From f78c1375339a291cba492a70eaf12ec501d28a8e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 14 Feb 2024 20:08:35 +0100 Subject: [PATCH 01/73] wifi: nl80211: reject iftype change with mesh ID change It's currently possible to change the mesh ID when the interface isn't yet in mesh mode, at the same time as changing it into mesh mode. This leads to an overwrite of data in the wdev->u union for the interface type it currently has, causing cfg80211_change_iface() to do wrong things when switching. We could probably allow setting an interface to mesh while setting the mesh ID at the same time by doing a different order of operations here, but realistically there's no userspace that's going to do this, so just disallow changes in iftype when setting mesh ID. Cc: stable@vger.kernel.org Fixes: 29cbe68c516a ("cfg80211/mac80211: add mesh join/leave commands") Reported-by: syzbot+dd4779978217b1973180@syzkaller.appspotmail.com 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 b09700400d09..bd54a928bab4 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -4197,6 +4197,8 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) if (ntype != NL80211_IFTYPE_MESH_POINT) return -EINVAL; + if (otype != NL80211_IFTYPE_MESH_POINT) + return -EINVAL; if (netif_running(dev)) return -EBUSY; From e8335ef57c6816d81b24173ba88cc9b3f043687f Mon Sep 17 00:00:00 2001 From: Yochai Hagvi Date: Thu, 25 Jan 2024 15:40:55 +0200 Subject: [PATCH 02/73] ice: fix connection state of DPLL and out pin Fix the connection state between source DPLL and output pin, updating the attribute 'state' of 'parent_device'. Previously, the connection state was broken, and didn't reflect the correct state. When 'state_on_dpll_set' is called with the value 'DPLL_PIN_STATE_CONNECTED' (1), the output pin will switch to the given DPLL, and the state of the given DPLL will be set to connected. E.g.: --do pin-set --json '{"id":2, "parent-device":{"parent-id":1, "state": 1 }}' This command will connect DPLL device with id 1 to output pin with id 2. When 'state_on_dpll_set' is called with the value 'DPLL_PIN_STATE_DISCONNECTED' (2) and the given DPLL is currently connected, then the output pin will be disabled. E.g: --do pin-set --json '{"id":2, "parent-device":{"parent-id":1, "state": 2 }}' This command will disable output pin with id 2 if DPLL device with ID 1 is connected to it; otherwise, the command is ignored. Fixes: d7999f5ea64b ("ice: implement dpll interface to control cgu") Reviewed-by: Wojciech Drewek Reviewed-by: Arkadiusz Kubalewski Signed-off-by: Yochai Hagvi Tested-by: Sunitha Mekala (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dpll.c | 43 +++++++++++++++++------ 1 file changed, 32 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index b9c5eced6326..9c0d739be1e9 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -254,6 +254,7 @@ ice_dpll_output_frequency_get(const struct dpll_pin *pin, void *pin_priv, * ice_dpll_pin_enable - enable a pin on dplls * @hw: board private hw structure * @pin: pointer to a pin + * @dpll_idx: dpll index to connect to output pin * @pin_type: type of pin being enabled * @extack: error reporting * @@ -266,7 +267,7 @@ ice_dpll_output_frequency_get(const struct dpll_pin *pin, void *pin_priv, */ static int ice_dpll_pin_enable(struct ice_hw *hw, struct ice_dpll_pin *pin, - enum ice_dpll_pin_type pin_type, + u8 dpll_idx, enum ice_dpll_pin_type pin_type, struct netlink_ext_ack *extack) { u8 flags = 0; @@ -280,10 +281,12 @@ ice_dpll_pin_enable(struct ice_hw *hw, struct ice_dpll_pin *pin, ret = ice_aq_set_input_pin_cfg(hw, pin->idx, 0, flags, 0, 0); break; case ICE_DPLL_PIN_TYPE_OUTPUT: + flags = ICE_AQC_SET_CGU_OUT_CFG_UPDATE_SRC_SEL; if (pin->flags[0] & ICE_AQC_GET_CGU_OUT_CFG_ESYNC_EN) flags |= ICE_AQC_SET_CGU_OUT_CFG_ESYNC_EN; flags |= ICE_AQC_SET_CGU_OUT_CFG_OUT_EN; - ret = ice_aq_set_output_pin_cfg(hw, pin->idx, flags, 0, 0, 0); + ret = ice_aq_set_output_pin_cfg(hw, pin->idx, flags, dpll_idx, + 0, 0); break; default: return -EINVAL; @@ -398,14 +401,27 @@ ice_dpll_pin_state_update(struct ice_pf *pf, struct ice_dpll_pin *pin, break; case ICE_DPLL_PIN_TYPE_OUTPUT: ret = ice_aq_get_output_pin_cfg(&pf->hw, pin->idx, - &pin->flags[0], NULL, + &pin->flags[0], &parent, &pin->freq, NULL); if (ret) goto err; - if (ICE_AQC_SET_CGU_OUT_CFG_OUT_EN & pin->flags[0]) - pin->state[0] = DPLL_PIN_STATE_CONNECTED; - else - pin->state[0] = DPLL_PIN_STATE_DISCONNECTED; + + parent &= ICE_AQC_GET_CGU_OUT_CFG_DPLL_SRC_SEL; + if (ICE_AQC_SET_CGU_OUT_CFG_OUT_EN & pin->flags[0]) { + pin->state[pf->dplls.eec.dpll_idx] = + parent == pf->dplls.eec.dpll_idx ? + DPLL_PIN_STATE_CONNECTED : + DPLL_PIN_STATE_DISCONNECTED; + pin->state[pf->dplls.pps.dpll_idx] = + parent == pf->dplls.pps.dpll_idx ? + DPLL_PIN_STATE_CONNECTED : + DPLL_PIN_STATE_DISCONNECTED; + } else { + pin->state[pf->dplls.eec.dpll_idx] = + DPLL_PIN_STATE_DISCONNECTED; + pin->state[pf->dplls.pps.dpll_idx] = + DPLL_PIN_STATE_DISCONNECTED; + } break; case ICE_DPLL_PIN_TYPE_RCLK_INPUT: for (parent = 0; parent < pf->dplls.rclk.num_parents; @@ -570,7 +586,8 @@ ice_dpll_pin_state_set(const struct dpll_pin *pin, void *pin_priv, mutex_lock(&pf->dplls.lock); if (enable) - ret = ice_dpll_pin_enable(&pf->hw, p, pin_type, extack); + ret = ice_dpll_pin_enable(&pf->hw, p, d->dpll_idx, pin_type, + extack); else ret = ice_dpll_pin_disable(&pf->hw, p, pin_type, extack); if (!ret) @@ -603,6 +620,11 @@ ice_dpll_output_state_set(const struct dpll_pin *pin, void *pin_priv, struct netlink_ext_ack *extack) { bool enable = state == DPLL_PIN_STATE_CONNECTED; + struct ice_dpll_pin *p = pin_priv; + struct ice_dpll *d = dpll_priv; + + if (!enable && p->state[d->dpll_idx] == DPLL_PIN_STATE_DISCONNECTED) + return 0; return ice_dpll_pin_state_set(pin, pin_priv, dpll, dpll_priv, enable, extack, ICE_DPLL_PIN_TYPE_OUTPUT); @@ -669,10 +691,9 @@ ice_dpll_pin_state_get(const struct dpll_pin *pin, void *pin_priv, ret = ice_dpll_pin_state_update(pf, p, pin_type, extack); if (ret) goto unlock; - if (pin_type == ICE_DPLL_PIN_TYPE_INPUT) + if (pin_type == ICE_DPLL_PIN_TYPE_INPUT || + pin_type == ICE_DPLL_PIN_TYPE_OUTPUT) *state = p->state[d->dpll_idx]; - else if (pin_type == ICE_DPLL_PIN_TYPE_OUTPUT) - *state = p->state[0]; ret = 0; unlock: mutex_unlock(&pf->dplls.lock); From 3b14430c65b4f510b2a310ca4f18ed6ca7184b00 Mon Sep 17 00:00:00 2001 From: Arkadiusz Kubalewski Date: Thu, 8 Feb 2024 23:56:31 +0100 Subject: [PATCH 03/73] ice: fix dpll input pin phase_adjust value updates The value of phase_adjust for input pin shall be updated in ice_dpll_pin_state_update(..). Fix by adding proper argument to the firmware query function call - a pin's struct field pointer where the phase_adjust value during driver runtime is stored. Previously the phase_adjust used to misinform user about actual phase_adjust value. I.e., if phase_adjust was set to a non zero value and if driver was reloaded, the user would see the value equal 0, which is not correct - the actual value is equal to value set before driver reload. Fixes: 90e1c90750d7 ("ice: dpll: implement phase related callbacks") Reviewed-by: Alan Brady Signed-off-by: Arkadiusz Kubalewski Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dpll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index 9c0d739be1e9..2beaeb9c336d 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -373,7 +373,7 @@ ice_dpll_pin_state_update(struct ice_pf *pf, struct ice_dpll_pin *pin, case ICE_DPLL_PIN_TYPE_INPUT: ret = ice_aq_get_input_pin_cfg(&pf->hw, pin->idx, NULL, NULL, NULL, &pin->flags[0], - &pin->freq, NULL); + &pin->freq, &pin->phase_adjust); if (ret) goto err; if (ICE_AQC_GET_CGU_IN_CFG_FLG2_INPUT_EN & pin->flags[0]) { From fc7fd1a10a9d2d38378b42e9a508da4c68018453 Mon Sep 17 00:00:00 2001 From: Arkadiusz Kubalewski Date: Fri, 9 Feb 2024 22:24:30 +0100 Subject: [PATCH 04/73] ice: fix dpll and dpll_pin data access on PF reset Do not allow to acquire data or alter configuration of dpll and pins through firmware if PF reset is in progress, this would cause confusing netlink extack errors as the firmware cannot respond or process the request properly during the reset time. Return (-EBUSY) and extack error for the user who tries access/modify the config of dpll/pin through firmware during the reset time. The PF reset and kernel access to dpll data are both asynchronous. It is not possible to guard all the possible reset paths with any determinictic approach. I.e., it is possible that reset starts after reset check is performed (or if the reset would be checked after mutex is locked), but at the same time it is not possible to wait for dpll mutex unlock in the reset flow. This is best effort solution to at least give a clue to the user what is happening in most of the cases, knowing that there are possible race conditions where the user could see a different error received from firmware due to reset unexpectedly starting. Test by looping execution of below steps until netlink error appears: - perform PF reset $ echo 1 > /sys/class/net//device/reset - i.e. try to alter/read dpll/pin config: $ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/dpll.yaml \ --dump pin-get Fixes: d7999f5ea64b ("ice: implement dpll interface to control cgu") Reviewed-by: Aleksandr Loktionov Reviewed-by: Przemek Kitszel Signed-off-by: Arkadiusz Kubalewski Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dpll.c | 38 +++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index 2beaeb9c336d..343b2a668959 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -30,6 +30,26 @@ static const char * const pin_type_name[] = { [ICE_DPLL_PIN_TYPE_RCLK_INPUT] = "rclk-input", }; +/** + * ice_dpll_is_reset - check if reset is in progress + * @pf: private board structure + * @extack: error reporting + * + * If reset is in progress, fill extack with error. + * + * Return: + * * false - no reset in progress + * * true - reset in progress + */ +static bool ice_dpll_is_reset(struct ice_pf *pf, struct netlink_ext_ack *extack) +{ + if (ice_is_reset_in_progress(pf->state)) { + NL_SET_ERR_MSG(extack, "PF reset in progress"); + return true; + } + return false; +} + /** * ice_dpll_pin_freq_set - set pin's frequency * @pf: private board structure @@ -109,6 +129,9 @@ ice_dpll_frequency_set(const struct dpll_pin *pin, void *pin_priv, struct ice_pf *pf = d->pf; int ret; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); ret = ice_dpll_pin_freq_set(pf, p, pin_type, frequency, extack); mutex_unlock(&pf->dplls.lock); @@ -584,6 +607,9 @@ ice_dpll_pin_state_set(const struct dpll_pin *pin, void *pin_priv, struct ice_pf *pf = d->pf; int ret; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); if (enable) ret = ice_dpll_pin_enable(&pf->hw, p, d->dpll_idx, pin_type, @@ -687,6 +713,9 @@ ice_dpll_pin_state_get(const struct dpll_pin *pin, void *pin_priv, struct ice_pf *pf = d->pf; int ret; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); ret = ice_dpll_pin_state_update(pf, p, pin_type, extack); if (ret) @@ -811,6 +840,9 @@ ice_dpll_input_prio_set(const struct dpll_pin *pin, void *pin_priv, struct ice_pf *pf = d->pf; int ret; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); ret = ice_dpll_hw_input_prio_set(pf, d, p, prio, extack); mutex_unlock(&pf->dplls.lock); @@ -1090,6 +1122,9 @@ ice_dpll_rclk_state_on_pin_set(const struct dpll_pin *pin, void *pin_priv, int ret = -EINVAL; u32 hw_idx; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); hw_idx = parent->idx - pf->dplls.base_rclk_idx; if (hw_idx >= pf->dplls.num_inputs) @@ -1144,6 +1179,9 @@ ice_dpll_rclk_state_on_pin_get(const struct dpll_pin *pin, void *pin_priv, int ret = -EINVAL; u32 hw_idx; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); hw_idx = parent->idx - pf->dplls.base_rclk_idx; if (hw_idx >= pf->dplls.num_inputs) From 9a8385fe14bcb250a3889e744dc54e9c411d8400 Mon Sep 17 00:00:00 2001 From: Arkadiusz Kubalewski Date: Fri, 9 Feb 2024 22:24:31 +0100 Subject: [PATCH 05/73] ice: fix dpll periodic work data updates on PF reset Do not allow dpll periodic work function to acquire data from firmware if PF reset is in progress. Acquiring data will cause dmesg errors as the firmware cannot respond or process the request properly during the reset time. Test by looping execution of below step until dmesg error appears: - perform PF reset $ echo 1 > /sys/class/net//device/reset Fixes: d7999f5ea64b ("ice: implement dpll interface to control cgu") Reviewed-by: Igor Bagnucki Signed-off-by: Arkadiusz Kubalewski Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dpll.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index 343b2a668959..395e10c246f7 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -1364,8 +1364,10 @@ static void ice_dpll_periodic_work(struct kthread_work *work) struct ice_pf *pf = container_of(d, struct ice_pf, dplls); struct ice_dpll *de = &pf->dplls.eec; struct ice_dpll *dp = &pf->dplls.pps; - int ret; + int ret = 0; + if (ice_is_reset_in_progress(pf->state)) + goto resched; mutex_lock(&pf->dplls.lock); ret = ice_dpll_update_state(pf, de, false); if (!ret) @@ -1385,6 +1387,7 @@ static void ice_dpll_periodic_work(struct kthread_work *work) ice_dpll_notify_changes(de); ice_dpll_notify_changes(dp); +resched: /* Run twice a second or reschedule if update failed */ kthread_queue_delayed_work(d->kworker, &d->work, ret ? msecs_to_jiffies(10) : From ee89921da471edcb4b1e67f5bbfedddf39749782 Mon Sep 17 00:00:00 2001 From: Arkadiusz Kubalewski Date: Fri, 9 Feb 2024 22:24:32 +0100 Subject: [PATCH 06/73] ice: fix pin phase adjust updates on PF reset Do not allow to set phase adjust value for a pin if PF reset is in progress, this would cause confusing netlink extack errors as the firmware cannot process the request properly during the reset time. Return (-EBUSY) and report extack error for the user who tries configure pin phase adjust during the reset time. Test by looping execution of below steps until netlink error appears: - perform PF reset $ echo 1 > /sys/class/net//device/reset - change pin phase adjust value: $ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/dpll.yaml \ --do pin-set --json '{"id":0, "phase-adjust":1000}' Fixes: 90e1c90750d7 ("ice: dpll: implement phase related callbacks") Reviewed-by: Igor Bagnucki Signed-off-by: Arkadiusz Kubalewski Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_dpll.