-off-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 3 ++- include/linux/phy.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index d76e038cf2cb..355bfdef48d2 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -727,7 +727,8 @@ static irqreturn_t phy_interrupt(int irq, void *phy_dat) phy_trigger_machine(phydev); } - if (phy_clear_interrupt(phydev)) + /* did_interrupt() may have cleared the interrupt already */ + if (!phydev->drv->did_interrupt && phy_clear_interrupt(phydev)) goto phy_err; return IRQ_HANDLED; diff --git a/include/linux/phy.h b/include/linux/phy.h index c570e162e05e..22f5e763e894 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -557,6 +557,7 @@ struct phy_driver { /* * Checks if the PHY generated an interrupt. * For multi-PHY devices with shared PHY interrupt pin + * Set interrupt bits have to be cleared. */ int (*did_interrupt)(struct phy_device *phydev); -- cgit From a9b952d267e59a3b405e644930f46d252cea7122 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Sun, 1 Mar 2020 22:07:17 -0500 Subject: bnxt_en: reinitialize IRQs when MTU is modified MTU changes may affect the number of IRQs so we must call bnxt_close_nic()/bnxt_open_nic() with the irq_re_init parameter set to true. The reason is that a larger MTU may require aggregation rings not needed with smaller MTU. We may not be able to allocate the required number of aggregation rings and so we reduce the number of channels which will change the number of IRQs. Without this patch, it may crash eventually in pci_disable_msix() when the IRQs are not properly unwound. Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index f9a8151f092c..c5c8effc0139 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -10982,13 +10982,13 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu) struct bnxt *bp = netdev_priv(dev); if (netif_running(dev)) - bnxt_close_nic(bp, false, false); + bnxt_close_nic(bp, true, false); dev->mtu = new_mtu; bnxt_set_ring_params(bp); if (netif_running(dev)) - return bnxt_open_nic(bp, false, false); + return bnxt_open_nic(bp, true, false); return 0; } -- cgit From 22630e28f9c2b55abd217869cc0696def89f2284 Mon Sep 17 00:00:00 2001 From: Edwin Peer Date: Sun, 1 Mar 2020 22:07:18 -0500 Subject: bnxt_en: fix error handling when flashing from file After bnxt_hwrm_do_send_message() was updated to return standard error codes in a recent commit, a regression in bnxt_flash_package_from_file() was introduced. The return value does not properly reflect all possible firmware errors when calling firmware to flash the package. Fix it by consolidating all errors in one local variable rc instead of having 2 variables for different errors. Fixes: d4f1420d3656 ("bnxt_en: Convert error code in firmware message response to standard code.") Signed-off-by: Edwin Peer Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 24 +++++++++++------------ 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index e8fc1671c581..1f67e6729a2c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -2007,8 +2007,8 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, struct hwrm_nvm_install_update_output *resp = bp->hwrm_cmd_resp_addr; struct hwrm_nvm_install_update_input install = {0}; const struct firmware *fw; - int rc, hwrm_err = 0; u32 item_len; + int rc = 0; u16 index; bnxt_hwrm_fw_set_time(bp); @@ -2052,15 +2052,14 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, memcpy(kmem, fw->data, fw->size); modify.host_src_addr = cpu_to_le64(dma_handle); - hwrm_err = hwrm_send_message(bp, &modify, - sizeof(modify), - FLASH_PACKAGE_TIMEOUT); + rc = hwrm_send_message(bp, &modify, sizeof(modify), + FLASH_PACKAGE_TIMEOUT); dma_free_coherent(&bp->pdev->dev, fw->size, kmem, dma_handle); } } release_firmware(fw); - if (rc || hwrm_err) + if (rc) goto err_exit; if ((install_type & 0xffff) == 0) @@ -2069,20 +2068,19 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, install.install_type = cpu_to_le32(install_type); mutex_lock(&bp->hwrm_cmd_lock); - hwrm_err = _hwrm_send_message(bp, &install, sizeof(install), - INSTALL_PACKAGE_TIMEOUT); - if (hwrm_err) { + rc = _hwrm_send_message(bp, &install, sizeof(install), + INSTALL_PACKAGE_TIMEOUT); + if (rc) { u8 error_code = ((struct hwrm_err_output *)resp)->cmd_err; if (resp->error_code && error_code == NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR) { install.flags |= cpu_to_le16( NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG); - hwrm_err = _hwrm_send_message(bp, &install, - sizeof(install), - INSTALL_PACKAGE_TIMEOUT); + rc = _hwrm_send_message(bp, &install, sizeof(install), + INSTALL_PACKAGE_TIMEOUT); } - if (hwrm_err) + if (rc) goto flash_pkg_exit; } @@ -2094,7 +2092,7 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, flash_pkg_exit: mutex_unlock(&bp->hwrm_cmd_lock); err_exit: - if (hwrm_err == -EACCES) + if (rc == -EACCES) bnxt_print_admin_err(bp); return rc; } -- cgit From b102f0c522cf668c8382c56a4f771b37d011cda2 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 20 Feb 2020 12:41:39 +0100 Subject: mt76: fix array overflow on receiving too many fragments for a packet If the hardware receives an oversized packet with too many rx fragments, skb_shinfo(skb)->frags can overflow and corrupt memory of adjacent pages. This becomes especially visible if it corrupts the freelist pointer of a slab page. Cc: stable@vger.kernel.org Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo --- drivers/net/wireless/mediatek/mt76/dma.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 6173c80189ba..1847f55e199b 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -447,10 +447,13 @@ mt76_add_fragment(struct mt76_dev *dev, struct mt76_queue *q, void *data, struct page *page = virt_to_head_page(data); int offset = data - page_address(page); struct sk_buff *skb = q->rx_head; + struct skb_shared_info *shinfo = skb_shinfo(skb); - offset += q->buf_offset; - skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, offset, len, - q->buf_size); + if (shinfo->nr_frags < ARRAY_SIZE(shinfo->frags)) { + offset += q->buf_offset; + skb_add_rx_frag(skb, shinfo->nr_frags, page, offset, len, + q->buf_size); + } if (more) return; -- cgit From 8750939b6ad86abc3f53ec8a9683a1cded4a5654 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:11 -0800 Subject: devlink: validate length of param values DEVLINK_ATTR_PARAM_VALUE_DATA may have different types so it's not checked by the normal netlink policy. Make sure the attribute length is what we expect. Fixes: e3b7ca18ad7b ("devlink: Add param set command") Signed-off-by: Jakub Kicinski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/devlink.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/net/core/devlink.c b/net/core/devlink.c index 5e220809844c..8e44dc5cde73 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -3352,34 +3352,41 @@ devlink_param_value_get_from_info(const struct devlink_param *param, struct genl_info *info, union devlink_param_value *value) { + struct nlattr *param_data; int len; - if (param->type != DEVLINK_PARAM_TYPE_BOOL && - !info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]) + param_data = info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]; + + if (param->type != DEVLINK_PARAM_TYPE_BOOL && !param_data) return -EINVAL; switch (param->type) { case DEVLINK_PARAM_TYPE_U8: - value->vu8 = nla_get_u8(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]); + if (nla_len(param_data) != sizeof(u8)) + return -EINVAL; + value->vu8 = nla_get_u8(param_data); break; case DEVLINK_PARAM_TYPE_U16: - value->vu16 = nla_get_u16(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]); + if (nla_len(param_data) != sizeof(u16)) + return -EINVAL; + value->vu16 = nla_get_u16(param_data); break; case DEVLINK_PARAM_TYPE_U32: - value->vu32 = nla_get_u32(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]); + if (nla_len(param_data) != sizeof(u32)) + return -EINVAL; + value->vu32 = nla_get_u32(param_data); break; case DEVLINK_PARAM_TYPE_STRING: - len = strnlen(nla_data(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]), - nla_len(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA])); - if (len == nla_len(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA]) || + len = strnlen(nla_data(param_data), nla_len(param_data)); + if (len == nla_len(param_data) || len >= __DEVLINK_PARAM_MAX_STRING_VALUE) return -EINVAL; - strcpy(value->vstr, - nla_data(info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA])); + strcpy(value->vstr, nla_data(param_data)); break; case DEVLINK_PARAM_TYPE_BOOL: - value->vbool = info->attrs[DEVLINK_ATTR_PARAM_VALUE_DATA] ? - true : false; + if (param_data && nla_len(param_data)) + return -EINVAL; + value->vbool = nla_get_flag(param_data); break; } return 0; -- cgit From ff3b63b8c299b73ac599b120653b47e275407656 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:12 -0800 Subject: devlink: validate length of region addr/len DEVLINK_ATTR_REGION_CHUNK_ADDR and DEVLINK_ATTR_REGION_CHUNK_LEN lack entries in the netlink policy. Corresponding nla_get_u64()s may read beyond the end of the message. Fixes: 4e54795a27f5 ("devlink: Add support for region snapshot read command") Signed-off-by: Jakub Kicinski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/devlink.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/core/devlink.c b/net/core/devlink.c index 8e44dc5cde73..b831c5545d6a 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -5958,6 +5958,8 @@ static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_PARAM_VALUE_CMODE] = { .type = NLA_U8 }, [DEVLINK_ATTR_REGION_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_REGION_SNAPSHOT_ID] = { .type = NLA_U32 }, + [DEVLINK_ATTR_REGION_CHUNK_ADDR] = { .type = NLA_U64 }, + [DEVLINK_ATTR_REGION_CHUNK_LEN] = { .type = NLA_U64 }, [DEVLINK_ATTR_HEALTH_REPORTER_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_HEALTH_REPORTER_GRACEFUL_PERIOD] = { .type = NLA_U64 }, [DEVLINK_ATTR_HEALTH_REPORTER_AUTO_RECOVER] = { .type = NLA_U8 }, -- cgit From 4c16d64ea04056f1b1b324ab6916019f6a064114 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:13 -0800 Subject: fib: add missing attribute validation for tun_id Add missing netlink policy entry for FRA_TUN_ID. Fixes: e7030878fc84 ("fib: Add fib rule match on tunnel id") Signed-off-by: Jakub Kicinski Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/net/fib_rules.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 54e227e6b06a..a259050f84af 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -108,6 +108,7 @@ struct fib_rule_notifier_info { [FRA_OIFNAME] = { .type = NLA_STRING, .len = IFNAMSIZ - 1 }, \ [FRA_PRIORITY] = { .type = NLA_U32 }, \ [FRA_FWMARK] = { .type = NLA_U32 }, \ + [FRA_TUN_ID] = { .type = NLA_U64 }, \ [FRA_FWMASK] = { .type = NLA_U32 }, \ [FRA_TABLE] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_PREFIXLEN] = { .type = NLA_U32 }, \ -- cgit From 9322cd7c4af2ccc7fe7c5f01adb53f4f77949e92 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:14 -0800 Subject: nl802154: add missing attribute validation Add missing attribute validation for several u8 types. Fixes: 2c21d11518b6 ("net: add NL802154 interface for configuration of 802.15.4 devices") Signed-off-by: Jakub Kicinski Acked-by: Stefan Schmidt Signed-off-by: David S. Miller --- net/ieee802154/nl_policy.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/ieee802154/nl_policy.c b/net/ieee802154/nl_policy.c index 2c7a38d76a3a..824e7e84014c 100644 --- a/net/ieee802154/nl_policy.c +++ b/net/ieee802154/nl_policy.c @@ -21,6 +21,11 @@ const struct nla_policy ieee802154_policy[IEEE802154_ATTR_MAX + 1] = { [IEEE802154_ATTR_HW_ADDR] = { .type = NLA_HW_ADDR, }, [IEEE802154_ATTR_PAN_ID] = { .type = NLA_U16, }, [IEEE802154_ATTR_CHANNEL] = { .type = NLA_U8, }, + [IEEE802154_ATTR_BCN_ORD] = { .type = NLA_U8, }, + [IEEE802154_ATTR_SF_ORD] = { .type = NLA_U8, }, + [IEEE802154_ATTR_PAN_COORD] = { .type = NLA_U8, }, + [IEEE802154_ATTR_BAT_EXT] = { .type = NLA_U8, }, + [IEEE802154_ATTR_COORD_REALIGN] = { .type = NLA_U8, }, [IEEE802154_ATTR_PAGE] = { .type = NLA_U8, }, [IEEE802154_ATTR_COORD_SHORT_ADDR] = { .type = NLA_U16, }, [IEEE802154_ATTR_COORD_HW_ADDR] = { .type = NLA_HW_ADDR, }, -- cgit From b60673c4c418bef7550d02faf53c34fbfeb366bf Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:15 -0800 Subject: nl802154: add missing attribute validation for dev_type Add missing attribute type validation for IEEE802154_ATTR_DEV_TYPE to the netlink policy. Fixes: 90c049b2c6ae ("ieee802154: interface type to be added") Signed-off-by: Jakub Kicinski Acked-by: Stefan Schmidt Signed-off-by: David S. Miller --- net/ieee802154/nl_policy.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ieee802154/nl_policy.c b/net/ieee802154/nl_policy.c index 824e7e84014c..0672b2f01586 100644 --- a/net/ieee802154/nl_policy.c +++ b/net/ieee802154/nl_policy.c @@ -27,6 +27,7 @@ const struct nla_policy ieee802154_policy[IEEE802154_ATTR_MAX + 1] = { [IEEE802154_ATTR_BAT_EXT] = { .type = NLA_U8, }, [IEEE802154_ATTR_COORD_REALIGN] = { .type = NLA_U8, }, [IEEE802154_ATTR_PAGE] = { .type = NLA_U8, }, + [IEEE802154_ATTR_DEV_TYPE] = { .type = NLA_U8, }, [IEEE802154_ATTR_COORD_SHORT_ADDR] = { .type = NLA_U16, }, [IEEE802154_ATTR_COORD_HW_ADDR] = { .type = NLA_HW_ADDR, }, [IEEE802154_ATTR_COORD_PAN_ID] = { .type = NLA_U16, }, -- cgit From ab02ad660586b94f5d08912a3952b939cf4c4430 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:16 -0800 Subject: can: add missing attribute validation for termination Add missing attribute validation for IFLA_CAN_TERMINATION to the netlink policy. Fixes: 12a6075cabc0 ("can: dev: add CAN interface termination API") Signed-off-by: Jakub Kicinski Acked-by: Oliver Hartkopp Signed-off-by: David S. Miller --- drivers/net/can/dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index 6ee06a49fb4c..68834a2853c9 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -883,6 +883,7 @@ static const struct nla_policy can_policy[IFLA_CAN_MAX + 1] = { = { .len = sizeof(struct can_bittiming) }, [IFLA_CAN_DATA_BITTIMING_CONST] = { .len = sizeof(struct can_bittiming_const) }, + [IFLA_CAN_TERMINATION] = { .type = NLA_U16 }, }; static int can_validate(struct nlattr *tb[], struct nlattr *data[], -- cgit From 31d9a1c524964bac77b7f9d0a1ac140dc6b57461 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:17 -0800 Subject: macsec: add missing attribute validation for port Add missing attribute validation for IFLA_MACSEC_PORT to the netlink policy. Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- drivers/net/macsec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 45bfd99f17fa..5af424eeea86 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3342,6 +3342,7 @@ static const struct device_type macsec_type = { static const struct nla_policy macsec_rtnl_policy[IFLA_MACSEC_MAX + 1] = { [IFLA_MACSEC_SCI] = { .type = NLA_U64 }, + [IFLA_MACSEC_PORT] = { .type = NLA_U16 }, [IFLA_MACSEC_ICV_LEN] = { .type = NLA_U8 }, [IFLA_MACSEC_CIPHER_SUITE] = { .type = NLA_U64 }, [IFLA_MACSEC_WINDOW] = { .type = NLA_U32 }, -- cgit From b5ab1f1be6180a2e975eede18731804b5164a05d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:18 -0800 Subject: openvswitch: add missing attribute validation for hash Add missing attribute validation for OVS_PACKET_ATTR_HASH to the netlink policy. Fixes: bd1903b7c459 ("net: openvswitch: add hash info to upcall") Signed-off-by: Jakub Kicinski Reviewed-by: Greg Rose Signed-off-by: David S. Miller --- net/openvswitch/datapath.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index c047afd12116..07a7dd185995 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -645,6 +645,7 @@ static const struct nla_policy packet_policy[OVS_PACKET_ATTR_MAX + 1] = { [OVS_PACKET_ATTR_ACTIONS] = { .type = NLA_NESTED }, [OVS_PACKET_ATTR_PROBE] = { .type = NLA_FLAG }, [OVS_PACKET_ATTR_MRU] = { .type = NLA_U16 }, + [OVS_PACKET_ATTR_HASH] = { .type = NLA_U64 }, }; static const struct genl_ops dp_packet_genl_ops[] = { -- cgit From 7e6dc03eeb023e18427a373522f1d247b916a641 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:19 -0800 Subject: net: fq: add missing attribute validation for orphan mask Add missing attribute validation for TCA_FQ_ORPHAN_MASK to the netlink policy. Fixes: 06eb395fa985 ("pkt_sched: fq: better control of DDOS traffic") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/sched/sch_fq.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index a5a295477ecc..371ad84def3b 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -744,6 +744,7 @@ static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = { [TCA_FQ_FLOW_MAX_RATE] = { .type = NLA_U32 }, [TCA_FQ_BUCKETS_LOG] = { .type = NLA_U32 }, [TCA_FQ_FLOW_REFILL_DELAY] = { .type = NLA_U32 }, + [TCA_FQ_ORPHAN_MASK] = { .type = NLA_U32 }, [TCA_FQ_LOW_RATE_THRESHOLD] = { .type = NLA_U32 }, [TCA_FQ_CE_THRESHOLD] = { .type = NLA_U32 }, }; -- cgit From e13aaa0643da10006ec35715954e7f92a62899a5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:20 -0800 Subject: net: taprio: add missing attribute validation for txtime delay Add missing attribute validation for TCA_TAPRIO_ATTR_TXTIME_DELAY to the netlink policy. Fixes: 4cfd5779bd6e ("taprio: Add support for txtime-assist mode") Signed-off-by: Jakub Kicinski Reviewed-by: Vinicius Costa Gomes Signed-off-by: David S. Miller --- net/sched/sch_taprio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index 660fc45ee40f..ee717e05372b 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -768,6 +768,7 @@ static const struct nla_policy taprio_policy[TCA_TAPRIO_ATTR_MAX + 1] = { [TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME] = { .type = NLA_S64 }, [TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME_EXTENSION] = { .type = NLA_S64 }, [TCA_TAPRIO_ATTR_FLAGS] = { .type = NLA_U32 }, + [TCA_TAPRIO_ATTR_TXTIME_DELAY] = { .type = NLA_U32 }, }; static int fill_sched_entry(struct nlattr **tb, struct sched_entry *entry, -- cgit From dd25cb272ccce4db67dc8509278229099e4f5e99 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:21 -0800 Subject: team: add missing attribute validation for port ifindex Add missing attribute validation for TEAM_ATTR_OPTION_PORT_IFINDEX to the netlink policy. Fixes: 80f7c6683fe0 ("team: add support for per-port options") Signed-off-by: Jakub Kicinski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/team/team.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index ca70a1d840eb..44dd26a62a6d 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2240,6 +2240,7 @@ team_nl_option_policy[TEAM_ATTR_OPTION_MAX + 1] = { [TEAM_ATTR_OPTION_CHANGED] = { .type = NLA_FLAG }, [TEAM_ATTR_OPTION_TYPE] = { .type = NLA_U8 }, [TEAM_ATTR_OPTION_DATA] = { .type = NLA_BINARY }, + [TEAM_ATTR_OPTION_PORT_IFINDEX] = { .type = NLA_U32 }, }; static int team_nl_cmd_noop(struct sk_buff *skb, struct genl_info *info) -- cgit From 669fcd7795900cd1880237cbbb57a7db66cb9ac8 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:22 -0800 Subject: team: add missing attribute validation for array index Add missing attribute validation for TEAM_ATTR_OPTION_ARRAY_INDEX to the netlink policy. Fixes: b13033262d24 ("team: introduce array options") Signed-off-by: Jakub Kicinski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/team/team.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 44dd26a62a6d..4004f98e50d9 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2241,6 +2241,7 @@ team_nl_option_policy[TEAM_ATTR_OPTION_MAX + 1] = { [TEAM_ATTR_OPTION_TYPE] = { .type = NLA_U8 }, [TEAM_ATTR_OPTION_DATA] = { .type = NLA_BINARY }, [TEAM_ATTR_OPTION_PORT_IFINDEX] = { .type = NLA_U32 }, + [TEAM_ATTR_OPTION_ARRAY_INDEX] = { .type = NLA_U32 }, }; static int team_nl_cmd_noop(struct sk_buff *skb, struct genl_info *info) -- cgit From 213320a67962ff6e7b83b704d55cbebc341426db Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:23 -0800 Subject: tipc: add missing attribute validation for MTU property Add missing attribute validation for TIPC_NLA_PROP_MTU to the netlink policy. Fixes: 901271e0403a ("tipc: implement configuration of UDP media MTU") Signed-off-by: Jakub Kicinski Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/tipc/netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 7c35094c20b8..bb9862410e68 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -116,6 +116,7 @@ const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = { [TIPC_NLA_PROP_PRIO] = { .type = NLA_U32 }, [TIPC_NLA_PROP_TOL] = { .type = NLA_U32 }, [TIPC_NLA_PROP_WIN] = { .type = NLA_U32 }, + [TIPC_NLA_PROP_MTU] = { .type = NLA_U32 }, [TIPC_NLA_PROP_BROADCAST] = { .type = NLA_U32 }, [TIPC_NLA_PROP_BROADCAST_RATIO] = { .type = NLA_U32 } }; -- cgit From 361d23e41ca6e504033f7e66a03b95788377caae Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:24 -0800 Subject: nfc: add missing attribute validation for SE API Add missing attribute validation for NFC_ATTR_SE_INDEX to the netlink policy. Fixes: 5ce3f32b5264 ("NFC: netlink: SE API implementation") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/nfc/netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index eee0dddb7749..842407a48f96 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -43,6 +43,7 @@ static const struct nla_policy nfc_genl_policy[NFC_ATTR_MAX + 1] = { [NFC_ATTR_LLC_SDP] = { .type = NLA_NESTED }, [NFC_ATTR_FIRMWARE_NAME] = { .type = NLA_STRING, .len = NFC_FIRMWARE_NAME_MAXSIZE }, + [NFC_ATTR_SE_INDEX] = { .type = NLA_U32 }, [NFC_ATTR_SE_APDU] = { .type = NLA_BINARY }, [NFC_ATTR_VENDOR_DATA] = { .type = NLA_BINARY }, -- cgit From 88e706d5168b07df4792dbc3d1bc37b83e4bd74d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:25 -0800 Subject: nfc: add missing attribute validation for deactivate target Add missing attribute validation for NFC_ATTR_TARGET_INDEX to the netlink policy. Fixes: 4d63adfe12dd ("NFC: Add NFC_CMD_DEACTIVATE_TARGET support") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/nfc/netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index 842407a48f96..e988ca486d66 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -32,6 +32,7 @@ static const struct nla_policy nfc_genl_policy[NFC_ATTR_MAX + 1] = { [NFC_ATTR_DEVICE_NAME] = { .type = NLA_STRING, .len = NFC_DEVICE_NAME_MAXSIZE }, [NFC_ATTR_PROTOCOLS] = { .type = NLA_U32 }, + [NFC_ATTR_TARGET_INDEX] = { .type = NLA_U32 }, [NFC_ATTR_COMM_MODE] = { .type = NLA_U8 }, [NFC_ATTR_RF_MODE] = { .type = NLA_U8 }, [NFC_ATTR_DEVICE_POWERED] = { .type = NLA_U8 }, -- cgit From 6ba3da446551f2150fadbf8c7788edcb977683d3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:05:26 -0800 Subject: nfc: add missing attribute validation for vendor subcommand Add missing attribute validation for vendor subcommand attributes to the netlink policy. Fixes: 9e58095f9660 ("NFC: netlink: Implement vendor command support") Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/nfc/netlink.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index e988ca486d66..e894254c17d4 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -46,6 +46,8 @@ static const struct nla_policy nfc_genl_policy[NFC_ATTR_MAX + 1] = { .len = NFC_FIRMWARE_NAME_MAXSIZE }, [NFC_ATTR_SE_INDEX] = { .type = NLA_U32 }, [NFC_ATTR_SE_APDU] = { .type = NLA_BINARY }, + [NFC_ATTR_VENDOR_ID] = { .type = NLA_U32 }, + [NFC_ATTR_VENDOR_SUBCMD] = { .type = NLA_U32 }, [NFC_ATTR_VENDOR_DATA] = { .type = NLA_BINARY }, }; -- cgit From 617940123e0140521f3080d2befc2bf55bcda094 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 3 Mar 2020 14:37:34 +0800 Subject: net/ipv6: need update peer route when modify metric When we modify the route metric, the peer address's route need also be updated. Before the fix: + ip addr add dev dummy1 2001:db8::1 peer 2001:db8::2 metric 60 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 60 pref medium 2001:db8::2 proto kernel metric 60 pref medium + ip addr change dev dummy1 2001:db8::1 peer 2001:db8::2 metric 61 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 61 pref medium 2001:db8::2 proto kernel metric 60 pref medium After the fix: + ip addr change dev dummy1 2001:db8::1 peer 2001:db8::2 metric 61 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 61 pref medium 2001:db8::2 proto kernel metric 61 pref medium Fixes: 8308f3ff1753 ("net/ipv6: Add support for specifying metric of connected routes") Signed-off-by: Hangbin Liu Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 164c71c54b5c..4fb72028ca45 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4586,12 +4586,14 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, } static int modify_prefix_route(struct inet6_ifaddr *ifp, - unsigned long expires, u32 flags) + unsigned long expires, u32 flags, + bool modify_peer) { struct fib6_info *f6i; u32 prio; - f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len, + f6i = addrconf_get_prefix_route(modify_peer ? &ifp->peer_addr : &ifp->addr, + ifp->prefix_len, ifp->idev->dev, 0, RTF_DEFAULT, true); if (!f6i) return -ENOENT; @@ -4602,7 +4604,8 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp, ip6_del_rt(dev_net(ifp->idev->dev), f6i); /* add new one */ - addrconf_prefix_route(&ifp->addr, ifp->prefix_len, + addrconf_prefix_route(modify_peer ? &ifp->peer_addr : &ifp->addr, + ifp->prefix_len, ifp->rt_priority, ifp->idev->dev, expires, flags, GFP_KERNEL); } else { @@ -4678,7 +4681,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) int rc = -ENOENT; if (had_prefixroute) - rc = modify_prefix_route(ifp, expires, flags); + rc = modify_prefix_route(ifp, expires, flags, false); /* prefix route could have been deleted; if so restore it */ if (rc == -ENOENT) { @@ -4686,6 +4689,15 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) ifp->rt_priority, ifp->idev->dev, expires, flags, GFP_KERNEL); } + + if (had_prefixroute && !ipv6_addr_any(&ifp->peer_addr)) + rc = modify_prefix_route(ifp, expires, flags, true); + + if (rc == -ENOENT && !ipv6_addr_any(&ifp->peer_addr)) { + addrconf_prefix_route(&ifp->peer_addr, ifp->prefix_len, + ifp->rt_priority, ifp->idev->dev, + expires, flags, GFP_KERNEL); + } } else if (had_prefixroute) { enum cleanup_prefix_rt_t action; unsigned long rt_expires; -- cgit From d0098e4c6b83e502cc1cd96d67ca86bc79a6c559 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 3 Mar 2020 14:37:35 +0800 Subject: net/ipv6: remove the old peer route if change it to a new one When we modify the peer route and changed it to a new one, we should remove the old route first. Before the fix: + ip addr add dev dummy1 2001:db8::1 peer 2001:db8::2 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 256 pref medium 2001:db8::2 proto kernel metric 256 pref medium + ip addr change dev dummy1 2001:db8::1 peer 2001:db8::3 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 256 pref medium 2001:db8::2 proto kernel metric 256 pref medium After the fix: + ip addr change dev dummy1 2001:db8::1 peer 2001:db8::3 + ip -6 route show dev dummy1 2001:db8::1 proto kernel metric 256 pref medium 2001:db8::3 proto kernel metric 256 pref medium This patch depend on the previous patch "net/ipv6: need update peer route when modify metric" to update new peer route after delete old one. Signed-off-by: Hangbin Liu Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4fb72028ca45..e6e1290ea06f 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1226,11 +1226,13 @@ check_cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long *expires) } static void -cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, bool del_rt) +cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, + bool del_rt, bool del_peer) { struct fib6_info *f6i; - f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len, + f6i = addrconf_get_prefix_route(del_peer ? &ifp->peer_addr : &ifp->addr, + ifp->prefix_len, ifp->idev->dev, 0, RTF_DEFAULT, true); if (f6i) { if (del_rt) @@ -1293,7 +1295,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) if (action != CLEANUP_PREFIX_RT_NOP) { cleanup_prefix_route(ifp, expires, - action == CLEANUP_PREFIX_RT_DEL); + action == CLEANUP_PREFIX_RT_DEL, false); } /* clean up prefsrc entries */ @@ -4627,6 +4629,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) unsigned long timeout; bool was_managetempaddr; bool had_prefixroute; + bool new_peer = false; ASSERT_RTNL(); @@ -4658,6 +4661,13 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) cfg->preferred_lft = timeout; } + if (cfg->peer_pfx && + memcmp(&ifp->peer_addr, cfg->peer_pfx, sizeof(struct in6_addr))) { + if (!ipv6_addr_any(&ifp->peer_addr)) + cleanup_prefix_route(ifp, expires, true, true); + new_peer = true; + } + spin_lock_bh(&ifp->lock); was_managetempaddr = ifp->flags & IFA_F_MANAGETEMPADDR; had_prefixroute = ifp->flags & IFA_F_PERMANENT && @@ -4673,6 +4683,9 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) if (cfg->rt_priority && cfg->rt_priority != ifp->rt_priority) ifp->rt_priority = cfg->rt_priority; + if (new_peer) + ifp->peer_addr = *cfg->peer_pfx; + spin_unlock_bh(&ifp->lock); if (!