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/ice_dpll.c b/drivers/net/ethernet/intel/ice/ice_dpll.c index 395e10c246f7..adfa1f2a80a6 100644 --- a/drivers/net/ethernet/intel/ice/ice_dpll.c +++ b/drivers/net/ethernet/intel/ice/ice_dpll.c @@ -963,6 +963,9 @@ ice_dpll_pin_phase_adjust_set(const struct dpll_pin *pin, void *pin_priv, u8 flag, flags_en = 0; int ret; + if (ice_dpll_is_reset(pf, extack)) + return -EBUSY; + mutex_lock(&pf->dplls.lock); switch (type) { case ICE_DPLL_PIN_TYPE_INPUT: From 080b0c8d6d261b400f24bb1075fbab8c6daaf69e Mon Sep 17 00:00:00 2001 From: Amritha Nambiar Date: Tue, 13 Feb 2024 11:48:50 -0800 Subject: [PATCH 07/73] ice: Fix ASSERT_RTNL() warning during certain scenarios Commit 91fdbce7e8d6 ("ice: Add support in the driver for associating queue with napi") invoked the netif_queue_set_napi() call. This kernel function requires to be called with rtnl_lock taken, otherwise ASSERT_RTNL() warning will be triggered. ice_vsi_rebuild() initiating this call is under rtnl_lock when the rebuild is in response to configuration changes from external interfaces (such as tc, ethtool etc. which holds the lock). But, the VSI rebuild generated from service tasks and resets (PFR/CORER/GLOBR) is not under rtnl lock protection. Handle these cases as well to hold lock before the kernel call (by setting the 'locked' boolean to false). netif_queue_set_napi() is also used to clear previously set napi in the q_vector unroll flow. Handle this for locked/lockless execution paths. Fixes: 91fdbce7e8d6 ("ice: Add support in the driver for associating queue with napi") Signed-off-by: Amritha Nambiar Reviewed-by: Sridhar Samudrala Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_base.c | 10 ++- drivers/net/ethernet/intel/ice/ice_lib.c | 88 ++++++++++++++++++----- drivers/net/ethernet/intel/ice/ice_lib.h | 10 ++- drivers/net/ethernet/intel/ice/ice_main.c | 3 +- 4 files changed, 84 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 7ac847718882..c979192e44d1 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -190,15 +190,13 @@ static void ice_free_q_vector(struct ice_vsi *vsi, int v_idx) q_vector = vsi->q_vectors[v_idx]; ice_for_each_tx_ring(tx_ring, q_vector->tx) { - if (vsi->netdev) - netif_queue_set_napi(vsi->netdev, tx_ring->q_index, - NETDEV_QUEUE_TYPE_TX, NULL); + ice_queue_set_napi(vsi, tx_ring->q_index, NETDEV_QUEUE_TYPE_TX, + NULL); tx_ring->q_vector = NULL; } ice_for_each_rx_ring(rx_ring, q_vector->rx) { - if (vsi->netdev) - netif_queue_set_napi(vsi->netdev, rx_ring->q_index, - NETDEV_QUEUE_TYPE_RX, NULL); + ice_queue_set_napi(vsi, rx_ring->q_index, NETDEV_QUEUE_TYPE_RX, + NULL); rx_ring->q_vector = NULL; } diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 9be724291ef8..097bf8fd6bf0 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2426,7 +2426,7 @@ ice_vsi_cfg_def(struct ice_vsi *vsi, struct ice_vsi_cfg_params *params) ice_vsi_map_rings_to_vectors(vsi); /* Associate q_vector rings to napi */ - ice_vsi_set_napi_queues(vsi, true); + ice_vsi_set_napi_queues(vsi); vsi->stat_offsets_loaded = false; @@ -2904,19 +2904,19 @@ void ice_vsi_dis_irq(struct ice_vsi *vsi) } /** - * ice_queue_set_napi - Set the napi instance for the queue + * __ice_queue_set_napi - Set the napi instance for the queue * @dev: device to which NAPI and queue belong * @queue_index: Index of queue * @type: queue type as RX or TX * @napi: NAPI context * @locked: is the rtnl_lock already held * - * Set the napi instance for the queue + * Set the napi instance for the queue. Caller indicates the lock status. */ static void -ice_queue_set_napi(struct net_device *dev, unsigned int queue_index, - enum netdev_queue_type type, struct napi_struct *napi, - bool locked) +__ice_queue_set_napi(struct net_device *dev, unsigned int queue_index, + enum netdev_queue_type type, struct napi_struct *napi, + bool locked) { if (!locked) rtnl_lock(); @@ -2926,26 +2926,79 @@ ice_queue_set_napi(struct net_device *dev, unsigned int queue_index, } /** - * ice_q_vector_set_napi_queues - Map queue[s] associated with the napi + * ice_queue_set_napi - Set the napi instance for the queue + * @vsi: VSI being configured + * @queue_index: Index of queue + * @type: queue type as RX or TX + * @napi: NAPI context + * + * Set the napi instance for the queue. The rtnl lock state is derived from the + * execution path. + */ +void +ice_queue_set_napi(struct ice_vsi *vsi, unsigned int queue_index, + enum netdev_queue_type type, struct napi_struct *napi) +{ + struct ice_pf *pf = vsi->back; + + if (!vsi->netdev) + return; + + if (current_work() == &pf->serv_task || + test_bit(ICE_PREPARED_FOR_RESET, pf->state) || + test_bit(ICE_DOWN, pf->state) || + test_bit(ICE_SUSPENDED, pf->state)) + __ice_queue_set_napi(vsi->netdev, queue_index, type, napi, + false); + else + __ice_queue_set_napi(vsi->netdev, queue_index, type, napi, + true); +} + +/** + * __ice_q_vector_set_napi_queues - Map queue[s] associated with the napi * @q_vector: q_vector pointer * @locked: is the rtnl_lock already held * - * Associate the q_vector napi with all the queue[s] on the vector + * Associate the q_vector napi with all the queue[s] on the vector. + * Caller indicates the lock status. */ -void ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector, bool locked) +void __ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector, bool locked) { struct ice_rx_ring *rx_ring; struct ice_tx_ring *tx_ring; ice_for_each_rx_ring(rx_ring, q_vector->rx) - ice_queue_set_napi(q_vector->vsi->netdev, rx_ring->q_index, - NETDEV_QUEUE_TYPE_RX, &q_vector->napi, - locked); + __ice_queue_set_napi(q_vector->vsi->netdev, rx_ring->q_index, + NETDEV_QUEUE_TYPE_RX, &q_vector->napi, + locked); ice_for_each_tx_ring(tx_ring, q_vector->tx) - ice_queue_set_napi(q_vector->vsi->netdev, tx_ring->q_index, - NETDEV_QUEUE_TYPE_TX, &q_vector->napi, - locked); + __ice_queue_set_napi(q_vector->vsi->netdev, tx_ring->q_index, + NETDEV_QUEUE_TYPE_TX, &q_vector->napi, + locked); + /* Also set the interrupt number for the NAPI */ + netif_napi_set_irq(&q_vector->napi, q_vector->irq.virq); +} + +/** + * ice_q_vector_set_napi_queues - Map queue[s] associated with the napi + * @q_vector: q_vector pointer + * + * Associate the q_vector napi with all the queue[s] on the vector + */ +void ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector) +{ + struct ice_rx_ring *rx_ring; + struct ice_tx_ring *tx_ring; + + ice_for_each_rx_ring(rx_ring, q_vector->rx) + ice_queue_set_napi(q_vector->vsi, rx_ring->q_index, + NETDEV_QUEUE_TYPE_RX, &q_vector->napi); + + ice_for_each_tx_ring(tx_ring, q_vector->tx) + ice_queue_set_napi(q_vector->vsi, tx_ring->q_index, + NETDEV_QUEUE_TYPE_TX, &q_vector->napi); /* Also set the interrupt number for the NAPI */ netif_napi_set_irq(&q_vector->napi, q_vector->irq.virq); } @@ -2953,11 +3006,10 @@ void ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector, bool locked) /** * ice_vsi_set_napi_queues * @vsi: VSI pointer - * @locked: is the rtnl_lock already held * * Associate queue[s] with napi for all vectors */ -void ice_vsi_set_napi_queues(struct ice_vsi *vsi, bool locked) +void ice_vsi_set_napi_queues(struct ice_vsi *vsi) { int i; @@ -2965,7 +3017,7 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi, bool locked) return; ice_for_each_q_vector(vsi, i) - ice_q_vector_set_napi_queues(vsi->q_vectors[i], locked); + ice_q_vector_set_napi_queues(vsi->q_vectors[i]); } /** diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index 71bd27244941..bfcfc582a4c0 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -91,9 +91,15 @@ void ice_vsi_cfg_netdev_tc(struct ice_vsi *vsi, u8 ena_tc); struct ice_vsi * ice_vsi_setup(struct ice_pf *pf, struct ice_vsi_cfg_params *params); -void ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector, bool locked); +void +ice_queue_set_napi(struct ice_vsi *vsi, unsigned int queue_index, + enum netdev_queue_type type, struct napi_struct *napi); -void ice_vsi_set_napi_queues(struct ice_vsi *vsi, bool locked); +void __ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector, bool locked); + +void ice_q_vector_set_napi_queues(struct ice_q_vector *q_vector); + +void ice_vsi_set_napi_queues(struct ice_vsi *vsi); int ice_vsi_release(struct ice_vsi *vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index dd4a9bc0dfdc..59c7e37f175f 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3495,7 +3495,7 @@ static void ice_napi_add(struct ice_vsi *vsi) ice_for_each_q_vector(vsi, v_idx) { netif_napi_add(vsi->netdev, &vsi->q_vectors[v_idx]->napi, ice_napi_poll); - ice_q_vector_set_napi_queues(vsi->q_vectors[v_idx], false); + __ice_q_vector_set_napi_queues(vsi->q_vectors[v_idx], false); } } @@ -5447,6 +5447,7 @@ static int ice_reinit_interrupt_scheme(struct ice_pf *pf) if (ret) goto err_reinit; ice_vsi_map_rings_to_vectors(pf->vsi[v]); + ice_vsi_set_napi_queues(pf->vsi[v]); } ret = ice_req_irq_msix_misc(pf); From 78f65fbf421a61894c14a1b91fe2fb4437b3fe5f Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Sun, 18 Feb 2024 19:51:47 +0200 Subject: [PATCH 08/73] wifi: iwlwifi: mvm: ensure offloading TID queue exists The resume code path assumes that the TX queue for the offloading TID has been configured. At resume time it then tries to sync the write pointer as it may have been updated by the firmware. In the unusual event that no packets have been send on TID 0, the queue will not have been allocated and this causes a crash. Fix this by ensuring the queue exist at suspend time. Signed-off-by: Benjamin Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240218194912.6632e6dc7b35.Ie6e6a7488c9c7d4529f13d48f752b5439d8ac3c4@changeid Signed-off-by: Johannes Berg --- drivers/net/wireless/intel/iwlwifi/mvm/d3.c | 9 ++++++- drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 28 ++++++++++++++++++++ drivers/net/wireless/intel/iwlwifi/mvm/sta.h | 3 ++- 3 files changed, 38 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c index 4582afb149d7..05b64176859e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c @@ -1279,7 +1279,9 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw, mvm->net_detect = true; } else { - struct iwl_wowlan_config_cmd wowlan_config_cmd = {}; + struct iwl_wowlan_config_cmd wowlan_config_cmd = { + .offloading_tid = 0, + }; wowlan_config_cmd.sta_id = mvmvif->deflink.ap_sta_id; @@ -1291,6 +1293,11 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw, goto out_noreset; } + ret = iwl_mvm_sta_ensure_queue( + mvm, ap_sta->txq[wowlan_config_cmd.offloading_tid]); + if (ret) + goto out_noreset; + ret = iwl_mvm_get_wowlan_config(mvm, wowlan, &wowlan_config_cmd, vif, mvmvif, ap_sta); if (ret) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 2a3ca9785974..c2e0cff740e9 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -1502,6 +1502,34 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, return ret; } +int iwl_mvm_sta_ensure_queue(struct iwl_mvm *mvm, + struct ieee80211_txq *txq) +{ + struct iwl_mvm_txq *mvmtxq = iwl_mvm_txq_from_mac80211(txq); + int ret = -EINVAL; + + lockdep_assert_held(&mvm->mutex); + + if (likely(test_bit(IWL_MVM_TXQ_STATE_READY, &mvmtxq->state)) || + !txq->sta) { + return 0; + } + + if (!iwl_mvm_sta_alloc_queue(mvm, txq->sta, txq->ac, txq->tid)) { + set_bit(IWL_MVM_TXQ_STATE_READY, &mvmtxq->state); + ret = 0; + } + + local_bh_disable(); + spin_lock(&mvm->add_stream_lock); + if (!list_empty(&mvmtxq->list)) + list_del_init(&mvmtxq->list); + spin_unlock(&mvm->add_stream_lock); + local_bh_enable(); + + return ret; +} + void iwl_mvm_add_new_dqa_stream_wk(struct work_struct *wk) { struct iwl_mvm *mvm = container_of(wk, struct iwl_mvm, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h index b33a0ce096d4..3cf8a70274ce 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2012-2014, 2018-2023 Intel Corporation + * Copyright (C) 2012-2014, 2018-2024 Intel Corporation * Copyright (C) 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2015-2016 Intel Deutschland GmbH */ @@ -571,6 +571,7 @@ void iwl_mvm_modify_all_sta_disable_tx(struct iwl_mvm *mvm, bool disable); void iwl_mvm_csa_client_absent(struct iwl_mvm *mvm, struct ieee80211_vif *vif); +int iwl_mvm_sta_ensure_queue(struct iwl_mvm *mvm, struct ieee80211_txq *txq); void iwl_mvm_add_new_dqa_stream_wk(struct work_struct *wk); int iwl_mvm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, struct iwl_mvm_int_sta *sta, u8 *addr, u32 cipher, From d3433d1bb7bde449035f54b7000361ce151bad07 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 18 Feb 2024 19:51:50 +0200 Subject: [PATCH 09/73] wifi: iwlwifi: mvm: fix the TXF mapping for BZ devices Those devices' fifos are numbered differently. Because of that, we were looking at the size of the VO fifo size to determine the size of the A-MSDU which led to a lower throughput. Note that for those devices the only user of the AC -> fifo mapping is the size limitation of A-MSDU. Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://msgid.link/20240218194912.da336ca2fa0a.I73e44d5fc474ebb6f275b9008950e59c012f33b2@changeid Signed-off-by: Johannes Berg --- drivers/net/wireless/intel/iwlwifi/fw/api/txq.h | 12 +++++++++++- drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 11 +++++++++++ drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 8 ++++++-- 3 files changed, 28 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/txq.h b/drivers/net/wireless/intel/iwlwifi/fw/api/txq.h index 9c69d3674384..e6c0f928a6bb 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/txq.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/txq.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2005-2014, 2019-2021, 2023 Intel Corporation + * Copyright (C) 2005-2014, 2019-2021, 2023-2024 Intel Corporation * Copyright (C) 2013-2015 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -66,6 +66,16 @@ enum iwl_gen2_tx_fifo { IWL_GEN2_TRIG_TX_FIFO_VO, }; +enum iwl_bz_tx_fifo { + IWL_BZ_EDCA_TX_FIFO_BK, + IWL_BZ_EDCA_TX_FIFO_BE, + IWL_BZ_EDCA_TX_FIFO_VI, + IWL_BZ_EDCA_TX_FIFO_VO, + IWL_BZ_TRIG_TX_FIFO_BK, + IWL_BZ_TRIG_TX_FIFO_BE, + IWL_BZ_TRIG_TX_FIFO_VI, + IWL_BZ_TRIG_TX_FIFO_VO, +}; /** * enum iwl_tx_queue_cfg_actions - TXQ config options * @TX_QUEUE_CFG_ENABLE_QUEUE: enable a queue diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c index c4f96125cf33..25a5a31e63c2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c @@ -31,6 +31,17 @@ const u8 iwl_mvm_ac_to_gen2_tx_fifo[] = { IWL_GEN2_TRIG_TX_FIFO_BK, }; +const u8 iwl_mvm_ac_to_bz_tx_fifo[] = { + IWL_BZ_EDCA_TX_FIFO_VO, + IWL_BZ_EDCA_TX_FIFO_VI, + IWL_BZ_EDCA_TX_FIFO_BE, + IWL_BZ_EDCA_TX_FIFO_BK, + IWL_BZ_TRIG_TX_FIFO_VO, + IWL_BZ_TRIG_TX_FIFO_VI, + IWL_BZ_TRIG_TX_FIFO_BE, + IWL_BZ_TRIG_TX_FIFO_BK, +}; + struct iwl_mvm_mac_iface_iterator_data { struct iwl_mvm *mvm; struct ieee80211_vif *vif; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 40627961b834..81dbef6947f5 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -1581,12 +1581,16 @@ static inline int iwl_mvm_max_active_links(struct iwl_mvm *mvm, extern const u8 iwl_mvm_ac_to_tx_fifo[]; extern const u8 iwl_mvm_ac_to_gen2_tx_fifo[]; +extern const u8 iwl_mvm_ac_to_bz_tx_fifo[]; static inline u8 iwl_mvm_mac_ac_to_tx_fifo(struct iwl_mvm *mvm, enum ieee80211_ac_numbers ac) { - return iwl_mvm_has_new_tx_api(mvm) ? - iwl_mvm_ac_to_gen2_tx_fifo[ac] : iwl_mvm_ac_to_tx_fifo[ac]; + if (mvm->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) + return iwl_mvm_ac_to_bz_tx_fifo[ac]; + if (iwl_mvm_has_new_tx_api(mvm)) + return iwl_mvm_ac_to_gen2_tx_fifo[ac]; + return iwl_mvm_ac_to_tx_fifo[ac]; } struct iwl_rate_info { From 27dc4c6ee5feb87fd4967eb9978fd38e5711cb8b Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Sat, 17 Feb 2024 12:47:01 -0800 Subject: [PATCH 10/73] MAINTAINERS: wifi: update Jeff Johnson e-mail address I now have a @kernel.org e-mail address, so use that for my maintenance activities. Signed-off-by: Jeff Johnson Signed-off-by: Kalle Valo Link: https://msgid.link/20240217-update-maintainer-v1-1-1426cf7a8bb1@quicinc.com --- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 25d530ab90d7..d982db96a2ab 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -17973,7 +17973,7 @@ F: drivers/media/tuners/qt1010* QUALCOMM ATH12K WIRELESS DRIVER M: Kalle Valo -M: Jeff Johnson +M: Jeff Johnson L: ath12k@lists.infradead.org S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath12k @@ -17982,7 +17982,7 @@ F: drivers/net/wireless/ath/ath12k/ QUALCOMM ATHEROS ATH10K WIRELESS DRIVER M: Kalle Valo -M: Jeff Johnson +M: Jeff Johnson L: ath10k@lists.infradead.org S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath10k @@ -17992,7 +17992,7 @@ F: drivers/net/wireless/ath/ath10k/ QUALCOMM ATHEROS ATH11K WIRELESS DRIVER M: Kalle Valo -M: Jeff Johnson +M: Jeff Johnson L: ath11k@lists.infradead.org S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath11k From 1b7d9ab3e66655f19ee0c60cc667913da19a41f0 Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Sat, 17 Feb 2024 12:47:02 -0800 Subject: [PATCH 11/73] MAINTAINERS: wifi: Add N: ath1*k entries to match .yaml files In [1] it was observed that the ath12k maintainers were not added to the review of a new ath12k YAML file. Bartosz suggested "adding an N: ath12k entry to MAINTAINERS" to prevent this in the future. In the process it was noticed that one of the ath11k YAML files was also not explicitly referenced, so add N: entries to ath10k, ath11k, and ath12k, and remove the explicit F: entries. Link: https://lore.kernel.org/linux-wireless/20240216203215.40870-7-brgl@bgdev.pl/ [1] Suggested-by: Bartosz Golaszewski Signed-off-by: Jeff Johnson Signed-off-by: Kalle Valo Link: https://msgid.link/20240217-update-maintainer-v1-2-1426cf7a8bb1@quicinc.com --- MAINTAINERS | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index d982db96a2ab..de1c79c6718b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -17979,6 +17979,7 @@ S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath12k T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git F: drivers/net/wireless/ath/ath12k/ +N: ath12k QUALCOMM ATHEROS ATH10K WIRELESS DRIVER M: Kalle Valo @@ -17987,8 +17988,8 @@ L: ath10k@lists.infradead.org S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath10k T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git -F: Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml F: drivers/net/wireless/ath/ath10k/ +N: ath10k QUALCOMM ATHEROS ATH11K WIRELESS DRIVER M: Kalle Valo @@ -17998,8 +17999,8 @@ S: Supported W: https://wireless.wiki.kernel.org/en/users/Drivers/ath11k B: https://wireless.wiki.kernel.org/en/users/Drivers/ath11k/bugreport T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git -F: Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml F: drivers/net/wireless/ath/ath11k/ +N: ath11k QUALCOMM ATHEROS ATH9K WIRELESS DRIVER M: Toke Høiland-Jørgensen From 661779e1fcafe1b74b3f3fe8e980c1e207fea1fd Mon Sep 17 00:00:00 2001 From: Ryosuke Yasuoka Date: Wed, 21 Feb 2024 16:40:48 +0900 Subject: [PATCH 12/73] netlink: Fix kernel-infoleak-after-free in __skb_datagram_iter syzbot reported the following uninit-value access issue [1]: netlink_to_full_skb() creates a new `skb` and puts the `skb->data` passed as a 1st arg of netlink_to_full_skb() onto new `skb`. The data size is specified as `len` and passed to skb_put_data(). This `len` is based on `skb->end` that is not data offset but buffer offset. The `skb->end` contains data and tailroom. Since the tailroom is not initialized when the new `skb` created, KMSAN detects uninitialized memory area when copying the data. This patch resolved this issue by correct the len from `skb->end` to `skb->len`, which is the actual data offset. BUG: KMSAN: kernel-infoleak-after-free in instrument_copy_to_user include/linux/instrumented.h:114 [inline] BUG: KMSAN: kernel-infoleak-after-free in copy_to_user_iter lib/iov_iter.c:24 [inline] BUG: KMSAN: kernel-infoleak-after-free in iterate_ubuf include/linux/iov_iter.h:29 [inline] BUG: KMSAN: kernel-infoleak-after-free in iterate_and_advance2 include/linux/iov_iter.h:245 [inline] BUG: KMSAN: kernel-infoleak-after-free in iterate_and_advance include/linux/iov_iter.h:271 [inline] BUG: KMSAN: kernel-infoleak-after-free in _copy_to_iter+0x364/0x2520 lib/iov_iter.c:186 instrument_copy_to_user include/linux/instrumented.h:114 [inline] copy_to_user_iter lib/iov_iter.c:24 [inline] iterate_ubuf include/linux/iov_iter.h:29 [inline] iterate_and_advance2 include/linux/iov_iter.h:245 [inline] iterate_and_advance include/linux/iov_iter.h:271 [inline] _copy_to_iter+0x364/0x2520 lib/iov_iter.c:186 copy_to_iter include/linux/uio.h:197 [inline] simple_copy_to_iter+0x68/0xa0 net/core/datagram.c:532 __skb_datagram_iter+0x123/0xdc0 net/core/datagram.c:420 skb_copy_datagram_iter+0x5c/0x200 net/core/datagram.c:546 skb_copy_datagram_msg include/linux/skbuff.h:3960 [inline] packet_recvmsg+0xd9c/0x2000 net/packet/af_packet.c:3482 sock_recvmsg_nosec net/socket.c:1044 [inline] sock_recvmsg net/socket.c:1066 [inline] sock_read_iter+0x467/0x580 net/socket.c:1136 call_read_iter include/linux/fs.h:2014 [inline] new_sync_read fs/read_write.c:389 [inline] vfs_read+0x8f6/0xe00 fs/read_write.c:470 ksys_read+0x20f/0x4c0 fs/read_write.c:613 __do_sys_read fs/read_write.c:623 [inline] __se_sys_read fs/read_write.c:621 [inline] __x64_sys_read+0x93/0xd0 fs/read_write.c:621 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x44/0x110 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b Uninit was stored to memory at: skb_put_data include/linux/skbuff.h:2622 [inline] netlink_to_full_skb net/netlink/af_netlink.c:181 [inline] __netlink_deliver_tap_skb net/netlink/af_netlink.c:298 [inline] __netlink_deliver_tap+0x5be/0xc90 net/netlink/af_netlink.c:325 netlink_deliver_tap net/netlink/af_netlink.c:338 [inline] netlink_deliver_tap_kernel net/netlink/af_netlink.c:347 [inline] netlink_unicast_kernel net/netlink/af_netlink.c:1341 [inline] netlink_unicast+0x10f1/0x1250 net/netlink/af_netlink.c:1368 netlink_sendmsg+0x1238/0x13d0 net/netlink/af_netlink.c:1910 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg net/socket.c:745 [inline] ____sys_sendmsg+0x9c2/0xd60 net/socket.c:2584 ___sys_sendmsg+0x28d/0x3c0 net/socket.c:2638 __sys_sendmsg net/socket.c:2667 [inline] __do_sys_sendmsg net/socket.c:2676 [inline] __se_sys_sendmsg net/socket.c:2674 [inline] __x64_sys_sendmsg+0x307/0x490 net/socket.c:2674 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x44/0x110 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b Uninit was created at: free_pages_prepare mm/page_alloc.c:1087 [inline] free_unref_page_prepare+0xb0/0xa40 mm/page_alloc.c:2347 free_unref_page_list+0xeb/0x1100 mm/page_alloc.c:2533 release_pages+0x23d3/0x2410 mm/swap.c:1042 free_pages_and_swap_cache+0xd9/0xf0 mm/swap_state.c:316 tlb_batch_pages_flush mm/mmu_gather.c:98 [inline] tlb_flush_mmu_free mm/mmu_gather.c:293 [inline] tlb_flush_mmu+0x6f5/0x980 mm/mmu_gather.c:300 tlb_finish_mmu+0x101/0x260 mm/mmu_gather.c:392 exit_mmap+0x49e/0xd30 mm/mmap.c:3321 __mmput+0x13f/0x530 kernel/fork.c:1349 mmput+0x8a/0xa0 kernel/fork.c:1371 exit_mm+0x1b8/0x360 kernel/exit.c:567 do_exit+0xd57/0x4080 kernel/exit.c:858 do_group_exit+0x2fd/0x390 kernel/exit.c:1021 __do_sys_exit_group kernel/exit.c:1032 [inline] __se_sys_exit_group kernel/exit.c:1030 [inline] __x64_sys_exit_group+0x3c/0x50 kernel/exit.c:1030 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x44/0x110 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b Bytes 3852-3903 of 3904 are uninitialized Memory access of size 3904 starts at ffff88812ea1e000 Data copied to user address 0000000020003280 CPU: 1 PID: 5043 Comm: syz-executor297 Not tainted 6.7.0-rc5-syzkaller-00047-g5bd7ef53ffe5 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/10/2023 Fixes: 1853c9496460 ("netlink, mmap: transform mmap skb into full skb on taps") Reported-and-tested-by: syzbot+34ad5fab48f7bf510349@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=34ad5fab48f7bf510349 [1] Signed-off-by: Ryosuke Yasuoka Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240221074053.1794118-1-ryasuoka@redhat.com Signed-off-by: Jakub Kicinski --- 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 9c962347cf85..ff315351269f 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -167,7 +167,7 @@ static inline u32 netlink_group_mask(u32 group) static struct sk_buff *netlink_to_full_skb(const struct sk_buff *skb, gfp_t gfp_mask) { - unsigned int len = skb_end_offset(skb); + unsigned int len = skb->len; struct sk_buff *new; new = alloc_skb(len, gfp_mask); From 9a0d18853c280f6a0ee99f91619f2442a17a323a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 21 Feb 2024 18:27:33 +0100 Subject: [PATCH 13/73] netlink: add nla be16/32 types to minlen array BUG: KMSAN: uninit-value in nla_validate_range_unsigned lib/nlattr.c:222 [inline] BUG: KMSAN: uninit-value in nla_validate_int_range lib/nlattr.c:336 [inline] BUG: KMSAN: uninit-value in validate_nla lib/nlattr.c:575 [inline] BUG: KMSAN: uninit-value in __nla_validate_parse+0x2e20/0x45c0 lib/nlattr.c:631 nla_validate_range_unsigned lib/nlattr.c:222 [inline] nla_validate_int_range lib/nlattr.c:336 [inline] validate_nla lib/nlattr.c:575 [inline] ... The message in question matches this policy: [NFTA_TARGET_REV] = NLA_POLICY_MAX(NLA_BE32, 255), but because NLA_BE32 size in minlen array is 0, the validation code will read past the malformed (too small) attribute. Note: Other attributes, e.g. BITFIELD32, SINT, UINT.. are also missing: those likely should be added too. Reported-by: syzbot+3f497b07aa3baf2fb4d0@syzkaller.appspotmail.com Reported-by: xingwei lee Closes: https://lore.kernel.org/all/CABOYnLzFYHSnvTyS6zGa-udNX55+izqkOt2sB9WDqUcEGW6n8w@mail.gmail.com/raw Fixes: ecaf75ffd5f5 ("netlink: introduce bigendian integer types") Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20240221172740.5092-1-fw@strlen.de Signed-off-by: Jakub Kicinski --- lib/nlattr.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/lib/nlattr.c b/lib/nlattr.c index ed2ab43e1b22..be9c576b6e2d 100644 --- a/lib/nlattr.c +++ b/lib/nlattr.c @@ -30,6 +30,8 @@ static const u8 nla_attr_len[NLA_TYPE_MAX+1] = { [NLA_S16] = sizeof(s16), [NLA_S32] = sizeof(s32), [NLA_S64] = sizeof(s64), + [NLA_BE16] = sizeof(__be16), + [NLA_BE32] = sizeof(__be32), }; static const u8 nla_attr_minlen[NLA_TYPE_MAX+1] = { @@ -43,6 +45,8 @@ static const u8 nla_attr_minlen[NLA_TYPE_MAX+1] = { [NLA_S16] = sizeof(s16), [NLA_S32] = sizeof(s32), [NLA_S64] = sizeof(s64), + [NLA_BE16] = sizeof(__be16), + [NLA_BE32] = sizeof(__be32), }; /* From 45532b21dc2a692444b6ad5f71c253cca53e8103 Mon Sep 17 00:00:00 2001 From: Andre Werner Date: Mon, 19 Feb 2024 06:33:32 +0100 Subject: [PATCH 14/73] net: smsc95xx: add support for SYS TEC USB-SPEmodule1 This patch adds support for the SYS TEC USB-SPEmodule1 10Base-T1L ethernet device to the existing smsc95xx driver by adding the new USB VID/PID pair. Signed-off-by: Andre Werner Link: https://lore.kernel.org/r/20240219053413.4732-1-andre.werner@systec-electronic.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/smsc95xx.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c index a530f20ee257..2fa46baa589e 100644 --- a/drivers/net/usb/smsc95xx.c +++ b/drivers/net/usb/smsc95xx.c @@ -2104,6 +2104,11 @@ static const struct usb_device_id products[] = { USB_DEVICE(0x0424, 0x9E08), .driver_info = (unsigned long) &smsc95xx_info, }, + { + /* SYSTEC USB-SPEmodule1 10BASE-T1L Ethernet Device */ + USB_DEVICE(0x0878, 0x1400), + .driver_info = (unsigned long)&smsc95xx_info, + }, { /* Microchip's EVB-LAN8670-USB 10BASE-T1S Ethernet Device */ USB_DEVICE(0x184F, 0x0051), From 5ae1e9922bbdbaeb9cfbe91085ab75927488ac0f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 20 Feb 2024 14:56:02 +0100 Subject: [PATCH 15/73] net: ip_tunnel: prevent perpetual headroom growth syzkaller triggered following kasan splat: BUG: KASAN: use-after-free in __skb_flow_dissect+0x19d1/0x7a50 net/core/flow_dissector.c:1170 Read of size 1 at addr ffff88812fb4000e by task syz-executor183/5191 [..] kasan_report+0xda/0x110 mm/kasan/report.c:588 __skb_flow_dissect+0x19d1/0x7a50 net/core/flow_dissector.c:1170 skb_flow_dissect_flow_keys include/linux/skbuff.h:1514 [inline] ___skb_get_hash net/core/flow_dissector.c:1791 [inline] __skb_get_hash+0xc7/0x540 net/core/flow_dissector.c:1856 skb_get_hash include/linux/skbuff.h:1556 [inline] ip_tunnel_xmit+0x1855/0x33c0 net/ipv4/ip_tunnel.c:748 ipip_tunnel_xmit+0x3cc/0x4e0 net/ipv4/ipip.c:308 __netdev_start_xmit include/linux/netdevice.h:4940 [inline] netdev_start_xmit include/linux/netdevice.h:4954 [inline] xmit_one net/core/dev.c:3548 [inline] dev_hard_start_xmit+0x13d/0x6d0 net/core/dev.c:3564 __dev_queue_xmit+0x7c1/0x3d60 net/core/dev.c:4349 dev_queue_xmit include/linux/netdevice.h:3134 [inline] neigh_connected_output+0x42c/0x5d0 net/core/neighbour.c:1592 ... ip_finish_output2+0x833/0x2550 net/ipv4/ip_output.c:235 ip_finish_output+0x31/0x310 net/ipv4/ip_output.c:323 .. iptunnel_xmit+0x5b4/0x9b0 net/ipv4/ip_tunnel_core.c:82 ip_tunnel_xmit+0x1dbc/0x33c0 net/ipv4/ip_tunnel.c:831 ipgre_xmit+0x4a1/0x980 net/ipv4/ip_gre.c:665 __netdev_start_xmit include/linux/netdevice.h:4940 [inline] netdev_start_xmit include/linux/netdevice.h:4954 [inline] xmit_one net/core/dev.c:3548 [inline] dev_hard_start_xmit+0x13d/0x6d0 net/core/dev.c:3564 ... The splat occurs because skb->data points past skb->head allocated area. This is because neigh layer does: __skb_pull(skb, skb_network_offset(skb)); ... but skb_network_offset() returns a negative offset and __skb_pull() arg is unsigned. IOW, we skb->data gets "adjusted" by a huge value. The negative value is returned because skb->head and skb->data distance is more than 64k and skb->network_header (u16) has wrapped around. The bug is in the ip_tunnel infrastructure, which can cause dev->needed_headroom to increment ad infinitum. The syzkaller reproducer consists of packets getting routed via a gre tunnel, and route of gre encapsulated packets pointing at another (ipip) tunnel. The ipip encapsulation finds gre0 as next output device. This results in the following pattern: 1). First packet is to be sent out via gre0. Route lookup found an output device, ipip0. 2). ip_tunnel_xmit for gre0 bumps gre0->needed_headroom based on the future output device, rt.dev->needed_headroom (ipip0). 3). ip output / start_xmit moves skb on to ipip0. which runs the same code path again (xmit recursion). 4). Routing step for the post-gre0-encap packet finds gre0 as output device to use for ipip0 encapsulated packet. tunl0->needed_headroom is then incremented based on the (already bumped) gre0 device headroom. This repeats for every future packet: gre0->needed_headroom gets inflated because previous packets' ipip0 step incremented rt->dev (gre0) headroom, and ipip0 incremented because gre0 needed_headroom was increased. For each subsequent packet, gre/ipip0->needed_headroom grows until post-expand-head reallocations result in a skb->head/data distance of more than 64k. Once that happens, skb->network_header (u16) wraps around when pskb_expand_head tries to make sure that skb_network_offset() is unchanged after the headroom expansion/reallocation. After this skb_network_offset(skb) returns a different (and negative) result post headroom expansion. The next trip to neigh layer (or anything else that would __skb_pull the network header) makes skb->data point to a memory location outside skb->head area. v2: Cap the needed_headroom update to an arbitarily chosen upperlimit to prevent perpetual increase instead of dropping the headroom increment completely. Reported-and-tested-by: syzbot+bfde3bef047a81b8fde6@syzkaller.appspotmail.com Closes: https://groups.google.com/g/syzkaller-bugs/c/fL9G6GtWskY/m/VKk_PR5FBAAJ Fixes: 243aad830e8a ("ip_gre: include route header_len in max_headroom calculation") Signed-off-by: Florian Westphal Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240220135606.4939-1-fw@strlen.de Signed-off-by: Jakub Kicinski --- net/ipv4/ip_tunnel.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index a4513ffb66cb..1b6981de3f29 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -554,6 +554,20 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb, return 0; } +static void ip_tunnel_adj_headroom(struct net_device *dev, unsigned int headroom) +{ + /* we must cap headroom to some upperlimit, else pskb_expand_head + * will overflow header offsets in skb_headers_offset_update(). + */ + static const unsigned int max_allowed = 512; + + if (headroom > max_allowed) + headroom = max_allowed; + + if (headroom > READ_ONCE(dev->needed_headroom)) + WRITE_ONCE(dev->needed_headroom, headroom); +} + void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, u8 proto, int tunnel_hlen) { @@ -632,13 +646,13 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, } headroom += LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len; - if (headroom > READ_ONCE(dev->needed_headroom)) - WRITE_ONCE(dev->needed_headroom, headroom); - - if (skb_cow_head(skb, READ_ONCE(dev->needed_headroom))) { + if (skb_cow_head(skb, headroom)) { ip_rt_put(rt); goto tx_dropped; } + + ip_tunnel_adj_headroom(dev, headroom); + iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, proto, tos, ttl, df, !net_eq(tunnel->net, dev_net(dev))); return; @@ -818,16 +832,16 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, max_headroom = LL_RESERVED_SPACE(rt->dst.dev) + sizeof(struct iphdr) + rt->dst.header_len + ip_encap_hlen(&tunnel->encap); - if (max_headroom > READ_ONCE(dev->needed_headroom)) - WRITE_ONCE(dev->needed_headroom, max_headroom); - if (skb_cow_head(skb, READ_ONCE(dev->needed_headroom))) { + if (skb_cow_head(skb, max_headroom)) { ip_rt_put(rt); DEV_STATS_INC(dev, tx_dropped); kfree_skb(skb); return; } + ip_tunnel_adj_headroom(dev, max_headroom); + iptunnel_xmit(NULL, rt, skb, fl4.saddr, fl4.daddr, protocol, tos, ttl, df, !net_eq(tunnel->net, dev_net(dev))); return; From 3773d65ae5154ed7df404b050fd7387a36ab5ef3 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Tue, 20 Feb 2024 16:10:53 +0800 Subject: [PATCH 16/73] net: mctp: take ownership of skb in mctp_local_output Currently, mctp_local_output only takes ownership of skb on success, and we may leak an skb if mctp_local_output fails in specific states; the skb ownership isn't transferred until the actual output routing occurs. Instead, make mctp_local_output free the skb on all error paths up to the route action, so it always consumes the passed skb. Fixes: 833ef3b91de6 ("mctp: Populate socket implementation") Signed-off-by: Jeremy Kerr Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240220081053.1439104-1-jk@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- include/net/mctp.h | 1 + net/mctp/route.c | 10 ++++++++-- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/include/net/mctp.h b/include/net/mctp.h index da86e106c91d..2bff5f47ce82 100644 --- a/include/net/mctp.h +++ b/include/net/mctp.h @@ -249,6 +249,7 @@ struct mctp_route { struct mctp_route *mctp_route_lookup(struct net *net, unsigned int dnet, mctp_eid_t daddr); +/* always takes ownership of skb */ int mctp_local_output(struct sock *sk, struct mctp_route *rt, struct sk_buff *skb, mctp_eid_t daddr, u8 req_tag); diff --git a/net/mctp/route.c b/net/mctp/route.c index 6218dcd07e18..ceee44ea09d9 100644 --- a/net/mctp/route.c +++ b/net/mctp/route.c @@ -888,7 +888,7 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt, dev = dev_get_by_index_rcu(sock_net(sk), cb->ifindex); if (!dev) { rcu_read_unlock(); - return rc; + goto out_free; } rt->dev = __mctp_dev_get(dev); rcu_read_unlock(); @@ -903,7 +903,8 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt, rt->mtu = 0; } else { - return -EINVAL; + rc = -EINVAL; + goto out_free; } spin_lock_irqsave(&rt->dev->addrs_lock, flags); @@ -966,12 +967,17 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt, rc = mctp_do_fragment_route(rt, skb, mtu, tag); } + /* route output functions consume the skb, even on error */ + skb = NULL; + out_release: if (!ext_rt) mctp_route_release(rt); mctp_dev_put(tmp_rt.dev); +out_free: + kfree_skb(skb); return rc; } From 413dafc8170fcb925fb17af8842f06af305f8e0b Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Wed, 21 Feb 2024 15:05:35 +0100 Subject: [PATCH 17/73] wifi: mac80211: only call drv_sta_rc_update for uploaded stations When a station has not been uploaded yet, receiving SMPS or channel width notification action frames can lead to rate_control_rate_update calling drv_sta_rc_update with uninitialized driver private data. Fix this by adding a missing check for sta->uploaded. Signed-off-by: Felix Fietkau Link: https://msgid.link/20240221140535.16102-1-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rate.