(ifp->flags&IFA_F_TENTATIVE)) ipv6_ifa_notify(0, ifp); @@ -4708,7 +4721,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, struct ifa6_config *cfg) if (action != CLEANUP_PREFIX_RT_NOP) { cleanup_prefix_route(ifp, rt_expires, - action == CLEANUP_PREFIX_RT_DEL); + action == CLEANUP_PREFIX_RT_DEL, false); } } -- cgit From 0d29169a708bf730ede287248e429d579f432d1d Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 3 Mar 2020 14:37:36 +0800 Subject: selftests/net/fib_tests: update addr_metric_test for peer route testing This patch update {ipv4, ipv6}_addr_metric_test with 1. Set metric of address with peer route and see if the route added correctly. 2. Modify metric and peer address for peer route and see if the route changed correctly. Signed-off-by: Hangbin Liu Reviewed-by: David Ahern Signed-off-by: David S. Miller --- tools/testing/selftests/net/fib_tests.sh | 34 +++++++++++++++++++++++++++++--- 1 file changed, 31 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/net/fib_tests.sh b/tools/testing/selftests/net/fib_tests.sh index 60273f1bc7d9..b7616704b55e 100755 --- a/tools/testing/selftests/net/fib_tests.sh +++ b/tools/testing/selftests/net/fib_tests.sh @@ -1041,6 +1041,27 @@ ipv6_addr_metric_test() fi log_test $rc 0 "Prefix route with metric on link up" + # verify peer metric added correctly + set -e + run_cmd "$IP -6 addr flush dev dummy2" + run_cmd "$IP -6 addr add dev dummy2 2001:db8:104::1 peer 2001:db8:104::2 metric 260" + set +e + + check_route6 "2001:db8:104::1 dev dummy2 proto kernel metric 260" + log_test $? 0 "Set metric with peer route on local side" + log_test $? 0 "User specified metric on local address" + check_route6 "2001:db8:104::2 dev dummy2 proto kernel metric 260" + log_test $? 0 "Set metric with peer route on peer side" + + set -e + run_cmd "$IP -6 addr change dev dummy2 2001:db8:104::1 peer 2001:db8:104::3 metric 261" + set +e + + check_route6 "2001:db8:104::1 dev dummy2 proto kernel metric 261" + log_test $? 0 "Modify metric and peer address on local side" + check_route6 "2001:db8:104::3 dev dummy2 proto kernel metric 261" + log_test $? 0 "Modify metric and peer address on peer side" + $IP li del dummy1 $IP li del dummy2 cleanup @@ -1457,13 +1478,20 @@ ipv4_addr_metric_test() run_cmd "$IP addr flush dev dummy2" run_cmd "$IP addr add dev dummy2 172.16.104.1/32 peer 172.16.104.2 metric 260" - run_cmd "$IP addr change dev dummy2 172.16.104.1/32 peer 172.16.104.2 metric 261" rc=$? if [ $rc -eq 0 ]; then - check_route "172.16.104.2 dev dummy2 proto kernel scope link src 172.16.104.1 metric 261" + check_route "172.16.104.2 dev dummy2 proto kernel scope link src 172.16.104.1 metric 260" + rc=$? + fi + log_test $rc 0 "Set metric of address with peer route" + + run_cmd "$IP addr change dev dummy2 172.16.104.1/32 peer 172.16.104.3 metric 261" + rc=$? + if [ $rc -eq 0 ]; then + check_route "172.16.104.3 dev dummy2 proto kernel scope link src 172.16.104.1 metric 261" rc=$? fi - log_test $rc 0 "Modify metric of address with peer route" + log_test $rc 0 "Modify metric and peer address for peer route" $IP li del dummy1 $IP li del dummy2 -- cgit From 8640f8dc6d657ebfb4e67c202ad32c5457858a13 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 3 Mar 2020 15:01:46 +0000 Subject: net: dsa: fix phylink_start()/phylink_stop() calls Place phylink_start()/phylink_stop() inside dsa_port_enable() and dsa_port_disable(), which ensures that we call phylink_stop() before tearing down phylink - which is a documented requirement. Failure to do so can cause use-after-free bugs. Fixes: 0e27921816ad ("net: dsa: Use PHYLINK for the CPU/DSA ports") Signed-off-by: Russell King Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- net/dsa/dsa_priv.h | 2 ++ net/dsa/port.c | 32 ++++++++++++++++++++++++++------ net/dsa/slave.c | 8 ++------ 3 files changed, 30 insertions(+), 12 deletions(-) diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index a7662e7a691d..760e6ea3178a 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -117,7 +117,9 @@ static inline struct net_device *dsa_master_find_slave(struct net_device *dev, /* port.c */ int dsa_port_set_state(struct dsa_port *dp, u8 state, struct switchdev_trans *trans); +int dsa_port_enable_rt(struct dsa_port *dp, struct phy_device *phy); int dsa_port_enable(struct dsa_port *dp, struct phy_device *phy); +void dsa_port_disable_rt(struct dsa_port *dp); void dsa_port_disable(struct dsa_port *dp); int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br); void dsa_port_bridge_leave(struct dsa_port *dp, struct net_device *br); diff --git a/net/dsa/port.c b/net/dsa/port.c index 774facb8d547..ed7dabb57985 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -63,7 +63,7 @@ static void dsa_port_set_state_now(struct dsa_port *dp, u8 state) pr_err("DSA: failed to set STP state %u (%d)\n", state, err); } -int dsa_port_enable(struct dsa_port *dp, struct phy_device *phy) +int dsa_port_enable_rt(struct dsa_port *dp, struct phy_device *phy) { struct dsa_switch *ds = dp->ds; int port = dp->index; @@ -78,14 +78,31 @@ int dsa_port_enable(struct dsa_port *dp, struct phy_device *phy) if (!dp->bridge_dev) dsa_port_set_state_now(dp, BR_STATE_FORWARDING); + if (dp->pl) + phylink_start(dp->pl); + return 0; } -void dsa_port_disable(struct dsa_port *dp) +int dsa_port_enable(struct dsa_port *dp, struct phy_device *phy) +{ + int err; + + rtnl_lock(); + err = dsa_port_enable_rt(dp, phy); + rtnl_unlock(); + + return err; +} + +void dsa_port_disable_rt(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; int port = dp->index; + if (dp->pl) + phylink_stop(dp->pl); + if (!dp->bridge_dev) dsa_port_set_state_now(dp, BR_STATE_DISABLED); @@ -93,6 +110,13 @@ void dsa_port_disable(struct dsa_port *dp) ds->ops->port_disable(ds, port); } +void dsa_port_disable(struct dsa_port *dp) +{ + rtnl_lock(); + dsa_port_disable_rt(dp); + rtnl_unlock(); +} + int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br) { struct dsa_notifier_bridge_info info = { @@ -614,10 +638,6 @@ static int dsa_port_phylink_register(struct dsa_port *dp) goto err_phy_connect; } - rtnl_lock(); - phylink_start(dp->pl); - rtnl_unlock(); - return 0; err_phy_connect: diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 088c886e609e..ddc0f9236928 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -88,12 +88,10 @@ static int dsa_slave_open(struct net_device *dev) goto clear_allmulti; } - err = dsa_port_enable(dp, dev->phydev); + err = dsa_port_enable_rt(dp, dev->phydev); if (err) goto clear_promisc; - phylink_start(dp->pl); - return 0; clear_promisc: @@ -114,9 +112,7 @@ static int dsa_slave_close(struct net_device *dev) struct net_device *master = dsa_slave_to_master(dev); struct dsa_port *dp = dsa_slave_to_port(dev); - phylink_stop(dp->pl); - - dsa_port_disable(dp); + dsa_port_disable_rt(dp); dev_mc_unsync(master, dev); dev_uc_unsync(master, dev); -- cgit From f8a0fea9518c5ff7c37679504bd9eeabeae8ee36 Mon Sep 17 00:00:00 2001 From: Jonathan Neuschäfer Date: Tue, 3 Mar 2020 21:22:05 +0100 Subject: docs: networking: net_failover: Fix a few typos MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Jonathan Neuschäfer Signed-off-by: David S. Miller --- Documentation/networking/net_failover.rst | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/networking/net_failover.rst b/Documentation/networking/net_failover.rst index 06c97dcb57ca..e143ab79a960 100644 --- a/Documentation/networking/net_failover.rst +++ b/Documentation/networking/net_failover.rst @@ -8,9 +8,9 @@ Overview ======== The net_failover driver provides an automated failover mechanism via APIs -to create and destroy a failover master netdev and mananges a primary and +to create and destroy a failover master netdev and manages a primary and standby slave netdevs that get registered via the generic failover -infrastructrure. +infrastructure. The failover netdev acts a master device and controls 2 slave devices. The original paravirtual interface is registered as 'standby' slave netdev and @@ -29,7 +29,7 @@ virtio-net accelerated datapath: STANDBY mode ============================================= net_failover enables hypervisor controlled accelerated datapath to virtio-net -enabled VMs in a transparent manner with no/minimal guest userspace chanages. +enabled VMs in a transparent manner with no/minimal guest userspace changes. To support this, the hypervisor needs to enable VIRTIO_NET_F_STANDBY feature on the virtio-net interface and assign the same MAC address to both -- cgit From dc15af8e9dbd039ebb06336597d2c491ef46ab74 Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Tue, 25 Feb 2020 10:05:47 +0300 Subject: netfilter: nf_conntrack: ct_cpu_seq_next should increase position index If .next function does not change position index, following .show function will repeat output related to current position index. Cc: stable@vger.kernel.org Fixes: 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration code ...") Link: https://bugzilla.kernel.org/show_bug.cgi?id=206283 Signed-off-by: Vasily Averin Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_standalone.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 410809c669e1..4912069627b6 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -411,7 +411,7 @@ static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) *pos = cpu + 1; return per_cpu_ptr(net->ct.stat, cpu); } - + (*pos)++; return NULL; } -- cgit From bb71f846a0002239f7058c84f1496648ff4a5c20 Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Tue, 25 Feb 2020 10:05:59 +0300 Subject: netfilter: synproxy: synproxy_cpu_seq_next should increase position index If .next function does not change position index, following .show function will repeat output related to current position index. Cc: stable@vger.kernel.org Fixes: 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration code ...") Link: https://bugzilla.kernel.org/show_bug.cgi?id=206283 Signed-off-by: Vasily Averin Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_synproxy_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index b0930d4aba22..b9cbe1e2453e 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -267,7 +267,7 @@ static void *synproxy_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) *pos = cpu + 1; return per_cpu_ptr(snet->stats, cpu); } - + (*pos)++; return NULL; } -- cgit From db25517a550926f609c63054b12ea9ad515e1a10 Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Tue, 25 Feb 2020 10:06:29 +0300 Subject: netfilter: xt_recent: recent_seq_next should increase position index If .next function does not change position index, following .show function will repeat output related to current position index. Without the patch: # dd if=/proc/net/xt_recent/SSH # original file outpt src=127.0.0.4 ttl: 0 last_seen: 6275444819 oldest_pkt: 1 6275444819 src=127.0.0.2 ttl: 0 last_seen: 6275438906 oldest_pkt: 1 6275438906 src=127.0.0.3 ttl: 0 last_seen: 6275441953 oldest_pkt: 1 6275441953 0+1 records in 0+1 records out 204 bytes copied, 6.1332e-05 s, 3.3 MB/s Read after lseek into middle of last line (offset 140 in example below) generates expected end of last line and then unexpected whole last line once again # dd if=/proc/net/xt_recent/SSH bs=140 skip=1 dd: /proc/net/xt_recent/SSH: cannot skip to specified offset 127.0.0.3 ttl: 0 last_seen: 6275441953 oldest_pkt: 1 6275441953 src=127.0.0.3 ttl: 0 last_seen: 6275441953 oldest_pkt: 1 6275441953 0+1 records in 0+1 records out 132 bytes copied, 6.2487e-05 s, 2.1 MB/s Cc: stable@vger.kernel.org Fixes: 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration code ...") Link: https://bugzilla.kernel.org/show_bug.cgi?id=206283 Signed-off-by: Vasily Averin Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_recent.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 0a9708004e20..225a7ab6d79a 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -492,12 +492,12 @@ static void *recent_seq_next(struct seq_file *seq, void *v, loff_t *pos) const struct recent_entry *e = v; const struct list_head *head = e->list.next; + (*pos)++; while (head == &t->iphash[st->bucket]) { if (++st->bucket >= ip_list_hash_size) return NULL; head = t->iphash[st->bucket].next; } - (*pos)++; return list_entry(head, struct recent_entry, list); } -- cgit From ee84f19cbbe9cf7cba2958acb03163fed3ecbb0f Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Tue, 25 Feb 2020 10:07:12 +0300 Subject: netfilter: x_tables: xt_mttg_seq_next should increase position index If .next function does not change position index, following .show function will repeat output related to current position index. Without patch: # dd if=/proc/net/ip_tables_matches # original file output conntrack conntrack conntrack recent recent icmp udplite udp tcp 0+1 records in 0+1 records out 65 bytes copied, 5.4074e-05 s, 1.2 MB/s # dd if=/proc/net/ip_tables_matches bs=62 skip=1 dd: /proc/net/ip_tables_matches: cannot skip to specified offset cp <<< end of last line tcp <<< and then unexpected whole last line once again 0+1 records in 0+1 records out 7 bytes copied, 0.000102447 s, 68.3 kB/s Cc: stable@vger.kernel.org Fixes: 1f4aace60b0e ("fs/seq_file.c: simplify seq_file iteration code ...") Link: https://bugzilla.kernel.org/show_bug.cgi?id=206283 Signed-off-by: Vasily Averin Signed-off-by: Pablo Neira Ayuso --- net/netfilter/x_tables.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index e27c6c5ba9df..cd2b034eef59 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1551,6 +1551,9 @@ static void *xt_mttg_seq_next(struct seq_file *seq, void *v, loff_t *ppos, uint8_t nfproto = (unsigned long)PDE_DATA(file_inode(seq->file)); struct nf_mttg_trav *trav = seq->private; + if (ppos != NULL) + ++(*ppos); + switch (trav->class) { case MTTG_TRAV_INIT: trav->class = MTTG_TRAV_NFP_UNSPEC; @@ -1576,9 +1579,6 @@ static void *xt_mttg_seq_next(struct seq_file *seq, void *v, loff_t *ppos, default: return NULL; } - - if (ppos != NULL) - ++*ppos; return trav; } -- cgit From 2d285f26ecd072800a29c5b71e63437f21ef830a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 2 Mar 2020 21:58:50 +0100 Subject: netfilter: nf_tables: free flowtable hooks on hook register error If hook registration fails, the hooks allocated via nft_netdev_hook_alloc need to be freed. We can't change the goto label to 'goto 5' -- while it does fix the memleak it does cause a warning splat from the netfilter core (the hooks were not registered). Fixes: 3f0465a9ef02 ("netfilter: nf_tables: dynamically allocate hooks per net_device in flowtables") Reported-by: syzbot+a2ff6fa45162a5ed4dd3@syzkaller.appspotmail.com Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index d1318bdf49ca..bb064aa4154b 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -6300,8 +6300,13 @@ static int nf_tables_newflowtable(struct net *net, struct sock *nlsk, goto err4; err = nft_register_flowtable_net_hooks(ctx.net, table, flowtable); - if (err < 0) + if (err < 0) { + list_for_each_entry_safe(hook, next, &flowtable->hook_list, list) { + list_del_rcu(&hook->list); + kfree_rcu(hook, rcu); + } goto err4; + } err = nft_trans_flowtable_add(&ctx, NFT_MSG_NEWFLOWTABLE, flowtable); if (err < 0) -- cgit From c049b3450072b8e3998053490e025839fecfef31 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:08:31 -0800 Subject: netfilter: cthelper: add missing attribute validation for cthelper Add missing attribute validation for cthelper to the netlink policy. Fixes: 12f7a505331e ("netfilter: add user-space connection tracking helper infrastructure") Signed-off-by: Jakub Kicinski Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_cthelper.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/netfilter/nfnetlink_cthelper.c b/net/netfilter/nfnetlink_cthelper.c index de3a9596b7f1..a5f294aa8e4c 100644 --- a/net/netfilter/nfnetlink_cthelper.c +++ b/net/netfilter/nfnetlink_cthelper.c @@ -742,6 +742,8 @@ static const struct nla_policy nfnl_cthelper_policy[NFCTH_MAX+1] = { [NFCTH_NAME] = { .type = NLA_NUL_STRING, .len = NF_CT_HELPER_NAME_LEN-1 }, [NFCTH_QUEUE_NUM] = { .type = NLA_U32, }, + [NFCTH_PRIV_DATA_LEN] = { .type = NLA_U32, }, + [NFCTH_STATUS] = { .type = NLA_U32, }, }; static const struct nfnl_callback nfnl_cthelper_cb[NFNL_MSG_CTHELPER_MAX] = { -- cgit From 9d6effb2f1523eb84516e44213c00f2fd9e6afff Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:08:32 -0800 Subject: netfilter: nft_payload: add missing attribute validation for payload csum flags Add missing attribute validation for NFTA_PAYLOAD_CSUM_FLAGS to the netlink policy. Fixes: 1814096980bb ("netfilter: nft_payload: layer 4 checksum adjustment for pseudoheader fields") Signed-off-by: Jakub Kicinski Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_payload.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index 1993af3a2979..a7de3a58f553 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -129,6 +129,7 @@ static const struct nla_policy nft_payload_policy[NFTA_PAYLOAD_MAX + 1] = { [NFTA_PAYLOAD_LEN] = { .type = NLA_U32 }, [NFTA_PAYLOAD_CSUM_TYPE] = { .type = NLA_U32 }, [NFTA_PAYLOAD_CSUM_OFFSET] = { .type = NLA_U32 }, + [NFTA_PAYLOAD_CSUM_FLAGS] = { .type = NLA_U32 }, }; static int nft_payload_init(const struct nft_ctx *ctx, -- cgit From 88a637719a1570705c02cacb3297af164b1714e7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:08:33 -0800 Subject: netfilter: nft_tunnel: add missing attribute validation for tunnels Add missing attribute validation for tunnel source and destination ports to the netlink policy. Fixes: af308b94a2a4 ("netfilter: nf_tables: add tunnel support") Signed-off-by: Jakub Kicinski Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_tunnel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c index 4c3f2e24c7cb..764e88682a81 100644 --- a/net/netfilter/nft_tunnel.c +++ b/net/netfilter/nft_tunnel.c @@ -339,6 +339,8 @@ static const struct nla_policy nft_tunnel_key_policy[NFTA_TUNNEL_KEY_MAX + 1] = [NFTA_TUNNEL_KEY_FLAGS] = { .type = NLA_U32, }, [NFTA_TUNNEL_KEY_TOS] = { .type = NLA_U8, }, [NFTA_TUNNEL_KEY_TTL] = { .type = NLA_U8, }, + [NFTA_TUNNEL_KEY_SPORT] = { .type = NLA_U16, }, + [NFTA_TUNNEL_KEY_DPORT] = { .type = NLA_U16, }, [NFTA_TUNNEL_KEY_OPTS] = { .type = NLA_NESTED, }, }; -- cgit From 707518348ae7337db34bdeb544edf254f66e20f2 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Mon, 2 Mar 2020 14:21:19 -0800 Subject: devlink: remove trigger command from devlink-region.rst The devlink trigger command does not exist. While rewriting the documentation for devlink into the reStructuredText format, documentation for the trigger command was accidentally merged in. This occurred because the author was also working on a potential extension to devlink regions which included this trigger command, and accidentally squashed the documentation incorrectly. Further review eventually settled on using the previously unused "new" command instead of creating a new trigger command. Fix this by removing mention of the trigger command from the documentation. Fixes: 0b0f945f5458 ("devlink: add a file documenting devlink regions", 2020-01-10) Noticed-by: Jiri Pirko Signed-off-by: Jacob Keller Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- Documentation/networking/devlink/devlink-region.rst | 3 --- 1 file changed, 3 deletions(-) diff --git a/Documentation/networking/devlink/devlink-region.rst b/Documentation/networking/devlink/devlink-region.rst index 1a7683e7acb2..8b46e8591fe0 100644 --- a/Documentation/networking/devlink/devlink-region.rst +++ b/Documentation/networking/devlink/devlink-region.rst @@ -40,9 +40,6 @@ example usage # Delete a snapshot using: $ devlink region del pci/0000:00:05.0/cr-space snapshot 1 - # Trigger (request) a snapshot be taken: - $ devlink region trigger pci/0000:00:05.0/cr-space - # Dump a snapshot: $ devlink region dump pci/0000:00:05.0/fw-health snapshot 1 0000000000000000 0014 95dc 0014 9514 0035 1670 0034 db30 -- cgit From 43de81b0601df7d7988d3f5617ee0987df65c883 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Mon, 2 Mar 2020 20:46:57 +0100 Subject: net: phy: bcm63xx: fix OOPS due to missing driver name 719655a14971 ("net: phy: Replace phy driver features u32 with link_mode bitmap") was a bit over-eager and also removed the second phy driver's name, resulting in a nasty OOPS on registration: [ 1.319854] CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == 804dd50c, ra == 804dd4f0 [ 1.330859] Oops[#1]: [ 1.333138] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.4.22 #0 [ 1.339217] $ 0 : 00000000 00000001 87ca7f00 805c1874 [ 1.344590] $ 4 : 00000000 00000047 00585000 8701f800 [ 1.349965] $ 8 : 8701f800 804f4a5c 00000003 64726976 [ 1.355341] $12 : 00000001 00000000 00000000 00000114 [ 1.360718] $16 : 87ca7f80 00000000 00000000 80639fe4 [ 1.366093] $20 : 00000002 00000000 806441d0 80b90000 [ 1.371470] $24 : 00000000 00000000 [ 1.376847] $28 : 87c1e000 87c1fda0 80b90000 804dd4f0 [ 1.382224] Hi : d1c8f8da [ 1.385180] Lo : 5518a480 [ 1.388182] epc : 804dd50c kset_find_obj+0x3c/0x114 [ 1.393345] ra : 804dd4f0 kset_find_obj+0x20/0x114 [ 1.398530] Status: 10008703 KERNEL EXL IE [ 1.402833] Cause : 00800008 (ExcCode 02) [ 1.406952] BadVA : 00000000 [ 1.409913] PrId : 0002a075 (Broadcom BMIPS4350) [ 1.414745] Modules linked in: [ 1.417895] Process swapper/0 (pid: 1, threadinfo=(ptrval), task=(ptrval), tls=00000000) [ 1.426214] Stack : 87cec000 80630000 80639370 80640658 80640000 80049af4 80639fe4 8063a0d8 [ 1.434816] 8063a0d8 802ef078 00000002 00000000 806441d0 80b90000 8063a0d8 802ef114 [ 1.443417] 87cea0de 87c1fde0 00000000 804de488 87cea000 8063a0d8 8063a0d8 80334e48 [ 1.452018] 80640000 8063984c 80639bf4 00000000 8065de48 00000001 8063a0d8 80334ed0 [ 1.460620] 806441d0 80b90000 80b90000 802ef164 8065dd70 80620000 80b90000 8065de58 [ 1.469222] ... [ 1.471734] Call Trace: [ 1.474255] [<804dd50c>] kset_find_obj+0x3c/0x114 [ 1.479141] [<802ef078>] driver_find+0x1c/0x44 [ 1.483665] [<802ef114>] driver_register+0x74/0x148 [ 1.488719] [<80334e48>] phy_driver_register+0x9c/0xd0 [ 1.493968] [<80334ed0>] phy_drivers_register+0x54/0xe8 [ 1.499345] [<8001061c>] do_one_initcall+0x7c/0x1f4 [ 1.504374] [<80644ed8>] kernel_init_freeable+0x1d4/0x2b4 [ 1.509940] [<804f4e24>] kernel_init+0x10/0xf8 [ 1.514502] [<80018e68>] ret_from_kernel_thread+0x14/0x1c [ 1.520040] Code: 1060000c 02202025 90650000 <90810000> 24630001 14250004 24840001 14a0fffb 90650000 [ 1.530061] [ 1.531698] ---[ end trace d52f1717cd29bdc8 ]--- Fix it by readding the name. Fixes: 719655a14971 ("net: phy: Replace phy driver features u32 with link_mode bitmap") Signed-off-by: Jonas Gorski Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm63xx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/bcm63xx.c b/drivers/net/phy/bcm63xx.c index 23f1958ba6ad..459fb2069c7e 100644 --- a/drivers/net/phy/bcm63xx.c +++ b/drivers/net/phy/bcm63xx.c @@ -73,6 +73,7 @@ static struct phy_driver bcm63xx_driver[] = { /* same phy as above, with just a different OUI */ .phy_id = 0x002bdc00, .phy_id_mask = 0xfffffc00, + .name = "Broadcom BCM63XX (2)", /* PHY_BASIC_FEATURES */ .flags = PHY_IS_INTERNAL, .config_init = bcm63xx_config_init, -- cgit From 20d8bb0d172d87dcc52727cb7174ae9994de8978 Mon Sep 17 00:00:00 2001 From: Hauke Mehrtens Date: Mon, 2 Mar 2020 00:55:02 +0100 Subject: phylink: Improve error message when validate failed This should improve the error message when the PHY validate in the MAC driver failed. I ran into this problem multiple times that I put wrong interface values into the device tree and was searching why it is failing with -22 (-EINVAL). This should make it easier to spot the problem. Signed-off-by: Hauke Mehrtens Acked-by: Russell King Signed-off-by: David S. Miller --- drivers/net/phy/phylink.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index 70b9a143db84..6e66b8e77ec7 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -761,8 +761,14 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy, config.interface = interface; ret = phylink_validate(pl, supported, &config); - if (ret) + if (ret) { + phylink_warn(pl, "validation of %s with support %*pb and advertisement %*pb failed: %d\n", + phy_modes(config.interface), + __ETHTOOL_LINK_MODE_MASK_NBITS, phy->supported, + __ETHTOOL_LINK_MODE_MASK_NBITS, config.advertising, + ret); return ret; + } phy->phylink = pl; phy->phy_link_change = phylink_phy_change; -- cgit From 116ca924aea664141afa86a1425edc3fcda0d06f Mon Sep 17 00:00:00 2001 From: Vishal Kulkarni Date: Mon, 2 Mar 2020 10:54:13 +0530 Subject: cxgb4: fix checks for max queues to allocate Hardware can support more than 8 queues currently limited by netif_get_num_default_rss_queues(). So, rework and fix checks for max number of queues to allocate. The checks should be based on how many are actually supported by hardware, OR the number of online cpus; whichever is lower. Fixes: 5952dde72307 ("cxgb4: set maximal number of default RSS queues") Signed-off-by: Vishal Kulkarni " Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 49 ++++++++++++++----------- 1 file changed, 27 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 649842a8aa28..97f90edbc068 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -5381,12 +5381,11 @@ static inline bool is_x_10g_port(const struct link_config *lc) static int cfg_queues(struct adapter *adap) { u32 avail_qsets, avail_eth_qsets, avail_uld_qsets; + u32 i, n10g = 0, qidx = 0, n1g = 0; + u32 ncpus = num_online_cpus(); u32 niqflint, neq, num_ulds; struct sge *s = &adap->sge; - u32 i, n10g = 0, qidx = 0; -#ifndef CONFIG_CHELSIO_T4_DCB - int q10g = 0; -#endif + u32 q10g = 0, q1g; /* Reduce memory usage in kdump environment, disable all offload. */ if (is_kdump_kernel() || (is_uld(adap) && t4_uld_mem_alloc(adap))) { @@ -5424,44 +5423,50 @@ static int cfg_queues(struct adapter *adap) n10g += is_x_10g_port(&adap2pinfo(adap, i)->link_cfg); avail_eth_qsets = min_t(u32, avail_qsets, MAX_ETH_QSETS); + + /* We default to 1 queue per non-10G port and up to # of cores queues + * per 10G port. + */ + if (n10g) + q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g; + + n1g = adap->params.nports - n10g; #ifdef CONFIG_CHELSIO_T4_DCB /* For Data Center Bridging support we need to be able to support up * to 8 Traffic Priorities; each of which will be assigned to its * own TX Queue in order to prevent Head-Of-Line Blocking. */ + q1g = 8; if (adap->params.nports * 8 > avail_eth_qsets) { dev_err(adap->pdev_dev, "DCB avail_eth_qsets=%d < %d!\n", avail_eth_qsets, adap->params.nports * 8); return -ENOMEM; } - for_each_port(adap, i) { - struct port_info *pi = adap2pinfo(adap, i); + if (adap->params.nports * ncpus < avail_eth_qsets) + q10g = max(8U, ncpus); + else + q10g = max(8U, q10g); - pi->first_qset = qidx; - pi->nqsets = is_kdump_kernel() ? 