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index d5ea5f5bcf3a..9d33fd2377c8 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -119,7 +119,8 @@ void rate_control_rate_update(struct ieee80211_local *local, rcu_read_unlock(); } - drv_sta_rc_update(local, sta->sdata, &sta->sta, changed); + if (sta->uploaded) + drv_sta_rc_update(local, sta->sdata, &sta->sta, changed); } int ieee80211_rate_control_register(const struct rate_control_ops *ops) From 734f06db599f66d6a159c78abfdbadfea3b7d43b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 21 Feb 2024 00:34:42 +0200 Subject: [PATCH 18/73] net: dpaa: fman_memac: accept phy-interface-type = "10gbase-r" in the device tree Since commit 5d93cfcf7360 ("net: dpaa: Convert to phylink"), we support the "10gbase-r" phy-mode through a driver-based conversion of "xgmii", but we still don't actually support it when the device tree specifies "10gbase-r" proper. This is because boards such as LS1046A-RDB do not define pcs-handle-names (for whatever reason) in the ethernet@f0000 device tree node, and the code enters through this code path: err = of_property_match_string(mac_node, "pcs-handle-names", "xfi"); // code takes neither branch and falls through if (err >= 0) { (...) } else if (err != -EINVAL && err != -ENODATA) { goto _return_fm_mac_free; } (...) /* For compatibility, if pcs-handle-names is missing, we assume this * phy is the first one in pcsphy-handle */ err = of_property_match_string(mac_node, "pcs-handle-names", "sgmii"); if (err == -EINVAL || err == -ENODATA) pcs = memac_pcs_create(mac_node, 0); // code takes this branch else if (err < 0) goto _return_fm_mac_free; else pcs = memac_pcs_create(mac_node, err); // A default PCS is created and saved in "pcs" // This determination fails and mistakenly saves the default PCS // memac->sgmii_pcs instead of memac->xfi_pcs, because at this // stage, mac_dev->phy_if == PHY_INTERFACE_MODE_10GBASER. if (err && mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII) memac->xfi_pcs = pcs; else memac->sgmii_pcs = pcs; In other words, in the absence of pcs-handle-names, the default xfi_pcs assignment logic only works when in the device tree we have PHY_INTERFACE_MODE_XGMII. By reversing the order between the fallback xfi_pcs assignment and the "xgmii" overwrite with "10gbase-r", we are able to support both values in the device tree, with identical behavior. Currently, it is impossible to make the s/xgmii/10gbase-r/ device tree conversion, because it would break forward compatibility (new device tree with old kernel). The only way to modify existing device trees to phy-interface-mode = "10gbase-r" is to fix stable kernels to accept this value and handle it properly. One reason why the conversion is desirable is because with pre-phylink kernels, the Aquantia PHY driver used to warn about the improper use of PHY_INTERFACE_MODE_XGMII [1]. It is best to have a single (latest) device tree that works with all supported stable kernel versions. Note that the blamed commit does not constitute a regression per se. Older stable kernels like 6.1 still do not work with "10gbase-r", but for a different reason. That is a battle for another time. [1] https://lore.kernel.org/netdev/20240214-ls1046-dts-use-10gbase-r-v1-1-8c2d68547393@concurrent-rt.com/ Fixes: 5d93cfcf7360 ("net: dpaa: Convert to phylink") Signed-off-by: Vladimir Oltean Reviewed-by: Sean Anderson Acked-by: Madalin Bucur Signed-off-by: David S. Miller --- .../net/ethernet/freescale/fman/fman_memac.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index 9ba15d3183d7..758535adc9ff 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -1073,6 +1073,14 @@ int memac_initialization(struct mac_device *mac_dev, unsigned long capabilities; unsigned long *supported; + /* The internal connection to the serdes is XGMII, but this isn't + * really correct for the phy mode (which is the external connection). + * However, this is how all older device trees say that they want + * 10GBASE-R (aka XFI), so just convert it for them. + */ + if (mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII) + mac_dev->phy_if = PHY_INTERFACE_MODE_10GBASER; + mac_dev->phylink_ops = &memac_mac_ops; mac_dev->set_promisc = memac_set_promiscuous; mac_dev->change_addr = memac_modify_mac_address; @@ -1139,7 +1147,7 @@ int memac_initialization(struct mac_device *mac_dev, * (and therefore that xfi_pcs cannot be set). If we are defaulting to * XGMII, assume this is for XFI. Otherwise, assume it is for SGMII. */ - if (err && mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII) + if (err && mac_dev->phy_if == PHY_INTERFACE_MODE_10GBASER) memac->xfi_pcs = pcs; else memac->sgmii_pcs = pcs; @@ -1153,14 +1161,6 @@ int memac_initialization(struct mac_device *mac_dev, goto _return_fm_mac_free; } - /* The internal connection to the serdes is XGMII, but this isn't - * really correct for the phy mode (which is the external connection). - * However, this is how all older device trees say that they want - * 10GBASE-R (aka XFI), so just convert it for them. - */ - if (mac_dev->phy_if == PHY_INTERFACE_MODE_XGMII) - mac_dev->phy_if = PHY_INTERFACE_MODE_10GBASER; - /* TODO: The following interface modes are supported by (some) hardware * but not by this driver: * - 1000BASE-KX From b0b1210bc150fbd741b4b9fce8a24541306b40fc Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Wed, 21 Feb 2024 11:27:29 +0900 Subject: [PATCH 19/73] ps3/gelic: Fix SKB allocation Commit 3ce4f9c3fbb3 ("net/ps3_gelic_net: Add gelic_descr structures") of 6.8-rc1 had a copy-and-paste error where the pointer that holds the allocated SKB (struct gelic_descr.skb) was set to NULL after the SKB was allocated. This resulted in a kernel panic when the SKB pointer was accessed. This fix moves the initialization of the gelic_descr to before the SKB is allocated. Reported-by: sambat goson Fixes: 3ce4f9c3fbb3 ("net/ps3_gelic_net: Add gelic_descr structures") Signed-off-by: Geoff Levand Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/toshiba/ps3_gelic_net.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c index d5b75af163d3..c1b0d35c8d05 100644 --- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c +++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c @@ -384,18 +384,18 @@ static int gelic_descr_prepare_rx(struct gelic_card *card, if (gelic_descr_get_status(descr) != GELIC_DESCR_DMA_NOT_IN_USE) dev_info(ctodev(card), "%s: ERROR status\n", __func__); - descr->skb = netdev_alloc_skb(*card->netdev, rx_skb_size); - if (!descr->skb) { - descr->hw_regs.payload.dev_addr = 0; /* tell DMAC don't touch memory */ - return -ENOMEM; - } descr->hw_regs.dmac_cmd_status = 0; descr->hw_regs.result_size = 0; descr->hw_regs.valid_size = 0; descr->hw_regs.data_error = 0; descr->hw_regs.payload.dev_addr = 0; descr->hw_regs.payload.size = 0; - descr->skb = NULL; + + descr->skb = netdev_alloc_skb(*card->netdev, rx_skb_size); + if (!descr->skb) { + descr->hw_regs.payload.dev_addr = 0; /* tell DMAC don't touch memory */ + return -ENOMEM; + } offset = ((unsigned long)descr->skb->data) & (GELIC_NET_RXBUF_ALIGN - 1); From 2a770cdc4382b457ca3d43d03f0f0064f905a0d0 Mon Sep 17 00:00:00 2001 From: Yunjian Wang Date: Tue, 20 Feb 2024 11:12:07 +0800 Subject: [PATCH 20/73] tun: Fix xdp_rxq_info's queue_index when detaching When a queue(tfile) is detached, we only update tfile's queue_index, but do not update xdp_rxq_info's queue_index. This patch fixes it. Fixes: 8bf5c4ee1889 ("tun: setup xdp_rxq_info") Signed-off-by: Yunjian Wang Link: https://lore.kernel.org/r/1708398727-46308-1-git-send-email-wangyunjian@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 4a4f8c8e79fa..8f95a562b8d0 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -653,6 +653,7 @@ static void __tun_detach(struct tun_file *tfile, bool clean) tun->tfiles[tun->numqueues - 1]); ntfile = rtnl_dereference(tun->tfiles[index]); ntfile->queue_index = index; + ntfile->xdp_rxq.queue_index = index; rcu_assign_pointer(tun->tfiles[tun->numqueues - 1], NULL); From fe9f801355f0b47668419f30f1fac1cf4539e736 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 21 Feb 2024 15:12:10 -0800 Subject: [PATCH 21/73] net: veth: clear GRO when clearing XDP even when down MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit veth sets NETIF_F_GRO automatically when XDP is enabled, because both features use the same NAPI machinery. The logic to clear NETIF_F_GRO sits in veth_disable_xdp() which is called both on ndo_stop and when XDP is turned off. To avoid the flag from being cleared when the device is brought down, the clearing is skipped when IFF_UP is not set. Bringing the device down should indeed not modify its features. Unfortunately, this means that clearing is also skipped when XDP is disabled _while_ the device is down. And there's nothing on the open path to bring the device features back into sync. IOW if user enables XDP, disables it and then brings the device up we'll end up with a stray GRO flag set but no NAPI instances. We don't depend on the GRO flag on the datapath, so the datapath won't crash. We will crash (or hang), however, next time features are sync'ed (either by user via ethtool or peer changing its config). The GRO flag will go away, and veth will try to disable the NAPIs. But the open path never created them since XDP was off, the GRO flag was a stray. If NAPI was initialized before we'll hang in napi_disable(). If it never was we'll crash trying to stop uninitialized hrtimer. Move the GRO flag updates to the XDP enable / disable paths, instead of mixing them with the ndo_open / ndo_close paths. Fixes: d3256efd8e8b ("veth: allow enabling NAPI even without XDP") Reported-by: Thomas Gleixner Reported-by: syzbot+039399a9b96297ddedca@syzkaller.appspotmail.com Signed-off-by: Jakub Kicinski Reviewed-by: Toke Høiland-Jørgensen Signed-off-by: David S. Miller --- drivers/net/veth.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 578e36ea1589..a786be805709 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -1208,14 +1208,6 @@ static int veth_enable_xdp(struct net_device *dev) veth_disable_xdp_range(dev, 0, dev->real_num_rx_queues, true); return err; } - - if (!veth_gro_requested(dev)) { - /* user-space did not require GRO, but adding XDP - * is supposed to get GRO working - */ - dev->features |= NETIF_F_GRO; - netdev_features_change(dev); - } } } @@ -1235,18 +1227,9 @@ static void veth_disable_xdp(struct net_device *dev) for (i = 0; i < dev->real_num_rx_queues; i++) rcu_assign_pointer(priv->rq[i].xdp_prog, NULL); - if (!netif_running(dev) || !veth_gro_requested(dev)) { + if (!netif_running(dev) || !veth_gro_requested(dev)) veth_napi_del(dev); - /* if user-space did not require GRO, since adding XDP - * enabled it, clear it now - */ - if (!veth_gro_requested(dev) && netif_running(dev)) { - dev->features &= ~NETIF_F_GRO; - netdev_features_change(dev); - } - } - veth_disable_xdp_range(dev, 0, dev->real_num_rx_queues, false); } @@ -1654,6 +1637,14 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog, } if (!old_prog) { + if (!veth_gro_requested(dev)) { + /* user-space did not require GRO, but adding + * XDP is supposed to get GRO working + */ + dev->features |= NETIF_F_GRO; + netdev_features_change(dev); + } + peer->hw_features &= ~NETIF_F_GSO_SOFTWARE; peer->max_mtu = max_mtu; } @@ -1669,6 +1660,14 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog, if (dev->flags & IFF_UP) veth_disable_xdp(dev); + /* if user-space did not require GRO, since adding XDP + * enabled it, clear it now + */ + if (!veth_gro_requested(dev)) { + dev->features &= ~NETIF_F_GRO; + netdev_features_change(dev); + } + if (peer) { peer->hw_features |= NETIF_F_GSO_SOFTWARE; peer->max_mtu = ETH_MAX_MTU; From 1a825e4cdf457b7aef7ebbc2f1206654f5beb150 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 21 Feb 2024 15:12:11 -0800 Subject: [PATCH 22/73] selftests: net: veth: test syncing GRO and XDP state while device is down MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Test that we keep GRO flag in sync when XDP is disabled while the device is closed. Signed-off-by: Jakub Kicinski Reviewed-by: Toke Høiland-Jørgensen Signed-off-by: David S. Miller --- tools/testing/selftests/net/veth.sh | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/tools/testing/selftests/net/veth.sh b/tools/testing/selftests/net/veth.sh index 27574bbf2d63..5ae85def0739 100755 --- a/tools/testing/selftests/net/veth.sh +++ b/tools/testing/selftests/net/veth.sh @@ -246,6 +246,20 @@ ip netns exec $NS_DST ethtool -K veth$DST rx-udp-gro-forwarding on chk_gro " - aggregation with TSO off" 1 cleanup +create_ns +ip -n $NS_DST link set dev veth$DST up +ip -n $NS_DST link set dev veth$DST xdp object ${BPF_FILE} section xdp +chk_gro_flag "gro vs xdp while down - gro flag on" $DST on +ip -n $NS_DST link set dev veth$DST down +chk_gro_flag " - after down" $DST on +ip -n $NS_DST link set dev veth$DST xdp off +chk_gro_flag " - after xdp off" $DST off +ip -n $NS_DST link set dev veth$DST up +chk_gro_flag " - after up" $DST off +ip -n $NS_SRC link set dev veth$SRC xdp object ${BPF_FILE} section xdp +chk_gro_flag " - after peer xdp" $DST off +cleanup + create_ns chk_channels "default channels" $DST 1 1 From 10bfd453da64a057bcfd1a49fb6b271c48653cdb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 22 Feb 2024 12:17:47 +0000 Subject: [PATCH 23/73] ipv6: fix potential "struct net" leak in inet6_rtm_getaddr() It seems that if userspace provides a correct IFA_TARGET_NETNSID value but no IFA_ADDRESS and IFA_LOCAL attributes, inet6_rtm_getaddr() returns -EINVAL with an elevated "struct net" refcount. Fixes: 6ecf4c37eb3e ("ipv6: enable IFA_TARGET_NETNSID for RTM_GETADDR") Signed-off-by: Eric Dumazet Cc: Christian Brauner Cc: David Ahern Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5a839c5fb1a5..055230b669cf 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5509,9 +5509,10 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr *nlh, } addr = extract_addr(tb[IFA_ADDRESS], tb[IFA_LOCAL], &peer); - if (!addr) - return -EINVAL; - + if (!addr) { + err = -EINVAL; + goto errout; + } ifm = nlmsg_data(nlh); if (ifm->ifa_index) dev = dev_get_by_index(tgt_net, ifm->ifa_index); From 0e67899abfbfdea0c3c0ed3fd263ffc601c5c157 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Thu, 22 Feb 2024 13:38:38 +0100 Subject: [PATCH 24/73] lan78xx: enable auto speed configuration for LAN7850 if no EEPROM is detected Same as LAN7800, LAN7850 can be used without EEPROM. If EEPROM is not present or not flashed, LAN7850 will fail to sync the speed detected by the PHY with the MAC. In case link speed is 100Mbit, it will accidentally work, otherwise no data can be transferred. Better way would be to implement link_up callback, or set auto speed configuration unconditionally. But this changes would be more intrusive. So, for now, set it only if no EEPROM is found. Fixes: e69647a19c87 ("lan78xx: Set ASD in MAC_CR when EEE is enabled.") Signed-off-by: Oleksij Rempel Link: https://lore.kernel.org/r/20240222123839.2816561-1-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/usb/lan78xx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index a6d653ff552a..e06193e831a5 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -3033,7 +3033,8 @@ static int lan78xx_reset(struct lan78xx_net *dev) if (dev->chipid == ID_REV_CHIP_ID_7801_) buf &= ~MAC_CR_GMII_EN_; - if (dev->chipid == ID_REV_CHIP_ID_7800_) { + if (dev->chipid == ID_REV_CHIP_ID_7800_ || + dev->chipid == ID_REV_CHIP_ID_7850_) { ret = lan78xx_read_raw_eeprom(dev, 0, 1, &sig); if (!ret && sig != EEPROM_INDICATOR) { /* Implies there is no external eeprom. Set mac speed */ From 0d60d8df6f493bb46bf5db40d39dd60a1bafdd4e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 23 Feb 2024 12:32:08 +0000 Subject: [PATCH 25/73] dpll: rely on rcu for netdev_dpll_pin() This fixes a possible UAF in if_nlmsg_size(), which can run without RTNL. Add rcu protection to "struct dpll_pin" Move netdev_dpll_pin() from netdevice.h to dpll.h to decrease name pollution. Note: This looks possible to no longer acquire RTNL in netdev_dpll_pin_assign() later in net-next. v2: do not force rcu_read_lock() in rtnl_dpll_pin_size() (Jiri Pirko) Fixes: 5f1842692880 ("netdev: expose DPLL pin handle for netdevice") Signed-off-by: Eric Dumazet Cc: Arkadiusz Kubalewski Cc: Vadim Fedorenko Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20240223123208.3543319-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/dpll/dpll_core.c | 2 +- drivers/dpll/dpll_core.h | 2 ++ include/linux/dpll.h | 11 +++++++++++ include/linux/netdevice.h | 11 +---------- net/core/dev.c | 2 +- 5 files changed, 16 insertions(+), 12 deletions(-) diff --git a/drivers/dpll/dpll_core.c b/drivers/dpll/dpll_core.c index 5152bd1b0daf..4c2bb27c99fe 100644 --- a/drivers/dpll/dpll_core.c +++ b/drivers/dpll/dpll_core.c @@ -564,7 +564,7 @@ void dpll_pin_put(struct dpll_pin *pin) xa_destroy(&pin->parent_refs); xa_erase(&dpll_pin_xa, pin->id); dpll_pin_prop_free(&pin->prop); - kfree(pin); + kfree_rcu(pin, rcu); } mutex_unlock(&dpll_lock); } diff --git a/drivers/dpll/dpll_core.h b/drivers/dpll/dpll_core.h index 717f715015c7..2b6d8ef1cdf3 100644 --- a/drivers/dpll/dpll_core.h +++ b/drivers/dpll/dpll_core.h @@ -47,6 +47,7 @@ struct dpll_device { * @prop: pin properties copied from the registerer * @rclk_dev_name: holds name of device when pin can recover clock from it * @refcount: refcount + * @rcu: rcu_head for kfree_rcu() **/ struct dpll_pin { u32 id; @@ -57,6 +58,7 @@ struct dpll_pin { struct xarray parent_refs; struct dpll_pin_properties prop; refcount_t refcount; + struct rcu_head rcu; }; /** diff --git a/include/linux/dpll.h b/include/linux/dpll.h index 9cf896ea1d41..4ec2fe9caf5a 100644 --- a/include/linux/dpll.h +++ b/include/linux/dpll.h @@ -10,6 +10,8 @@ #include #include #include +#include +#include struct dpll_device; struct dpll_pin; @@ -167,4 +169,13 @@ int dpll_device_change_ntf(struct dpll_device *dpll); int dpll_pin_change_ntf(struct dpll_pin *pin); +static inline struct dpll_pin *netdev_dpll_pin(const struct net_device *dev) +{ +#if IS_ENABLED(CONFIG_DPLL) + return rcu_dereference_rtnl(dev->dpll_pin); +#else + return NULL; +#endif +} + #endif diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ef7bfbb98497..a9c973b92294 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2469,7 +2469,7 @@ struct net_device { struct devlink_port *devlink_port; #if IS_ENABLED(CONFIG_DPLL) - struct dpll_pin *dpll_pin; + struct dpll_pin __rcu *dpll_pin; #endif #if IS_ENABLED(CONFIG_PAGE_POOL) /** @page_pools: page pools created for this netdevice */ @@ -4035,15 +4035,6 @@ bool netdev_port_same_parent_id(struct net_device *a, struct net_device *b); void netdev_dpll_pin_set(struct net_device *dev, struct dpll_pin *dpll_pin); void netdev_dpll_pin_clear(struct net_device *dev); -static inline struct dpll_pin *netdev_dpll_pin(const struct net_device *dev) -{ -#if IS_ENABLED(CONFIG_DPLL) - return dev->dpll_pin; -#else - return NULL; -#endif -} - struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev, bool *again); struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq, int *ret); diff --git a/net/core/dev.c b/net/core/dev.c index 73a021973007..0230391c78f7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -9078,7 +9078,7 @@ static void netdev_dpll_pin_assign(struct net_device *dev, struct dpll_pin *dpll { #if IS_ENABLED(CONFIG_DPLL) rtnl_lock(); - dev->dpll_pin = dpll_pin; + rcu_assign_pointer(dev->dpll_pin, dpll_pin); rtnl_unlock(); #endif } From 535d620ea5ff1a033dc64ee3d912acadc7470619 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Fri, 23 Feb 2024 17:14:11 +0100 Subject: [PATCH 26/73] mptcp: map v4 address to v6 when destroying subflow Address family of server side mismatches with that of client side, like in "userspace pm add & remove address" test: userspace_pm_add_addr $ns1 10.0.2.1 10 userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $SUB_ESTABLISHED That's because on the server side, the family is set to AF_INET6 and the v4 address is mapped in a v6 one. This patch fixes this issue. In mptcp_pm_nl_subflow_destroy_doit(), before checking local address family with remote address family, map an IPv4 address to an IPv6 address if the pair is a v4-mapped address. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/387 Fixes: 702c2f646d42 ("mptcp: netlink: allow userspace-driven subflow establishment") Cc: stable@vger.kernel.org Signed-off-by: Geliang Tang Reviewed-by: Mat Martineau Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-1-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_userspace.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c index d396a5973429..bc97cc30f013 100644 --- a/net/mptcp/pm_userspace.c +++ b/net/mptcp/pm_userspace.c @@ -495,6 +495,16 @@ int mptcp_pm_nl_subflow_destroy_doit(struct sk_buff *skb, struct genl_info *info goto destroy_err; } +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + if (addr_l.family == AF_INET && ipv6_addr_v4mapped(&addr_r.addr6)) { + ipv6_addr_set_v4mapped(addr_l.addr.s_addr, &addr_l.addr6); + addr_l.family = AF_INET6; + } + if (addr_r.family == AF_INET && ipv6_addr_v4mapped(&addr_l.addr6)) { + ipv6_addr_set_v4mapped(addr_r.addr.s_addr, &addr_r.addr6); + addr_r.family = AF_INET6; + } +#endif if (addr_l.family != addr_r.family) { GENL_SET_ERR_MSG(info, "address families do not match"); err = -EINVAL; From 7092dbee23282b6fcf1313fc64e2b92649ee16e8 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Fri, 23 Feb 2024 17:14:12 +0100 Subject: [PATCH 27/73] selftests: mptcp: rm subflow with v4/v4mapped addr Now both a v4 address and a v4-mapped address are supported when destroying a userspace pm subflow, this patch adds a second subflow to "userspace pm add & remove address" test, and two subflows could be removed two different ways, one with the v4mapped and one with v4. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/387 Fixes: 48d73f609dcc ("selftests: mptcp: update userspace pm addr tests") Cc: stable@vger.kernel.org Signed-off-by: Geliang Tang Reviewed-by: Mat Martineau Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-2-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- .../testing/selftests/net/mptcp/mptcp_join.sh | 28 +++++++++++-------- .../testing/selftests/net/mptcp/mptcp_lib.sh | 4 +-- 2 files changed, 18 insertions(+), 14 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index c07386e21e0a..e68b1bc2c2e4 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -3333,16 +3333,17 @@ userspace_pm_rm_sf() { local evts=$evts_ns1 local t=${3:-1} - local ip=4 + local ip local tk da dp sp local cnt [ "$1" == "$ns2" ] && evts=$evts_ns2 - if mptcp_lib_is_v6 $2; then ip=6; fi + [ -n "$(mptcp_lib_evts_get_info "saddr4" "$evts" $t)" ] && ip=4 + [ -n "$(mptcp_lib_evts_get_info "saddr6" "$evts" $t)" ] && ip=6 tk=$(mptcp_lib_evts_get_info token "$evts") - da=$(mptcp_lib_evts_get_info "daddr$ip" "$evts" $t) - dp=$(mptcp_lib_evts_get_info dport "$evts" $t) - sp=$(mptcp_lib_evts_get_info sport "$evts" $t) + da=$(mptcp_lib_evts_get_info "daddr$ip" "$evts" $t $2) + dp=$(mptcp_lib_evts_get_info dport "$evts" $t $2) + sp=$(mptcp_lib_evts_get_info sport "$evts" $t $2) cnt=$(rm_sf_count ${1}) ip netns exec $1 ./pm_nl_ctl dsf lip $2 lport $sp \ @@ -3429,20 +3430,23 @@ userspace_tests() if reset_with_events "userspace pm add & remove address" && continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then set_userspace_pm $ns1 - pm_nl_set_limits $ns2 1 1 + pm_nl_set_limits $ns2 2 2 speed=5 \ run_tests $ns1 $ns2 10.0.1.1 & local tests_pid=$! wait_mpj $ns1 userspace_pm_add_addr $ns1 10.0.2.1 10 - chk_join_nr 1 1 1 - chk_add_nr 1 1 - chk_mptcp_info subflows 1 subflows 1 - chk_subflows_total 2 2 - chk_mptcp_info add_addr_signal 1 add_addr_accepted 1 + userspace_pm_add_addr $ns1 10.0.3.1 20 + chk_join_nr 2 2 2 + chk_add_nr 2 2 + chk_mptcp_info subflows 2 subflows 2 + chk_subflows_total 3 3 + chk_mptcp_info add_addr_signal 2 add_addr_accepted 2 userspace_pm_rm_addr $ns1 10 userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $SUB_ESTABLISHED - chk_rm_nr 1 1 invert + userspace_pm_rm_addr $ns1 20 + userspace_pm_rm_sf $ns1 10.0.3.1 $SUB_ESTABLISHED + chk_rm_nr 2 2 invert chk_mptcp_info subflows 0 subflows 0 chk_subflows_total 1 1 kill_events_pids diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index 3a2abae5993e..3777d66fc56d 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -213,9 +213,9 @@ mptcp_lib_get_info_value() { grep "${2}" | sed -n 's/.*\('"${1}"':\)\([0-9a-f:.]*\).*$/\2/p;q' } -# $1: info name ; $2: evts_ns ; $3: event type +# $1: info name ; $2: evts_ns ; [$3: event type; [$4: addr]] mptcp_lib_evts_get_info() { - mptcp_lib_get_info_value "${1}" "^type:${3:-1}," < "${2}" + grep "${4:-}" "${2}" | mptcp_lib_get_info_value "${1}" "^type:${3:-1}," } # $1: PID From 5b49c41ac8f27aa3a63a1712b1f54f91015c18f2 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 23 Feb 2024 17:14:13 +0100 Subject: [PATCH 28/73] mptcp: avoid printing warning once on client side After the 'Fixes' commit mentioned below, the client side might print the following warning once when a subflow is fully established at the reception of any valid additional ack: MPTCP: bogus mpc option on established client sk That's a normal situation, and no warning should be printed for that. We can then skip the check when the label is used. Fixes: e4a0fa47e816 ("mptcp: corner case locking for rx path fields initialization") Cc: stable@vger.kernel.org Suggested-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-3-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mptcp/options.c b/net/mptcp/options.c index e3e96a49f922..63fc0758c22d 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -981,10 +981,10 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk, if (mp_opt->deny_join_id0) WRITE_ONCE(msk->pm.remote_deny_join_id0, true); -set_fully_established: if (unlikely(!READ_ONCE(msk->pm.server_side))) pr_warn_once("bogus mpc option on established client sk"); +set_fully_established: mptcp_data_lock((struct sock *)msk); __mptcp_subflow_fully_established(msk, subflow, mp_opt); mptcp_data_unlock((struct sock *)msk); From b9cd26f640a308ea314ad23532de9a8592cd09d2 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 23 Feb 2024 17:14:14 +0100 Subject: [PATCH 29/73] mptcp: push at DSS boundaries when inserting not contiguous data in the subflow write queue, the protocol creates a new skb and prevent the TCP stack from merging it later with already queued skbs by setting the EOR marker. Still no push flag is explicitly set at the end of previous GSO packet, making the aggregation on the receiver side sub-optimal - and packetdrill self-tests less predictable. Explicitly mark the end of not contiguous DSS with the push flag. Fixes: 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-4-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 948606a537da..442fa7d9b57a 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1260,6 +1260,7 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk, mpext = mptcp_get_ext(skb); if (!mptcp_skb_can_collapse_to(data_seq, skb, mpext)) { TCP_SKB_CB(skb)->eor = 1; + tcp_mark_push(tcp_sk(ssk), skb); goto alloc_skb; } From adf1bb78dab55e36d4d557aa2fb446ebcfe9e5ce Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 23 Feb 2024 17:14:15 +0100 Subject: [PATCH 30/73] mptcp: fix snd_wnd initialization for passive socket Such value should be inherited from the first subflow, but passive sockets always used 'rsk_rcv_wnd'. Fixes: 6f8a612a33e4 ("mptcp: keep track of advertised windows right edge") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-5-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 442fa7d9b57a..2c8f931c6d5b 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3211,7 +3211,7 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk, msk->write_seq = subflow_req->idsn + 1; msk->snd_nxt = msk->write_seq; msk->snd_una = msk->write_seq; - msk->wnd_end = msk->snd_nxt + req->rsk_rcv_wnd; + msk->wnd_end = msk->snd_nxt + tcp_sk(ssk)->snd_wnd; msk->setsockopt_seq = mptcp_sk(sk)->setsockopt_seq; mptcp_init_sched(msk, mptcp_sk(sk)->sched); From b111d8fbd2cbc63e05f3adfbbe0d4df655dfcc5b Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 23 Feb 2024 17:14:16 +0100 Subject: [PATCH 31/73] mptcp: fix potential wake-up event loss After the blamed commit below, the send buffer auto-tuning can happen after that the mptcp_propagate_sndbuf() completes - via the delegated action infrastructure. We must check for write space even after such change or we risk missing the wake-up event. Fixes: 8005184fd1ca ("mptcp: refactor sndbuf auto-tuning") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-6-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.h | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 631a7f445f34..07f6242afc1a 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -790,6 +790,16 @@ static inline bool mptcp_data_fin_enabled(const struct mptcp_sock *msk) READ_ONCE(msk->write_seq) == READ_ONCE(msk->snd_nxt); } +static inline void mptcp_write_space(struct sock *sk) +{ + if (sk_stream_is_writeable(sk)) { + /* pairs with memory barrier in mptcp_poll */ + smp_mb(); + if (test_and_clear_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags)) + sk_stream_write_space(sk); + } +} + static inline void __mptcp_sync_sndbuf(struct sock *sk) { struct mptcp_subflow_context *subflow; @@ -808,6 +818,7 @@ static inline void __mptcp_sync_sndbuf(struct sock *sk) /* the msk max wmem limit is * tcp wmem[2] */ WRITE_ONCE(sk->sk_sndbuf, new_sndbuf); + mptcp_write_space(sk); } /* The called held both the msk socket and the subflow socket locks, @@ -838,16 +849,6 @@ static inline void mptcp_propagate_sndbuf(struct sock *sk, struct sock *ssk) local_bh_enable(); } -static inline void mptcp_write_space(struct sock *sk) -{ - if (sk_stream_is_writeable(sk)) { - /* pairs with memory barrier in mptcp_poll */ - smp_mb(); - if (test_and_clear_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags)) - sk_stream_write_space(sk); - } -} - void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags); #define MPTCP_TOKEN_MAX_RETRIES 4 From 9480f388a2ef54fba911d9325372abd69a328601 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Fri, 23 Feb 2024 17:14:17 +0100 Subject: [PATCH 32/73] selftests: mptcp: join: add ss mptcp support check Commands 'ss -M' are used in script mptcp_join.sh to display only MPTCP sockets. So it must be checked if ss tool supports MPTCP in this script. Fixes: e274f7154008 ("selftests: mptcp: add subflow limits test-cases") Cc: stable@vger.kernel.org Signed-off-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-7-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_join.sh | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index e68b1bc2c2e4..e4581b0dfb96 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -161,6 +161,11 @@ check_tools() exit $ksft_skip fi + if ! ss -h | grep -q MPTCP; then + echo "SKIP: ss tool does not support MPTCP" + exit $ksft_skip + fi + # Use the legacy version if available to support old kernel versions if iptables-legacy -V &> /dev/null; then iptables="iptables-legacy" From 10048689def7e40a4405acda16fdc6477d4ecc5c Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 23 Feb 2024 17:14:18 +0100 Subject: [PATCH 33/73] mptcp: fix double-free on socket dismantle when MPTCP server accepts an incoming connection, it clones