1 : 8; - qidx += pi->nqsets; - } -#else /* !CONFIG_CHELSIO_T4_DCB */ - /* We default to 1 queue per non-10G port and up to # of cores queues - * per 10G port. - */ - if (n10g) - q10g = (avail_eth_qsets - (adap->params.nports - n10g)) / n10g; - if (q10g > netif_get_num_default_rss_queues()) - q10g = netif_get_num_default_rss_queues(); + while ((q10g * n10g) > (avail_eth_qsets - n1g * q1g)) + q10g--; - if (is_kdump_kernel()) +#else /* !CONFIG_CHELSIO_T4_DCB */ + q1g = 1; + q10g = min(q10g, ncpus); +#endif /* !CONFIG_CHELSIO_T4_DCB */ + if (is_kdump_kernel()) { q10g = 1; + q1g = 1; + } for_each_port(adap, i) { struct port_info *pi = adap2pinfo(adap, i); pi->first_qset = qidx; - pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : 1; + pi->nqsets = is_x_10g_port(&pi->link_cfg) ? q10g : q1g; qidx += pi->nqsets; } -#endif /* !CONFIG_CHELSIO_T4_DCB */ s->ethqsets = qidx; s->max_ethqsets = qidx; /* MSI-X may lower it later */ @@ -5473,7 +5478,7 @@ static int cfg_queues(struct adapter *adap) * capped by the number of available cores. */ num_ulds = adap->num_uld + adap->num_ofld_uld; - i = min_t(u32, MAX_OFLD_QSETS, num_online_cpus()); + i = min_t(u32, MAX_OFLD_QSETS, ncpus); avail_uld_qsets = roundup(i, adap->params.nports); if (avail_qsets < num_ulds * adap->params.nports) { adap->params.offload = 0; -- cgit From 209c65b61d94344522c41a83cd6ce51aac5fd0a4 Mon Sep 17 00:00:00 2001 From: Dajun Jin Date: Mon, 2 Mar 2020 20:24:21 -0800 Subject: drivers/of/of_mdio.c:fix of_mdiobus_register() When registers a phy_device successful, should terminate the loop or the phy_device would be registered in other addr. If there are multiple PHYs without reg properties, it will go wrong. Signed-off-by: Dajun Jin Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/of/of_mdio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 8270bbf505fb..9f982c0627a0 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -306,6 +306,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) rc = of_mdiobus_register_phy(mdio, child, addr); if (rc && rc != -ENODEV) goto unregister; + break; } } } -- cgit From d78008de6103c708171baff9650a7862645d23b0 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 3 Mar 2020 15:02:45 +0100 Subject: netfilter: nf_tables: dump NFTA_CHAIN_FLAGS attribute Missing NFTA_CHAIN_FLAGS netlink attribute when dumping basechain definitions. Fixes: c9626a2cbdb2 ("netfilter: nf_tables: add hardware offload support") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index bb064aa4154b..f9e60981bd36 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1405,6 +1405,11 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, lockdep_commit_lock_is_held(net)); if (nft_dump_stats(skb, stats)) goto nla_put_failure; + + if ((chain->flags & NFT_CHAIN_HW_OFFLOAD) && + nla_put_be32(skb, NFTA_CHAIN_FLAGS, + htonl(NFT_CHAIN_HW_OFFLOAD))) + goto nla_put_failure; } if (nla_put_be32(skb, NFTA_CHAIN_USE, htonl(chain->use))) -- cgit From 1d305ba40eb8081ff21eeb8ca6ba5c70fd920934 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 5 Mar 2020 11:15:36 +0100 Subject: netfilter: nf_tables: fix infinite loop when expr is not available nft will loop forever if the kernel doesn't support an expression: 1. nft_expr_type_get() appends the family specific name to the module list. 2. -EAGAIN is returned to nfnetlink, nfnetlink calls abort path. 3. abort path sets ->done to true and calls request_module for the expression. 4. nfnetlink replays the batch, we end up in nft_expr_type_get() again. 5. nft_expr_type_get attempts to append family-specific name. This one already exists on the list, so we continue 6. nft_expr_type_get adds the generic expression name to the module list. -EAGAIN is returned, nfnetlink calls abort path. 7. abort path encounters the family-specific expression which has 'done' set, so it gets removed. 8. abort path requests the generic expression name, sets done to true. 9. batch is replayed. If the expression could not be loaded, then we will end up back at 1), because the family-specific name got removed and the cycle starts again. Note that userspace can SIGKILL the nft process to stop the cycle, but the desired behaviour is to return an error after the generic expr name fails to load the expression. Fixes: eb014de4fd418 ("netfilter: nf_tables: autoload modules from the abort path") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index f9e60981bd36..38c680f28f15 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7388,13 +7388,8 @@ static void nf_tables_module_autoload(struct net *net) list_splice_init(&net->nft.module_list, &module_list); mutex_unlock(&net->nft.commit_mutex); list_for_each_entry_safe(req, next, &module_list, list) { - if (req->done) { - list_del(&req->list); - kfree(req); - } else { - request_module("%s", req->module); - req->done = true; - } + request_module("%s", req->module); + req->done = true; } mutex_lock(&net->nft.commit_mutex); list_splice(&module_list, &net->nft.module_list); @@ -8177,6 +8172,7 @@ static void __net_exit nf_tables_exit_net(struct net *net) __nft_release_tables(net); mutex_unlock(&net->nft.commit_mutex); WARN_ON_ONCE(!list_empty(&net->nft.tables)); + WARN_ON_ONCE(!list_empty(&net->nft.module_list)); } static struct pernet_operations nf_tables_net_ops = { -- cgit From 110a40dfb708fe940a3f3704d470e431c368d256 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 4 Mar 2020 15:51:43 -0800 Subject: slip: make slhc_compress() more robust against malicious packets Before accessing various fields in IPV4 network header and TCP header, make sure the packet : - Has IP version 4 (ip->version == 4) - Has not a silly network length (ip->ihl >= 5) - Is big enough to hold network and transport headers - Has not a silly TCP header size (th->doff >= sizeof(struct tcphdr) / 4) syzbot reported : BUG: KMSAN: uninit-value in slhc_compress+0x5b9/0x2e60 drivers/net/slip/slhc.c:270 CPU: 0 PID: 11728 Comm: syz-executor231 Not tainted 5.6.0-rc2-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1c9/0x220 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:118 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:215 slhc_compress+0x5b9/0x2e60 drivers/net/slip/slhc.c:270 ppp_send_frame drivers/net/ppp/ppp_generic.c:1637 [inline] __ppp_xmit_process+0x1902/0x2970 drivers/net/ppp/ppp_generic.c:1495 ppp_xmit_process+0x147/0x2f0 drivers/net/ppp/ppp_generic.c:1516 ppp_write+0x6bb/0x790 drivers/net/ppp/ppp_generic.c:512 do_loop_readv_writev fs/read_write.c:717 [inline] do_iter_write+0x812/0xdc0 fs/read_write.c:1000 compat_writev+0x2df/0x5a0 fs/read_write.c:1351 do_compat_pwritev64 fs/read_write.c:1400 [inline] __do_compat_sys_pwritev fs/read_write.c:1420 [inline] __se_compat_sys_pwritev fs/read_write.c:1414 [inline] __ia32_compat_sys_pwritev+0x349/0x3f0 fs/read_write.c:1414 do_syscall_32_irqs_on arch/x86/entry/common.c:339 [inline] do_fast_syscall_32+0x3c7/0x6e0 arch/x86/entry/common.c:410 entry_SYSENTER_compat+0x68/0x77 arch/x86/entry/entry_64_compat.S:139 RIP: 0023:0xf7f7cd99 Code: 90 e8 0b 00 00 00 f3 90 0f ae e8 eb f9 8d 74 26 00 89 3c 24 c3 90 90 90 90 90 90 90 90 90 90 90 90 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 eb 0d 90 90 90 90 90 90 90 90 90 90 90 90 RSP: 002b:00000000ffdb84ac EFLAGS: 00000217 ORIG_RAX: 000000000000014e RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000200001c0 RDX: 0000000000000001 RSI: 0000000000000000 RDI: 0000000000000003 RBP: 0000000040047459 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Uninit was created at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:144 [inline] kmsan_internal_poison_shadow+0x66/0xd0 mm/kmsan/kmsan.c:127 kmsan_slab_alloc+0x8a/0xe0 mm/kmsan/kmsan_hooks.c:82 slab_alloc_node mm/slub.c:2793 [inline] __kmalloc_node_track_caller+0xb40/0x1200 mm/slub.c:4401 __kmalloc_reserve net/core/skbuff.c:142 [inline] __alloc_skb+0x2fd/0xac0 net/core/skbuff.c:210 alloc_skb include/linux/skbuff.h:1051 [inline] ppp_write+0x115/0x790 drivers/net/ppp/ppp_generic.c:500 do_loop_readv_writev fs/read_write.c:717 [inline] do_iter_write+0x812/0xdc0 fs/read_write.c:1000 compat_writev+0x2df/0x5a0 fs/read_write.c:1351 do_compat_pwritev64 fs/read_write.c:1400 [inline] __do_compat_sys_pwritev fs/read_write.c:1420 [inline] __se_compat_sys_pwritev fs/read_write.c:1414 [inline] __ia32_compat_sys_pwritev+0x349/0x3f0 fs/read_write.c:1414 do_syscall_32_irqs_on arch/x86/entry/common.c:339 [inline] do_fast_syscall_32+0x3c7/0x6e0 arch/x86/entry/common.c:410 entry_SYSENTER_compat+0x68/0x77 arch/x86/entry/entry_64_compat.S:139 Fixes: b5451d783ade ("slip: Move the SLIP drivers") Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: David S. Miller --- drivers/net/slip/slhc.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/net/slip/slhc.c b/drivers/net/slip/slhc.c index 58a69f830d29..f78ceba42e57 100644 --- a/drivers/net/slip/slhc.c +++ b/drivers/net/slip/slhc.c @@ -232,7 +232,7 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, struct cstate *cs = lcs->next; unsigned long deltaS, deltaA; short changes = 0; - int hlen; + int nlen, hlen; unsigned char new_seq[16]; unsigned char *cp = new_seq; struct iphdr *ip; @@ -248,6 +248,8 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, return isize; ip = (struct iphdr *) icp; + if (ip->version != 4 || ip->ihl < 5) + return isize; /* Bail if this packet isn't TCP, or is an IP fragment */ if (ip->protocol != IPPROTO_TCP || (ntohs(ip->frag_off) & 0x3fff)) { @@ -258,10 +260,14 @@ slhc_compress(struct slcompress *comp, unsigned char *icp, int isize, comp->sls_o_tcp++; return isize; } - /* Extract TCP header */ + nlen = ip->ihl * 4; + if (isize < nlen + sizeof(*th)) + return isize; - th = (struct tcphdr *)(((unsigned char *)ip) + ip->ihl*4); - hlen = ip->ihl*4 + th->doff*4; + th = (struct tcphdr *)(icp + nlen); + if (th->doff < sizeof(struct tcphdr) / 4) + return isize; + hlen = nlen + th->doff * 4; /* Bail if the TCP packet isn't `compressible' (i.e., ACK isn't set or * some other control bit is set). Also uncompressible if -- cgit From 68e1006f618e509fc7869259fe83ceec4a95dac3 Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 5 Mar 2020 09:47:53 +0800 Subject: net: hns3: fix a not link up issue when fibre port supports autoneg When fibre port supports auto-negotiation, the IMP(Intelligent Management Process) processes the speed of auto-negotiation and the user's speed separately. For below case, the port will get a not link up problem. step 1: disables auto-negotiation and sets speed to A, then the driver's MAC speed will be updated to A. step 2: enables auto-negotiation and MAC gets negotiated speed B, then the driver's MAC speed will be updated to B through querying in periodical task. step 3: MAC gets new negotiated speed A. step 4: disables auto-negotiation and sets speed to B before periodical task query new MAC speed A, the driver will ignore the speed configuration. This patch fixes it by skipping speed and duplex checking when fibre port supports auto-negotiation. Fixes: 22f48e24a23d ("net: hns3: add autoneg and change speed support for fibre port") Signed-off-by: Jian Shen Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 492bc9446463..acf0c29fcbcd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -2446,10 +2446,12 @@ static int hclge_cfg_mac_speed_dup_hw(struct hclge_dev *hdev, int speed, int hclge_cfg_mac_speed_dup(struct hclge_dev *hdev, int speed, u8 duplex) { + struct hclge_mac *mac = &hdev->hw.mac; int ret; duplex = hclge_check_speed_dup(duplex, speed); - if (hdev->hw.mac.speed == speed && hdev->hw.mac.duplex == duplex) + if (!mac->support_autoneg && mac->speed == speed && + mac->duplex == duplex) return 0; ret = hclge_cfg_mac_speed_dup_hw(hdev, speed, duplex); -- cgit From 3b4f06c715d0d3ecd6497275e3c85fe91462d0ee Mon Sep 17 00:00:00 2001 From: Tom Zhao Date: Thu, 5 Mar 2020 11:38:45 +0000 Subject: sfc: complete the next packet when we receive a timestamp We now ignore the "completion" event when using tx queue timestamping, and only pay attention to the two (high and low) timestamp events. The NIC will send a pair of timestamp events for every packet transmitted. The current firmware may merge the completion events, and it is possible that future versions may reorder the completion and timestamp events. As such the completion event is not useful. Without this patch in place a merged completion event on a queue with timestamping will cause a "spurious TX completion" error. This affects SFN8000-series adapters. Signed-off-by: Tom Zhao Acked-by: Martin Habets Signed-off-by: David S. Miller --- drivers/net/ethernet/sfc/ef10.c | 32 +++++++++++++++-------------- drivers/net/ethernet/sfc/efx.h | 1 + drivers/net/ethernet/sfc/net_driver.h | 3 --- drivers/net/ethernet/sfc/tx.c | 38 +++++++++++++++++++++++++++++++++++ drivers/net/ethernet/sfc/tx_common.c | 29 ++++++++++++++------------ drivers/net/ethernet/sfc/tx_common.h | 6 ++++++ 6 files changed, 78 insertions(+), 31 deletions(-) diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 52113b7529d6..3f16bd807c6e 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -2853,11 +2853,24 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) } /* Transmit timestamps are only available for 8XXX series. They result - * in three events per packet. These occur in order, and are: - * - the normal completion event + * in up to three events per packet. These occur in order, and are: + * - the normal completion event (may be omitted) * - the low part of the timestamp * - the high part of the timestamp * + * It's possible for multiple completion events to appear before the + * corresponding timestamps. So we can for example get: + * COMP N + * COMP N+1 + * TS_LO N + * TS_HI N + * TS_LO N+1 + * TS_HI N+1 + * + * In addition it's also possible for the adjacent completions to be + * merged, so we may not see COMP N above. As such, the completion + * events are not very useful here. + * * Each part of the timestamp is itself split across two 16 bit * fields in the event. */ @@ -2865,17 +2878,7 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) switch (tx_ev_type) { case TX_TIMESTAMP_EVENT_TX_EV_COMPLETION: - /* In case of Queue flush or FLR, we might have received - * the previous TX completion event but not the Timestamp - * events. - */ - if (tx_queue->completed_desc_ptr != tx_queue->ptr_mask) - efx_xmit_done(tx_queue, tx_queue->completed_desc_ptr); - - tx_ev_desc_ptr = EFX_QWORD_FIELD(*event, - ESF_DZ_TX_DESCR_INDX); - tx_queue->completed_desc_ptr = - tx_ev_desc_ptr & tx_queue->ptr_mask; + /* Ignore this event - see above. */ break; case TX_TIMESTAMP_EVENT_TX_EV_TSTAMP_LO: @@ -2887,8 +2890,7 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) ts_part = efx_ef10_extract_event_ts(event); tx_queue->completed_timestamp_major = ts_part; - efx_xmit_done(tx_queue, tx_queue->completed_desc_ptr); - tx_queue->completed_desc_ptr = tx_queue->ptr_mask; + efx_xmit_done_single(tx_queue); break; default: diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index f1bdb04efbe4..95395d67ea2d 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -20,6 +20,7 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, struct net_device *net_dev); netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb); void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index); +void efx_xmit_done_single(struct efx_tx_queue *tx_queue); int efx_setup_tc(struct net_device *net_dev, enum tc_setup_type type, void *type_data); extern unsigned int efx_piobuf_size; diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 9f9886f222c8..8164f0edcbf0 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -208,8 +208,6 @@ struct efx_tx_buffer { * avoid cache-line ping-pong between the xmit path and the * completion path. * @merge_events: Number of TX merged completion events - * @completed_desc_ptr: Most recent completed pointer - only used with - * timestamping. * @completed_timestamp_major: Top part of the most recent tx timestamp. * @completed_timestamp_minor: Low part of the most recent tx timestamp. * @insert_count: Current insert pointer @@ -269,7 +267,6 @@ struct efx_tx_queue { unsigned int merge_events; unsigned int bytes_compl; unsigned int pkts_compl; - unsigned int completed_desc_ptr; u32 completed_timestamp_major; u32 completed_timestamp_minor; diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c index 04d7f41d7ed9..8aafc54a4684 100644 --- a/drivers/net/ethernet/sfc/tx.c +++ b/drivers/net/ethernet/sfc/tx.c @@ -535,6 +535,44 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, return efx_enqueue_skb(tx_queue, skb); } +void efx_xmit_done_single(struct efx_tx_queue *tx_queue) +{ + unsigned int pkts_compl = 0, bytes_compl = 0; + unsigned int read_ptr; + bool finished = false; + + read_ptr = tx_queue->read_count & tx_queue->ptr_mask; + + while (!finished) { + struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr]; + + if (!efx_tx_buffer_in_use(buffer)) { + struct efx_nic *efx = tx_queue->efx; + + netif_err(efx, hw, efx->net_dev, + "TX queue %d spurious single TX completion\n", + tx_queue->queue); + efx_schedule_reset(efx, RESET_TYPE_TX_SKIP); + return; + } + + /* Need to check the flag before dequeueing. */ + if (buffer->flags & EFX_TX_BUF_SKB) + finished = true; + efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl); + + ++tx_queue->read_count; + read_ptr = tx_queue->read_count & tx_queue->ptr_mask; + } + + tx_queue->pkts_compl += pkts_compl; + tx_queue->bytes_compl += bytes_compl; + + EFX_WARN_ON_PARANOID(pkts_compl != 1); + + efx_xmit_done_check_empty(tx_queue); +} + void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue) { struct efx_nic *efx = tx_queue->efx; diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c index b1571e9789d0..70876df1da69 100644 --- a/drivers/net/ethernet/sfc/tx_common.c +++ b/drivers/net/ethernet/sfc/tx_common.c @@ -80,7 +80,6 @@ void efx_init_tx_queue(struct efx_tx_queue *tx_queue) tx_queue->xmit_more_available = false; tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps(efx) && tx_queue->channel == efx_ptp_channel(efx)); - tx_queue->completed_desc_ptr = tx_queue->ptr_mask; tx_queue->completed_timestamp_major = 0; tx_queue->completed_timestamp_minor = 0; @@ -210,10 +209,9 @@ static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue, while (read_ptr != stop_index) { struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr]; - if (!(buffer->flags & EFX_TX_BUF_OPTION) && - unlikely(buffer->len == 0)) { + if (!efx_tx_buffer_in_use(buffer)) { netif_err(efx, tx_err, efx->net_dev, - "TX queue %d spurious TX completion id %x\n", + "TX queue %d spurious TX completion id %d\n", tx_queue->queue, read_ptr); efx_schedule_reset(efx, RESET_TYPE_TX_SKIP); return; @@ -226,6 +224,19 @@ static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue, } } +void efx_xmit_done_check_empty(struct efx_tx_queue *tx_queue) +{ + if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) { + tx_queue->old_write_count = READ_ONCE(tx_queue->write_count); + if (tx_queue->read_count == tx_queue->old_write_count) { + /* Ensure that read_count is flushed. */ + smp_mb(); + tx_queue->empty_read_count = + tx_queue->read_count | EFX_EMPTY_COUNT_VALID; + } + } +} + void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) { unsigned int fill_level, pkts_compl = 0, bytes_compl = 0; @@ -256,15 +267,7 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) netif_tx_wake_queue(tx_queue->core_txq); } - /* Check whether the hardware queue is now empty */ - if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) { - tx_queue->old_write_count = READ_ONCE(tx_queue->write_count); - if (tx_queue->read_count == tx_queue->old_write_count) { - smp_mb(); - tx_queue->empty_read_count = - tx_queue->read_count | EFX_EMPTY_COUNT_VALID; - } - } + efx_xmit_done_check_empty(tx_queue); } /* Remove buffers put into a tx_queue for the current packet. diff --git a/drivers/net/ethernet/sfc/tx_common.h b/drivers/net/ethernet/sfc/tx_common.h index f92f1fe3a87f..99cf7ce2f36c 100644 --- a/drivers/net/ethernet/sfc/tx_common.h +++ b/drivers/net/ethernet/sfc/tx_common.h @@ -21,6 +21,12 @@ void efx_dequeue_buffer(struct efx_tx_queue *tx_queue, unsigned int *pkts_compl, unsigned int *bytes_compl); +static inline bool efx_tx_buffer_in_use(struct efx_tx_buffer *buffer) +{ + return buffer->len || (buffer->flags & EFX_TX_BUF_OPTION); +} + +void efx_xmit_done_check_empty(struct efx_tx_queue *tx_queue); void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index); void efx_enqueue_unwind(struct efx_tx_queue *tx_queue, -- cgit From e25d5dbcffae62c9a7fa03517dfa4b8e67670e3d Mon Sep 17 00:00:00 2001 From: Jiang Lidong Date: Wed, 4 Mar 2020 09:49:29 +0800 Subject: veth: ignore peer tx_dropped when counting local rx_dropped When local NET_RX backlog is full due to traffic overrun, peer veth tx_dropped counter increases. At that time, list local veth stats, rx_dropped has double value of peer tx_dropped, even bigger than transmit packets by peer. In NET_RX softirq process, if any packet drop case happens, it increases dev's rx_dropped counter and returns NET_RX_DROP. At veth tx side, it records any error returned from peer netif_rx into local dev tx_dropped counter. In veth get stats process, it puts local dev rx_dropped and peer dev tx_dropped into together as local rx_drpped value. So that it shows double value of real dropped packets number in this case. This patch ignores peer tx_dropped when counting local rx_dropped, since peer tx_dropped is duplicated to local rx_dropped at most cases. Signed-off-by: Jiang Lidong Signed-off-by: David S. Miller --- drivers/net/veth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 8cdc4415fa70..d4cbb9e8c63f 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -328,7 +328,7 @@ static void veth_get_stats64(struct net_device *dev, rcu_read_lock(); peer = rcu_dereference(priv->peer); if (peer) { - tot->rx_dropped += veth_stats_tx(peer, &packets, &bytes); + veth_stats_tx(peer, &packets, &bytes); tot->rx_bytes += bytes; tot->rx_packets += packets; -- cgit From a3aefbfe45751bf7b338c181b97608e276b5bb73 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 4 Mar 2020 17:24:31 +0300 Subject: net: nfc: fix bounds checking bugs on "pipe" This is similar to commit 674d9de02aa7 ("NFC: Fix possible memory corruption when handling SHDLC I-Frame commands") and commit d7ee81ad09f0 ("NFC: nci: Add some bounds checking in nci_hci_cmd_received()") which added range checks on "pipe". The "pipe" variable comes skb->data[0] in nfc_hci_msg_rx_work(). It's in the 0-255 range. We're using it as the array index into the hdev->pipes[] array which has NFC_HCI_MAX_PIPES (128) members. Fixes: 118278f20aa8 ("NFC: hci: Add pipes table to reference them with a tuple {gate, host}") Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller --- net/nfc/hci/core.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/net/nfc/hci/core.c b/net/nfc/hci/core.c index 6f1b096e601c..43811b5219b5 100644 --- a/net/nfc/hci/core.c +++ b/net/nfc/hci/core.c @@ -181,13 +181,20 @@ exit: void nfc_hci_cmd_received(struct nfc_hci_dev *hdev, u8 pipe, u8 cmd, struct sk_buff *skb) { - u8 gate = hdev->pipes[pipe].gate; u8 status = NFC_HCI_ANY_OK; struct hci_create_pipe_resp *create_info; struct hci_delete_pipe_noti *delete_info; struct hci_all_pipe_cleared_noti *cleared_info; + u8 gate; - pr_debug("from gate %x pipe %x cmd %x\n", gate, pipe, cmd); + pr_debug("from pipe %x cmd %x\n", pipe, cmd); + + if (pipe >= NFC_HCI_MAX_PIPES) { + status = NFC_HCI_ANY_E_NOK; + goto exit; + } + + gate = hdev->pipes[pipe].gate; switch (cmd) { case NFC_HCI_ADM_NOTIFY_PIPE_CREATED: @@ -375,8 +382,14 @@ void nfc_hci_event_received(struct nfc_hci_dev *hdev, u8 pipe, u8 event, struct sk_buff *skb) { int r = 0; - u8 gate = hdev->pipes[pipe].gate; + u8 gate; + + if (pipe >= NFC_HCI_MAX_PIPES) { + pr_err("Discarded event %x to invalid pipe %x\n", event, pipe); + goto exit; + } + gate = hdev->pipes[pipe].gate; if (gate == NFC_HCI_INVALID_GATE) { pr_err("Discarded event %x to unopened pipe %x\n", event, pipe); goto exit; -- cgit From 2398e3991bda7caa6b112a6f650fbab92f732b91 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 4 Mar 2020 16:51:07 +0100 Subject: mptcp: always include dack if possible. Currently passive MPTCP socket can skip including the DACK option - if the peer sends data before accept() completes. The above happens because the msk 'can_ack' flag is set only after the accept() call. Such missing DACK option may cause - as per RFC spec - unwanted fallback to TCP. This change addresses the issue using the key material available in the current subflow, if any, to create a suitable dack option when msk ack seq is not yet available. v1 -> v2: - adavance the generated ack after the initial MPC packet Fixes: d22f4988ffec ("mptcp: process MP_CAPABLE data option") Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: David S. Miller --- net/mptcp/options.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 45acd877bef3..fd2c3150e591 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -334,6 +334,8 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, struct mptcp_sock *msk; unsigned int ack_size; bool ret = false; + bool can_ack; + u64 ack_seq; u8 tcp_fin; if (skb) { @@ -360,9 +362,22 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, ret = true; } + /* passive sockets msk will set the 'can_ack' after accept(), even + * if the first subflow may have the already the remote key handy + */ + can_ack = true; opts->ext_copy.use_ack = 0; msk = mptcp_sk(subflow->conn); - if (!msk || !READ_ONCE(msk->can_ack)) { + if (likely(msk && READ_ONCE(msk->can_ack))) { + ack_seq = msk->ack_seq; + } else if (subflow->can_ack) { + mptcp_crypto_key_sha(subflow->remote_key, NULL, &ack_seq); + ack_seq++; + } else { + can_ack = false; + } + + if (unlikely(!can_ack)) { *size = ALIGN(dss_size, 4); return ret; } @@ -375,7 +390,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, dss_size += ack_size; - opts->ext_copy.data_ack = msk->ack_seq; + opts->ext_copy.data_ack = ack_seq; opts->ext_copy.ack64 = 1; opts->ext_copy.use_ack = 1; -- cgit From 6a42cefb25d8bdc1b391f4a53c78c32164eea2dd Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 6 Mar 2020 17:37:28 +0100 Subject: netfilter: nft_chain_nat: inet family is missing module ownership Set owner to THIS_MODULE, otherwise the nft_chain_nat module might be removed while there are still inet/nat chains in place. [ 117.942096] BUG: unable to handle page fault for address: ffffffffa0d5e040 [ 117.942101] #PF: supervisor read access in kernel mode [ 117.942103] #PF: error_code(0x0000) - not-present page [ 117.942106] PGD 200c067 P4D 200c067 PUD 200d063 PMD 3dc909067 PTE 0 [ 117.942113] Oops: 0000 [#1] PREEMPT SMP PTI [ 117.942118] CPU: 3 PID: 27 Comm: kworker/3:0 Not tainted 5.6.0-rc3+ #348 [ 117.942133] Workqueue: events nf_tables_trans_destroy_work [nf_tables] [ 117.942145] RIP: 0010:nf_tables_chain_destroy.isra.0+0x94/0x15a [nf_tables] [ 117.942149] Code: f6 45 54 01 0f 84 d1 00 00 00 80 3b 05 74 44 48 8b 75 e8 48 c7 c7 72 be de a0 e8 56 e6 2d e0 48 8b 45 e8 48 c7 c7 7f be de a0 <48> 8b 30 e8 43 e6 2d e0 48 8b 45 e8 48 8b 40 10 48 85 c0 74 5b 8b [ 117.942152] RSP: 0018:ffffc9000015be10 EFLAGS: 00010292 [ 117.942155] RAX: ffffffffa0d5e040 RBX: ffff88840be87fc2 RCX: 0000000000000007 [ 117.942158] RDX: 0000000000000007 RSI: 0000000000000086 RDI: ffffffffa0debe7f [ 117.942160] RBP: ffff888403b54b50 R08: 0000000000001482 R09: 0000000000000004 [ 117.942162] R10: 0000000000000000 R11: 0000000000000001 R12: ffff8883eda7e540 [ 117.942164] R13: dead000000000122 R14: dead000000000100 R15: ffff888403b3db80 [ 117.942167] FS: 0000000000000000(0000) GS:ffff88840e4c0000(0000) knlGS:0000000000000000 [ 117.942169] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 117.942172] CR2: ffffffffa0d5e040 CR3: 00000003e4c52002 CR4: 00000000001606e0 [ 117.942174] Call Trace: [ 117.942188] nf_tables_trans_destroy_work.cold+0xd/0x12 [nf_tables] [ 117.942196] process_one_work+0x1d6/0x3b0 [ 117.942200] worker_thread+0x45/0x3c0 [ 117.942203] ? process_one_work+0x3b0/0x3b0 [ 117.942210] kthread+0x112/0x130 [ 117.942214] ? kthread_create_worker_on_cpu+0x40/0x40 [ 117.942221] ret_from_fork+0x35/0x40 nf_tables_chain_destroy() crashes on module_put() because the module is gone. Fixes: d164385ec572 ("netfilter: nat: add inet family nat support") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_chain_nat.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nft_chain_nat.c b/net/netfilter/nft_chain_nat.c index ff9ac8ae0031..eac4a901233f 100644 --- a/net/netfilter/nft_chain_nat.c +++ b/net/netfilter/nft_chain_nat.c @@ -89,6 +89,7 @@ static const struct nft_chain_type nft_chain_nat_inet = { .name = "nat", .type = NFT_CHAIN_T_NAT, .family = NFPROTO_INET, + .owner = THIS_MODULE, .hook_mask = (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN) | (1 << NF_INET_LOCAL_OUT) | -- cgit From 26d5bb9e4c4b541c475751e015072eb2cbf70d15 Mon Sep 17 00:00:00 2001 From: Madalin Bucur Date: Wed, 4 Mar 2020 18:04:25 +0200 Subject: dt-bindings: net: FMan erratum A050385 FMAN DMA read or writes under heavy traffic load may cause FMAN internal resource leak; thus stopping further packet processing. The FMAN internal queue can overflow when FMAN splits single read or write transactions into multiple smaller transactions such that more than 17 AXI transactions are in flight from FMAN to interconnect. When the FMAN internal queue overflows, it can stall further packet processing. The issue can occur with any one of the following three conditions: 1. FMAN AXI transaction crosses 4K address boundary (Errata A010022) 2. FMAN DMA address for an AXI transaction is not 16 byte aligned, i.e. the last 4 bits of an address are non-zero 3. Scatter Gather (SG) frames have more than one SG buffer in the SG list and any one of the buffers, except the last buffer in the SG list has data size that is not a multiple of 16 bytes, i.e., other than 16, 32, 48, 64, etc. With any one of the above three conditions present, there is likelihood of stalled FMAN packet processing, especially under stress with multiple ports injecting line-rate traffic. To avoid situations that stall FMAN packet processing, all of the above three conditions must be avoided; therefore, configure the system with the following rules: 1. Frame buffers must not span a 4KB address boundary, unless the frame start address is 256 byte aligned 2. All FMAN DMA start addresses (for example, BMAN buffer address, FD[address] + FD[offset]) are 16B aligned 3. SG table and buffer addresses are 16B aligned and the size of SG buffers are multiple of 16 bytes, except for the last SG buffer that can be of any size. Additional workaround notes: - Address alignment of 64 bytes is recommended for maximally efficient system bus transactions (although 16 byte alignment is sufficient to avoid the stall condition) - To support frame sizes that are larger than 4K bytes, there are two options: 1. Large single buffer frames that span a 4KB page boundary can be converted into SG frames to avoid transaction splits at the 4KB boundary, 2. Align the large single buffer to 256B address boundaries, ensure that the frame address plus offset is 256B aligned. - If software generated SG frames have buffers that are unaligned and with random non-multiple of 16 byte lengths, before transmitting such frames via FMAN, frames will need to be copied into a new single buffer or multiple buffer SG frame that is compliant with the three rules listed above. Signed-off-by: Madalin Bucur Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/fsl-fman.txt | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/devicetree/bindings/net/fsl-fman.txt b/Documentation/devicetree/bindings/net/fsl-fman.txt index 250f8d8cdce4..c00fb0d22c7b 100644 --- a/Documentation/devicetree/bindings/net/fsl-fman.txt +++ b/Documentation/devicetree/bindings/net/fsl-fman.txt @@ -110,6 +110,13 @@ PROPERTIES Usage: required Definition: See soc/fsl/qman.txt and soc/fsl/bman.txt +- fsl,erratum-a050385 + Usage: optional + Value type: boolean + Definition: A boolean property. Indicates the presence of the + erratum A050385 which indicates that DMA transactions that are + split can result in a FMan lock. + ============================================================================= FMan MURAM Node -- cgit From b54d3900862374e1bb2846e6b39d79c896c0b200 Mon Sep 17 00:00:00 2001 From: Madalin Bucur Date: Wed, 4 Mar 2020 18:04:26 +0200 Subject: arm64: dts: ls1043a: FMan erratum A050385 The LS1043A SoC is affected by the A050385 erratum stating that FMAN DMA read or writes under heavy traffic load may cause FMAN internal resource leak thus stopping further packet processing. Signed-off-by: Madalin Bucur Signed-off-by: David S. Miller --- arch/arm64/boot/dts/freescale/fsl-ls1043-post.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1043-post.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls1043-post.dtsi index 6082ae022136..d237162a8744 100644 --- a/arch/arm64/boot/dts/freescale/fsl-ls1043-post.dtsi +++ b/arch/arm64/boot/dts/freescale/fsl-ls1043-post.dtsi @@ -20,6 +20,8 @@ }; &fman0 { + fsl,erratum-a050385; + /* these aliases provide the FMan ports mapping */ enet0: ethernet@e0000 { }; -- cgit From b281f7b93b258ce1419043bbd898a29254d5c9c7 Mon Sep 17 00:00:00 2001 From: Madalin Bucur Date: Wed, 4 Mar 2020 18:04:27 +0200 Subject: fsl/fman: detect FMan erratum A050385 Detect the presence of the A050385 erratum. Signed-off-by: Madalin Bucur Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fman/Kconfig | 28 ++++++++++++++++++++++++++++ drivers/net/ethernet/freescale/fman/fman.c | 18 ++++++++++++++++++ drivers/net/ethernet/freescale/fman/fman.h | 5 +++++ 3 files changed, 51 insertions(+) diff --git a/drivers/net/ethernet/freescale/fman/Kconfig b/drivers/net/ethernet/freescale/fman/Kconfig index 0139cb9042ec..34150182cc35 100644 --- a/drivers/net/ethernet/freescale/fman/Kconfig +++ b/drivers/net/ethernet/freescale/fman/Kconfig @@ -8,3 +8,31 @@ config FSL_FMAN help Freescale Data-Path Acceleration Architecture Frame Manager (FMan) support + +config DPAA_ERRATUM_A050385 + bool + depends on ARM64 && FSL_DPAA + default y + help + DPAA FMan erratum A050385 software workaround implementation: + align buffers, data start, SG fragment length to avoid FMan DMA + splits. + FMAN DMA read or writes under heavy traffic load may cause FMAN + internal resource leak thus stopping further packet processing. + The FMAN internal queue can overflow when FMAN splits single + read or write transactions into multiple smaller transactions + such that more than 17 AXI transactions are in flight from FMAN + to interconnect. When the FMAN internal queue overflows, it can + stall further packet processing. The issue can occur with any + one of the following three conditions: + 1. FMAN AXI transaction crosses 4K address boundary (Errata + A010022) + 2. FMAN DMA address for an AXI transaction is not 16 byte + aligned, i.e. the last 4 bits of an address are non-zero + 3. Scatter Gather (SG) frames have more than one SG buffer in + the SG list and any one of the buffers, except the last + buffer in the SG list has data size that is not a multiple + of 16 bytes, i.e., other than 16, 32, 48, 64, etc. + With any one of the above three conditions present, there is + likelihood of stalled FMAN packet processing, especially under + stress with multiple ports injecting line-rate traffic. diff --git a/drivers/net/ethernet/freescale/fman/fman.c b/drivers/net/ethernet/freescale/fman/fman.c index 934111def0be..f151d6e111dd 100644 --- a/drivers/net/ethernet/freescale/fman/fman.c +++ b/drivers/net/ethernet/freescale/fman/fman.c @@ -1,5 +1,6 @@ /* * Copyright 2008-2015 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -566,6 +567,10 @@ struct fman_cfg { u32 qmi_def_tnums_thresh; }; +#ifdef CONFIG_DPAA_ERRATUM_A050385 +static bool fman_has_err_a050385; +#endif + static irqreturn_t fman_exceptions(struct fman *fman, enum fman_exceptions exception) { @@ -2518,6 +2523,14 @@ struct fman *fman_bind(struct device *fm_dev) } EXPORT_SYMBOL(fman_bind); +#ifdef CONFIG_DPAA_ERRATUM_A050385 +bool fman_has_errata_a050385(void) +{ + return fman_has_err_a050385; +} +EXPORT_SYMBOL(fman_has_errata_a050385); +#endif + static irqreturn_t fman_err_irq(int irq, void *handle) { struct fman *fman = (struct fman *)handle; @@ -2845,6 +2858,11 @@ static struct fman *read_dts_node(struct platform_device *of_dev) goto fman_free; } +#ifdef CONFIG_DPAA_ERRATUM_A050385 + fman_has_err_a050385 = + of_property_read_bool(fm_node, "fsl,erratum-a050385"); +#endif + return fman; fman_node_put: diff --git a/drivers/net/ethernet/freescale/fman/fman.h b/drivers/net/ethernet/freescale/fman/fman.h index 935c317fa696..f2ede1360f03 100644 --- a/drivers/net/ethernet/freescale/fman/fman.h +++ b/drivers/net/ethernet/freescale/fman/fman.h @@ -1,5 +1,6 @@ /* * Copyright 2008-2015 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -398,6 +399,10 @@ u16 fman_get_max_frm(void); int fman_get_rx_extra_headroom(void); +#ifdef CONFIG_DPAA_ERRATUM_A050385 +bool fman_has_errata_a050385(void); +#endif + struct fman *fman_bind(struct device *dev); #endif /* __FM_H */ -- cgit From 3c68b8fffb48c0018c24e73c48f2bac768c6203e Mon Sep 17 00:00:00 2001 From: Madalin Bucur Date: Wed, 4 Mar 2020 18:04:28 +0200 Subject: dpaa_eth: FMan erratum A050385 workaround Align buffers, data start, SG fragment length to avoid DMA splits. These changes prevent the A050385 erratum to manifest itself: FMAN DMA read or writes under heavy traffic load may cause FMAN internal resource leak; thus stopping further packet processing. The FMAN internal queue can overflow when FMAN splits single read or write transactions into multiple smaller transactions such that more than 17 AXI transactions are in flight from FMAN to interconnect. When the FMAN internal queue overflows, it can stall further packet processing. The issue can occur with any one of the following three conditions: 1. FMAN AXI transaction crosses 4K address boundary (Errata A010022) 2. FMAN DMA address for an AXI transaction is not 16 byte aligned, i.e. the last 4 bits of an address are non-zero 3. Scatter Gather (SG) frames have more than one SG buffer in the SG list and any one of the buffers, except the last buffer in the SG list has data size that is not a multiple of 16 bytes, i.e., other than 16, 32, 48, 64, etc. With any one of the above three conditions present, there is likelihood of stalled FMAN packet processing, especially under stress with multiple ports injecting line-rate traffic. To avoid situations that stall FMAN packet processing, all of the above three conditions must be avoided; therefore, configure the system with the following rules: 1. Frame buffers must not span a 4KB address boundary, unless the frame start address is 256 byte aligned 2. All FMAN DMA start addresses (for example, BMAN buffer address, FD[address] + FD[offset]) are 16B aligned 3. SG table and buffer addresses are 16B aligned and the size of SG buffers are multiple of 16 bytes, except for the last SG buffer that can be of any size. Additional workaround notes: - Address alignment of 64 bytes is recommended for maximally efficient system bus transactions (although 16 byte alignment is sufficient to avoid the stall condition) - To support frame sizes that are larger than 4K bytes, there are two options: 1. Large single buffer frames that span a 4KB page boundary can be converted into SG frames to avoid transaction splits at the 4KB boundary, 2. Align the large single buffer to 256B address boundaries, ensure that the frame address plus offset is 256B aligned. - If software generated SG frames have buffers that are unaligned and with random non-multiple of 16 byte lengths, before transmitting such frames via FMAN, frames will need to be copied into a new single buffer or multiple buffer SG frame that is compliant with the three rules listed above. Signed-off-by: Madalin Bucur Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 110 ++++++++++++++++++++++++- 1 file changed, 107 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index fd93d542f497..e3ac9ec54c7c 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -1,4 +1,5 @@ /* Copyright 2008 - 2016 Freescale Semiconductor Inc. + * Copyright 2020 NXP * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -123,7 +124,22 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define FSL_QMAN_MAX_OAL 127 /* Default alignment for start of data in an Rx FD */ +#ifdef CONFIG_DPAA_ERRATUM_A050385 +/* aligning data start to 64 avoids DMA transaction splits, unless the buffer + * is crossing a 4k page boundary + */ +#define DPAA_FD_DATA_ALIGNMENT (fman_has_errata_a050385() ? 64 : 16) +/* aligning to 256 avoids DMA transaction splits caused by 4k page boundary + * crossings; also, all SG fragments except the last must have a size multiple + * of 256 to avoid DMA transaction splits + */ +#define DPAA_A050385_ALIGN 256 +#define DPAA_FD_RX_DATA_ALIGNMENT (fman_has_errata_a050385() ? \ + DPAA_A050385_ALIGN : 16) +#else #define DPAA_FD_DATA_ALIGNMENT 16 +#define DPAA_FD_RX_DATA_ALIGNMENT DPAA_FD_DATA_ALIGNMENT +#endif /* The DPAA requires 256 bytes reserved and mapped for the SGT */ #define DPAA_SGT_SIZE 256 @@ -158,8 +174,13 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define DPAA_PARSE_RESULTS_SIZE sizeof(struct fman_prs_result) #define DPAA_TIME_STAMP_SIZE 8 #define DPAA_HASH_RESULTS_SIZE 8 +#ifdef CONFIG_DPAA_ERRATUM_A050385 +#define DPAA_RX_PRIV_DATA_SIZE (DPAA_A050385_ALIGN - (DPAA_PARSE_RESULTS_SIZE\ + + DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE)) +#else #define DPAA_RX_PRIV_DATA_SIZE (u16)(DPAA_TX_PRIV_DATA_SIZE + \ dpaa_rx_extra_headroom) +#endif #define DPAA_ETH_PCD_RXQ_NUM 128 @@ -180,7 +201,12 @@ static struct dpaa_bp *dpaa_bp_array[BM_MAX_NUM_OF_POOLS]; #define DPAA_BP_RAW_SIZE 4096 +#ifdef CONFIG_DPAA_ERRATUM_A050385 +#define dpaa_bp_size(raw_size) (SKB_WITH_OVERHEAD(raw_size) & \ + ~(DPAA_A050385_ALIGN - 1)) +#else #define dpaa_bp_size(raw_size) SKB_WITH_OVERHEAD(raw_size) +#endif static int dpaa_max_frm; @@ -1192,7 +1218,7 @@ static int dpaa_eth_init_rx_port(struct fman_port *port, struct dpaa_bp *bp, buf_prefix_content.pass_prs_result = true; buf_prefix_content.pass_hash_result = true; buf_prefix_content.pass_time_stamp = true; - buf_prefix_content.data_align = DPAA_FD_DATA_ALIGNMENT; + buf_prefix_content.data_align = DPAA_FD_RX_DATA_ALIGNMENT; rx_p = ¶ms.specific_params.rx_params; rx_p->err_fqid = errq->fqid; @@ -1662,6 +1688,8 @@ static u8 rx_csum_offload(const struct dpaa_priv *priv, const struct qm_fd *fd) return CHECKSUM_NONE; } +#define PTR_IS_ALIGNED(x, a) (IS_ALIGNED((unsigned long)(x), (a))) + /* Build a linear skb around the received buffer. * We are guaranteed there is enough room at the end of the data buffer to * accommodate the shared info area of the skb. @@ -1733,8 +1761,7 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv, sg_addr = qm_sg_addr(&sgt[i]); sg_vaddr = phys_to_virt(sg_addr); - WARN_ON(!IS_ALIGNED((unsigned long)sg_vaddr, - SMP_CACHE_BYTES)); + WARN_ON(!PTR_IS_ALIGNED(sg_vaddr, SMP_CACHE_BYTES)); dma_unmap_page(priv->rx_dma_dev, sg_addr, DPAA_BP_RAW_SIZE, DMA_FROM_DEVICE); @@ -2022,6 +2049,75 @@ static inline int dpaa_xmit(struct dpaa_priv *priv, return 0; } +#ifdef CONFIG_DPAA_ERRATUM_A050385 +int dpaa_a050385_wa(struct net_device *net_dev, struct sk_buff **s) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct sk_buff *new_skb, *skb = *s; + unsigned char *start, i; + + /* check linear buffer alignment */ + if (!PTR_IS_ALIGNED(skb->data, DPAA_A050385_ALIGN)) + goto workaround; + + /* linear buffers just need to have an aligned start */ + if (!skb_is_nonlinear(skb)) + return 0; + + /* linear data size for nonlinear skbs needs to be aligned */ + if (!IS_ALIGNED(skb_headlen(skb), DPAA_A050385_ALIGN)) + goto workaround; + + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + /* all fragments need to have aligned start addresses */ + if (!IS_ALIGNED(skb_frag_off(frag), DPAA_A050385_ALIGN)) + goto workaround; + + /* all but last fragment need to have aligned sizes */ + if (!IS_ALIGNED(skb_frag_size(frag), DPAA_A050385_ALIGN) && + (i < skb_shinfo(skb)->nr_frags - 1)) + goto workaround; + } + + return 0; + +workaround: + /* copy all the skb content into a new linear buffer */ + new_skb = netdev_alloc_skb(net_dev, skb->len + DPAA_A050385_ALIGN - 1 + + priv->tx_headroom); + if (!new_skb) + return -ENOMEM; + + /* NET_SKB_PAD bytes already reserved, adding up to tx_headroom */ + skb_reserve(new_skb, priv->tx_headroom - NET_SKB_PAD); + + /* Workaround for DPAA_A050385 requires data start to be aligned */ + start = PTR_ALIGN(new_skb->data, DPAA_A050385_ALIGN); + if (start - new_skb->data != 0) + skb_reserve(new_skb, start - new_skb->data); + + skb_put(new_skb, skb->len); + skb_copy_bits(skb, 0, new_skb->data, skb->len); + skb_copy_header(new_skb, skb); + new_skb->dev = skb->dev; + + /* We move the headroom when we align it so we have to reset the + * network and transport header offsets relative to the new data + * pointer. The checksum offload relies on these offsets. + */ + skb_set_network_header(new_skb, skb_network_offset(skb)); + skb_set_transport_header(new_skb, skb_transport_offset(skb)); + + /* TODO: does timestamping need the result in the old skb? */ + dev_kfree_skb(skb); + *s = new_skb; + + return 0; +} +#endif + static netdev_tx_t dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) { @@ -2068,6 +2164,14 @@ dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) nonlinear = skb_is_nonlinear(skb); } +#ifdef CONFIG_DPAA_ERRATUM_A050385 + if (unlikely(fman_has_errata_a050385())) { + if (dpaa_a050385_wa(net_dev, &skb)) + goto enomem; + nonlinear = skb_is_nonlinear(skb); + } +#endif + if (nonlinear) { /* Just create a S/G fd based on the skb */ err = skb_to_sg_fd(priv, skb, &fd); -- cgit From b7469e83d2add567e4e0b063963db185f3167cea Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 4 Mar 2020 09:32:16 -0800 Subject: bonding/alb: make sure arp header is pulled before accessing it Similar to commit 38f88c454042 ("bonding/alb: properly access headers in bond_alb_xmit()"), we need to make sure arp header was pulled in skb->head before blindly accessing it in rlb_arp_xmit(). Remove arp_pkt() private helper, since it is more readable/obvious to have the following construct back to back : if (!pskb_network_may_pull(skb, sizeof(*arp))) return NULL; arp = (struct arp_pkt *)skb_network_header(skb); syzbot reported : BUG: KMSAN: uninit-value in bond_slave_has_mac_rx include/net/bonding.h:704 [inline] BUG: KMSAN: uninit-value in rlb_arp_xmit drivers/net/bonding/bond_alb.c:662 [inline] BUG: KMSAN: uninit-value in bond_alb_xmit+0x575/0x25e0 drivers/net/bonding/bond_alb.c:1477 CPU: 0 PID: 12743 Comm: syz-executor.4 Not tainted 5.6.0-rc2-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1c9/0x220 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:118 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:215 bond_slave_has_mac_rx include/net/bonding.h:704 [inline] rlb_arp_xmit drivers/net/bonding/bond_alb.c:662 [inline] bond_alb_xmit+0x575/0x25e0 drivers/net/bonding/bond_alb.c:1477 __bond_start_xmit drivers/net/bonding/bond_main.c:4257 [inline] bond_start_xmit+0x85d/0x2f70 drivers/net/bonding/bond_main.c:4282 __netdev_start_xmit include/linux/netdevice.h:4524 [inline] netdev_start_xmit include/linux/netdevice.h:4538 [inline] xmit_one net/core/dev.c:3470 [inline] dev_hard_start_xmit+0x531/0xab0 net/core/dev.c:3486 __dev_queue_xmit+0x37de/0x4220 net/core/dev.c:4063 dev_queue_xmit+0x4b/0x60 net/core/dev.c:4096 packet_snd net/packet/af_packet.c:2967 [inline] packet_sendmsg+0x8347/0x93b0 net/packet/af_packet.c:2992 sock_sendmsg_nosec net/socket.c:652 [inline] sock_sendmsg net/socket.c:672 [inline] __sys_sendto+0xc1b/0xc50 net/socket.c:1998 __do_sys_sendto net/socket.c:2010 [inline] __se_sys_sendto+0x107/0x130 net/socket.c:2006 __x64_sys_sendto+0x6e/0x90 net/socket.c:2006 do_syscall_64+0xb8/0x160 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x45c479 Code: ad b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 7b b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fc77ffbbc78 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 00007fc77ffbc6d4 RCX: 000000000045c479 RDX: 000000000000000e RSI: 00000000200004c0 RDI: 0000000000000003 RBP: 000000000076bf20 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff R13: 0000000000000a04 R14: 00000000004cc7b0 R15: 000000000076bf2c Uninit was created at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:144 [inline] kmsan_internal_poison_shadow+0x66/0xd0 mm/kmsan/kmsan.c:127 kmsan_slab_alloc+0x8a/0xe0 mm/kmsan/kmsan_hooks.c:82 slab_alloc_node mm/slub.c:2793 [inline] __kmalloc_node_track_caller+0xb40/0x1200 mm/slub.c:4401 __kmalloc_reserve net/core/skbuff.c:142 [inline] __alloc_skb+0x2fd/0xac0 net/core/skbuff.c:210 alloc_skb include/linux/skbuff.h:1051 [inline] alloc_skb_with_frags+0x18c/0xa70 net/core/skbuff.c:5766 sock_alloc_send_pskb+0xada/0xc60 net/core/sock.c:2242 packet_alloc_skb net/packet/af_packet.c:2815 [inline] packet_snd net/packet/af_packet.c:2910 [inline] packet_sendmsg+0x66a0/0x93b0 net/packet/af_packet.c:2992 sock_sendmsg_nosec net/socket.c:652 [inline] sock_sendmsg net/socket.c:672 [inline] __sys_sendto+0xc1b/0xc50 net/socket.c:1998 __do_sys_sendto net/socket.c:2010 [inline] __se_sys_sendto+0x107/0x130 net/socket.c:2006 __x64_sys_sendto+0x6e/0x90 net/socket.c:2006 do_syscall_64+0xb8/0x160 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: David S. Miller --- drivers/net/bonding/bond_alb.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index 1cc2cd894f87..c81698550e5a 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -50,11 +50,6 @@ struct arp_pkt { }; #pragma pack() -static inline struct arp_pkt *arp_pkt(const struct sk_buff *skb) -{ - return (struct arp_pkt *)skb_network_header(skb); -} - /* Forward declaration */ static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[], bool strict_match); @@ -553,10 +548,11 @@ static void rlb_req_update_subnet_clients(struct bonding *bond, __be32 src_ip) spin_unlock(&bond->mode_lock); } -static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bond) +static struct slave *rlb_choose_channel(struct sk_buff *skb, + struct bonding *bond, + const struct arp_pkt *arp) { struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond)); - struct arp_pkt *arp = arp_pkt(skb); struct slave *assigned_slave, *curr_active_slave; struct rlb_client_info *client_info; u32 hash_index = 0; @@ -653,8 +649,12 @@ static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bon */ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) { - struct arp_pkt *arp = arp_pkt(skb); struct slave *tx_slave = NULL; + struct arp_pkt *arp; + + if (!pskb_network_may_pull(skb, sizeof(*arp))) + return NULL; + arp = (struct arp_pkt *)skb_network_header(skb); /* Don't modify or load balance ARPs that do not originate locally * (e.g.,arrive via a bridge). @@ -664,7 +664,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) if (arp->op_code == htons(ARPOP_REPLY)) { /* the arp must be sent on the selected rx channel */ - tx_slave = rlb_choose_channel(skb, bond); + tx_slave = rlb_choose_channel(skb, bond, arp); if (tx_slave) bond_hw_addr_copy(arp->mac_src, tx_slave->dev->dev_addr, tx_slave->dev->addr_len); @@ -676,7 +676,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) * When the arp reply is received the entry will be updated * with the correct unicast address of the client. */ - tx_slave = rlb_choose_channel(skb, bond); + tx_slave = rlb_choose_channel(skb, bond, arp); /* The ARP reply packets must be delayed so that * they can cancel out the influence of the ARP request. -- cgit From e396ce5f4229388db5143fec0ec36af504bff12e Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Wed, 4 Mar 2020 09:42:10 -0800 Subject: ionic: fix vf op lock usage These are a couple of read locks that should be write locks. Fixes: fbb39807e9ae ("ionic: support sr-iov operations") Signed-off-by: Shannon Nelson Reviewed-by: Parav Pandit Signed-off-by: David S. Miller --- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 191271f6260d..c2f5b691e0fa 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -1688,7 +1688,7 @@ static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) if (!(is_zero_ether_addr(mac) || is_valid_ether_addr(mac))) return -EINVAL; - down_read(&ionic->vf_op_lock); + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { ret = -EINVAL; @@ -1698,7 +1698,7 @@ static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) ether_addr_copy(ionic->vfs[vf].macaddr, mac); } - up_read(&ionic->vf_op_lock); + up_write(&ionic->vf_op_lock); return ret; } @@ -1719,7 +1719,7 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, if (proto != htons(ETH_P_8021Q)) return -EPROTONOSUPPORT; - down_read(&ionic->vf_op_lock); + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { ret = -EINVAL; @@ -1730,7 +1730,7 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, ionic->vfs[vf].vlanid = vlan; } - up_read(&ionic->vf_op_lock); + up_write(&ionic->vf_op_lock); return ret; } -- cgit From 03138e2bf78b2e219b5f15b9bbd84e2d457e3b73 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 4 Mar 2020 10:17:53 -0800 Subject: MAINTAINERS: remove bouncing pkaustub@cisco.com from enic pkaustub@cisco.com is bouncing, remove it. Signed-off-by: Jakub Kicinski Acked-by: Christian Benvenuti Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8f27f40d22bb..b1935c2ae118 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4073,7 +4073,6 @@ F: drivers/scsi/snic/ CISCO VIC ETHERNET NIC DRIVER M: Christian Benvenuti M: Govindarajulu Varadarajan <_govind@gmx.com> -M: Parvi Kaustubhi S: Supported F: drivers/net/ethernet/cisco/enic/ -- cgit From aeaa925bff844d225f259215a250d2811e436499 Mon Sep 17 00:00:00 2001 From: Jonathan Neuschäfer Date: Thu, 5 Mar 2020 17:05:16 +0100 Subject: rhashtable: Document the right function parameters MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit rhashtable_lookup_get_insert_key doesn't have a parameter `data`. It does have a parameter `key`, however. Signed-off-by: Jonathan Neuschäfer Acked-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/rhashtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h index beb9a9da1699..70ebef866cc8 100644 --- a/include/linux/rhashtable.h +++ b/include/linux/rhashtable.h @@ -972,9 +972,9 @@ static inline int rhashtable_lookup_insert_key( /** * rhashtable_lookup_get_insert_key - lookup and insert object into hash table * @ht: hash table + * @key: key * @obj: pointer to hash head inside object * @params: hash table parameters - * @data: pointer to element data already in hashes * * Just like rhashtable_lookup_insert_key(), but this function returns the * object if it exists, NULL if it does not and the insertion was successful, -- cgit From 63aae7b17344d4b08a7d05cb07044de4c0f9dcc6 Mon Sep 17 00:00:00 2001 From: Jiri Wiesner Date: Sat, 7 Mar 2020 13:31:57 +0100 Subject: ipvlan: do not add hardware address of master to its unicast filter list There is a problem when ipvlan slaves are created on a master device that is a vmxnet3 device (ipvlan in VMware guests). The vmxnet3 driver does not support unicast address filtering. When an ipvlan device is brought up in ipvlan_open(), the ipvlan driver calls dev_uc_add() to add the hardware address of the vmxnet3 master device to the unicast address list of the master device, phy_dev->uc. This inevitably leads to the vmxnet3 master device being forced into promiscuous mode by __dev_set_rx_mode(). Promiscuous mode is switched on the master despite the fact that there is still only one hardware address that the master device should use for filtering in order for the ipvlan device to be able to receive packets. The comment above struct net_device describes the uc_promisc member as a "counter, that indicates, that promiscuous mode has been enabled due to the need to listen to additional unicast addresses in a device that does not implement ndo_set_rx_mode()". Moreover, the design of ipvlan guarantees that only the hardware address of a master device, phy_dev->dev_addr, will be used to transmit and receive all packets from its ipvlan slaves. Thus, the unicast address list of the master device should not be modified by ipvlan_open() and ipvlan_stop() in order to make ipvlan a workable option on masters that do not support unicast address filtering. Fixes: 2ad7bf3638411 ("ipvlan: Initial check-in of the IPVLAN driver") Reported-by: Per Sundstrom Signed-off-by: Jiri Wiesner Reviewed-by: Eric Dumazet Acked-by: Mahesh Bandewar Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_main.