its listener socket. However, the pointer to 'inet_opt' for the new socket has the same value as the original one: as a consequence, on program exit it's possible to observe the following splat: BUG: KASAN: double-free in inet_sock_destruct+0x54f/0x8b0 Free of addr ffff888485950880 by task swapper/25/0 CPU: 25 PID: 0 Comm: swapper/25 Kdump: loaded Not tainted 6.8.0-rc1+ #609 Hardware name: Supermicro SYS-6027R-72RF/X9DRH-7TF/7F/iTF/iF, BIOS 3.0 07/26/2013 Call Trace: dump_stack_lvl+0x32/0x50 print_report+0xca/0x620 kasan_report_invalid_free+0x64/0x90 __kasan_slab_free+0x1aa/0x1f0 kfree+0xed/0x2e0 inet_sock_destruct+0x54f/0x8b0 __sk_destruct+0x48/0x5b0 rcu_do_batch+0x34e/0xd90 rcu_core+0x559/0xac0 __do_softirq+0x183/0x5a4 irq_exit_rcu+0x12d/0x170 sysvec_apic_timer_interrupt+0x6b/0x80 asm_sysvec_apic_timer_interrupt+0x16/0x20 RIP: 0010:cpuidle_enter_state+0x175/0x300 Code: 30 00 0f 84 1f 01 00 00 83 e8 01 83 f8 ff 75 e5 48 83 c4 18 44 89 e8 5b 5d 41 5c 41 5d 41 5e 41 5f c3 cc cc cc cc fb 45 85 ed <0f> 89 60 ff ff ff 48 c1 e5 06 48 c7 43 18 00 00 00 00 48 83 44 2b RSP: 0018:ffff888481cf7d90 EFLAGS: 00000202 RAX: 0000000000000000 RBX: ffff88887facddc8 RCX: 0000000000000000 RDX: 1ffff1110ff588b1 RSI: 0000000000000019 RDI: ffff88887fac4588 RBP: 0000000000000004 R08: 0000000000000002 R09: 0000000000043080 R10: 0009b02ea273363f R11: ffff88887fabf42b R12: ffffffff932592e0 R13: 0000000000000004 R14: 0000000000000000 R15: 00000022c880ec80 cpuidle_enter+0x4a/0xa0 do_idle+0x310/0x410 cpu_startup_entry+0x51/0x60 start_secondary+0x211/0x270 secondary_startup_64_no_verify+0x184/0x18b Allocated by task 6853: kasan_save_stack+0x1c/0x40 kasan_save_track+0x10/0x30 __kasan_kmalloc+0xa6/0xb0 __kmalloc+0x1eb/0x450 cipso_v4_sock_setattr+0x96/0x360 netlbl_sock_setattr+0x132/0x1f0 selinux_netlbl_socket_post_create+0x6c/0x110 selinux_socket_post_create+0x37b/0x7f0 security_socket_post_create+0x63/0xb0 __sock_create+0x305/0x450 __sys_socket_create.part.23+0xbd/0x130 __sys_socket+0x37/0xb0 __x64_sys_socket+0x6f/0xb0 do_syscall_64+0x83/0x160 entry_SYSCALL_64_after_hwframe+0x6e/0x76 Freed by task 6858: kasan_save_stack+0x1c/0x40 kasan_save_track+0x10/0x30 kasan_save_free_info+0x3b/0x60 __kasan_slab_free+0x12c/0x1f0 kfree+0xed/0x2e0 inet_sock_destruct+0x54f/0x8b0 __sk_destruct+0x48/0x5b0 subflow_ulp_release+0x1f0/0x250 tcp_cleanup_ulp+0x6e/0x110 tcp_v4_destroy_sock+0x5a/0x3a0 inet_csk_destroy_sock+0x135/0x390 tcp_fin+0x416/0x5c0 tcp_data_queue+0x1bc8/0x4310 tcp_rcv_state_process+0x15a3/0x47b0 tcp_v4_do_rcv+0x2c1/0x990 tcp_v4_rcv+0x41fb/0x5ed0 ip_protocol_deliver_rcu+0x6d/0x9f0 ip_local_deliver_finish+0x278/0x360 ip_local_deliver+0x182/0x2c0 ip_rcv+0xb5/0x1c0 __netif_receive_skb_one_core+0x16e/0x1b0 process_backlog+0x1e3/0x650 __napi_poll+0xa6/0x500 net_rx_action+0x740/0xbb0 __do_softirq+0x183/0x5a4 The buggy address belongs to the object at ffff888485950880 which belongs to the cache kmalloc-64 of size 64 The buggy address is located 0 bytes inside of 64-byte region [ffff888485950880, ffff8884859508c0) The buggy address belongs to the physical page: page:0000000056d1e95e refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff888485950700 pfn:0x485950 flags: 0x57ffffc0000800(slab|node=1|zone=2|lastcpupid=0x1fffff) page_type: 0xffffffff() raw: 0057ffffc0000800 ffff88810004c640 ffffea00121b8ac0 dead000000000006 raw: ffff888485950700 0000000000200019 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff888485950780: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc ffff888485950800: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc >ffff888485950880: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc ^ ffff888485950900: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc ffff888485950980: 00 00 00 00 00 01 fc fc fc fc fc fc fc fc fc fc Something similar (a refcount underflow) happens with CALIPSO/IPv6. Fix this by duplicating IP / IPv6 options after clone, so that ip{,6}_sock_destruct() doesn't end up freeing the same memory area twice. Fixes: cf7da0d66cc1 ("mptcp: Create SUBFLOW socket for incoming connections") Cc: stable@vger.kernel.org Signed-off-by: Davide Caratti Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-8-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 49 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 2c8f931c6d5b..7833a49f6214 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3178,8 +3178,50 @@ static struct ipv6_pinfo *mptcp_inet6_sk(const struct sock *sk) return (struct ipv6_pinfo *)(((u8 *)sk) + offset); } + +static void mptcp_copy_ip6_options(struct sock *newsk, const struct sock *sk) +{ + const struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6_txoptions *opt; + struct ipv6_pinfo *newnp; + + newnp = inet6_sk(newsk); + + rcu_read_lock(); + opt = rcu_dereference(np->opt); + if (opt) { + opt = ipv6_dup_options(newsk, opt); + if (!opt) + net_warn_ratelimited("%s: Failed to copy ip6 options\n", __func__); + } + RCU_INIT_POINTER(newnp->opt, opt); + rcu_read_unlock(); +} #endif +static void mptcp_copy_ip_options(struct sock *newsk, const struct sock *sk) +{ + struct ip_options_rcu *inet_opt, *newopt = NULL; + const struct inet_sock *inet = inet_sk(sk); + struct inet_sock *newinet; + + newinet = inet_sk(newsk); + + rcu_read_lock(); + inet_opt = rcu_dereference(inet->inet_opt); + if (inet_opt) { + newopt = sock_kmalloc(newsk, sizeof(*inet_opt) + + inet_opt->opt.optlen, GFP_ATOMIC); + if (newopt) + memcpy(newopt, inet_opt, sizeof(*inet_opt) + + inet_opt->opt.optlen); + else + net_warn_ratelimited("%s: Failed to copy ip options\n", __func__); + } + RCU_INIT_POINTER(newinet->inet_opt, newopt); + rcu_read_unlock(); +} + struct sock *mptcp_sk_clone_init(const struct sock *sk, const struct mptcp_options_received *mp_opt, struct sock *ssk, @@ -3200,6 +3242,13 @@ struct sock *mptcp_sk_clone_init(const struct sock *sk, __mptcp_init_sock(nsk); +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + if (nsk->sk_family == AF_INET6) + mptcp_copy_ip6_options(nsk, sk); + else +#endif + mptcp_copy_ip_options(nsk, sk); + msk = mptcp_sk(nsk); msk->local_key = subflow_req->local_key; msk->token = subflow_req->token; From d6a9608af9a75d13243d217f6ce1e30e57d56ffe Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 23 Feb 2024 17:14:19 +0100 Subject: [PATCH 34/73] mptcp: fix possible deadlock in subflow diag Syzbot and Eric reported a lockdep splat in the subflow diag: WARNING: possible circular locking dependency detected 6.8.0-rc4-syzkaller-00212-g40b9385dd8e6 #0 Not tainted syz-executor.2/24141 is trying to acquire lock: ffff888045870130 (k-sk_lock-AF_INET6){+.+.}-{0:0}, at: tcp_diag_put_ulp net/ipv4/tcp_diag.c:100 [inline] ffff888045870130 (k-sk_lock-AF_INET6){+.+.}-{0:0}, at: tcp_diag_get_aux+0x738/0x830 net/ipv4/tcp_diag.c:137 but task is already holding lock: ffffc9000135e488 (&h->lhash2[i].lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline] ffffc9000135e488 (&h->lhash2[i].lock){+.+.}-{2:2}, at: inet_diag_dump_icsk+0x39f/0x1f80 net/ipv4/inet_diag.c:1038 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&h->lhash2[i].lock){+.+.}-{2:2}: lock_acquire+0x1e3/0x530 kernel/locking/lockdep.c:5754 __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline] _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154 spin_lock include/linux/spinlock.h:351 [inline] __inet_hash+0x335/0xbe0 net/ipv4/inet_hashtables.c:743 inet_csk_listen_start+0x23a/0x320 net/ipv4/inet_connection_sock.c:1261 __inet_listen_sk+0x2a2/0x770 net/ipv4/af_inet.c:217 inet_listen+0xa3/0x110 net/ipv4/af_inet.c:239 rds_tcp_listen_init+0x3fd/0x5a0 net/rds/tcp_listen.c:316 rds_tcp_init_net+0x141/0x320 net/rds/tcp.c:577 ops_init+0x352/0x610 net/core/net_namespace.c:136 __register_pernet_operations net/core/net_namespace.c:1214 [inline] register_pernet_operations+0x2cb/0x660 net/core/net_namespace.c:1283 register_pernet_device+0x33/0x80 net/core/net_namespace.c:1370 rds_tcp_init+0x62/0xd0 net/rds/tcp.c:735 do_one_initcall+0x238/0x830 init/main.c:1236 do_initcall_level+0x157/0x210 init/main.c:1298 do_initcalls+0x3f/0x80 init/main.c:1314 kernel_init_freeable+0x42f/0x5d0 init/main.c:1551 kernel_init+0x1d/0x2a0 init/main.c:1441 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1b/0x30 arch/x86/entry/entry_64.S:242 -> #0 (k-sk_lock-AF_INET6){+.+.}-{0:0}: check_prev_add kernel/locking/lockdep.c:3134 [inline] check_prevs_add kernel/locking/lockdep.c:3253 [inline] validate_chain+0x18ca/0x58e0 kernel/locking/lockdep.c:3869 __lock_acquire+0x1345/0x1fd0 kernel/locking/lockdep.c:5137 lock_acquire+0x1e3/0x530 kernel/locking/lockdep.c:5754 lock_sock_fast include/net/sock.h:1723 [inline] subflow_get_info+0x166/0xd20 net/mptcp/diag.c:28 tcp_diag_put_ulp net/ipv4/tcp_diag.c:100 [inline] tcp_diag_get_aux+0x738/0x830 net/ipv4/tcp_diag.c:137 inet_sk_diag_fill+0x10ed/0x1e00 net/ipv4/inet_diag.c:345 inet_diag_dump_icsk+0x55b/0x1f80 net/ipv4/inet_diag.c:1061 __inet_diag_dump+0x211/0x3a0 net/ipv4/inet_diag.c:1263 inet_diag_dump_compat+0x1c1/0x2d0 net/ipv4/inet_diag.c:1371 netlink_dump+0x59b/0xc80 net/netlink/af_netlink.c:2264 __netlink_dump_start+0x5df/0x790 net/netlink/af_netlink.c:2370 netlink_dump_start include/linux/netlink.h:338 [inline] inet_diag_rcv_msg_compat+0x209/0x4c0 net/ipv4/inet_diag.c:1405 sock_diag_rcv_msg+0xe7/0x410 netlink_rcv_skb+0x1e3/0x430 net/netlink/af_netlink.c:2543 sock_diag_rcv+0x2a/0x40 net/core/sock_diag.c:280 netlink_unicast_kernel net/netlink/af_netlink.c:1341 [inline] netlink_unicast+0x7ea/0x980 net/netlink/af_netlink.c:1367 netlink_sendmsg+0xa3b/0xd70 net/netlink/af_netlink.c:1908 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x221/0x270 net/socket.c:745 ____sys_sendmsg+0x525/0x7d0 net/socket.c:2584 ___sys_sendmsg net/socket.c:2638 [inline] __sys_sendmsg+0x2b0/0x3a0 net/socket.c:2667 do_syscall_64+0xf9/0x240 entry_SYSCALL_64_after_hwframe+0x6f/0x77 As noted by Eric we can break the lock dependency chain avoid dumping any extended info for the mptcp subflow listener: nothing actually useful is presented there. Fixes: b8adb69a7d29 ("mptcp: fix lockless access in subflow ULP diag") Cc: stable@vger.kernel.org Reported-by: Eric Dumazet Closes: https://lore.kernel.org/netdev/CANn89iJ=Oecw6OZDwmSYc9HJKQ_G32uN11L+oUcMu+TOD5Xiaw@mail.gmail.com/ Suggested-by: Eric Dumazet Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-9-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/diag.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/mptcp/diag.c b/net/mptcp/diag.c index 6ff6f14674aa..7017dd60659d 100644 --- a/net/mptcp/diag.c +++ b/net/mptcp/diag.c @@ -21,6 +21,9 @@ static int subflow_get_info(struct sock *sk, struct sk_buff *skb) bool slow; int err; + if (inet_sk_state_load(sk) == TCP_LISTEN) + return 0; + start = nla_nest_start_noflag(skb, INET_ULP_INFO_MPTCP); if (!start) return -EMSGSIZE; From b4b51d36bbaa3ddb93b3e1ca3a1ef0aa629d6521 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 23 Feb 2024 17:14:20 +0100 Subject: [PATCH 35/73] selftests: mptcp: explicitly trigger the listener diag code-path The mptcp diag interface already experienced a few locking bugs that lockdep and appropriate coverage have detected in advance. Let's add a test-case triggering the relevant code path, to prevent similar issues in the future. Be careful to cope with very slow environments. Note that we don't need an explicit timeout on the mptcp_connect subprocess to cope with eventual bug/hang-up as the final cleanup terminating the child processes will take care of that. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240223-upstream-net-20240223-misc-fixes-v1-10-162e87e48497@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/diag.sh | 30 ++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/mptcp/diag.sh b/tools/testing/selftests/net/mptcp/diag.sh index 0a58ebb8b04c..f300f4e1eb59 100755 --- a/tools/testing/selftests/net/mptcp/diag.sh +++ b/tools/testing/selftests/net/mptcp/diag.sh @@ -20,7 +20,7 @@ flush_pids() ip netns pids "${ns}" | xargs --no-run-if-empty kill -SIGUSR1 &>/dev/null - for _ in $(seq 10); do + for _ in $(seq $((timeout_poll * 10))); do [ -z "$(ip netns pids "${ns}")" ] && break sleep 0.1 done @@ -91,6 +91,15 @@ chk_msk_nr() __chk_msk_nr "grep -c token:" "$@" } +chk_listener_nr() +{ + local expected=$1 + local msg="$2" + + __chk_nr "ss -inmlHMON $ns | wc -l" "$expected" "$msg - mptcp" 0 + __chk_nr "ss -inmlHtON $ns | wc -l" "$expected" "$msg - subflows" +} + wait_msk_nr() { local condition="grep -c token:" @@ -289,5 +298,24 @@ flush_pids chk_msk_inuse 0 "many->0" chk_msk_cestab 0 "many->0" +chk_listener_nr 0 "no listener sockets" +NR_SERVERS=100 +for I in $(seq 1 $NR_SERVERS); do + ip netns exec $ns ./mptcp_connect -p $((I + 20001)) \ + -t ${timeout_poll} -l 0.0.0.0 >/dev/null 2>&1 & +done + +for I in $(seq 1 $NR_SERVERS); do + mptcp_lib_wait_local_port_listen $ns $((I + 20001)) +done + +chk_listener_nr $NR_SERVERS "many listener sockets" + +# graceful termination +for I in $(seq 1 $NR_SERVERS); do + echo a | ip netns exec $ns ./mptcp_connect -p $((I + 20001)) 127.0.0.1 >/dev/null 2>&1 & +done +flush_pids + mptcp_lib_result_print_all_tap exit $ret From a36b0787f074d7441f66c172745653570e09c320 Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Fri, 23 Feb 2024 14:27:40 -0800 Subject: [PATCH 36/73] ionic: check before releasing pci regions AER recovery handler can trigger a PCI Reset after tearing down the device setup in the error detection handler. The PCI Reset handler will also attempt to tear down the device setup, and this second tear down needs to know that it doesn't need to call pci_release_regions() a second time. We can clear num_bars on tear down and use that to decide later if we need to clear the resources. This prevents a harmless but disturbing warning message resource: Trying to free nonexistent resource <0xXXXXXXXXXX-0xXXXXXXXXXX> Fixes: c3a910e1c47a ("ionic: fill out pci error handlers") Reviewed-by: Brett Creeley Signed-off-by: Shannon Nelson Signed-off-by: Paolo Abeni --- .../net/ethernet/pensando/ionic/ionic_bus_pci.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c index 10a9d80db32c..6ba8d4aca0a0 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c @@ -93,6 +93,7 @@ static void ionic_unmap_bars(struct ionic *ionic) bars[i].len = 0; } } + ionic->num_bars = 0; } void __iomem *ionic_bus_map_dbpage(struct ionic *ionic, int page_num) @@ -215,13 +216,15 @@ static int ionic_sriov_configure(struct pci_dev *pdev, int num_vfs) static void ionic_clear_pci(struct ionic *ionic) { - ionic->idev.dev_info_regs = NULL; - ionic->idev.dev_cmd_regs = NULL; - ionic->idev.intr_status = NULL; - ionic->idev.intr_ctrl = NULL; + if (ionic->num_bars) { + ionic->idev.dev_info_regs = NULL; + ionic->idev.dev_cmd_regs = NULL; + ionic->idev.intr_status = NULL; + ionic->idev.intr_ctrl = NULL; - ionic_unmap_bars(ionic); - pci_release_regions(ionic->pdev); + ionic_unmap_bars(ionic); + pci_release_regions(ionic->pdev); + } if (pci_is_enabled(ionic->pdev)) pci_disable_device(ionic->pdev); From 7662fad348ac54120e9e6443cb0bbe4f3b582219 Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Fri, 23 Feb 2024 14:27:41 -0800 Subject: [PATCH 37/73] ionic: check cmd_regs before copying in or out Since we now have potential cases of NULL cmd_regs and info_regs during a reset recovery, and left NULL if a reset recovery has failed, we need to check that they exist before we use them. Most of the cases were covered in the original patch where we verify before doing the ioreadb() for health or cmd status. However, we need to protect a few uses of io mem that could be hit in error recovery or asynchronous threads calls as well (e.g. ethtool or devlink handlers). Fixes: 219e183272b4 ("ionic: no fw read when PCI reset failed") Reviewed-by: Brett Creeley Signed-off-by: Shannon Nelson Signed-off-by: Paolo Abeni --- drivers/net/ethernet/pensando/ionic/ionic_dev.c | 10 ++++++++++ drivers/net/ethernet/pensando/ionic/ionic_ethtool.c | 7 ++++++- drivers/net/ethernet/pensando/ionic/ionic_fw.c | 5 +++++ drivers/net/ethernet/pensando/ionic/ionic_main.c | 3 +++ 4 files changed, 24 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index 1e7c71f7f081..746072b4dbd0 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -319,22 +319,32 @@ int ionic_heartbeat_check(struct ionic *ionic) u8 ionic_dev_cmd_status(struct ionic_dev *idev) { + if (!idev->dev_cmd_regs) + return (u8)PCI_ERROR_RESPONSE; return ioread8(&idev->dev_cmd_regs->comp.comp.status); } bool ionic_dev_cmd_done(struct ionic_dev *idev) { + if (!idev->dev_cmd_regs) + return false; return ioread32(&idev->dev_cmd_regs->done) & IONIC_DEV_CMD_DONE; } void ionic_dev_cmd_comp(struct ionic_dev *idev, union ionic_dev_cmd_comp *comp) { + if (!idev->dev_cmd_regs) + return; memcpy_fromio(comp, &idev->dev_cmd_regs->comp, sizeof(*comp)); } void ionic_dev_cmd_go(struct ionic_dev *idev, union ionic_dev_cmd *cmd) { idev->opcode = cmd->cmd.opcode; + + if (!idev->dev_cmd_regs) + return; + memcpy_toio(&idev->dev_cmd_regs->cmd, cmd, sizeof(*cmd)); iowrite32(0, &idev->dev_cmd_regs->done); iowrite32(1, &idev->dev_cmd_regs->doorbell); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c index cd3c0b01402e..0ffc9c4904ac 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c @@ -90,18 +90,23 @@ static void ionic_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) { struct ionic_lif *lif = netdev_priv(netdev); + struct ionic_dev *idev; unsigned int offset; unsigned int size; regs->version = IONIC_DEV_CMD_REG_VERSION; + idev = &lif->ionic->idev; + if (!idev->dev_info_regs) + return; + offset = 0; size = IONIC_DEV_INFO_REG_COUNT * sizeof(u32); memcpy_fromio(p + offset, lif->ionic->idev.dev_info_regs->words, size); offset += size; size = IONIC_DEV_CMD_REG_COUNT * sizeof(u32); - memcpy_fromio(p + offset, lif->ionic->idev.dev_cmd_regs->words, size); + memcpy_fromio(p + offset, idev->dev_cmd_regs->words, size); } static void ionic_get_link_ext_stats(struct net_device *netdev, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_fw.c b/drivers/net/ethernet/pensando/ionic/ionic_fw.c index 5f40324cd243..3c209c1a2337 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_fw.