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index a70662261a5a..f195f278a83a 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -164,7 +164,6 @@ static void ipvlan_uninit(struct net_device *dev) static int ipvlan_open(struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); - struct net_device *phy_dev = ipvlan->phy_dev; struct ipvl_addr *addr; if (ipvlan->port->mode == IPVLAN_MODE_L3 || @@ -178,7 +177,7 @@ static int ipvlan_open(struct net_device *dev) ipvlan_ht_addr_add(ipvlan, addr); rcu_read_unlock(); - return dev_uc_add(phy_dev, phy_dev->dev_addr); + return 0; } static int ipvlan_stop(struct net_device *dev) @@ -190,8 +189,6 @@ static int ipvlan_stop(struct net_device *dev) dev_uc_unsync(phy_dev, dev); dev_mc_unsync(phy_dev, dev); - dev_uc_del(phy_dev, phy_dev->dev_addr); - rcu_read_lock(); list_for_each_entry_rcu(addr, &ipvlan->addrs, anode) ipvlan_ht_addr_del(addr); -- cgit From 17c25cafd4d3e74c83dce56b158843b19c40b414 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 7 Mar 2020 22:05:14 -0800 Subject: gre: fix uninit-value in __iptunnel_pull_header syzbot found an interesting case of the kernel reading an uninit-value [1] Problem is in the handling of ETH_P_WCCP in gre_parse_header() We look at the byte following GRE options to eventually decide if the options are four bytes longer. Use skb_header_pointer() to not pull bytes if we found that no more bytes were needed. All callers of gre_parse_header() are properly using pskb_may_pull() anyway before proceeding to next header. [1] BUG: KMSAN: uninit-value in pskb_may_pull include/linux/skbuff.h:2303 [inline] BUG: KMSAN: uninit-value in __iptunnel_pull_header+0x30c/0xbd0 net/ipv4/ip_tunnel_core.c:94 CPU: 1 PID: 11784 Comm: syz-executor940 Not tainted 5.6.0-rc2-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1c9/0x220 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:118 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:215 pskb_may_pull include/linux/skbuff.h:2303 [inline] __iptunnel_pull_header+0x30c/0xbd0 net/ipv4/ip_tunnel_core.c:94 iptunnel_pull_header include/net/ip_tunnels.h:411 [inline] gre_rcv+0x15e/0x19c0 net/ipv6/ip6_gre.c:606 ip6_protocol_deliver_rcu+0x181b/0x22c0 net/ipv6/ip6_input.c:432 ip6_input_finish net/ipv6/ip6_input.c:473 [inline] NF_HOOK include/linux/netfilter.h:307 [inline] ip6_input net/ipv6/ip6_input.c:482 [inline] ip6_mc_input+0xdf2/0x1460 net/ipv6/ip6_input.c:576 dst_input include/net/dst.h:442 [inline] ip6_rcv_finish net/ipv6/ip6_input.c:76 [inline] NF_HOOK include/linux/netfilter.h:307 [inline] ipv6_rcv+0x683/0x710 net/ipv6/ip6_input.c:306 __netif_receive_skb_one_core net/core/dev.c:5198 [inline] __netif_receive_skb net/core/dev.c:5312 [inline] netif_receive_skb_internal net/core/dev.c:5402 [inline] netif_receive_skb+0x66b/0xf20 net/core/dev.c:5461 tun_rx_batched include/linux/skbuff.h:4321 [inline] tun_get_user+0x6aef/0x6f60 drivers/net/tun.c:1997 tun_chr_write_iter+0x1f2/0x360 drivers/net/tun.c:2026 call_write_iter include/linux/fs.h:1901 [inline] new_sync_write fs/read_write.c:483 [inline] __vfs_write+0xa5a/0xca0 fs/read_write.c:496 vfs_write+0x44a/0x8f0 fs/read_write.c:558 ksys_write+0x267/0x450 fs/read_write.c:611 __do_sys_write fs/read_write.c:623 [inline] __se_sys_write fs/read_write.c:620 [inline] __ia32_sys_write+0xdb/0x120 fs/read_write.c:620 do_syscall_32_irqs_on arch/x86/entry/common.c:339 [inline] do_fast_syscall_32+0x3c7/0x6e0 arch/x86/entry/common.c:410 entry_SYSENTER_compat+0x68/0x77 arch/x86/entry/entry_64_compat.S:139 RIP: 0023:0xf7f62d99 Code: 90 e8 0b 00 00 00 f3 90 0f ae e8 eb f9 8d 74 26 00 89 3c 24 c3 90 90 90 90 90 90 90 90 90 90 90 90 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 eb 0d 90 90 90 90 90 90 90 90 90 90 90 90 RSP: 002b:00000000fffedb2c EFLAGS: 00000217 ORIG_RAX: 0000000000000004 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000020002580 RDX: 0000000000000fca RSI: 0000000000000036 RDI: 0000000000000004 RBP: 0000000000008914 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Uninit was created at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:144 [inline] kmsan_internal_poison_shadow+0x66/0xd0 mm/kmsan/kmsan.c:127 kmsan_slab_alloc+0x8a/0xe0 mm/kmsan/kmsan_hooks.c:82 slab_alloc_node mm/slub.c:2793 [inline] __kmalloc_node_track_caller+0xb40/0x1200 mm/slub.c:4401 __kmalloc_reserve net/core/skbuff.c:142 [inline] __alloc_skb+0x2fd/0xac0 net/core/skbuff.c:210 alloc_skb include/linux/skbuff.h:1051 [inline] alloc_skb_with_frags+0x18c/0xa70 net/core/skbuff.c:5766 sock_alloc_send_pskb+0xada/0xc60 net/core/sock.c:2242 tun_alloc_skb drivers/net/tun.c:1529 [inline] tun_get_user+0x10ae/0x6f60 drivers/net/tun.c:1843 tun_chr_write_iter+0x1f2/0x360 drivers/net/tun.c:2026 call_write_iter include/linux/fs.h:1901 [inline] new_sync_write fs/read_write.c:483 [inline] __vfs_write+0xa5a/0xca0 fs/read_write.c:496 vfs_write+0x44a/0x8f0 fs/read_write.c:558 ksys_write+0x267/0x450 fs/read_write.c:611 __do_sys_write fs/read_write.c:623 [inline] __se_sys_write fs/read_write.c:620 [inline] __ia32_sys_write+0xdb/0x120 fs/read_write.c:620 do_syscall_32_irqs_on arch/x86/entry/common.c:339 [inline] do_fast_syscall_32+0x3c7/0x6e0 arch/x86/entry/common.c:410 entry_SYSENTER_compat+0x68/0x77 arch/x86/entry/entry_64_compat.S:139 Fixes: 95f5c64c3c13 ("gre: Move utility functions to common headers") Fixes: c54419321455 ("GRE: Refactor GRE tunneling code.") Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: David S. Miller --- net/ipv4/gre_demux.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/net/ipv4/gre_demux.c b/net/ipv4/gre_demux.c index 5fd6e8ed02b5..66fdbfe5447c 100644 --- a/net/ipv4/gre_demux.c +++ b/net/ipv4/gre_demux.c @@ -56,7 +56,9 @@ int gre_del_protocol(const struct gre_protocol *proto, u8 version) } EXPORT_SYMBOL_GPL(gre_del_protocol); -/* Fills in tpi and returns header length to be pulled. */ +/* Fills in tpi and returns header length to be pulled. + * Note that caller must use pskb_may_pull() before pulling GRE header. + */ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, bool *csum_err, __be16 proto, int nhs) { @@ -110,8 +112,14 @@ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header */ if (greh->flags == 0 && tpi->proto == htons(ETH_P_WCCP)) { + u8 _val, *val; + + val = skb_header_pointer(skb, nhs + hdr_len, + sizeof(_val), &_val); + if (!val) + return -EINVAL; tpi->proto = proto; - if ((*(u8 *)options & 0xF0) != 0x40) + if ((*val & 0xF0) != 0x40) hdr_len += 4; } tpi->hdr_len = hdr_len; -- cgit From b723bd933980f4956dabc8a8d84b3e83be8d094c Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Sun, 8 Mar 2020 10:25:56 +0100 Subject: net: stmmac: dwmac1000: Disable ACS if enhanced descs are not used ACS (auto PAD/FCS stripping) removes FCS off 802.3 packets (LLC) so that there is no need to manually strip it for such packets. The enhanced DMA descriptors allow to flag LLC packets so that the receiving callback can use that to strip FCS manually or not. On the other hand, normal descriptors do not support that. Thus in order to not truncate LLC packet ACS should be disabled when using normal DMA descriptors. Fixes: 47dd7a540b8a0 ("net: add support for STMicroelectronics Ethernet controllers.") Signed-off-by: Remi Pommarel Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index d0356fbd1e43..542784300620 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -24,6 +24,7 @@ static void dwmac1000_core_init(struct mac_device_info *hw, struct net_device *dev) { + struct stmmac_priv *priv = netdev_priv(dev); void __iomem *ioaddr = hw->pcsr; u32 value = readl(ioaddr + GMAC_CONTROL); int mtu = dev->mtu; @@ -35,7 +36,7 @@ static void dwmac1000_core_init(struct mac_device_info *hw, * Broadcom tags can look like invalid LLC/SNAP packets and cause the * hardware to truncate packets on reception. */ - if (netdev_uses_dsa(dev)) + if (netdev_uses_dsa(dev) || !priv->plat->enh_desc) value &= ~GMAC_CONTROL_ACS; if (mtu > 1500) -- cgit From 83f73c5bb7b9a9135173f0ba2b1aa00c06664ff9 Mon Sep 17 00:00:00 2001 From: Dmitry Yakunin Date: Thu, 5 Mar 2020 15:33:12 +0300 Subject: inet_diag: return classid for all socket types In commit 1ec17dbd90f8 ("inet_diag: fix reporting cgroup classid and fallback to priority") croup classid reporting was fixed. But this works only for TCP sockets because for other socket types icsk parameter can be NULL and classid code path is skipped. This change moves classid handling to inet_diag_msg_attrs_fill() function. Also inet_diag_msg_attrs_size() helper was added and addends in nlmsg_new() were reordered to save order from inet_sk_diag_fill(). Fixes: 1ec17dbd90f8 ("inet_diag: fix reporting cgroup classid and fallback to priority") Signed-off-by: Dmitry Yakunin Reviewed-by: Konstantin Khlebnikov Signed-off-by: David S. Miller --- include/linux/inet_diag.h | 18 ++++++++++++------ net/ipv4/inet_diag.c | 44 ++++++++++++++++++++------------------------ net/ipv4/raw_diag.c | 5 +++-- net/ipv4/udp_diag.c | 5 +++-- net/sctp/diag.c | 8 ++------ 5 files changed, 40 insertions(+), 40 deletions(-) diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h index 39faaaf843e1..c91cf2dee12a 100644 --- a/include/linux/inet_diag.h +++ b/include/linux/inet_diag.h @@ -2,15 +2,10 @@ #ifndef _INET_DIAG_H_ #define _INET_DIAG_H_ 1 +#include #include -struct net; -struct sock; struct inet_hashinfo; -struct nlattr; -struct nlmsghdr; -struct sk_buff; -struct netlink_callback; struct inet_diag_handler { void (*dump)(struct sk_buff *skb, @@ -62,6 +57,17 @@ int inet_diag_bc_sk(const struct nlattr *_bc, struct sock *sk); void inet_diag_msg_common_fill(struct inet_diag_msg *r, struct sock *sk); +static inline size_t inet_diag_msg_attrs_size(void) +{ + return nla_total_size(1) /* INET_DIAG_SHUTDOWN */ + + nla_total_size(1) /* INET_DIAG_TOS */ +#if IS_ENABLED(CONFIG_IPV6) + + nla_total_size(1) /* INET_DIAG_TCLASS */ + + nla_total_size(1) /* INET_DIAG_SKV6ONLY */ +#endif + + nla_total_size(4) /* INET_DIAG_MARK */ + + nla_total_size(4); /* INET_DIAG_CLASS_ID */ +} int inet_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb, struct inet_diag_msg *r, int ext, struct user_namespace *user_ns, bool net_admin); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index f11e997e517b..8c8377568a78 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -100,13 +100,9 @@ static size_t inet_sk_attr_size(struct sock *sk, aux = handler->idiag_get_aux_size(sk, net_admin); return nla_total_size(sizeof(struct tcp_info)) - + nla_total_size(1) /* INET_DIAG_SHUTDOWN */ - + nla_total_size(1) /* INET_DIAG_TOS */ - + nla_total_size(1) /* INET_DIAG_TCLASS */ - + nla_total_size(4) /* INET_DIAG_MARK */ - + nla_total_size(4) /* INET_DIAG_CLASS_ID */ - + nla_total_size(sizeof(struct inet_diag_meminfo)) + nla_total_size(sizeof(struct inet_diag_msg)) + + inet_diag_msg_attrs_size() + + nla_total_size(sizeof(struct inet_diag_meminfo)) + nla_total_size(SK_MEMINFO_VARS * sizeof(u32)) + nla_total_size(TCP_CA_NAME_MAX) + nla_total_size(sizeof(struct tcpvegas_info)) @@ -147,6 +143,24 @@ int inet_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb, if (net_admin && nla_put_u32(skb, INET_DIAG_MARK, sk->sk_mark)) goto errout; + if (ext & (1 << (INET_DIAG_CLASS_ID - 1)) || + ext & (1 << (INET_DIAG_TCLASS - 1))) { + u32 classid = 0; + +#ifdef CONFIG_SOCK_CGROUP_DATA + classid = sock_cgroup_classid(&sk->sk_cgrp_data); +#endif + /* Fallback to socket priority if class id isn't set. + * Classful qdiscs use it as direct reference to class. + * For cgroup2 classid is always zero. + */ + if (!classid) + classid = sk->sk_priority; + + if (nla_put_u32(skb, INET_DIAG_CLASS_ID, classid)) + goto errout; + } + r->idiag_uid = from_kuid_munged(user_ns, sock_i_uid(sk)); r->idiag_inode = sock_i_ino(sk); @@ -284,24 +298,6 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, goto errout; } - if (ext & (1 << (INET_DIAG_CLASS_ID - 1)) || - ext & (1 << (INET_DIAG_TCLASS - 1))) { - u32 classid = 0; - -#ifdef CONFIG_SOCK_CGROUP_DATA - classid = sock_cgroup_classid(&sk->sk_cgrp_data); -#endif - /* Fallback to socket priority if class id isn't set. - * Classful qdiscs use it as direct reference to class. - * For cgroup2 classid is always zero. - */ - if (!classid) - classid = sk->sk_priority; - - if (nla_put_u32(skb, INET_DIAG_CLASS_ID, classid)) - goto errout; - } - out: nlmsg_end(skb, nlh); return 0; diff --git a/net/ipv4/raw_diag.c b/net/ipv4/raw_diag.c index e35736b99300..a93e7d1e1251 100644 --- a/net/ipv4/raw_diag.c +++ b/net/ipv4/raw_diag.c @@ -100,8 +100,9 @@ static int raw_diag_dump_one(struct sk_buff *in_skb, if (IS_ERR(sk)) return PTR_ERR(sk); - rep = nlmsg_new(sizeof(struct inet_diag_msg) + - sizeof(struct inet_diag_meminfo) + 64, + rep = nlmsg_new(nla_total_size(sizeof(struct inet_diag_msg)) + + inet_diag_msg_attrs_size() + + nla_total_size(sizeof(struct inet_diag_meminfo)) + 64, GFP_KERNEL); if (!rep) { sock_put(sk); diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c index 910555a4d9fe..dccd2286bc28 100644 --- a/net/ipv4/udp_diag.c +++ b/net/ipv4/udp_diag.c @@ -64,8 +64,9 @@ static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb, goto out; err = -ENOMEM; - rep = nlmsg_new(sizeof(struct inet_diag_msg) + - sizeof(struct inet_diag_meminfo) + 64, + rep = nlmsg_new(nla_total_size(sizeof(struct inet_diag_msg)) + + inet_diag_msg_attrs_size() + + nla_total_size(sizeof(struct inet_diag_meminfo)) + 64, GFP_KERNEL); if (!rep) goto out; diff --git a/net/sctp/diag.c b/net/sctp/diag.c index 8a15146faaeb..1069d7af3672 100644 --- a/net/sctp/diag.c +++ b/net/sctp/diag.c @@ -237,15 +237,11 @@ static size_t inet_assoc_attr_size(struct sctp_association *asoc) addrcnt++; return nla_total_size(sizeof(struct sctp_info)) - + nla_total_size(1) /* INET_DIAG_SHUTDOWN */ - + nla_total_size(1) /* INET_DIAG_TOS */ - + nla_total_size(1) /* INET_DIAG_TCLASS */ - + nla_total_size(4) /* INET_DIAG_MARK */ - + nla_total_size(4) /* INET_DIAG_CLASS_ID */ + nla_total_size(addrlen * asoc->peer.transport_count) + nla_total_size(addrlen * addrcnt) - + nla_total_size(sizeof(struct inet_diag_meminfo)) + nla_total_size(sizeof(struct inet_diag_msg)) + + inet_diag_msg_attrs_size() + + nla_total_size(sizeof(struct inet_diag_meminfo)) + 64; } -- cgit From 661388f9b0004421b2ecdeac17248bd92a3bb818 Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Mon, 9 Mar 2020 19:43:56 +0900 Subject: linux-next: DOC: RDS: Fix a typo in rds.txt This patch fix a spelling typo in rds.txt Signed-off-by: Masanari Iida Signed-off-by: David S. Miller --- Documentation/networking/rds.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/rds.txt b/Documentation/networking/rds.txt index f2a0147c933d..eec61694e894 100644 --- a/Documentation/networking/rds.txt +++ b/Documentation/networking/rds.txt @@ -159,7 +159,7 @@ Socket Interface set SO_RDS_TRANSPORT on a socket for which the transport has been previously attached explicitly (by SO_RDS_TRANSPORT) or implicitly (via bind(2)) will return an error of EOPNOTSUPP. - An attempt to set SO_RDS_TRANSPPORT to RDS_TRANS_NONE will + An attempt to set SO_RDS_TRANSPORT to RDS_TRANS_NONE will always return EINVAL. RDMA for RDS -- cgit From 4b1bd9db078f7d5332c8601a2f5bd43cf0458fd4 Mon Sep 17 00:00:00 2001 From: Edward Cree Date: Mon, 9 Mar 2020 18:16:24 +0000 Subject: sfc: detach from cb_page in efx_copy_channel() It's a resource, not a parameter, so we can't copy it into the new channel's TX queues, otherwise aliasing will lead to resource- management bugs if the channel is subsequently torn down without being initialised. Before the Fixes:-tagged commit there was a similar bug with tsoh_page, but I'm not sure it's worth doing another fix for such old kernels. Fixes: e9117e5099ea ("sfc: Firmware-Assisted TSO version 2") Suggested-by: Derek Shute Signed-off-by: Edward Cree Signed-off-by: David S. Miller --- drivers/net/ethernet/sfc/efx_channels.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c index aeb5e8aa2f2a..73d4e39b5b16 100644 --- a/drivers/net/ethernet/sfc/efx_channels.c +++ b/drivers/net/ethernet/sfc/efx_channels.c @@ -583,6 +583,7 @@ struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel) if (tx_queue->channel) tx_queue->channel = channel; tx_queue->buffer = NULL; + tx_queue->cb_page = NULL; memset(&tx_queue->txd, 0, sizeof(tx_queue->txd)); } -- cgit From ad8192767c9f9cf97da57b9ffcea70fb100febef Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Mon, 9 Mar 2020 15:56:56 -0700 Subject: ipvlan: don't deref eth hdr before checking it's set MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit IPvlan in L3 mode discards outbound multicast packets but performs the check before ensuring the ether-header is set or not. This is an error that Eric found through code browsing. Fixes: 2ad7bf363841 (“ipvlan: Initial check-in of the IPVLAN driver.”) Signed-off-by: Mahesh Bandewar Reported-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_core.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 30cd0c4f0be0..53dac397db37 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -498,19 +498,21 @@ static int ipvlan_process_outbound(struct sk_buff *skb) struct ethhdr *ethh = eth_hdr(skb); int ret = NET_XMIT_DROP; - /* In this mode we dont care about multicast and broadcast traffic */ - if (is_multicast_ether_addr(ethh->h_dest)) { - pr_debug_ratelimited("Dropped {multi|broad}cast of type=[%x]\n", - ntohs(skb->protocol)); - kfree_skb(skb); - goto out; - } - /* The ipvlan is a pseudo-L2 device, so the packets that we receive * will have L2; which need to discarded and processed further * in the net-ns of the main-device. */ if (skb_mac_header_was_set(skb)) { + /* In this mode we dont care about + * multicast and broadcast traffic */ + if (is_multicast_ether_addr(ethh->h_dest)) { + pr_debug_ratelimited( + "Dropped {multi|broad}cast of type=[%x]\n", + ntohs(skb->protocol)); + kfree_skb(skb); + goto out; + } + skb_pull(skb, sizeof(*ethh)); skb->mac_header = (typeof(skb->mac_header))~0U; skb_reset_network_header(skb); -- cgit From e18b353f102e371580f3f01dd47567a25acc3c1d Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Mon, 9 Mar 2020 15:57:02 -0700 Subject: ipvlan: add cond_resched_rcu() while processing muticast backlog MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If there are substantial number of slaves created as simulated by Syzbot, the backlog processing could take much longer and result into the issue found in the Syzbot report. INFO: rcu_sched detected stalls on CPUs/tasks: (detected by 1, t=10502 jiffies, g=5049, c=5048, q=752) All QSes seen, last rcu_sched kthread activity 10502 (4294965563-4294955061), jiffies_till_next_fqs=1, root ->qsmask 0x0 syz-executor.1 R running task on cpu 1 10984 11210 3866 0x30020008 179034491270 Call Trace: [] _sched_show_task kernel/sched/core.c:8063 [inline] [] _sched_show_task.cold+0x2fd/0x392 kernel/sched/core.c:8030 [] sched_show_task+0xb/0x10 kernel/sched/core.c:8073 [] print_other_cpu_stall kernel/rcu/tree.c:1577 [inline] [] check_cpu_stall kernel/rcu/tree.c:1695 [inline] [] __rcu_pending kernel/rcu/tree.c:3478 [inline] [] rcu_pending kernel/rcu/tree.c:3540 [inline] [] rcu_check_callbacks.cold+0xbb4/0xc29 kernel/rcu/tree.c:2876 [] update_process_times+0x32/0x80 kernel/time/timer.c:1635 [] tick_sched_handle+0xa0/0x180 kernel/time/tick-sched.c:161 [] tick_sched_timer+0x44/0x130 kernel/time/tick-sched.c:1193 [] __run_hrtimer kernel/time/hrtimer.c:1393 [inline] [] __hrtimer_run_queues+0x307/0xd90 kernel/time/hrtimer.c:1455 [] hrtimer_interrupt+0x2ea/0x730 kernel/time/hrtimer.c:1513 [] local_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1031 [inline] [] smp_apic_timer_interrupt+0x144/0x5e0 arch/x86/kernel/apic/apic.c:1056 [] apic_timer_interrupt+0x8e/0xa0 arch/x86/entry/entry_64.S:778 RIP: 0010:do_raw_read_lock+0x22/0x80 kernel/locking/spinlock_debug.c:153 RSP: 0018:ffff8801dad07ab8 EFLAGS: 00000a02 ORIG_RAX: ffffffffffffff12 RAX: 0000000000000000 RBX: ffff8801c4135680 RCX: 0000000000000000 RDX: 1ffff10038826afe RSI: ffff88019d816bb8 RDI: ffff8801c41357f0 RBP: ffff8801dad07ac0 R08: 0000000000004b15 R09: 0000000000310273 R10: ffff88019d816bb8 R11: 0000000000000001 R12: ffff8801c41357e8 R13: 0000000000000000 R14: ffff8801cfb19850 R15: ffff8801cfb198b0 [] __raw_read_lock_bh include/linux/rwlock_api_smp.h:177 [inline] [] _raw_read_lock_bh+0x3e/0x50 kernel/locking/spinlock.c:240 [] ipv6_chk_mcast_addr+0x11a/0x6f0 net/ipv6/mcast.c:1006 [] ip6_mc_input+0x319/0x8e0 net/ipv6/ip6_input.c:482 [] dst_input include/net/dst.h:449 [inline] [] ip6_rcv_finish+0x408/0x610 net/ipv6/ip6_input.c:78 [] NF_HOOK include/linux/netfilter.h:292 [inline] [] NF_HOOK include/linux/netfilter.h:286 [inline] [] ipv6_rcv+0x10e/0x420 net/ipv6/ip6_input.c:278 [] __netif_receive_skb_one_core+0x12a/0x1f0 net/core/dev.c:5303 [] __netif_receive_skb+0x2c/0x1b0 net/core/dev.c:5417 [] process_backlog+0x216/0x6c0 net/core/dev.c:6243 [] napi_poll net/core/dev.c:6680 [inline] [] net_rx_action+0x47b/0xfb0 net/core/dev.c:6748 [] __do_softirq+0x2c8/0x99a kernel/softirq.c:317 [] invoke_softirq kernel/softirq.c:399 [inline] [] irq_exit+0x16a/0x1a0 kernel/softirq.c:439 [] exiting_irq arch/x86/include/asm/apic.h:561 [inline] [] smp_apic_timer_interrupt+0x165/0x5e0 arch/x86/kernel/apic/apic.c:1058 [] apic_timer_interrupt+0x8e/0xa0 arch/x86/entry/entry_64.S:778 RIP: 0010:__sanitizer_cov_trace_pc+0x26/0x50 kernel/kcov.c:102 RSP: 0018:ffff880196033bd8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff12 RAX: ffff88019d8161c0 RBX: 00000000ffffffff RCX: ffffc90003501000 RDX: 0000000000000002 RSI: ffffffff816236d1 RDI: 0000000000000005 RBP: ffff880196033bd8 R08: ffff88019d8161c0 R09: 0000000000000000 R10: 1ffff10032c067f0 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000080 R14: 0000000000000000 R15: 0000000000000000 [] do_futex+0x151/0x1d50 kernel/futex.c:3548 [] C_SYSC_futex kernel/futex_compat.c:201 [inline] [] compat_SyS_futex+0x270/0x3b0 kernel/futex_compat.c:175 [] do_syscall_32_irqs_on arch/x86/entry/common.c:353 [inline] [] do_fast_syscall_32+0x357/0xe1c arch/x86/entry/common.c:415 [] entry_SYSENTER_compat+0x8b/0x9d arch/x86/entry/entry_64_compat.S:139 RIP: 0023:0xf7f23c69 RSP: 002b:00000000f5d1f12c EFLAGS: 00000282 ORIG_RAX: 00000000000000f0 RAX: ffffffffffffffda RBX: 000000000816af88 RCX: 0000000000000080 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000000816af8c RBP: 00000000f5d1f228 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 rcu_sched kthread starved for 10502 jiffies! g5049 c5048 f0x2 RCU_GP_WAIT_FQS(3) ->state=0x0 ->cpu=1 rcu_sched R running task on cpu 1 13048 8 2 0x90000000 179099587640 Call Trace: [] context_switch+0x60f/0xa60 kernel/sched/core.c:3209 [] __schedule+0x5aa/0x1da0 kernel/sched/core.c:3934 [] schedule+0x8f/0x1b0 kernel/sched/core.c:4011 [] schedule_timeout+0x50d/0xee0 kernel/time/timer.c:1803 [] rcu_gp_kthread+0xda1/0x3b50 kernel/rcu/tree.c:2327 [] kthread+0x348/0x420 kernel/kthread.c:246 [] ret_from_fork+0x56/0x70 arch/x86/entry/entry_64.S:393 Fixes: ba35f8588f47 (“ipvlan: Defer multicast / broadcast processing to a work-queue”) Signed-off-by: Mahesh Bandewar Reported-by: syzbot Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 53dac397db37..5759e91dec71 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -277,6 +277,7 @@ void ipvlan_process_multicast(struct work_struct *work) } ipvlan_count_rx(ipvlan, len, ret == NET_RX_SUCCESS, true); local_bh_enable(); + cond_resched_rcu(); } rcu_read_unlock(); -- cgit From ce9a4186f9ac475c415ffd20348176a4ea366670 Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Mon, 9 Mar 2020 15:57:07 -0700 Subject: macvlan: add cond_resched() during multicast processing The Rx bound multicast packets are deferred to a workqueue and macvlan can also suffer from the same attack that was discovered by Syzbot for IPvlan. This solution is not as effective as in IPvlan. IPvlan defers all (Tx and Rx) multicast packet processing to a workqueue while macvlan does this way only for the Rx. This fix should address the Rx codition to certain extent. Tx is still suseptible. Tx multicast processing happens when .ndo_start_xmit is called, hence we cannot add cond_resched(). However, it's not that severe since the user which is generating / flooding will be affected the most. Fixes: 412ca1550cbe ("macvlan: Move broadcasts into a work queue") Signed-off-by: Mahesh Bandewar Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 81aa7adf4801..e7289d67268f 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -334,6 +334,8 @@ static void macvlan_process_broadcast(struct work_struct *w) if (src) dev_put(src->dev); consume_skb(skb); + + cond_resched(); } } -- cgit From 018d26fcd12a75fb9b5fe233762aa3f2f0854b88 Mon Sep 17 00:00:00 2001 From: Dmitry Yakunin Date: Thu, 5 Mar 2020 17:45:57 +0300 Subject: cgroup, netclassid: periodically release file_lock on classid updating MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In our production environment we have faced with problem that updating classid in cgroup with heavy tasks cause long freeze of the file tables in this tasks. By heavy tasks we understand tasks with many threads and opened sockets (e.g. balancers). This freeze leads to an increase number of client timeouts. This patch implements following logic to fix this issue: аfter iterating 1000 file descriptors file table lock will be released thus providing a time gap for socket creation/deletion. Now update is non atomic and socket may be skipped using calls: dup2(oldfd, newfd); close(oldfd); But this case is not typical. Moreover before this patch skip is possible too by hiding socket fd in unix socket buffer. New sockets will be allocated with updated classid because cgroup state is updated before start of the file descriptors iteration. So in common cases this patch has no side effects. Signed-off-by: Dmitry Yakunin Reviewed-by: Konstantin Khlebnikov Signed-off-by: David S. Miller --- net/core/netclassid_cgroup.c | 47 ++++++++++++++++++++++++++++++++++---------- 1 file changed, 37 insertions(+), 10 deletions(-) diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c index 0642f91c4038..b4c87fe31be2 100644 --- a/net/core/netclassid_cgroup.c +++ b/net/core/netclassid_cgroup.c @@ -53,30 +53,60 @@ static void cgrp_css_free(struct cgroup_subsys_state *css) kfree(css_cls_state(css)); } +/* + * To avoid freezing of sockets creation for tasks with big number of threads + * and opened sockets lets release file_lock every 1000 iterated descriptors. + * New sockets will already have been created with new classid. + */ + +struct update_classid_context { + u32 classid; + unsigned int batch; +}; + +#define UPDATE_CLASSID_BATCH 1000 + static int update_classid_sock(const void *v, struct file *file, unsigned n) { int err; + struct update_classid_context *ctx = (void *)v; struct socket *sock = sock_from_file(file, &err); if (sock) { spin_lock(&cgroup_sk_update_lock); - sock_cgroup_set_classid(&sock->sk->sk_cgrp_data, - (unsigned long)v); + sock_cgroup_set_classid(&sock->sk->sk_cgrp_data, ctx->classid); spin_unlock(&cgroup_sk_update_lock); } + if (--ctx->batch == 0) { + ctx->batch = UPDATE_CLASSID_BATCH; + return n + 1; + } return 0; } +static void update_classid_task(struct task_struct *p, u32 classid) +{ + struct update_classid_context ctx = { + .classid = classid, + .batch = UPDATE_CLASSID_BATCH + }; + unsigned int fd = 0; + + do { + task_lock(p); + fd = iterate_fd(p->files, fd, update_classid_sock, &ctx); + task_unlock(p); + cond_resched(); + } while (fd); +} + static void cgrp_attach(struct cgroup_taskset *tset) { struct cgroup_subsys_state *css; struct task_struct *p; cgroup_taskset_for_each(p, css, tset) { - task_lock(p); - iterate_fd(p->files, 0, update_classid_sock, - (void *)(unsigned long)css_cls_state(css)->classid); - task_unlock(p); + update_classid_task(p, css_cls_state(css)->classid); } } @@ -98,10 +128,7 @@ static int write_classid(struct cgroup_subsys_state *css, struct cftype *cft, css_task_iter_start(css, 0, &it); while ((p = css_task_iter_next(&it))) { - task_lock(p); - iterate_fd(p->files, 0, update_classid_sock, - (void *)(unsigned long)cs->classid); - task_unlock(p); + update_classid_task(p, cs->classid); cond_resched(); } css_task_iter_end(&it); -- cgit From afe207d80a61e4d6e7cfa0611a4af46d0ba95628 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 9 Mar 2020 18:22:58 -0700 Subject: ipvlan: do not use cond_resched_rcu() in ipvlan_process_multicast() Commit e18b353f102e ("ipvlan: add cond_resched_rcu() while processing muticast backlog") added a cond_resched_rcu() in a loop using rcu protection to iterate over slaves. This is breaking rcu rules, so lets instead use cond_resched() at a point we can reschedule Fixes: e18b353f102e ("ipvlan: add cond_resched_rcu() while processing muticast backlog") Signed-off-by: Eric Dumazet Cc: Mahesh Bandewar Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index 5759e91dec71..8801d093135c 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -277,7 +277,6 @@ void ipvlan_process_multicast(struct work_struct *work) } ipvlan_count_rx(ipvlan, len, ret == NET_RX_SUCCESS, true); local_bh_enable(); - cond_resched_rcu(); } rcu_read_unlock(); @@ -294,6 +293,7 @@ void ipvlan_process_multicast(struct work_struct *work) } if (dev) dev_put(dev); + cond_resched(); } } -- cgit From a8015ded89ad740d21355470d41879c5bd82aab7 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 10 Mar 2020 03:28:18 +0200 Subject: net: mscc: ocelot: properly account for VLAN header length when setting MRU What the driver writes into MAC_MAXLEN_CFG does not actually represent VLAN_ETH_FRAME_LEN but instead ETH_FRAME_LEN + ETH_FCS_LEN. Yes they are numerically equal, but the difference is important, as the switch treats VLAN-tagged traffic specially and knows to increase the maximum accepted frame size automatically. So it is always wrong to account for VLAN in the MAC_MAXLEN_CFG register. Unconditionally increase the maximum allowed frame size for double-tagged traffic. Accounting for the additional length does not mean that the other VLAN membership checks aren't performed, so there's no harm done. Also, stop abusing the MTU name for configuring the MRU. There is no support for configuring the MRU on an interface at the moment. Fixes: a556c76adc05 ("net: mscc: Add initial Ocelot switch support") Fixes: fa914e9c4d94 ("net: mscc: ocelot: create a helper for changing the port MTU") Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- drivers/net/ethernet/mscc/ocelot.c | 28 +++++++++++++++++----------- include/soc/mscc/ocelot_dev.h | 2 +- 2 files changed, 18 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index 86d543ab1ab9..d3b7373c5961 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -2176,24 +2176,29 @@ static int ocelot_init_timestamp(struct ocelot *ocelot) return 0; } -static void ocelot_port_set_mtu(struct ocelot *ocelot, int port, size_t mtu) +/* Configure the maximum SDU (L2 payload) on RX to the value specified in @sdu. + * The length of VLAN tags is accounted for automatically via DEV_MAC_TAGS_CFG. + */ +static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu) { struct ocelot_port *ocelot_port = ocelot->ports[port]; + int maxlen = sdu + ETH_HLEN + ETH_FCS_LEN; int atop_wm; - ocelot_port_writel(ocelot_port, mtu, DEV_MAC_MAXLEN_CFG); + ocelot_port_writel(ocelot_port, maxlen, DEV_MAC_MAXLEN_CFG); /* Set Pause WM hysteresis - * 152 = 6 * mtu / OCELOT_BUFFER_CELL_SZ - * 101 = 4 * mtu / OCELOT_BUFFER_CELL_SZ + * 152 = 6 * maxlen / OCELOT_BUFFER_CELL_SZ + * 101 = 4 * maxlen / OCELOT_BUFFER_CELL_SZ */ ocelot_write_rix(ocelot, SYS_PAUSE_CFG_PAUSE_ENA | SYS_PAUSE_CFG_PAUSE_STOP(101) | SYS_PAUSE_CFG_PAUSE_START(152), SYS_PAUSE_CFG, port); /* Tail dropping watermark */ - atop_wm = (ocelot->shared_queue_sz - 9 * mtu) / OCELOT_BUFFER_CELL_SZ; - ocelot_write_rix(ocelot, ocelot_wm_enc(9 * mtu), + atop_wm = (ocelot->shared_queue_sz - 9 * maxlen) / + OCELOT_BUFFER_CELL_SZ; + ocelot_write_rix(ocelot, ocelot_wm_enc(9 * maxlen), SYS_ATOP, port); ocelot_write(ocelot, ocelot_wm_enc(atop_wm), SYS_ATOP_TOT_CFG); } @@ -2222,9 +2227,10 @@ void ocelot_init_port(struct ocelot *ocelot, int port) DEV_MAC_HDX_CFG); /* Set Max Length and maximum tags allowed */ - ocelot_port_set_mtu(ocelot, port, VLAN_ETH_FRAME_LEN); + ocelot_port_set_maxlen(ocelot, port, ETH_DATA_LEN); ocelot_port_writel(ocelot_port, DEV_MAC_TAGS_CFG_TAG_ID(ETH_P_8021AD) | DEV_MAC_TAGS_CFG_VLAN_AWR_ENA | + DEV_MAC_TAGS_CFG_VLAN_DBL_AWR_ENA | DEV_MAC_TAGS_CFG_VLAN_LEN_AWR_ENA, DEV_MAC_TAGS_CFG); @@ -2310,18 +2316,18 @@ void ocelot_set_cpu_port(struct ocelot *ocelot, int cpu, * Only one port can be an NPI at the same time. */ if (cpu < ocelot->num_phys_ports) { - int mtu = VLAN_ETH_FRAME_LEN + OCELOT_TAG_LEN; + int sdu = ETH_DATA_LEN + OCELOT_TAG_LEN; ocelot_write(ocelot, QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M | QSYS_EXT_CPU_CFG_EXT_CPU_PORT(cpu), QSYS_EXT_CPU_CFG); if (injection == OCELOT_TAG_PREFIX_SHORT) - mtu += OCELOT_SHORT_PREFIX_LEN; + sdu += OCELOT_SHORT_PREFIX_LEN; else if (injection == OCELOT_TAG_PREFIX_LONG) - mtu += OCELOT_LONG_PREFIX_LEN; + sdu += OCELOT_LONG_PREFIX_LEN; - ocelot_port_set_mtu(ocelot, cpu, mtu); + ocelot_port_set_maxlen(ocelot, cpu, sdu); } /* CPU port Injection/Extraction configuration */ diff --git a/include/soc/mscc/ocelot_dev.h b/include/soc/mscc/ocelot_dev.h index 0a50d53bbd3f..7c08437061fc 100644 --- a/include/soc/mscc/ocelot_dev.h +++ b/include/soc/mscc/ocelot_dev.h @@ -74,7 +74,7 @@ #define DEV_MAC_TAGS_CFG_TAG_ID_M GENMASK(31, 16) #define DEV_MAC_TAGS_CFG_TAG_ID_X(x) (((x) & GENMASK(31, 16)) >> 16) #define DEV_MAC_TAGS_CFG_VLAN_LEN_AWR_ENA BIT(2) -#define DEV_MAC_TAGS_CFG_PB_ENA BIT(1) +#define DEV_MAC_TAGS_CFG_VLAN_DBL_AWR_ENA BIT(1) #define DEV_MAC_TAGS_CFG_VLAN_AWR_ENA BIT(0) #define DEV_MAC_ADV_CHK_CFG 0x2c -- cgit From 65dfcf08072b072b633ba9a693ed48a53919b45f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 9 Mar 2020 20:11:42 -0700 Subject: MAINTAINERS: update cxgb4vf maintainer to Vishal Casey Leedomn is bouncing, Vishal indicated he's happy to take the role. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index b1935c2ae118..6693ce1b9e21 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4571,7 +4571,7 @@ F: drivers/infiniband/hw/cxgb4/ F: include/uapi/rdma/cxgb4-abi.h CXGB4VF ETHERNET DRIVER (CXGB4VF) -M: Casey Leedom +M: Vishal Kulkarni L: netdev@vger.kernel.org W: http://www.chelsio.com S: Supported -- cgit From e876ecc67db80dfdb8e237f71e5b43bb88ae549c Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Mon, 9 Mar 2020 22:16:05 -0700 Subject: cgroup: memcg: net: do not associate sock with unrelated cgroup We are testing network memory accounting in our setup and noticed inconsistent network memory usage and often unrelated cgroups network usage correlates with testing workload. On further inspection, it seems like mem_cgroup_sk_alloc() and cgroup_sk_alloc() are broken in irq context specially for cgroup v1. mem_cgroup_sk_alloc() and cgroup_sk_alloc() can be called in irq context and kind of assumes that this can only happen from sk_clone_lock() and the source sock object has already associated cgroup. However in cgroup v1, where network memory accounting is opt-in, the source sock can be unassociated with any cgroup and the new cloned sock can get associated with unrelated interrupted cgroup. Cgroup v2 can also suffer if the source sock object was created by process in the root cgroup or if sk_alloc() is called in irq context. The fix is to just do nothing in interrupt. WARNING: Please note that about half of the TCP sockets are allocated from the IRQ context, so, memory used by such sockets will not be accouted by the memcg. The stack trace of mem_cgroup_sk_alloc() from IRQ-context: CPU: 70 PID: 12720 Comm: ssh Tainted: 5.6.0-smp-DEV #1 Hardware name: ... Call Trace: dump_stack+0x57/0x75 mem_cgroup_sk_alloc+0xe9/0xf0 sk_clone_lock+0x2a7/0x420 inet_csk_clone_lock+0x1b/0x110 tcp_create_openreq_child+0x23/0x3b0 tcp_v6_syn_recv_sock+0x88/0x730 tcp_check_req+0x429/0x560 tcp_v6_rcv+0x72d/0xa40 ip6_protocol_deliver_rcu+0xc9/0x400 ip6_input+0x44/0xd0 ? ip6_protocol_deliver_rcu+0x400/0x400 ip6_rcv_finish+0x71/0x80 ipv6_rcv+0x5b/0xe0 ? ip6_sublist_rcv+0x2e0/0x2e0 process_backlog+0x108/0x1e0 net_rx_action+0x26b/0x460 __do_softirq+0x104/0x2a6 do_softirq_own_stack+0x2a/0x40 do_softirq.part.19+0x40/0x50 __local_bh_enable_ip+0x51/0x60 ip6_finish_output2+0x23d/0x520 ? ip6table_mangle_hook+0x55/0x160 __ip6_finish_output+0xa1/0x100 ip6_finish_output+0x30/0xd0 ip6_output+0x73/0x120 ? __ip6_finish_output+0x100/0x100 ip6_xmit+0x2e3/0x600 ? ipv6_anycast_cleanup+0x50/0x50 ? inet6_csk_route_socket+0x136/0x1e0 ? skb_free_head+0x1e/0x30 inet6_csk_xmit+0x95/0xf0 __tcp_transmit_skb+0x5b4/0xb20 __tcp_send_ack.part.60+0xa3/0x110 tcp_send_ack+0x1d/0x20 tcp_rcv_state_process+0xe64/0xe80 ? tcp_v6_connect+0x5d1/0x5f0 tcp_v6_do_rcv+0x1b1/0x3f0 ? tcp_v6_do_rcv+0x1b1/0x3f0 __release_sock+0x7f/0xd0 release_sock+0x30/0xa0 __inet_stream_connect+0x1c3/0x3b0 ? prepare_to_wait+0xb0/0xb0 inet_stream_connect+0x3b/0x60 __sys_connect+0x101/0x120 ? __sys_getsockopt+0x11b/0x140 __x64_sys_connect+0x1a/0x20 do_syscall_64+0x51/0x200 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The stack trace of mem_cgroup_sk_alloc() from IRQ-context: Fixes: 2d7580738345 ("mm: memcontrol: consolidate cgroup socket tracking") Fixes: d979a39d7242 ("cgroup: duplicate cgroup reference when cloning sockets") Signed-off-by: Shakeel Butt Reviewed-by: Roman Gushchin Signed-off-by: David S. Miller --- kernel/cgroup/cgroup.c | 4 ++++ mm/memcontrol.c | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 75f687301bbf..6b2fc56b2201 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -6258,6 +6258,10 @@ void cgroup_sk_alloc(struct sock_cgroup_data *skcd) return; } + /* Don't associate the sock with unrelated interrupted task's cgroup. */ + if (in_interrupt()) + return; + rcu_read_lock(); while (true) { diff --git a/mm/memcontrol.c b/mm/memcontrol.c index d09776cd6e10..1cf41ee22258 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -6696,6 +6696,10 @@ void mem_cgroup_sk_alloc(struct sock *sk) return; } + /* Do not associate the sock with unrelated interrupted task's memcg. */ + if (in_interrupt()) + return; + rcu_read_lock(); memcg = mem_cgroup_from_task(current); if (memcg == root_mem_cgroup) -- cgit From d752a4986532cb6305dfd5290a614cde8072769d Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Mon, 9 Mar 2020 22:16:06 -0700 Subject: net: memcg: late association of sock to memcg If a TCP socket is allocated in IRQ context or cloned from unassociated (i.e. not associated to a memcg) in IRQ context then it will remain unassociated for its whole life. Almost half of the TCPs created on the system are created in IRQ context, so, memory used by such sockets will not be accounted by the memcg. This issue is more widespread in cgroup v1 where network memory accounting is opt-in but it can happen in cgroup v2 if the source socket for the cloning was created in root memcg. To fix the issue, just do the association of the sockets at the accept() time in the process context and then force charge the memory buffer already used and reserved by the socket. Signed-off-by: Shakeel Butt Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- mm/memcontrol.c | 14 -------------- net/core/sock.c | 5 ++++- net/ipv4/inet_connection_sock.c | 20 ++++++++++++++++++++ 3 files changed, 24 insertions(+), 15 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 1cf41ee22258..2058b8da18db 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -6682,20 +6682,6 @@ void mem_cgroup_sk_alloc(struct sock *sk) if (!mem_cgroup_sockets_enabled) return; - /* - * Socket cloning can throw us here with sk_memcg already - * filled. It won't however, necessarily happen from - * process context. So the test for root memcg given - * the current task's memcg won't help us in this case. - * - * Respecting the original socket's memcg is a better - * decision in this case. - */ - if (sk->sk_memcg) { - css_get(&sk->sk_memcg->css); - return; - } - /* Do not associate the sock with unrelated interrupted task's memcg. */ if (in_interrupt()) return; diff --git a/net/core/sock.c b/net/core/sock.c index a4c8fac781ff..8f71684305c3 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1830,7 +1830,10 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) atomic_set(&newsk->sk_zckey, 0); sock_reset_flag(newsk, SOCK_DONE); - mem_cgroup_sk_alloc(newsk); + + /* sk->sk_memcg will be populated at accept() time */ + newsk->sk_memcg = NULL; + cgroup_sk_alloc(&newsk->sk_cgrp_data); rcu_read_lock(); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index a4db79b1b643..65a3b2565102 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -482,6 +482,26 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) } spin_unlock_bh(&queue->fastopenq.lock); } + + if (mem_cgroup_sockets_enabled) { + int amt; + + /* atomically get the memory usage, set and charge the + * sk->sk_memcg. + */ + lock_sock(newsk); + + /* The sk has not been accepted yet, no need to look at + * sk->sk_wmem_queued. + */ + amt = sk_mem_pages(newsk->sk_forward_alloc + + atomic_read(&sk->sk_rmem_alloc)); + mem_cgroup_sk_alloc(newsk); + if (newsk->sk_memcg && amt) + mem_cgroup_charge_skmem(newsk->sk_memcg, amt); + + release_sock(newsk); + } out: release_sock(sk); if (req) -- cgit From 60380488e4e0b95e9e82aa68aa9705baa86de84c Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 10 Mar 2020 15:27:37 +0800 Subject: ipv6/addrconf: call ipv6_mc_up() for non-Ethernet interface MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Rafał found an issue that for non-Ethernet interface, if we down and up frequently, the memory will be consumed slowly. The reason is we add allnodes/allrouters addressed in multicast list in ipv6_add_dev(). When link down, we call ipv6_mc_down(), store all multicast addresses via mld_add_delrec(). But when link up, we don't call ipv6_mc_up() for non-Ethernet interface to remove the addresses. This makes idev->mc_tomb getting bigger and bigger. The call stack looks like: addrconf_notify(NETDEV_REGISTER) ipv6_add_dev ipv6_dev_mc_inc(ff01::1) ipv6_dev_mc_inc(ff02::1) ipv6_dev_mc_inc(ff02::2) addrconf_notify(NETDEV_UP) addrconf_dev_config /* Alas, we support only Ethernet autoconfiguration. */ return; addrconf_notify(NETDEV_DOWN) addrconf_ifdown ipv6_mc_down igmp6_group_dropped(ff02::2) mld_add_delrec(ff02::2) igmp6_group_dropped(ff02::1) igmp6_group_dropped(ff01::1) After investigating, I can't found a rule to disable multicast on non-Ethernet interface. In RFC2460, the link could be Ethernet, PPP, ATM, tunnels, etc. In IPv4, it doesn't check the dev type when calls ip_mc_up() in inetdev_event(). Even for IPv6, we don't check the dev type and call ipv6_add_dev(), ipv6_dev_mc_inc() after register device. So I think it's OK to fix this memory consumer by calling ipv6_mc_up() for non-Ethernet interface. v2: Also check IFF_MULTICAST flag to make sure the interface supports multicast Reported-by: Rafał Miłecki Tested-by: Rafał Miłecki Fixes: 74235a25c673 ("[IPV6] addrconf: Fix IPv6 on tuntap tunnels") Fixes: 1666d49e1d41 ("mld: do not remove mld souce list info when set link down") Signed-off-by: Hangbin Liu Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index e6e1290ea06f..46d614b611db 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3347,6 +3347,10 @@ static void addrconf_dev_config(struct net_device *dev) (dev->type != ARPHRD_NONE) && (dev->type != ARPHRD_RAWIP)) { /* Alas, we support only Ethernet autoconfiguration. */ + idev = __in6_dev_get(dev); + if (!IS_ERR_OR_NULL(idev) && dev->flags & IFF_UP && + dev->flags & IFF_MULTICAST) + ipv6_mc_up(idev); return; } -- cgit From ece0d7bd74615773268475b6b64d6f1ebbd4b4c6 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Tue, 10 Mar 2020 09:33:30 +0100 Subject: net/smc: cancel event worker during device removal During IB device removal, cancel the event worker before the device structure is freed. Fixes: a4cf0443c414 ("smc: introduce SMC as an IB-client") Reported-by: syzbot+b297c6825752e7a07272@syzkaller.appspotmail.com Signed-off-by: Karsten Graul Reviewed-by: Ursula Braun Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- net/smc/smc_ib.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index d6ba186f67e2..05b825b3cfa4 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -582,6 +582,7 @@ static void smc_ib_remove_dev(struct ib_device *ibdev, void *client_data) smc_smcr_terminate_all(smcibdev); smc_ib_cleanup_per_ibdev(smcibdev); ib_unregister_event_handler(&smcibdev->event_handler); + cancel_work_sync(&smcibdev->port_event_work); kfree(smcibdev); } -- cgit From 7d7195a026bac47ac9943f11f84b7546276209dd Mon Sep 17 00:00:00 2001 From: Juliet Kim Date: Tue, 10 Mar 2020 09:23:58 -0500 Subject: ibmvnic: Do not process device remove during device reset The ibmvnic driver does not check the device state when the device is removed. If the device is removed while a device reset is being processed, the remove may free structures needed by the reset, causing an oops. Fix this by checking the device state before processing device remove. Signed-off-by: Juliet Kim Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 24 ++++++++++++++++++++++-- drivers/net/ethernet/ibm/ibmvnic.h | 6 +++++- 2 files changed, 27 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index c75239d8820f..4bd33245bad6 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -2142,6 +2142,8 @@ static void __ibmvnic_reset(struct work_struct *work) { struct ibmvnic_rwi *rwi; struct ibmvnic_adapter *adapter; + bool saved_state = false; + unsigned long flags; u32 reset_state; int rc = 0; @@ -2153,17 +2155,25 @@ static void __ibmvnic_reset(struct work_struct *work) return; } - reset_state = adapter->state; - rwi = get_next_rwi(adapter); while (rwi) { + spin_lock_irqsave(&adapter->state_lock, flags); + if (adapter->state == VNIC_REMOVING || adapter->state == VNIC_REMOVED) { + spin_unlock_irqrestore(&adapter->state_lock, flags); kfree(rwi); rc = EBUSY; break; } + if (!saved_state) { + reset_state = adapter->state; + adapter->state = VNIC_RESETTING; + saved_state = true; + } + spin_unlock_irqrestore(&adapter->state_lock, flags); + if (rwi->reset_reason == VNIC_RESET_CHANGE_PARAM) { /* CHANGE_PARAM requestor holds rtnl_lock */ rc = do_change_param_reset(adapter, rwi, reset_state); @@ -5091,6 +5101,7 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id) __ibmvnic_delayed_reset); INIT_LIST_HEAD(&adapter->rwi_list); spin_lock_init(&adapter->rwi_lock); + spin_lock_init(&adapter->state_lock); mutex_init(&adapter->fw_lock); init_completion(&adapter->init_done); init_completion(&adapter->fw_done); @@ -5163,8 +5174,17 @@ static int ibmvnic_remove(struct vio_dev *dev) { struct net_device *netdev = dev_get_drvdata(&dev->dev); struct ibmvnic_adapter *adapter = netdev_priv(netdev); + unsigned long flags; + + spin_lock_irqsave(&adapter->state_lock, flags); + if (adapter->state == VNIC_RESETTING) { + spin_unlock_irqrestore(&adapter->state_lock, flags); + return -EBUSY; + } adapter->state = VNIC_REMOVING; + spin_unlock_irqrestore(&adapter->state_lock, flags); + rtnl_lock(); unregister_netdevice(netdev); diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index 60eccaf91b12..f8416e1d4cf0 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -941,7 +941,8 @@ enum vnic_state {VNIC_PROBING = 1, VNIC_CLOSING, VNIC_CLOSED, VNIC_REMOVING, - VNIC_REMOVED}; + VNIC_REMOVED, + VNIC_RESETTING}; enum ibmvnic_reset_reason {VNIC_RESET_FAILOVER = 1, VNIC_RESET_MOBILITY, @@ -1090,4 +1091,7 @@ struct ibmvnic_adapter { struct ibmvnic_tunables desired; struct ibmvnic_tunables fallback; + + /* Used for serializatin of state field */ + spinlock_t state_lock; }; -- cgit From 6fc498bc82929ee23aa2f35a828c6178dfd3f823 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Tue, 10 Mar 2020 18:22:24 +0300 Subject: net: macsec: update SCI upon MAC address change. SCI should be updated, because it contains MAC in its first 6 octets. Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Dmitry Bogdanov Signed-off-by: Mark Starovoytov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- drivers/net/macsec.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 5af424eeea86..66c6392251bc 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -424,6 +424,11 @@ static struct macsec_eth_header *macsec_ethhdr(struct sk_buff *skb) return (struct macsec_eth_header *)skb_mac_header(skb); } +static sci_t dev_to_sci(struct net_device *dev, __be16 port) +{ + return make_sci(dev->dev_addr, port); +} + static void __macsec_pn_wrapped(struct macsec_secy *secy, struct macsec_tx_sa *tx_sa) { @@ -3268,6 +3273,7 @@ static int macsec_set_mac_address(struct net_device *dev, void *p) out: ether_addr_copy(dev->dev_addr, addr->sa_data); + macsec->secy.sci = dev_to_sci(dev, MACSEC_PORT_ES); return 0; } @@ -3593,11 +3599,6 @@ static bool sci_exists(struct net_device *dev, sci_t sci) return false; } -static sci_t dev_to_sci(struct net_device *dev, __be16 port) -{ - return make_sci(dev->dev_addr, port); -} - static int macsec_add_dev(struct net_device *dev, sci_t sci, u8 icv_len) { struct macsec_dev *macsec = macsec_priv(dev); -- cgit From 09f4136c5d6b4c5144a965bd086009863d58ff08 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Tue, 10 Mar 2020 18:22:25 +0300 Subject: net: macsec: invoke mdo_upd_secy callback when mac address changed Notify the offload engine about MAC address change to reconfigure it accordingly. Fixes: 3cf3227a21d1 ("net: macsec: hardware offloading infrastructure") Signed-off-by: Dmitry Bogdanov Signed-off-by: Mark Starovoytov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- drivers/net/macsec.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 66c6392251bc..6ec6fc191a6e 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3274,6 +3274,19 @@ static int macsec_set_mac_address(struct net_device *dev, void *p) out: ether_addr_copy(dev->dev_addr, addr->sa_data); macsec->secy.sci = dev_to_sci(dev, MACSEC_PORT_ES); + + /* If h/w offloading is available, propagate to the device */ + if (macsec_is_offloaded(macsec)) { + const struct macsec_ops *ops; + struct macsec_context ctx; + + ops = macsec_get_ops(macsec, &ctx); + if (ops) { + ctx.secy = &macsec->secy; + macsec_offload(ops->mdo_upd_secy, &ctx); + } + } + return 0; } -- cgit From 240c1948491b81cfe40f84ea040a8f2a4966f101 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Tue, 10 Mar 2020 18:38:01 +0100 Subject: s390/qeth: don't reset default_out_queue When an OSA device in prio-queue setup is reduced to 1 TX queue due to HW restrictions, we reset its the default_out_queue to 0. In the old code this was needed so that qeth_get_priority_queue() gets the queue selection right. But with proper multiqueue support we already reduced dev->real_num_tx_queues to 1, and so the stack puts all traffic on txq 0 without even calling .ndo_select_queue. Thus we can preserve the user's configuration, and apply it if the OSA device later re-gains support for multiple TX queues. Fixes: 73dc2daf110f ("s390/qeth: add TX multiqueue support for OSA devices") Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core_main.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 8ca85c8a01a1..2a05d13a0e79 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -1204,7 +1204,6 @@ static int qeth_osa_set_output_queues(struct qeth_card *card, bool single) if (count == 1) dev_info(&card->gdev->dev, "Priority Queueing not supported\n"); - card->qdio.default_out_queue = single ? 0 : QETH_DEFAULT_QUEUE; card->qdio.no_out_queues = count; return 0; } -- cgit From 17413852804d7e86e6f0576cca32c1541817800e Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Tue, 10 Mar 2020 18:38:02 +0100 Subject: s390/qeth: handle error when backing RX buffer qeth_init_qdio_queues() fills the RX ring with an initial set of RX buffers. If qeth_init_input_buffer() fails to back one of the RX buffers with memory, we need to bail out and report the error. Fixes: 4a71df50047f ("qeth: new qeth device driver") Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core_main.