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_fw.c @@ -109,6 +109,11 @@ int ionic_firmware_update(struct ionic_lif *lif, const struct firmware *fw, dl = priv_to_devlink(ionic); devlink_flash_update_status_notify(dl, "Preparing to flash", NULL, 0, 0); + if (!idev->dev_cmd_regs) { + err = -ENXIO; + goto err_out; + } + buf_sz = sizeof(idev->dev_cmd_regs->data); netdev_dbg(netdev, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c index 165ab08ad2dd..2f479de329fe 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_main.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c @@ -416,6 +416,9 @@ static void ionic_dev_cmd_clean(struct ionic *ionic) { struct ionic_dev *idev = &ionic->idev; + if (!idev->dev_cmd_regs) + return; + iowrite32(0, &idev->dev_cmd_regs->doorbell); memset_io(&idev->dev_cmd_regs->cmd, 0, sizeof(idev->dev_cmd_regs->cmd)); } From 155a1efc9b96a39857714aff49a11ebc93022c8c Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Fri, 23 Feb 2024 14:27:42 -0800 Subject: [PATCH 38/73] ionic: restore netdev feature bits after reset When rebuilding the lif after an FLR, be sure to restore the current netdev features, not do the usual first time feature init. This prevents losing user changes to things like TSO or vlan tagging states. Fixes: 45b84188a0a4 ("ionic: keep filters across FLR") Reviewed-by: Brett Creeley Signed-off-by: Shannon Nelson Signed-off-by: Paolo Abeni --- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index cf2d5ad7b68c..fcb44ceeb6aa 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -3559,7 +3559,10 @@ int ionic_lif_init(struct ionic_lif *lif) goto err_out_notifyq_deinit; } - err = ionic_init_nic_features(lif); + if (test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + err = ionic_set_nic_features(lif, lif->netdev->features); + else + err = ionic_init_nic_features(lif); if (err) goto err_out_notifyq_deinit; From 1ce7d306ea63f3e379557c79abd88052e0483813 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 23 Feb 2024 15:59:08 -0800 Subject: [PATCH 39/73] veth: try harder when allocating queue memory struct veth_rq is pretty large, 832B total without debug options enabled. Since commit under Fixes we try to pre-allocate enough queues for every possible CPU. Miao Wang reports that this may lead to order-5 allocations which will fail in production. Let the allocation fallback to vmalloc() and try harder. These are the same flags we pass to netdev queue allocation. Reported-and-tested-by: Miao Wang Fixes: 9d3684c24a52 ("veth: create by default nr_possible_cpus queues") Link: https://lore.kernel.org/all/5F52CAE2-2FB7-4712-95F1-3312FBBFA8DD@gmail.com/ Signed-off-by: Jakub Kicinski Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240223235908.693010-1-kuba@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/veth.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index a786be805709..cd4a6fe458f9 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -1461,7 +1461,8 @@ static int veth_alloc_queues(struct net_device *dev) struct veth_priv *priv = netdev_priv(dev); int i; - priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL_ACCOUNT); + priv->rq = kvcalloc(dev->num_rx_queues, sizeof(*priv->rq), + GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL); if (!priv->rq) return -ENOMEM; @@ -1477,7 +1478,7 @@ static void veth_free_queues(struct net_device *dev) { struct veth_priv *priv = netdev_priv(dev); - kfree(priv->rq); + kvfree(priv->rq); } static int veth_dev_init(struct net_device *dev) From c68b2c9eba38ec3f60f4894b189090febf4d8d22 Mon Sep 17 00:00:00 2001 From: Javier Carrasco Date: Sun, 25 Feb 2024 00:20:06 +0100 Subject: [PATCH 40/73] net: usb: dm9601: fix wrong return value in dm9601_mdio_read The MII code does not check the return value of mdio_read (among others), and therefore no error code should be sent. A previous fix to the use of an uninitialized variable propagates negative error codes, that might lead to wrong operations by the MII library. An example of such issues is the use of mii_nway_restart by the dm9601 driver. The mii_nway_restart function does not check the value returned by mdio_read, which in this case might be a negative number which could contain the exact bit the function checks (BMCR_ANENABLE = 0x1000). Return zero in case of error, as it is common practice in users of mdio_read to avoid wrong uses of the return value. Fixes: 8f8abb863fa5 ("net: usb: dm9601: fix uninitialized variable use in dm9601_mdio_read") Signed-off-by: Javier Carrasco Reviewed-by: Simon Horman Reviewed-by: Peter Korsgaard Link: https://lore.kernel.org/r/20240225-dm9601_ret_err-v1-1-02c1d959ea59@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/dm9601.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c index 99ec1d4a972d..8b6d6a1b3c2e 100644 --- a/drivers/net/usb/dm9601.c +++ b/drivers/net/usb/dm9601.c @@ -232,7 +232,7 @@ static int dm9601_mdio_read(struct net_device *netdev, int phy_id, int loc) err = dm_read_shared_word(dev, 1, loc, &res); if (err < 0) { netdev_err(dev->net, "MDIO read error: %d\n", err); - return err; + return 0; } netdev_dbg(dev->net, From f72a1994698ebe1a9eeadba7bfe2aa01afcde0a4 Mon Sep 17 00:00:00 2001 From: Kurt Kanzenbach Date: Sun, 25 Feb 2024 12:38:37 +0100 Subject: [PATCH 41/73] net: stmmac: Complete meta data only when enabled Currently using plain XDP/ZC sockets on stmmac results in a kernel crash: |[ 255.822584] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 |[...] |[ 255.822764] Call trace: |[ 255.822766] stmmac_tx_clean.constprop.0+0x848/0xc38 The program counter indicates xsk_tx_metadata_complete(). It works on compl->tx_timestamp, which is not set by xsk_tx_metadata_to_compl() due to missing meta data. Therefore, call xsk_tx_metadata_complete() only when meta data is actually used. Tested on imx93 without XDP, with XDP and with XDP/ZC. Fixes: 1347b419318d ("net: stmmac: Add Tx HWTS support to XDP ZC") Suggested-by: Serge Semin Tested-by: Serge Semin Link: https://lore.kernel.org/netdev/87r0h7wg8u.fsf@kurt.kurt.home/ Acked-by: Stanislav Fomichev Signed-off-by: Kurt Kanzenbach Link: https://lore.kernel.org/r/20240222-stmmac_xdp-v2-1-4beee3a037e4@linutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index e80d77bd9f1f..8b77c0952071 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2672,7 +2672,8 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue, } if (skb) { stmmac_get_tx_hwtstamp(priv, p, skb); - } else { + } else if (tx_q->xsk_pool && + xp_tx_metadata_enabled(tx_q->xsk_pool)) { struct stmmac_xsk_tx_complete tx_compl = { .priv = priv, .desc = p, From e3d5d70cb483df8296dd44e9ae3b6355ef86494c Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 26 Feb 2024 12:08:20 +0100 Subject: [PATCH 42/73] net: lan78xx: fix "softirq work is pending" error Disable BH around the call to napi_schedule() to avoid following error: NOHZ tick-stop error: local softirq work is pending, handler #08!!! Fixes: ec4c7e12396b ("lan78xx: Introduce NAPI polling support") Signed-off-by: Oleksij Rempel Link: https://lore.kernel.org/r/20240226110820.2113584-1-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/usb/lan78xx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index e06193e831a5..ba6c8ac2a736 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -1501,7 +1501,9 @@ static int lan78xx_link_reset(struct lan78xx_net *dev) lan78xx_rx_urb_submit_all(dev); + local_bh_disable(); napi_schedule(&dev->napi); + local_bh_enable(); } return 0; From 6a2008641920a9c6fe1abbeb9acbec463215d505 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Mon, 26 Feb 2024 13:49:21 +0100 Subject: [PATCH 43/73] uapi: in6: replace temporary label with rfc9486 Not really a fix per se, but IPV6_TLV_IOAM is still tagged as "TEMPORARY IANA allocation for IOAM", while RFC 9486 is available for some time now. Just update the reference. Fixes: 9ee11f0fff20 ("ipv6: ioam: Data plane support for Pre-allocated Trace") Signed-off-by: Justin Iurman Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240226124921.9097-1-justin.iurman@uliege.be Signed-off-by: Jakub Kicinski --- include/uapi/linux/in6.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/in6.h b/include/uapi/linux/in6.h index c4c53a9ab959..ff8d21f9e95b 100644 --- a/include/uapi/linux/in6.h +++ b/include/uapi/linux/in6.h @@ -145,7 +145,7 @@ struct in6_flowlabel_req { #define IPV6_TLV_PADN 1 #define IPV6_TLV_ROUTERALERT 5 #define IPV6_TLV_CALIPSO 7 /* RFC 5570 */ -#define IPV6_TLV_IOAM 49 /* TEMPORARY IANA allocation for IOAM */ +#define IPV6_TLV_IOAM 49 /* RFC 9486 */ #define IPV6_TLV_JUMBO 194 #define IPV6_TLV_HAO 201 /* home address option */ From 943d4bd67950685901addfa7b07aa3408ce17e7f Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 25 Feb 2024 23:48:20 -0800 Subject: [PATCH 44/73] net: ethernet: adi: move PHYLIB from vendor to driver symbol In a previous patch I added "select PHYLIB" at the wrong place for the ADIN1110 driver symbol, so move it to its correct place under the ADIN1110 kconfig symbol. Fixes: a9f80df4f514 ("net: ethernet: adi: requires PHYLIB support") Signed-off-by: Randy Dunlap Reported-by: Michal Kubecek Closes: https://lore.kernel.org/lkml/77012b38-4b49-47f4-9a88-d773d52909ad@infradead.org/T/#m8ba397484738711edc0ad607b2c63ca02244e3c3 Cc: Lennart Franzen Cc: "David S. Miller" Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Cc: netdev@vger.kernel.org Cc: Nuno Sa Tested-by: Michal Kubecek Signed-off-by: David S. Miller --- drivers/net/ethernet/adi/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/adi/Kconfig b/drivers/net/ethernet/adi/Kconfig index c91b4dcef4ec..760a9a60bc15 100644 --- a/drivers/net/ethernet/adi/Kconfig +++ b/drivers/net/ethernet/adi/Kconfig @@ -7,7 +7,6 @@ config NET_VENDOR_ADI bool "Analog Devices devices" default y depends on SPI - select PHYLIB help If you have a network (Ethernet) card belonging to this class, say Y. @@ -22,6 +21,7 @@ config ADIN1110 tristate "Analog Devices ADIN1110 MAC-PHY" depends on SPI && NET_SWITCHDEV select CRC8 + select PHYLIB help Say yes here to build support for Analog Devices ADIN1110 Low Power 10BASE-T1L Ethernet MAC-PHY. From 995161edfdb830fd83a652495324e26e0a4bd694 Mon Sep 17 00:00:00 2001 From: Lukasz Majewski Date: Mon, 26 Feb 2024 16:09:54 +0100 Subject: [PATCH 45/73] net: hsr: Fix typo in the hsr_forward_do() function comment Correct type in the hsr_forward_do() comment. Signed-off-by: Lukasz Majewski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- net/hsr/hsr_forward.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 80cdc6f6b34c..2afe28712a7a 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -435,7 +435,7 @@ static void hsr_forward_do(struct hsr_frame_info *frame) continue; /* Don't send frame over port where it has been sent before. - * Also fro SAN, this shouldn't be done. + * Also for SAN, this shouldn't be done. */ if (!frame->is_from_san && hsr_register_frame_out(port, frame->node_src, From 8af411bbba1f457c33734795f024d0ef26d0963f Mon Sep 17 00:00:00 2001 From: Jakub Raczynski Date: Mon, 26 Feb 2024 17:42:32 +0100 Subject: [PATCH 46/73] stmmac: Clear variable when destroying workqueue Currently when suspending driver and stopping workqueue it is checked whether workqueue is not NULL and if so, it is destroyed. Function destroy_workqueue() does drain queue and does clear variable, but it does not set workqueue variable to NULL. This can cause kernel/module panic if code attempts to clear workqueue that was not initialized. This scenario is possible when resuming suspended driver in stmmac_resume(), because there is no handling for failed stmmac_hw_setup(), which can fail and return if DMA engine has failed to initialize, and workqueue is initialized after DMA engine. Should DMA engine fail to initialize, resume will proceed normally, but interface won't work and TX queue will eventually timeout, causing 'Reset adapter' error. This then does destroy workqueue during reset process. And since workqueue is initialized after DMA engine and can be skipped, it will cause kernel/module panic. To secure against this possible crash, set workqueue variable to NULL when destroying workqueue. Log/backtrace from crash goes as follows: [88.031977]------------[ cut here ]------------ [88.031985]NETDEV WATCHDOG: eth0 (sxgmac): transmit queue 1 timed out [88.032017]WARNING: CPU: 0 PID: 0 at net/sched/sch_generic.c:477 dev_watchdog+0x390/0x398 [88.032251]---[ end trace e70de432e4d5c2c0 ]--- [88.032282]sxgmac 16d88000.ethernet eth0: Reset adapter. [88.036359]------------[ cut here ]------------ [88.036519]Call trace: [88.036523] flush_workqueue+0x3e4/0x430 [88.036528] drain_workqueue+0xc4/0x160 [88.036533] destroy_workqueue+0x40/0x270 [88.036537] stmmac_fpe_stop_wq+0x4c/0x70 [88.036541] stmmac_release+0x278/0x280 [88.036546] __dev_close_many+0xcc/0x158 [88.036551] dev_close_many+0xbc/0x190 [88.036555] dev_close.part.0+0x70/0xc0 [88.036560] dev_close+0x24/0x30 [88.036564] stmmac_service_task+0x110/0x140 [88.036569] process_one_work+0x1d8/0x4a0 [88.036573] worker_thread+0x54/0x408 [88.036578] kthread+0x164/0x170 [88.036583] ret_from_fork+0x10/0x20 [88.036588]---[ end trace e70de432e4d5c2c1 ]--- [88.036597]Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004 Fixes: 5a5586112b929 ("net: stmmac: support FPE link partner hand-shaking procedure") Signed-off-by: Jakub Raczynski Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 8b77c0952071..7c6aef033a45 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4006,8 +4006,10 @@ static void stmmac_fpe_stop_wq(struct stmmac_priv *priv) { set_bit(__FPE_REMOVING, &priv->fpe_task_state); - if (priv->fpe_wq) + if (priv->fpe_wq) { destroy_workqueue(priv->fpe_wq); + priv->fpe_wq = NULL; + } netdev_info(priv->dev, "FPE workqueue stop"); } From 4adfc94d4aeca1177e1188ba83c20ed581523fe1 Mon Sep 17 00:00:00 2001 From: Haiyue Wang Date: Tue, 27 Feb 2024 01:09:16 +0800 Subject: [PATCH 47/73] Documentations: correct net_cachelines title for struct inet_sock The fast path usage breakdown describes the detail for 'inet_sock', fix the markup title. Signed-off-by: Haiyue Wang Signed-off-by: David S. Miller --- Documentation/networking/net_cachelines/inet_sock.rst | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/networking/net_cachelines/inet_sock.rst b/Documentation/networking/net_cachelines/inet_sock.rst index a2babd0d7954..595d7ef5fc8b 100644 --- a/Documentation/networking/net_cachelines/inet_sock.rst +++ b/Documentation/networking/net_cachelines/inet_sock.rst @@ -1,9 +1,9 @@ .. SPDX-License-Identifier: GPL-2.0 .. Copyright (C) 2023 Google LLC -===================================================== -inet_connection_sock struct fast path usage breakdown -===================================================== +========================================== +inet_sock struct fast path usage breakdown +========================================== Type Name fastpath_tx_access fastpath_rx_access comment ..struct ..inet_sock From c17d2a7b216e168c3ba62d93482179c01b369ac7 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 27 Dec 2023 11:10:03 +0100 Subject: [PATCH 48/73] Bluetooth: hci_bcm4377: do not mark valid bd_addr as invalid A recent commit restored the original (and still documented) semantics for the HCI_QUIRK_USE_BDADDR_PROPERTY quirk so that the device address is considered invalid unless an address is provided by firmware. This specifically means that this flag must only be set for devices with invalid addresses, but the Broadcom BCM4377 driver has so far been setting this flag unconditionally. Fortunately the driver already checks for invalid addresses during setup and sets the HCI_QUIRK_INVALID_BDADDR flag, which can simply be replaced with HCI_QUIRK_USE_BDADDR_PROPERTY to indicate that the default address is invalid but can be overridden by firmware (long term, this should probably just always be allowed). Fixes: 6945795bc81a ("Bluetooth: fix use-bdaddr-property quirk") Cc: stable@vger.kernel.org # 6.5 Reported-by: Felix Zhang Link: https://lore.kernel.org/r/77419ffacc5b4875e920e038332575a2a5bff29f.camel@mrman314.tech/ Signed-off-by: Johan Hovold Reported-by: Felix Zhang Reviewed-by: Neal Gompa Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_bcm4377.