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 2a05d13a0e79..d8f0c610396e 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -2624,12 +2624,12 @@ static int qeth_init_input_buffer(struct qeth_card *card, ETH_HLEN + sizeof(struct ipv6hdr)); if (!buf->rx_skb) - return 1; + return -ENOMEM; } pool_entry = qeth_find_free_buffer_pool_entry(card); if (!pool_entry) - return 1; + return -ENOBUFS; /* * since the buffer is accessed only from the input_tasklet @@ -2674,10 +2674,15 @@ static int qeth_init_qdio_queues(struct qeth_card *card) /* inbound queue */ qdio_reset_buffers(card->qdio.in_q->qdio_bufs, QDIO_MAX_BUFFERS_PER_Q); memset(&card->rx, 0, sizeof(struct qeth_rx)); + qeth_initialize_working_pool_list(card); /*give only as many buffers to hardware as we have buffer pool entries*/ - for (i = 0; i < card->qdio.in_buf_pool.buf_count - 1; ++i) - qeth_init_input_buffer(card, &card->qdio.in_q->bufs[i]); + for (i = 0; i < card->qdio.in_buf_pool.buf_count - 1; i++) { + rc = qeth_init_input_buffer(card, &card->qdio.in_q->bufs[i]); + if (rc) + return rc; + } + card->qdio.in_q->next_buf_to_init = card->qdio.in_buf_pool.buf_count - 1; rc = do_QDIO(CARD_DDEV(card), QDIO_FLAG_SYNC_INPUT, 0, 0, -- cgit From 0e635c2a8713e3e013b4adcb7cb7e28a048b7c42 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Tue, 10 Mar 2020 18:38:03 +0100 Subject: s390/qeth: cancel RX reclaim work earlier When qeth's napi poll code fails to refill an entirely empty RX ring, it kicks off buffer_reclaim_work to try again later. Make sure that this worker is cancelled when setting the qeth device offline. Otherwise a RX refill action can unexpectedly end up running concurrently to bigger re-configurations (eg. resizing the buffer pool), without any locking. Fixes: b333293058aa ("qeth: add support for af_iucv HiperSockets transport") Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core_main.c | 1 - drivers/s390/net/qeth_l2_main.c | 1 + drivers/s390/net/qeth_l3_main.c | 1 + 3 files changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index d8f0c610396e..a599801d7727 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -2392,7 +2392,6 @@ static void qeth_free_qdio_queues(struct qeth_card *card) return; qeth_free_cq(card); - cancel_delayed_work_sync(&card->buffer_reclaim_work); for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; ++j) { if (card->qdio.in_q->bufs[j].rx_skb) dev_kfree_skb_any(card->qdio.in_q->bufs[j].rx_skb); diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 9972d96820f3..8fb29371788b 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -284,6 +284,7 @@ static void qeth_l2_stop_card(struct qeth_card *card) if (card->state == CARD_STATE_SOFTSETUP) { qeth_clear_ipacmd_list(card); qeth_drain_output_queues(card); + cancel_delayed_work_sync(&card->buffer_reclaim_work); card->state = CARD_STATE_DOWN; } diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 317d56647a4a..82f800d1d7b3 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -1178,6 +1178,7 @@ static void qeth_l3_stop_card(struct qeth_card *card) qeth_l3_clear_ip_htable(card, 1); qeth_clear_ipacmd_list(card); qeth_drain_output_queues(card); + cancel_delayed_work_sync(&card->buffer_reclaim_work); card->state = CARD_STATE_DOWN; } -- cgit From 0e1a1d853ecedc99da9d27f9f5c376935547a0e2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:10:56 -0800 Subject: nl80211: add missing attribute validation for critical protocol indication Add missing attribute validation for critical protocol fields to the netlink policy. Fixes: 5de17984898c ("cfg80211: introduce critical protocol indication from user-space") Signed-off-by: Jakub Kicinski Link: https://lore.kernel.org/r/20200303051058.4089398-2-kuba@kernel.org Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 5b19e9fac4aa..cd0e024d7cb6 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -531,6 +531,8 @@ const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_MDID] = { .type = NLA_U16 }, [NL80211_ATTR_IE_RIC] = { .type = NLA_BINARY, .len = IEEE80211_MAX_DATA_LEN }, + [NL80211_ATTR_CRIT_PROT_ID] = { .type = NLA_U16 }, + [NL80211_ATTR_MAX_CRIT_PROT_DURATION] = { .type = NLA_U16 }, [NL80211_ATTR_PEER_AID] = NLA_POLICY_RANGE(NLA_U16, 1, IEEE80211_MAX_AID), [NL80211_ATTR_CH_SWITCH_COUNT] = { .type = NLA_U32 }, -- cgit From 056e9375e1f3c4bf2fd49b70258c7daf788ecd9d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:10:57 -0800 Subject: nl80211: add missing attribute validation for beacon report scanning Add missing attribute validation for beacon report scanning to the netlink policy. Fixes: 1d76250bd34a ("nl80211: support beacon report scanning") Signed-off-by: Jakub Kicinski Link: https://lore.kernel.org/r/20200303051058.4089398-3-kuba@kernel.org Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index cd0e024d7cb6..48e6508aba52 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -470,6 +470,8 @@ const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_WOWLAN_TRIGGERS] = { .type = NLA_NESTED }, [NL80211_ATTR_STA_PLINK_STATE] = NLA_POLICY_MAX(NLA_U8, NUM_NL80211_PLINK_STATES - 1), + [NL80211_ATTR_MEASUREMENT_DURATION] = { .type = NLA_U16 }, + [NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY] = { .type = NLA_FLAG }, [NL80211_ATTR_MESH_PEER_AID] = NLA_POLICY_RANGE(NLA_U16, 1, IEEE80211_MAX_AID), [NL80211_ATTR_SCHED_SCAN_INTERVAL] = { .type = NLA_U32 }, -- cgit From 5cde05c61cbe13cbb3fa66d52b9ae84f7975e5e6 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 2 Mar 2020 21:10:58 -0800 Subject: nl80211: add missing attribute validation for channel switch Add missing attribute validation for NL80211_ATTR_OPER_CLASS to the netlink policy. Fixes: 1057d35ede5d ("cfg80211: introduce TDLS channel switch commands") Signed-off-by: Jakub Kicinski Link: https://lore.kernel.org/r/20200303051058.4089398-4-kuba@kernel.org Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 48e6508aba52..ec5d67794aab 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -565,6 +565,7 @@ const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { NLA_POLICY_MAX(NLA_U8, IEEE80211_NUM_UPS - 1), [NL80211_ATTR_ADMITTED_TIME] = { .type = NLA_U16 }, [NL80211_ATTR_SMPS_MODE] = { .type = NLA_U8 }, + [NL80211_ATTR_OPER_CLASS] = { .type = NLA_U8 }, [NL80211_ATTR_MAC_MASK] = { .type = NLA_EXACT_LEN_WARN, .len = ETH_ALEN -- cgit From ba32679cac50c38fdf488296f96b1f3175532b8e Mon Sep 17 00:00:00 2001 From: Nicolas Cavallari Date: Thu, 5 Mar 2020 15:04:09 +0100 Subject: mac80211: Do not send mesh HWMP PREQ if HWMP is disabled When trying to transmit to an unknown destination, the mesh code would unconditionally transmit a HWMP PREQ even if HWMP is not the current path selection algorithm. Signed-off-by: Nicolas Cavallari Link: https://lore.kernel.org/r/20200305140409.12204-1-cavallar@lri.fr Signed-off-by: Johannes Berg --- net/mac80211/mesh_hwmp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c index d69983370381..38a0383dfbcf 100644 --- a/net/mac80211/mesh_hwmp.c +++ b/net/mac80211/mesh_hwmp.c @@ -1152,7 +1152,8 @@ int mesh_nexthop_resolve(struct ieee80211_sub_if_data *sdata, } } - if (!(mpath->flags & MESH_PATH_RESOLVING)) + if (!(mpath->flags & MESH_PATH_RESOLVING) && + mesh_path_sel_is_hwmp(sdata)) mesh_queue_preq(mpath, PREQ_Q_F_START); if (skb_queue_len(&mpath->frame_queue) >= MESH_FRAME_QUEUE_LEN) -- cgit From 7395f62d95aafacdb9bd4996ec2f95b4a655d7e6 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 10 Mar 2020 15:06:54 -0700 Subject: dpaa_eth: Remove unnecessary boolean expression in dpaa_get_headroom Clang warns: drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:2860:9: warning: converting the result of '?:' with integer constants to a boolean always evaluates to 'true' [-Wtautological-constant-compare] return DPAA_FD_DATA_ALIGNMENT ? ALIGN(headroom, ^ drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:131:34: note: expanded from macro 'DPAA_FD_DATA_ALIGNMENT' \#define DPAA_FD_DATA_ALIGNMENT (fman_has_errata_a050385() ? 64 : 16) ^ 1 warning generated. This was exposed by commit 3c68b8fffb48 ("dpaa_eth: FMan erratum A050385 workaround") even though it appears to have been an issue since the introductory commit 9ad1a3749333 ("dpaa_eth: add support for DPAA Ethernet") since DPAA_FD_DATA_ALIGNMENT has never been able to be zero. Just replace the whole boolean expression with the true branch, as it is always been true. Link: https://github.com/ClangBuiltLinux/linux/issues/928 Signed-off-by: Nathan Chancellor Reviewed-by: Madalin Bucur Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index e3ac9ec54c7c..ca74a684a904 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -2845,9 +2845,7 @@ static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl) headroom = (u16)(bl->priv_data_size + DPAA_PARSE_RESULTS_SIZE + DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE); - return DPAA_FD_DATA_ALIGNMENT ? ALIGN(headroom, - DPAA_FD_DATA_ALIGNMENT) : - headroom; + return ALIGN(headroom, DPAA_FD_DATA_ALIGNMENT); } static int dpaa_eth_probe(struct platform_device *pdev) -- cgit From ab14961d10d02d20767612c78ce148f6eb85bd58 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 10 Mar 2020 20:36:16 -0700 Subject: net: fec: validate the new settings in fec_enet_set_coalesce() fec_enet_set_coalesce() validates the previously set params and if they are within range proceeds to apply the new ones. The new ones, however, are not validated. This seems backwards, probably a copy-paste error? Compile tested only. Fixes: d851b47b22fc ("net: fec: add interrupt coalescence feature support") Signed-off-by: Jakub Kicinski Acked-by: Fugang Duan Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fec_main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 4432a59904c7..23c5fef2f1ad 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2529,15 +2529,15 @@ fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) return -EINVAL; } - cycle = fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr); + cycle = fec_enet_us_to_itr_clock(ndev, ec->rx_coalesce_usecs); if (cycle > 0xFFFF) { dev_err(dev, "Rx coalesced usec exceed hardware limitation\n"); return -EINVAL; } - cycle = fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr); + cycle = fec_enet_us_to_itr_clock(ndev, ec->tx_coalesce_usecs); if (cycle > 0xFFFF) { - dev_err(dev, "Rx coalesced usec exceed hardware limitation\n"); + dev_err(dev, "Tx coalesced usec exceed hardware limitation\n"); return -EINVAL; } -- cgit From eecba79e694eda24ba698fe1bdf0466d2f797b42 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 11 Mar 2020 16:37:02 -0700 Subject: MAINTAINERS: remove Sathya Perla as Emulex NIC maintainer Remove Sathya Perla, sathya.perla@broadcom.com is bouncing. The driver has 3 more maintainers. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 6693ce1b9e21..39937a4d608e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6197,7 +6197,6 @@ S: Supported F: drivers/scsi/be2iscsi/ Emulex 10Gbps NIC BE2, BE3-R, Lancer, Skyhawk-R DRIVER (be2net) -M: Sathya Perla M: Ajit Khaparde M: Sriharsha Basavapatna M: Somnath Kotur -- cgit From f9fc28a8de2fb367f1ab76f0cf176ca545db3d6f Mon Sep 17 00:00:00 2001 From: Amol Grover Date: Thu, 12 Mar 2020 11:04:20 +0530 Subject: net: caif: Add lockdep expression to RCU traversal primitive caifdevs->list is traversed using list_for_each_entry_rcu() outside an RCU read-side critical section but under the protection of rtnl_mutex. Hence, add the corresponding lockdep expression to silence the following false-positive warning: [ 10.868467] ============================= [ 10.869082] WARNING: suspicious RCU usage [ 10.869817] 5.6.0-rc1-00177-g06ec0a154aae4 #1 Not tainted [ 10.870804] ----------------------------- [ 10.871557] net/caif/caif_dev.c:115 RCU-list traversed in non-reader section!! Reported-by: kernel test robot Signed-off-by: Amol Grover Signed-off-by: David S. Miller --- net/caif/caif_dev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c index 03c7cdd8e4cb..195d2d67be8a 100644 --- a/net/caif/caif_dev.c +++ b/net/caif/caif_dev.c @@ -112,7 +112,8 @@ static struct caif_device_entry *caif_get(struct net_device *dev) caif_device_list(dev_net(dev)); struct caif_device_entry *caifd; - list_for_each_entry_rcu(caifd, &caifdevs->list, list) { + list_for_each_entry_rcu(caifd, &caifdevs->list, list, + lockdep_rtnl_is_held()) { if (caifd->netdev == dev) return caifd; } -- cgit From f3cc008bf6d59b8d93b4190e01d3e557b0040e15 Mon Sep 17 00:00:00 2001 From: Dominik Czarnota Date: Mon, 9 Mar 2020 16:22:50 +0100 Subject: sxgbe: Fix off by one in samsung driver strncpy size arg This patch fixes an off-by-one error in strncpy size argument in drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c. The issue is that in: strncmp(opt, "eee_timer:", 6) the passed string literal: "eee_timer:" has 10 bytes (without the NULL byte) and the passed size argument is 6. As a result, the logic will also accept other, malformed strings, e.g. "eee_tiXXX:". This bug doesn't seem to have any security impact since its present in module's cmdline parsing code. Signed-off-by: Dominik Czarnota Signed-off-by: David S. Miller --- drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index c705743d69f7..2cc8184b7e6b 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -2277,7 +2277,7 @@ static int __init sxgbe_cmdline_opt(char *str) if (!str || !*str) return -EINVAL; while ((opt = strsep(&str, ",")) != NULL) { - if (!strncmp(opt, "eee_timer:", 6)) { + if (!strncmp(opt, "eee_timer:", 10)) { if (kstrtoint(opt + 10, 0, &eee_timer)) goto err; } -- cgit From 46e4c421a053c36bf7a33dda2272481bcaf3eed3 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Mon, 9 Mar 2020 11:34:35 -0400 Subject: net/packet: tpacket_rcv: do not increment ring index on drop In one error case, tpacket_rcv drops packets after incrementing the ring producer index. If this happens, it does not update tp_status to TP_STATUS_USER and thus the reader is stalled for an iteration of the ring, causing out of order arrival. The only such error path is when virtio_net_hdr_from_skb fails due to encountering an unknown GSO type. Signed-off-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/packet/af_packet.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 30c6879d6774..e5b0986215d2 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2274,6 +2274,13 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, TP_STATUS_KERNEL, (macoff+snaplen)); if (!h.raw) goto drop_n_account; + + if (do_vnet && + virtio_net_hdr_from_skb(skb, h.raw + macoff - + sizeof(struct virtio_net_hdr), + vio_le(), true, 0)) + goto drop_n_account; + if (po->tp_version <= TPACKET_V2) { packet_increment_rx_head(po, &po->rx_ring); /* @@ -2286,12 +2293,6 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, status |= TP_STATUS_LOSING; } - if (do_vnet && - virtio_net_hdr_from_skb(skb, h.raw + macoff - - sizeof(struct virtio_net_hdr), - vio_le(), true, 0)) - goto drop_n_account; - po->stats.stats1.tp_packets++; if (copy_skb) { status |= TP_STATUS_COPY; -- cgit From a20f997010c4ec76eaa55b8cc047d76dcac69f70 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 11 Mar 2020 16:24:24 +0100 Subject: net: dsa: Don't instantiate phylink for CPU/DSA ports unless needed By default, DSA drivers should configure CPU and DSA ports to their maximum speed. In many configurations this is sufficient to make the link work. In some cases it is necessary to configure the link to run slower, e.g. because of limitations of the SoC it is connected to. Or back to back PHYs are used and the PHY needs to be driven in order to establish link. In this case, phylink is used. Only instantiate phylink if it is required. If there is no PHY, or no fixed link properties, phylink can upset a link which works in the default configuration. Fixes: 0e27921816ad ("net: dsa: Use PHYLINK for the CPU/DSA ports") Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- net/dsa/port.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/net/dsa/port.c b/net/dsa/port.c index ed7dabb57985..ec13dc666788 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -648,9 +648,14 @@ err_phy_connect: int dsa_port_link_register_of(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; + struct device_node *phy_np; - if (!ds->ops->adjust_link) - return dsa_port_phylink_register(dp); + if (!ds->ops->adjust_link) { + phy_np = of_parse_phandle(dp->dn, "phy-handle", 0); + if (of_phy_is_fixed_link(dp->dn) || phy_np) + return dsa_port_phylink_register(dp); + return 0; + } dev_warn(ds->dev, "Using legacy PHYLIB callbacks. Please migrate to PHYLINK!\n"); @@ -665,11 +670,12 @@ void dsa_port_link_unregister_of(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; - if (!ds->ops->adjust_link) { + if (!ds->ops->adjust_link && dp->pl) { rtnl_lock(); phylink_disconnect_phy(dp->pl); rtnl_unlock(); phylink_destroy(dp->pl); + dp->pl = NULL; return; } -- cgit From 2677625387056136e256c743e3285b4fe3da87bb Mon Sep 17 00:00:00 2001 From: Paolo Lungaroni Date: Wed, 11 Mar 2020 17:54:06 +0100 Subject: seg6: fix SRv6 L2 tunnels to use IANA-assigned protocol number The Internet Assigned Numbers Authority (IANA) has recently assigned a protocol number value of 143 for Ethernet [1]. Before this assignment, encapsulation mechanisms such as Segment Routing used the IPv6-NoNxt protocol number (59) to indicate that the encapsulated payload is an Ethernet frame. In this patch, we add the definition of the Ethernet protocol number to the kernel headers and update the SRv6 L2 tunnels to use it. [1] https://www.iana.org/assignments/protocol-numbers/protocol-numbers.xhtml Signed-off-by: Paolo Lungaroni Reviewed-by: Andrea Mayer Acked-by: Ahmed Abdelsalam Signed-off-by: David S. Miller --- include/uapi/linux/in.h | 2 ++ net/ipv6/seg6_iptunnel.c | 2 +- net/ipv6/seg6_local.c | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h index 1521073b6348..8533bf07450f 100644 --- a/include/uapi/linux/in.h +++ b/include/uapi/linux/in.h @@ -74,6 +74,8 @@ enum { #define IPPROTO_UDPLITE IPPROTO_UDPLITE IPPROTO_MPLS = 137, /* MPLS in IP (RFC 4023) */ #define IPPROTO_MPLS IPPROTO_MPLS + IPPROTO_ETHERNET = 143, /* Ethernet-within-IPv6 Encapsulation */ +#define IPPROTO_ETHERNET IPPROTO_ETHERNET IPPROTO_RAW = 255, /* Raw IP packets */ #define IPPROTO_RAW IPPROTO_RAW IPPROTO_MPTCP = 262, /* Multipath TCP connection */ diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index ab7f124ff5d7..8c52efe299cc 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -268,7 +268,7 @@ static int seg6_do_srh(struct sk_buff *skb) skb_mac_header_rebuild(skb); skb_push(skb, skb->mac_len); - err = seg6_do_srh_encap(skb, tinfo->srh, NEXTHDR_NONE); + err = seg6_do_srh_encap(skb, tinfo->srh, IPPROTO_ETHERNET); if (err) return err; diff --git a/net/ipv6/seg6_local.c b/net/ipv6/seg6_local.c index 7cbc19731997..8165802d8e05 100644 --- a/net/ipv6/seg6_local.c +++ b/net/ipv6/seg6_local.c @@ -282,7 +282,7 @@ static int input_action_end_dx2(struct sk_buff *skb, struct net_device *odev; struct ethhdr *eth; - if (!decap_and_validate(skb, NEXTHDR_NONE)) + if (!decap_and_validate(skb, IPPROTO_ETHERNET)) goto drop; if (!pskb_may_pull(skb, ETH_HLEN)) -- cgit From f81649dfa5343eef7e579eb6f8dd8bd6d300ec31 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Wed, 11 Mar 2020 18:07:09 +0100 Subject: s390/qeth: use page pointers to manage RX buffer pool The RX buffer elements are always backed with full pages, reflect this in the pointer type. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core.h | 2 +- drivers/s390/net/qeth_core_main.c | 35 +++++++++++++++++------------------ 2 files changed, 18 insertions(+), 19 deletions(-) diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index 9575a627a1e1..242b05f644eb 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -369,7 +369,7 @@ enum qeth_qdio_info_states { struct qeth_buffer_pool_entry { struct list_head list; struct list_head init_list; - void *elements[QDIO_MAX_ELEMENTS_PER_BUFFER]; + struct page *elements[QDIO_MAX_ELEMENTS_PER_BUFFER]; }; struct qeth_qdio_buffer_pool { diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index a599801d7727..8f682fc178a9 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -215,7 +215,6 @@ EXPORT_SYMBOL_GPL(qeth_clear_working_pool_list); static int qeth_alloc_buffer_pool(struct qeth_card *card) { struct qeth_buffer_pool_entry *pool_entry; - void *ptr; int i, j; QETH_CARD_TEXT(card, 5, "alocpool"); @@ -225,17 +224,18 @@ static int qeth_alloc_buffer_pool(struct qeth_card *card) qeth_free_buffer_pool(card); return -ENOMEM; } + for (j = 0; j < QETH_MAX_BUFFER_ELEMENTS(card); ++j) { - ptr = (void *) __get_free_page(GFP_KERNEL); - if (!ptr) { + struct page *page = alloc_page(GFP_KERNEL); + + if (!page) { while (j > 0) - free_page((unsigned long) - pool_entry->elements[--j]); + __free_page(pool_entry->elements[--j]); kfree(pool_entry); qeth_free_buffer_pool(card); return -ENOMEM; } - pool_entry->elements[j] = ptr; + pool_entry->elements[j] = page; } list_add(&pool_entry->init_list, &card->qdio.init_pool.entry_list); @@ -1177,7 +1177,7 @@ static void qeth_free_buffer_pool(struct qeth_card *card) list_for_each_entry_safe(pool_entry, tmp, &card->qdio.init_pool.entry_list, init_list){ for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) - free_page((unsigned long)pool_entry->elements[i]); + __free_page(pool_entry->elements[i]); list_del(&pool_entry->init_list); kfree(pool_entry); } @@ -2573,7 +2573,6 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( struct list_head *plh; struct qeth_buffer_pool_entry *entry; int i, free; - struct page *page; if (list_empty(&card->qdio.in_buf_pool.entry_list)) return NULL; @@ -2582,7 +2581,7 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( entry = list_entry(plh, struct qeth_buffer_pool_entry, list); free = 1; for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { - if (page_count(virt_to_page(entry->elements[i])) > 1) { + if (page_count(entry->elements[i]) > 1) { free = 0; break; } @@ -2597,15 +2596,15 @@ static struct qeth_buffer_pool_entry *qeth_find_free_buffer_pool_entry( entry = list_entry(card->qdio.in_buf_pool.entry_list.next, struct qeth_buffer_pool_entry, list); for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { - if (page_count(virt_to_page(entry->elements[i])) > 1) { - page = alloc_page(GFP_ATOMIC); - if (!page) { + if (page_count(entry->elements[i]) > 1) { + struct page *page = alloc_page(GFP_ATOMIC); + + if (!page) return NULL; - } else { - free_page((unsigned long)entry->elements[i]); - entry->elements[i] = page_address(page); - QETH_CARD_STAT_INC(card, rx_sg_alloc_page); - } + + __free_page(entry->elements[i]); + entry->elements[i] = page; + QETH_CARD_STAT_INC(card, rx_sg_alloc_page); } } list_del_init(&entry->list); @@ -2641,7 +2640,7 @@ static int qeth_init_input_buffer(struct qeth_card *card, for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { buf->buffer->element[i].length = PAGE_SIZE; buf->buffer->element[i].addr = - virt_to_phys(pool_entry->elements[i]); + page_to_phys(pool_entry->elements[i]); if (i == QETH_MAX_BUFFER_ELEMENTS(card) - 1) buf->buffer->element[i].eflags = SBAL_EFLAGS_LAST_ENTRY; else -- cgit From 0f75e149298bedc48db03259a38a303611d247b1 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Wed, 11 Mar 2020 18:07:10 +0100 Subject: s390/qeth: refactor buffer pool code In preparation for a subsequent fix, split out helpers to allocate/free individual pool entries. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core_main.c | 83 ++++++++++++++++++++++++--------------- 1 file changed, 51 insertions(+), 32 deletions(-) diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 8f682fc178a9..ceab3d0c4dfa 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -65,7 +65,6 @@ static struct lock_class_key qdio_out_skb_queue_key; static void qeth_issue_next_read_cb(struct qeth_card *card, struct qeth_cmd_buffer *iob, unsigned int data_length); -static void qeth_free_buffer_pool(struct qeth_card *); static int qeth_qdio_establish(struct qeth_card *); static void qeth_free_qdio_queues(struct qeth_card *card); static void qeth_notify_skbs(struct qeth_qdio_out_q *queue, @@ -212,33 +211,66 @@ void qeth_clear_working_pool_list(struct qeth_card *card) } EXPORT_SYMBOL_GPL(qeth_clear_working_pool_list); +static void qeth_free_pool_entry(struct qeth_buffer_pool_entry *entry) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(entry->elements); i++) { + if (entry->elements[i]) + __free_page(entry->elements[i]); + } + + kfree(entry); +} + +static void qeth_free_buffer_pool(struct qeth_card *card) +{ + struct qeth_buffer_pool_entry *entry, *tmp; + + list_for_each_entry_safe(entry, tmp, &card->qdio.init_pool.entry_list, + init_list) { + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + } +} + +static struct qeth_buffer_pool_entry *qeth_alloc_pool_entry(unsigned int pages) +{ + struct qeth_buffer_pool_entry *entry; + unsigned int i; + + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return NULL; + + for (i = 0; i < pages; i++) { + entry->elements[i] = alloc_page(GFP_KERNEL); + + if (!entry->elements[i]) { + qeth_free_pool_entry(entry); + return NULL; + } + } + + return entry; +} + static int qeth_alloc_buffer_pool(struct qeth_card *card) { - struct qeth_buffer_pool_entry *pool_entry; - int i, j; + unsigned int buf_elements = QETH_MAX_BUFFER_ELEMENTS(card); + unsigned int i; QETH_CARD_TEXT(card, 5, "alocpool"); for (i = 0; i < card->qdio.init_pool.buf_count; ++i) { - pool_entry = kzalloc(sizeof(*pool_entry), GFP_KERNEL); - if (!pool_entry) { + struct qeth_buffer_pool_entry *entry; + + entry = qeth_alloc_pool_entry(buf_elements); + if (!entry) { qeth_free_buffer_pool(card); return -ENOMEM; } - for (j = 0; j < QETH_MAX_BUFFER_ELEMENTS(card); ++j) { - struct page *page = alloc_page(GFP_KERNEL); - - if (!page) { - while (j > 0) - __free_page(pool_entry->elements[--j]); - kfree(pool_entry); - qeth_free_buffer_pool(card); - return -ENOMEM; - } - pool_entry->elements[j] = page; - } - list_add(&pool_entry->init_list, - &card->qdio.init_pool.entry_list); + list_add(&entry->init_list, &card->qdio.init_pool.entry_list); } return 0; } @@ -1170,19 +1202,6 @@ void qeth_drain_output_queues(struct qeth_card *card) } EXPORT_SYMBOL_GPL(qeth_drain_output_queues); -static void qeth_free_buffer_pool(struct qeth_card *card) -{ - struct qeth_buffer_pool_entry *pool_entry, *tmp; - int i = 0; - list_for_each_entry_safe(pool_entry, tmp, - &card->qdio.init_pool.entry_list, init_list){ - for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) - __free_page(pool_entry->elements[i]); - list_del(&pool_entry->init_list); - kfree(pool_entry); - } -} - static int qeth_osa_set_output_queues(struct qeth_card *card, bool single) { unsigned int count = single ? 1 : card->dev->num_tx_queues; -- cgit From 5d4f78564c9ca38146102c4f8998edcad5afbc22 Mon Sep 17 00:00:00 2001 From: Julian Wiedmann Date: Wed, 11 Mar 2020 18:07:11 +0100 Subject: s390/qeth: implement smarter resizing of the RX buffer pool The RX buffer pool is allocated in qeth_alloc_qdio_queues(). A subsequent pool resizing is then handled in a very simple way: first free the current pool, then allocate a new pool of the requested size. There's two ways where this can go wrong: 1. if the resize action happens _before_ the initial pool was allocated, then a subsequent initialization will call qeth_alloc_qdio_queues() and fill the pool with a second(!) set of pages. We consume twice the planned amount of memory. This is easy to fix - just skip the resizing if the queues haven't been allocated yet. 2. if the initial pool was created by qeth_alloc_qdio_queues() but a subsequent resizing fails, then the device has no(!) RX buffer pool. The next initialization will _not_ call qeth_alloc_qdio_queues(), and attempting to back the RX buffers with pages in qeth_init_qdio_queues() will fail. Not very difficult to fix either - instead of re-allocating the whole pool, just allocate/free as many entries to match the desired size. Fixes: 4a71df50047f ("qeth: new qeth device driver") Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core.h | 2 +- drivers/s390/net/qeth_core_main.c | 55 +++++++++++++++++++++++++++++++++------ drivers/s390/net/qeth_core_sys.c | 9 +++---- drivers/s390/net/qeth_l3_sys.c | 9 +++---- 4 files changed, 56 insertions(+), 19 deletions(-) diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index 242b05f644eb..468cada49e72 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -983,7 +983,7 @@ extern const struct attribute_group qeth_device_blkt_group; extern const struct device_type qeth_generic_devtype; const char *qeth_get_cardname_short(struct qeth_card *); -int qeth_realloc_buffer_pool(struct qeth_card *, int); +int qeth_resize_buffer_pool(struct qeth_card *card, unsigned int count); int qeth_core_load_discipline(struct qeth_card *, enum qeth_discipline_id); void qeth_core_free_discipline(struct qeth_card *); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index ceab3d0c4dfa..6d3f2f14b414 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -275,18 +275,57 @@ static int qeth_alloc_buffer_pool(struct qeth_card *card) return 0; } -int qeth_realloc_buffer_pool(struct qeth_card *card, int bufcnt) +int qeth_resize_buffer_pool(struct qeth_card *card, unsigned int count) { + unsigned int buf_elements = QETH_MAX_BUFFER_ELEMENTS(card); + struct qeth_qdio_buffer_pool *pool = &card->qdio.init_pool; + struct qeth_buffer_pool_entry *entry, *tmp; + int delta = count - pool->buf_count; + LIST_HEAD(entries); + QETH_CARD_TEXT(card, 2, "realcbp"); - /* TODO: steel/add buffers from/to a running card's buffer pool (?) */ - qeth_clear_working_pool_list(card); - qeth_free_buffer_pool(card); - card->qdio.in_buf_pool.buf_count = bufcnt; - card->qdio.init_pool.buf_count = bufcnt; - return qeth_alloc_buffer_pool(card); + /* Defer until queue is allocated: */ + if (!card->qdio.in_q) + goto out; + + /* Remove entries from the pool: */ + while (delta < 0) { + entry = list_first_entry(&pool->entry_list, + struct qeth_buffer_pool_entry, + init_list); + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + + delta++; + } + + /* Allocate additional entries: */ + while (delta > 0) { + entry = qeth_alloc_pool_entry(buf_elements); + if (!entry) { + list_for_each_entry_safe(entry, tmp, &entries, + init_list) { + list_del(&entry->init_list); + qeth_free_pool_entry(entry); + } + + return -ENOMEM; + } + + list_add(&entry->init_list, &entries); + + delta--; + } + + list_splice(&entries, &pool->entry_list); + +out: + card->qdio.in_buf_pool.buf_count = count; + pool->buf_count = count; + return 0; } -EXPORT_SYMBOL_GPL(qeth_realloc_buffer_pool); +EXPORT_SYMBOL_GPL(qeth_resize_buffer_pool); static void qeth_free_qdio_queue(struct qeth_qdio_q *q) { diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c index 2bd9993aa60b..78cae61bc924 100644 --- a/drivers/s390/net/qeth_core_sys.c +++ b/drivers/s390/net/qeth_core_sys.c @@ -247,8 +247,8 @@ static ssize_t qeth_dev_bufcnt_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct qeth_card *card = dev_get_drvdata(dev); + unsigned int cnt; char *tmp; - int cnt, old_cnt; int rc = 0; mutex_lock(&card->conf_mutex); @@ -257,13 +257,12 @@ static ssize_t qeth_dev_bufcnt_store(struct device *dev, goto out; } - old_cnt = card->qdio.in_buf_pool.buf_count; cnt = simple_strtoul(buf, &tmp, 10); cnt = (cnt < QETH_IN_BUF_COUNT_MIN) ? QETH_IN_BUF_COUNT_MIN : ((cnt > QETH_IN_BUF_COUNT_MAX) ? QETH_IN_BUF_COUNT_MAX : cnt); - if (old_cnt != cnt) { - rc = qeth_realloc_buffer_pool(card, cnt); - } + + rc = qeth_resize_buffer_pool(card, cnt); + out: mutex_unlock(&card->conf_mutex); return rc ? rc : count; diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c index 29f2517d2a31..a3d1c3bdfadb 100644 --- a/drivers/s390/net/qeth_l3_sys.c +++ b/drivers/s390/net/qeth_l3_sys.c @@ -206,12 +206,11 @@ static ssize_t qeth_l3_dev_sniffer_store(struct device *dev, qdio_get_ssqd_desc(CARD_DDEV(card), &card->ssqd); if (card->ssqd.qdioac2 & CHSC_AC2_SNIFFER_AVAILABLE) { card->options.sniffer = i; - if (card->qdio.init_pool.buf_count != - QETH_IN_BUF_COUNT_MAX) - qeth_realloc_buffer_pool(card, - QETH_IN_BUF_COUNT_MAX); - } else + qeth_resize_buffer_pool(card, QETH_IN_BUF_COUNT_MAX); + } else { rc = -EPERM; + } + break; default: rc = -EINVAL; -- cgit From 06669ea346e476a5339033d77ef175566a40efbb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Mar 2020 11:44:26 -0700 Subject: net: memcg: fix lockdep splat in inet_csk_accept() Locking newsk while still holding the listener lock triggered a lockdep splat [1] We can simply move the memcg code after we release the listener lock, as this can also help if multiple threads are sharing a common listener. Also fix a typo while reading socket sk_rmem_alloc. [1] WARNING: possible recursive locking detected 5.6.0-rc3-syzkaller #0 Not tainted -------------------------------------------- syz-executor598/9524 is trying to acquire lock: ffff88808b5b8b90 (sk_lock-AF_INET6){+.+.}, at: lock_sock include/net/sock.h:1541 [inline] ffff88808b5b8b90 (sk_lock-AF_INET6){+.+.}, at: inet_csk_accept+0x69f/0xd30 net/ipv4/inet_connection_sock.c:492 but task is already holding lock: ffff88808b5b9590 (sk_lock-AF_INET6){+.+.}, at: lock_sock include/net/sock.h:1541 [inline] ffff88808b5b9590 (sk_lock-AF_INET6){+.+.}, at: inet_csk_accept+0x8d/0xd30 net/ipv4/inet_connection_sock.c:445 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(sk_lock-AF_INET6); lock(sk_lock-AF_INET6); *** DEADLOCK *** May be due to missing lock nesting notation 1 lock held by syz-executor598/9524: #0: ffff88808b5b9590 (sk_lock-AF_INET6){+.+.}, at: lock_sock include/net/sock.h:1541 [inline] #0: ffff88808b5b9590 (sk_lock-AF_INET6){+.+.}, at: inet_csk_accept+0x8d/0xd30 net/ipv4/inet_connection_sock.c:445 stack backtrace: CPU: 0 PID: 9524 Comm: syz-executor598 Not tainted 5.6.0-rc3-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x188/0x20d lib/dump_stack.c:118 print_deadlock_bug kernel/locking/lockdep.c:2370 [inline] check_deadlock kernel/locking/lockdep.c:2411 [inline] validate_chain kernel/locking/lockdep.c:2954 [inline] __lock_acquire.cold+0x114/0x288 kernel/locking/lockdep.c:3954 lock_acquire+0x197/0x420 kernel/locking/lockdep.c:4484 lock_sock_nested+0xc5/0x110 net/core/sock.c:2947 lock_sock include/net/sock.h:1541 [inline] inet_csk_accept+0x69f/0xd30 net/ipv4/inet_connection_sock.c:492 inet_accept+0xe9/0x7c0 net/ipv4/af_inet.c:734 __sys_accept4_file+0x3ac/0x5b0 net/socket.c:1758 __sys_accept4+0x53/0x90 net/socket.c:1809 __do_sys_accept4 net/socket.c:1821 [inline] __se_sys_accept4 net/socket.c:1818 [inline] __x64_sys_accept4+0x93/0xf0 net/socket.c:1818 do_syscall_64+0xf6/0x790 arch/x86/entry/common.c:294 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x4445c9 Code: e8 0c 0d 03 00 48 83 c4 18 c3 0f 1f 80 00 00 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 eb 08 fc ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007ffc35b37608 EFLAGS: 00000246 ORIG_RAX: 0000000000000120 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004445c9 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003 RBP: 0000000000000000 R08: 0000000000306777 R09: 0000000000306777 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00000000004053d0 R14: 0000000000000000 R15: 0000000000000000 Fixes: d752a4986532 ("net: memcg: late association of sock to memcg") Signed-off-by: Eric Dumazet Cc: Shakeel Butt Reported-by: syzbot Signed-off-by: David S. Miller --- net/ipv4/inet_connection_sock.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 65a3b2565102..d545fb99a8a1 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -483,27 +483,27 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) spin_unlock_bh(&queue->fastopenq.lock); } - if (mem_cgroup_sockets_enabled) { +out: + release_sock(sk); + if (newsk && mem_cgroup_sockets_enabled) { int amt; /* atomically get the memory usage, set and charge the - * sk->sk_memcg. + * newsk->sk_memcg. */ lock_sock(newsk); - /* The sk has not been accepted yet, no need to look at - * sk->sk_wmem_queued. + /* The socket has not been accepted yet, no need to look at + * newsk->sk_wmem_queued. */ amt = sk_mem_pages(newsk->sk_forward_alloc + - atomic_read(&sk->sk_rmem_alloc)); + atomic_read(&newsk->sk_rmem_alloc)); mem_cgroup_sk_alloc(newsk); if (newsk->sk_memcg && amt) mem_cgroup_charge_skmem(newsk->sk_memcg, amt); release_sock(newsk); } -out: - release_sock(sk); if (req) reqsk_put(req); return newsk; -- cgit From 012fc74517b25177dfede2ed45cd108258564e4a Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 11 Mar 2020 21:02:31 +0100 Subject: net: dsa: mv88e6xxx: Add missing mask of ATU occupancy register Only the bottom 12 bits contain the ATU bin occupancy statistics. The upper bits need masking off. Fixes: e0c69ca7dfbb ("net: dsa: mv88e6xxx: Add ATU occupancy via devlink resources") Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx/chip.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 8c9289549688..2f993e673ec7 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -2769,6 +2769,8 @@ static u64 mv88e6xxx_devlink_atu_bin_get(struct mv88e6xxx_chip *chip, goto unlock; } + occupancy &= MV88E6XXX_G2_ATU_STATS_MASK; + unlock: mv88e6xxx_reg_unlock(chip); -- cgit From e1f550dc44a4d535da4e25ada1b0eaf8f3417929 Mon Sep 17 00:00:00 2001 From: Chris Packham Date: Thu, 12 Mar 2020 09:05:46 +1300 Subject: net: mvmdio: avoid error message for optional IRQ Per the dt-binding the interrupt is optional so use platform_get_irq_optional() instead of platform_get_irq(). Since commit 7723f4c5ecdb ("driver core: platform: Add an error message to platform_get_irq*()") platform_get_irq() produces an error message orion-mdio f1072004.mdio: IRQ index 0 not found which is perfectly normal if one hasn't specified the optional property in the device tree. Signed-off-by: Chris Packham Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvmdio.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/marvell/mvmdio.c b/drivers/net/ethernet/marvell/mvmdio.c index 0b9e851f3da4..d2e2dc538428 100644 --- a/drivers/net/ethernet/marvell/mvmdio.c +++ b/drivers/net/ethernet/marvell/mvmdio.c @@ -347,7 +347,7 @@ static int orion_mdio_probe(struct platform_device *pdev) } - dev->err_interrupt = platform_get_irq(pdev, 0); + dev->err_interrupt = platform_get_irq_optional(pdev, 0); if (dev->err_interrupt > 0 && resource_size(r) < MVMDIO_ERR_INT_MASK + 4) { dev_err(&pdev->dev, @@ -364,8 +364,8 @@ static int orion_mdio_probe(struct platform_device *pdev) writel(MVMDIO_ERR_INT_SMI_DONE, dev->regs + MVMDIO_ERR_INT_MASK); - } else if (dev->err_interrupt == -EPROBE_DEFER) { - ret = -EPROBE_DEFER; + } else if (dev->err_interrupt < 0) { + ret = dev->err_interrupt; goto out_mdio; } -- cgit From b09fe70ef520e011ba4a64f4b93f948a8f14717b Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Mon, 9 Mar 2020 10:39:53 -0700 Subject: taprio: Fix sending packets without dequeueing them There was a bug that was causing packets to be sent to the driver without first calling dequeue() on the "child" qdisc. And the KASAN report below shows that sending a packet without calling dequeue() leads to bad results. The problem is that when checking the last qdisc "child" we do not set the returned skb to NULL, which can cause it to be sent to the driver, and so after the skb is sent, it may be freed, and in some situations a reference to it may still be in the child qdisc, because it was never dequeued. The crash log looks like this: [ 19.937538] ================================================================== [ 19.938300] BUG: KASAN: use-after-free in taprio_dequeue_soft+0x620/0x780 [ 19.938968] Read of size 4 at addr ffff8881128628cc by task swapper/1/0 [ 19.939612] [ 19.939772] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.6.0-rc3+ #97 [ 19.940397] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qe4 [ 19.941523] Call Trace: [ 19.941774] [ 19.941985] dump_stack+0x97/0xe0 [ 19.942323] print_address_description.constprop.0+0x3b/0x60 [ 19.942884] ? taprio_dequeue_soft+0x620/0x780 [ 19.943325] ? taprio_dequeue_soft+0x620/0x780 [ 19.943767] __kasan_report.cold+0x1a/0x32 [ 19.944173] ? taprio_dequeue_soft+0x620/0x780 [ 19.944612] kasan_report+0xe/0x20 [ 19.944954] taprio_dequeue_soft+0x620/0x780 [ 19.945380] __qdisc_run+0x164/0x18d0 [ 19.945749] net_tx_action+0x2c4/0x730 [ 19.946124] __do_softirq+0x268/0x7bc [ 19.946491] irq_exit+0x17d/0x1b0 [ 19.946824] smp_apic_timer_interrupt+0xeb/0x380 [ 19.947280] apic_timer_interrupt+0xf/0x20 [ 19.947687] [ 19.947912] RIP: 0010:default_idle+0x2d/0x2d0 [ 19.948345] Code: 00 00 41 56 41 55 65 44 8b 2d 3f 8d 7c 7c 41 54 55 53 0f 1f 44 00 00 e8 b1 b2 c5 fd e9 07 00 3 [ 19.950166] RSP: 0018:ffff88811a3efda0 EFLAGS: 00000282 ORIG_RAX: ffffffffffffff13 [ 19.950909] RAX: 0000000080000000 RBX: ffff88811a3a9600 RCX: ffffffff8385327e [ 19.951608] RDX: 1ffff110234752c0 RSI: 0000000000000000 RDI: ffffffff8385262f [ 19.952309] RBP: ffffed10234752c0 R08: 0000000000000001 R09: ffffed10234752c1 [ 19.953009] R10: ffffed10234752c0 R11: ffff88811a3a9607 R12: 0000000000000001 [ 19.953709] R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000000 [ 19.954408] ? default_idle_call+0x2e/0x70 [ 19.954816] ? default_idle+0x1f/0x2d0 [ 19.955192] default_idle_call+0x5e/0x70 [ 19.955584] do_idle+0x3d4/0x500 [ 19.955909] ? arch_cpu_idle_exit+0x40/0x40 [ 19.956325] ? _raw_spin_unlock_irqrestore+0x23/0x30 [ 19.956829] ? trace_hardirqs_on+0x30/0x160 [ 19.957242] cpu_startup_entry+0x19/0x20 [ 19.957633] start_secondary+0x2a6/0x380 [ 19.958026] ? set_cpu_sibling_map+0x18b0/0x18b0 [ 19.958486] secondary_startup_64+0xa4/0xb0 [ 19.958921] [ 19.959078] Allocated by task 33: [ 19.959412] save_stack+0x1b/0x80 [ 19.959747] __kasan_kmalloc.constprop.0+0xc2/0xd0 [ 19.960222] kmem_cache_alloc+0xe4/0x230 [ 19.960617] __alloc_skb+0x91/0x510 [ 19.960967] ndisc_alloc_skb+0x133/0x330 [ 19.961358] ndisc_send_ns+0x134/0x810 [ 19.961735] addrconf_dad_work+0xad5/0xf80 [ 19.962144] process_one_work+0x78e/0x13a0 [ 19.962551] worker_thread+0x8f/0xfa0 [ 19.962919] kthread+0x2ba/0x3b0 [ 19.963242] ret_from_fork+0x3a/0x50 [ 19.963596] [ 19.963753] Freed by task 33: [ 19.964055] save_stack+0x1b/0x80 [ 19.964386] __kasan_slab_free+0x12f/0x180 [ 19.964830] kmem_cache_free+0x80/0x290 [ 19.965231] ip6_mc_input+0x38a/0x4d0 [ 19.965617] ipv6_rcv+0x1a4/0x1d0 [ 19.965948] __netif_receive_skb_one_core+0xf2/0x180 [ 19.966437] netif_receive_skb+0x8c/0x3c0 [ 19.966846] br_handle_frame_finish+0x779/0x1310 [ 19.967302] br_handle_frame+0x42a/0x830 [ 19.967694] __netif_receive_skb_core+0xf0e/0x2a90 [ 19.968167] __netif_receive_skb_one_core+0x96/0x180 [ 19.968658] process_backlog+0x198/0x650 [ 19.969047] net_rx_action+0x2fa/0xaa0 [ 19.969420] __do_softirq+0x268/0x7bc [ 19.969785] [ 19.969940] The buggy address belongs to the object at ffff888112862840 [ 19.969940] which belongs to the cache skbuff_head_cache of size 224 [ 19.971202] The buggy address is located 140 bytes inside of [ 19.971202] 224-byte region [ffff888112862840, ffff888112862920) [ 19.972344] The buggy address belongs to the page: [ 19.972820] page:ffffea00044a1800 refcount:1 mapcount:0 mapping:ffff88811a2bd1c0 index:0xffff8881128625c0 compo0 [ 19.973930] flags: 0x8000000000010200(slab|head) [ 19.974388] raw: 8000000000010200 ffff88811a2ed650 ffff88811a2ed650 ffff88811a2bd1c0 [ 19.975151] raw: ffff8881128625c0 0000000000190013 00000001ffffffff 0000000000000000 [ 19.975915] page dumped because: kasan: bad access detected [ 19.976461] page_owner tracks the page as allocated [ 19.976946] page last allocated via order 2, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NO) [ 19.978332] prep_new_page+0x24b/0x330 [ 19.978707] get_page_from_freelist+0x2057/0x2c90 [ 19.979170] __alloc_pages_nodemask+0x218/0x590 [ 19.979619] new_slab+0x9d/0x300 [ 19.979948] ___slab_alloc.constprop.0+0x2f9/0x6f0 [ 19.980421] __slab_alloc.constprop.0+0x30/0x60 [ 19.980870] kmem_cache_alloc+0x201/0x230 [ 19.981269] __alloc_skb+0x91/0x510 [ 19.981620] alloc_skb_with_frags+0x78/0x4a0 [ 19.982043] sock_alloc_send_pskb+0x5eb/0x750 [ 19.982476] unix_stream_sendmsg+0x399/0x7f0 [ 19.982904] sock_sendmsg+0xe2/0x110 [ 19.983262] ____sys_sendmsg+0x4de/0x6d0 [ 19.983660] ___sys_sendmsg+0xe4/0x160 [ 19.984032] __sys_sendmsg+0xab/0x130 [ 19.984396] do_syscall_64+0xe7/0xae0 [ 19.984761] page last free stack trace: [ 19.985142] __free_pages_ok+0x432/0xbc0 [ 19.985533] qlist_free_all+0x56/0xc0 [ 19.985907] quarantine_reduce+0x149/0x170 [ 19.986315] __kasan_kmalloc.constprop.0+0x9e/0xd0 [ 19.986791] kmem_cache_alloc+0xe4/0x230 [ 19.987182] prepare_creds+0x24/0x440 [ 19.987548] do_faccessat+0x80/0x590 [ 19.987906] do_syscall_64+0xe7/0xae0 [ 19.988276] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 19.988775] [ 19.988930] Memory state around the buggy address: [ 19.989402] ffff888112862780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 19.990111] ffff888112862800: fc fc fc fc fc fc fc fc fb fb fb fb fb fb fb fb [ 19.990822] >ffff888112862880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 19.991529] ^ [ 19.992081] ffff888112862900: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc [ 19.992796] ffff888112862980: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc Fixes: 5a781ccbd19e ("tc: Add support for configuring the taprio scheduler") Reported-by: Michael Schmidt Signed-off-by: Vinicius Costa Gomes Acked-by: Andre Guedes Signed-off-by: David S. Miller --- net/sched/sch_taprio.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index ee717e05372b..b1eb12d33b9a 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -564,8 +564,10 @@ static struct sk_buff *taprio_dequeue_soft(struct Qdisc *sch) prio = skb->priority; tc = netdev_get_prio_tc_map(dev, prio); - if (!(gate_mask & BIT(tc))) + if (!(gate_mask & BIT(tc))) { + skb = NULL; continue; + } len = qdisc_pkt_len(skb); guard = ktime_add_ns(taprio_get_time(q), @@ -575,13 +577,17 @@ static struct sk_buff *taprio_dequeue_soft(struct Qdisc *sch) * guard band ... */ if (gate_mask != TAPRIO_ALL_GATES_OPEN && - ktime_after(guard, entry->close_time)) + ktime_after(guard, entry->close_time)) { + skb = NULL; continue; + } /* ... and no budget. */ if (gate_mask != TAPRIO_ALL_GATES_OPEN && - atomic_sub_return(len, &entry->budget) < 0) + atomic_sub_return(len, &entry->budget) < 0) { + skb = NULL; continue; + } skb = child->ops->dequeue(child); if (unlikely(!skb)) -- cgit From 5eb01ddfcfb25e6ebc404a41deae946bde776731 Mon Sep 17 00:00:00 2001 From: Yonglong Liu Date: Thu, 12 Mar 2020 15:11:03 +0800 Subject: net: hns3: fix "tc qdisc del" failed issue The HNS3 driver supports to configure TC numbers and TC to priority map via "tc" tool. But when delete the rule, will fail, because the HNS3 driver needs at least one TC, but the "tc" tool sets TC number to zero when delete. This patch makes sure that the TC number is at least one. Fixes: 30d240dfa2e8 ("net: hns3: Add mqprio hardware offload support in hns3 driver") Signed-off-by: Yonglong Liu Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index acb796cc10d0..a7f40aa1a0ea 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -1711,7 +1711,7 @@ static int hns3_setup_tc(struct net_device *netdev, void *type_data) netif_dbg(h, drv, netdev, "setup tc: num_tc=%u\n", tc); return (kinfo->dcb_ops && kinfo->dcb_ops->setup_tc) ? - kinfo->dcb_ops->setup_tc(h, tc, prio_tc) : -EOPNOTSUPP; + kinfo->dcb_ops->setup_tc(h, tc ? tc : 1, prio_tc) : -EOPNOTSUPP; } static int hns3_nic_setup_tc(struct net_device *dev, enum tc_setup_type type, -- cgit From 23b4201dead410c4a5108a8e723240419ab75257 Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 12 Mar 2020 15:11:04 +0800 Subject: net: hns3: fix VF VLAN table entries inconsistent issue Currently, if VF is loaded on the host side, the host doesn't clear the VF's VLAN table entries when VF removing. In this case, when doing reset and disabling sriov at the same time the VLAN device over VF will be removed, but the VLAN table entries in hardware are remained. This patch fixes it by asking PF to clear the VLAN table entries for VF when VF is removing. It also clears the VLAN table full bit after VF VLAN table entries being cleared. Fixes: c6075b193462 ("net: hns3: Record VF vlan tables") Signed-off-by: Jian Shen Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h | 1 + drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 1 + drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c | 1 + drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 3 +++ 4 files changed, 6 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h index 1b0313900f98..d87158acdf6f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h +++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h @@ -46,6 +46,7 @@ enum HCLGE_MBX_OPCODE { HCLGE_MBX_PUSH_VLAN_INFO, /* (PF -> VF) push port base vlan */ HCLGE_MBX_GET_MEDIA_TYPE, /* (VF -> PF) get media type */ HCLGE_MBX_PUSH_PROMISC_INFO, /* (PF -> VF) push vf promisc info */ + HCLGE_MBX_VF_UNINIT, /* (VF -> PF) vf is unintializing */ HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf flr status */ HCLGE_MBX_PUSH_LINK_STATUS, /* (M7 -> PF) get port link status */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index acf0c29fcbcd..6deeb964ea6b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -8272,6 +8272,7 @@ void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list) kfree(vlan); } } + clear_bit(vport->vport_id, hdev->vf_vlan_full); } void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index a3c0822191a9..3d850f6b1e37 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -799,6 +799,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) hclge_get_link_mode(vport, req); break; case HCLGE_MBX_GET_VF_FLR_STATUS: + case HCLGE_MBX_VF_UNINIT: hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_UC); hclge_rm_vport_all_mac_table(vport, true, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index d6597206e692..0510d85a7f6a 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -2803,6 +2803,9 @@ static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev) { hclgevf_state_uninit(hdev); + hclgevf_send_mbx_msg(hdev, HCLGE_MBX_VF_UNINIT, 0, NULL, 0, + false, NULL, 0); + if (test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) { hclgevf_misc_irq_uninit(hdev); hclgevf_uninit_msi(hdev); -- cgit From 903b85d3adce99a5301d5959c4d3c9d14a7974d4 Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 12 Mar 2020 15:11:05 +0800 Subject: net: hns3: fix RMW issue for VLAN filter switch According to the user manual, the ingress and egress VLAN filter are configured at the same time. Currently, hclge_init_vlan_config() and hclge_set_vlan_spoofchk() will both change the VLAN filter switch. So it's necessary to read the old configuration before modifying it. Fixes: 22044f95faa0 ("net: hns3: add support for spoof check setting") Signed-off-by: Jian Shen Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- .../net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 6deeb964ea6b..06d0ed0fe4c9 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -7745,16 +7745,27 @@ static int hclge_set_vlan_filter_ctrl(struct hclge_dev *hdev, u8 vlan_type, struct hclge_desc desc; int ret; - hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_CTRL, false); - + /* read current vlan filter parameter */ + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_VLAN_FILTER_CTRL, true); req = (struct hclge_vlan_filter_ctrl_cmd *)desc.data; req->vlan_type = vlan_type; - req->vlan_fe = filter_en ? fe_type : 0; req->vf_id = vf_id; + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to get vlan filter config, ret = %d.\n", ret); + return ret; + } + + /* modify and write new config parameter */ + hclge_cmd_reuse_desc(&desc, false); + req->vlan_fe = filter_en ? + (req->vlan_fe | fe_type) : (req->vlan_fe & ~fe_type); + ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) - dev_err(&hdev->pdev->dev, "set vlan filter fail, ret =%d.\n", + dev_err(&hdev->pdev->dev, "failed to set vlan filter, ret = %d.\n", ret); return ret; -- cgit From 59359fc8a2f7af062777692e6a7aae73483729ec Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 12 Mar 2020 15:11:06 +0800 Subject: net: hns3: clear port base VLAN when unload PF Currently, PF missed to clear the port base VLAN for VF when unload. In this case, the VLAN id will remain in the VLAN table. This patch fixes it. Fixes: 92f11ea177cd ("net: hns3: fix set port based VLAN issue for VF") Signed-off-by: Jian Shen Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 23 ++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 06d0ed0fe4c9..d3b0cd74ecd2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -8500,6 +8500,28 @@ static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, } } +static void hclge_clear_vf_vlan(struct hclge_dev *hdev) +{ + struct hclge_vlan_info *vlan_info; + struct hclge_vport *vport; + int ret; + int vf; + + /* clear port base vlan for all vf */ + for (vf = HCLGE_VF_VPORT_START_NUM; vf < hdev->num_alloc_vport; vf++) { + vport = &hdev->vport[vf]; + vlan_info = &vport->port_base_vlan_cfg.vlan_info; + + ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q), + vport->vport_id, + vlan_info->vlan_tag, true); + if (ret) + dev_err(&hdev->pdev->dev, + "failed to clear vf vlan for vf%d, ret = %d\n", + vf - HCLGE_VF_VPORT_START_NUM, ret); + } +} + int hclge_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, u16 vlan_id, bool is_kill) { @@ -9909,6 +9931,7 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev) struct hclge_mac *mac = &hdev->hw.mac; hclge_reset_vf_rate(hdev); + hclge_clear_vf_vlan(hdev); hclge_misc_affinity_teardown(hdev); hclge_state_uninit(hdev); -- cgit From 611d779af7cad2b87487ff58e4931a90c20b113c Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 12 Mar 2020 22:25:20 +0100 Subject: net: phy: fix MDIO bus PM PHY resuming So far we have the unfortunate situation that mdio_bus_phy_may_suspend() is called in suspend AND resume path, assuming that function result is the same. After the original change this is no longer the case, resulting in broken resume as reported by Geert. To fix this call mdio_bus_phy_may_suspend() in the suspend path only, and let the phy_device store the info whether it was suspended by MDIO bus PM. Fixes: 503ba7c69610 ("net: phy: Avoid multiple suspends") Reported-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven Signed-off-by: Heiner Kallweit Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 6 +++++- include/linux/phy.h | 2 ++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index c8b0c34030d3..28e3c5c0e3c3 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -286,6 +286,8 @@ static int mdio_bus_phy_suspend(struct device *dev) if (!mdio_bus_phy_may_suspend(phydev)) return 0; + phydev->suspended_by_mdio_bus = 1; + return phy_suspend(phydev); } @@ -294,9 +296,11 @@ static int mdio_bus_phy_resume(struct device *dev) struct phy_device *phydev = to_phy_device(dev); int ret; - if (!mdio_bus_phy_may_suspend(phydev)) + if (!phydev->suspended_by_mdio_bus) goto no_resume; + phydev->suspended_by_mdio_bus = 0; + ret = phy_resume(phydev); if (ret < 0) return ret; diff --git a/include/linux/phy.h b/include/linux/phy.h index 22f5e763e894..452e8ba8665f 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -357,6 +357,7 @@ struct macsec_ops; * is_gigabit_capable: Set to true if PHY supports 1000Mbps * has_fixups: Set to true if this phy has fixups/quirks. * suspended: Set to true if this phy has been suspended successfully. + * suspended_by_mdio_bus: Set to true if this phy was suspended by MDIO bus. * sysfs_links: Internal boolean tracking sysfs symbolic links setup/removal. * loopback_enabled: Set true if this phy has been loopbacked successfully. * state: state of the PHY for management purposes @@ -396,6 +397,7 @@ struct phy_device { unsigned is_gigabit_capable:1; unsigned has_fixups:1; unsigned suspended:1; + unsigned suspended_by_mdio_bus:1; unsigned sysfs_links:1; unsigned loopback_enabled:1; -- cgit From 9d0e0cd9a529ba68b2acda68b5075d4a5ea7118a Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Thu, 12 Mar 2020 17:51:45 +0100 Subject: tc-testing: add ETS scheduler to tdc build configuration add CONFIG_NET_SCH_ETS to 'config', otherwise test suites using this file to perform a full tdc run will encounter the following warning: ok 645 e90e - Add ETS qdisc using bands # skipped - "-----> teardown stage" did not complete successfully Fixes: 82c664b69c8b ("selftests: qdiscs: Add test coverage for ETS Qdisc") Reported-by: Jamal Hadi Salim Signed-off-by: Davide Caratti Signed-off-by: David S. Miller --- tools/testing/selftests/tc-testing/config | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/tc-testing/config b/tools/testing/selftests/tc-testing/config index 477bc61b374a..c03af4600281 100644 --- a/tools/testing/selftests/tc-testing/config +++ b/tools/testing/selftests/tc-testing/config @@ -57,3 +57,4 @@ CONFIG_NET_IFE_SKBMARK=m CONFIG_NET_IFE_SKBPRIO=m CONFIG_NET_IFE_SKBTCINDEX=m CONFIG_NET_SCH_FIFO=y +CONFIG_NET_SCH_ETS=m -- cgit From c0368595c1639947839c0db8294ee96aca0b3b86 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 12 Mar 2020 15:04:30 +0000 Subject: net: systemport: fix index check to avoid an array out of bounds access Currently the bounds check on index is off by one and can lead to an out of bounds access on array priv->filters_loc when index is RXCHK_BRCM_TAG_MAX. Fixes: bb9051a2b230 ("net: systemport: Add support for WAKE_FILTER") Signed-off-by: Colin Ian King Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index e0611cba87f9..15b31cddc054 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -2135,7 +2135,7 @@ static int bcm_sysport_rule_set(struct bcm_sysport_priv *priv, return -ENOSPC; index = find_first_zero_bit(priv->filters, RXCHK_BRCM_TAG_MAX); - if (index > RXCHK_BRCM_TAG_MAX) + if (index >= RXCHK_BRCM_TAG_MAX) return -ENOSPC; /* Location is the classification ID, and index is the position -- cgit