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c index a61757835695..9a7243d5db71 100644 --- a/drivers/bluetooth/hci_bcm4377.c +++ b/drivers/bluetooth/hci_bcm4377.c @@ -1417,7 +1417,7 @@ static int bcm4377_check_bdaddr(struct bcm4377_data *bcm4377) bda = (struct hci_rp_read_bd_addr *)skb->data; if (!bcm4377_is_valid_bdaddr(bcm4377, &bda->bdaddr)) - set_bit(HCI_QUIRK_INVALID_BDADDR, &bcm4377->hdev->quirks); + set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &bcm4377->hdev->quirks); kfree_skb(skb); return 0; @@ -2368,7 +2368,6 @@ static int bcm4377_probe(struct pci_dev *pdev, const struct pci_device_id *id) hdev->set_bdaddr = bcm4377_hci_set_bdaddr; hdev->setup = bcm4377_hci_setup; - set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); if (bcm4377->hw->broken_mws_transport_config) set_bit(HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG, &hdev->quirks); if (bcm4377->hw->broken_ext_scan) From 6b3899be24b16ff8ee0cb25f0bd59b01b15ba1d1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jonas=20Dre=C3=9Fler?= Date: Tue, 2 Jan 2024 19:08:08 +0100 Subject: [PATCH 49/73] Bluetooth: hci_sync: Check the correct flag before starting a scan MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There's a very confusing mistake in the code starting a HCI inquiry: We're calling hci_dev_test_flag() to test for HCI_INQUIRY, but hci_dev_test_flag() checks hdev->dev_flags instead of hdev->flags. HCI_INQUIRY is a bit that's set on hdev->flags, not on hdev->dev_flags though. HCI_INQUIRY equals the integer 7, and in hdev->dev_flags, 7 means HCI_BONDABLE, so we were actually checking for HCI_BONDABLE here. The mistake is only present in the synchronous code for starting an inquiry, not in the async one. Also devices are typically bondable while doing an inquiry, so that might be the reason why nobody noticed it so far. Fixes: abfeea476c68 ("Bluetooth: hci_sync: Convert MGMT_OP_START_DISCOVERY") Signed-off-by: Jonas Dreßler Reviewed-by: Simon Horman Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index a6fc8a2a5c67..b3141e3f9cf6 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5559,7 +5559,7 @@ static int hci_inquiry_sync(struct hci_dev *hdev, u8 length) bt_dev_dbg(hdev, ""); - if (hci_dev_test_flag(hdev, HCI_INQUIRY)) + if (test_bit(HCI_INQUIRY, &hdev->flags)) return 0; hci_dev_lock(hdev); From 2449007d3f73b2842c9734f45f0aadb522daf592 Mon Sep 17 00:00:00 2001 From: Ying Hsu Date: Thu, 4 Jan 2024 11:56:32 +0000 Subject: [PATCH 50/73] Bluetooth: Avoid potential use-after-free in hci_error_reset While handling the HCI_EV_HARDWARE_ERROR event, if the underlying BT controller is not responding, the GPIO reset mechanism would free the hci_dev and lead to a use-after-free in hci_error_reset. Here's the call trace observed on a ChromeOS device with Intel AX201: queue_work_on+0x3e/0x6c __hci_cmd_sync_sk+0x2ee/0x4c0 [bluetooth ] ? init_wait_entry+0x31/0x31 __hci_cmd_sync+0x16/0x20 [bluetooth ] hci_error_reset+0x4f/0xa4 [bluetooth ] process_one_work+0x1d8/0x33f worker_thread+0x21b/0x373 kthread+0x13a/0x152 ? pr_cont_work+0x54/0x54 ? kthread_blkcg+0x31/0x31 ret_from_fork+0x1f/0x30 This patch holds the reference count on the hci_dev while processing a HCI_EV_HARDWARE_ERROR event to avoid potential crash. Fixes: c7741d16a57c ("Bluetooth: Perform a power cycle when receiving hardware error event") Signed-off-by: Ying Hsu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 65601aa52e0d..2821a42cefdc 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1049,6 +1049,7 @@ static void hci_error_reset(struct work_struct *work) { struct hci_dev *hdev = container_of(work, struct hci_dev, error_reset); + hci_dev_hold(hdev); BT_DBG("%s", hdev->name); if (hdev->hw_error) @@ -1056,10 +1057,10 @@ static void hci_error_reset(struct work_struct *work) else bt_dev_err(hdev, "hardware error 0x%2.2x", hdev->hw_error_code); - if (hci_dev_do_close(hdev)) - return; + if (!hci_dev_do_close(hdev)) + hci_dev_do_open(hdev); - hci_dev_do_open(hdev); + hci_dev_put(hdev); } void hci_uuids_clear(struct hci_dev *hdev) From e5469adb2a7e930d96813316592302d9f8f1df4e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 5 Jan 2024 10:43:26 -0500 Subject: [PATCH 51/73] Bluetooth: hci_sync: Fix accept_list when attempting to suspend During suspend, only wakeable devices can be in acceptlist, so if the device was previously added it needs to be removed otherwise the device can end up waking up the system prematurely. Fixes: 3b42055388c3 ("Bluetooth: hci_sync: Fix attempting to suspend with unfiltered passive scan") Signed-off-by: Clancy Shang Signed-off-by: Luiz Augusto von Dentz Reviewed-by: Paul Menzel --- net/bluetooth/hci_sync.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index b3141e3f9cf6..5716345a26df 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -2206,8 +2206,11 @@ static int hci_le_add_accept_list_sync(struct hci_dev *hdev, /* During suspend, only wakeable devices can be in acceptlist */ if (hdev->suspended && - !(params->flags & HCI_CONN_FLAG_REMOTE_WAKEUP)) + !(params->flags & HCI_CONN_FLAG_REMOTE_WAKEUP)) { + hci_le_del_accept_list_sync(hdev, ¶ms->addr, + params->addr_type); return 0; + } /* Select filter policy to accept all advertising */ if (*num_entries >= hdev->le_accept_list_size) From 2535b848fa0f42ddff3e5255cf5e742c9b77bb26 Mon Sep 17 00:00:00 2001 From: Yuxuan Hu <20373622@buaa.edu.cn> Date: Wed, 3 Jan 2024 17:10:43 +0800 Subject: [PATCH 52/73] Bluetooth: rfcomm: Fix null-ptr-deref in rfcomm_check_security During our fuzz testing of the connection and disconnection process at the RFCOMM layer, we discovered this bug. By comparing the packets from a normal connection and disconnection process with the testcase that triggered a KASAN report. We analyzed the cause of this bug as follows: 1. In the packets captured during a normal connection, the host sends a `Read Encryption Key Size` type of `HCI_CMD` packet (Command Opcode: 0x1408) to the controller to inquire the length of encryption key.After receiving this packet, the controller immediately replies with a Command Completepacket (Event Code: 0x0e) to return the Encryption Key Size. 2. In our fuzz test case, the timing of the controller's response to this packet was delayed to an unexpected point: after the RFCOMM and L2CAP layers had disconnected but before the HCI layer had disconnected. 3. After receiving the Encryption Key Size Response at the time described in point 2, the host still called the rfcomm_check_security function. However, by this time `struct l2cap_conn *conn = l2cap_pi(sk)->chan->conn;` had already been released, and when the function executed `return hci_conn_security(conn->hcon, d->sec_level, auth_type, d->out);`, specifically when accessing `conn->hcon`, a null-ptr-deref error occurred. To fix this bug, check if `sk->sk_state` is BT_CLOSED before calling rfcomm_recv_frame in rfcomm_process_rx. Signed-off-by: Yuxuan Hu <20373622@buaa.edu.cn> Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/rfcomm/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 053ef8f25fae..1d34d8497033 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -1941,7 +1941,7 @@ static struct rfcomm_session *rfcomm_process_rx(struct rfcomm_session *s) /* Get data directly from socket receive queue without copying it. */ while ((skb = skb_dequeue(&sk->sk_receive_queue))) { skb_orphan(skb); - if (!skb_linearize(skb)) { + if (!skb_linearize(skb) && sk->sk_state != BT_CLOSED) { s = rfcomm_recv_frame(s, skb); if (!s) break; From 61a5ab72edea7ebc3ad2c6beea29d966f528ebfb Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Tue, 9 Jan 2024 19:03:23 +0800 Subject: [PATCH 53/73] Bluetooth: hci_event: Fix wrongly recorded wakeup BD_ADDR hci_store_wake_reason() wrongly parses event HCI_Connection_Request as HCI_Connection_Complete and HCI_Connection_Complete as HCI_Connection_Request, so causes recording wakeup BD_ADDR error and potential stability issue, fix it by using the correct field. Fixes: 2f20216c1d6f ("Bluetooth: Emit controller suspend and resume events") Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index ef8c3bed7361..22b22c264c2a 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -7420,10 +7420,10 @@ static void hci_store_wake_reason(struct hci_dev *hdev, u8 event, * keep track of the bdaddr of the connection event that woke us up. */ if (event == HCI_EV_CONN_REQUEST) { - bacpy(&hdev->wake_addr, &conn_complete->bdaddr); + bacpy(&hdev->wake_addr, &conn_request->bdaddr); hdev->wake_addr_type = BDADDR_BREDR; } else if (event == HCI_EV_CONN_COMPLETE) { - bacpy(&hdev->wake_addr, &conn_request->bdaddr); + bacpy(&hdev->wake_addr, &conn_complete->bdaddr); hdev->wake_addr_type = BDADDR_BREDR; } else if (event == HCI_EV_LE_META) { struct hci_ev_le_meta *le_ev = (void *)skb->data; From 0bd1fb586235224048c726922db048d1bce6354a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fr=C3=A9d=C3=A9ric=20Danis?= Date: Mon, 22 Jan 2024 17:59:55 +0100 Subject: [PATCH 54/73] Bluetooth: mgmt: Fix limited discoverable off timeout MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit LIMITED_DISCOVERABLE flag is not reset from Class of Device and advertisement on limited discoverable timeout. This prevents to pass PTS test GAP/DISC/LIMM/BV-02-C Calling set_discoverable_sync as when the limited discovery is set correctly update the Class of Device and advertisement. Signed-off-by: Frédéric Danis Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index bb72ff6eb22f..ee3b4aad8bd8 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1045,6 +1045,8 @@ static void rpa_expired(struct work_struct *work) hci_cmd_sync_queue(hdev, rpa_expired_sync, NULL, NULL); } +static int set_discoverable_sync(struct hci_dev *hdev, void *data); + static void discov_off(struct work_struct *work) { struct hci_dev *hdev = container_of(work, struct hci_dev, @@ -1063,7 +1065,7 @@ static void discov_off(struct work_struct *work) hci_dev_clear_flag(hdev, HCI_DISCOVERABLE); hdev->discov_timeout = 0; - hci_update_discoverable(hdev); + hci_cmd_sync_queue(hdev, set_discoverable_sync, NULL, NULL); mgmt_new_settings(hdev); From 7e74aa53a68bf60f6019bd5d9a9a1406ec4d4865 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 22 Jan 2024 09:02:47 -0500 Subject: [PATCH 55/73] Bluetooth: hci_event: Fix handling of HCI_EV_IO_CAPA_REQUEST If we received HCI_EV_IO_CAPA_REQUEST while HCI_OP_READ_REMOTE_EXT_FEATURES is yet to be responded assume the remote does support SSP since otherwise this event shouldn't be generated. Link: https://lore.kernel.org/linux-bluetooth/CABBYNZ+9UdG1cMZVmdtN3U2aS16AKMCyTARZZyFX7xTEDWcMOw@mail.gmail.com/T/#t Fixes: c7f59461f5a7 ("Bluetooth: Fix a refcnt underflow problem for hci_conn") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 22b22c264c2a..613f2fd0bcc1 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -5329,9 +5329,12 @@ static void hci_io_capa_request_evt(struct hci_dev *hdev, void *data, hci_dev_lock(hdev); conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, &ev->bdaddr); - if (!conn || !hci_conn_ssp_enabled(conn)) + if (!conn || !hci_dev_test_flag(hdev, HCI_SSP_ENABLED)) goto unlock; + /* Assume remote supports SSP since it has triggered this event */ + set_bit(HCI_CONN_SSP_ENABLED, &conn->flags); + hci_conn_hold(conn); if (!hci_dev_test_flag(hdev, HCI_MGMT)) From e4b019515f950b4e6e5b74b2e1bb03a90cb33039 Mon Sep 17 00:00:00 2001 From: Kai-Heng Feng Date: Thu, 25 Jan 2024 14:50:28 +0800 Subject: [PATCH 56/73] Bluetooth: Enforce validation on max value of connection interval Right now Linux BT stack cannot pass test case "GAP/CONN/CPUP/BV-05-C 'Connection Parameter Update Procedure Invalid Parameters Central Responder'" in Bluetooth Test Suite revision GAP.TS.p44. [0] That was revoled by commit c49a8682fc5d ("Bluetooth: validate BLE connection interval updates"), but later got reverted due to devices like keyboards and mice may require low connection interval. So only validate the max value connection interval to pass the Test Suite, and let devices to request low connection interval if needed. [0] https://www.bluetooth.org/docman/handlers/DownloadDoc.ashx?doc_id=229869 Fixes: 68d19d7d9957 ("Revert "Bluetooth: validate BLE connection interval updates"") Signed-off-by: Kai-Heng Feng Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 4 ++++ net/bluetooth/l2cap_core.c | 8 +++++++- 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 613f2fd0bcc1..2a5f5a7d2412 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -6797,6 +6797,10 @@ static void hci_le_remote_conn_param_req_evt(struct hci_dev *hdev, void *data, return send_conn_param_neg_reply(hdev, handle, HCI_ERROR_UNKNOWN_CONN_ID); + if (max > hcon->le_conn_max_interval) + return send_conn_param_neg_reply(hdev, handle, + HCI_ERROR_INVALID_LL_PARAMS); + if (hci_check_conn_params(min, max, latency, timeout)) return send_conn_param_neg_reply(hdev, handle, HCI_ERROR_INVALID_LL_PARAMS); diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 60298975d5c4..656f49b299d2 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5613,7 +5613,13 @@ static inline int l2cap_conn_param_update_req(struct l2cap_conn *conn, memset(&rsp, 0, sizeof(rsp)); - err = hci_check_conn_params(min, max, latency, to_multiplier); + if (max > hcon->le_conn_max_interval) { + BT_DBG("requested connection interval exceeds current bounds."); + err = -EINVAL; + } else { + err = hci_check_conn_params(min, max, latency, to_multiplier); + } + if (err) rsp.result = cpu_to_le16(L2CAP_CONN_PARAM_REJECTED); else From c0dbc56077ae759f2dd602c7561480bc2b1b712c Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Fri, 19 Jan 2024 17:45:30 +0800 Subject: [PATCH 57/73] Bluetooth: qca: Fix wrong event type for patch config command Vendor-specific command patch config has HCI_Command_Complete event as response, but qca_send_patch_config_cmd() wrongly expects vendor-specific event for the command, fixed by using right event type. Btmon log for the vendor-specific command are shown below: < HCI Command: Vendor (0x3f|0x0000) plen 5 28 01 00 00 00 > HCI Event: Command Complete (0x0e) plen 5 Vendor (0x3f|0x0000) ncmd 1 Status: Success (0x00) 28 Fixes: 4fac8a7ac80b ("Bluetooth: btqca: sequential validation") Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btqca.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c index fdb0fae88d1c..b40b32fa7f1c 100644 --- a/drivers/bluetooth/btqca.c +++ b/drivers/bluetooth/btqca.c @@ -152,7 +152,7 @@ static int qca_send_patch_config_cmd(struct hci_dev *hdev) bt_dev_dbg(hdev, "QCA Patch config"); skb = __hci_cmd_sync_ev(hdev, EDL_PATCH_CMD_OPCODE, sizeof(cmd), - cmd, HCI_EV_VENDOR, HCI_INIT_TIMEOUT); + cmd, 0, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); bt_dev_err(hdev, "Sending QCA Patch config failed (%d)", err); From 7dcd3e014aa7faeeaf4047190b22d8a19a0db696 Mon Sep 17 00:00:00 2001 From: Janaki Ramaiah Thota Date: Wed, 24 Jan 2024 20:00:42 +0530 Subject: [PATCH 58/73] Bluetooth: hci_qca: Set BDA quirk bit if fwnode exists in DT BT adapter going into UNCONFIGURED state during BT turn ON when devicetree has no local-bd-address node. Bluetooth will not work out of the box on such devices, to avoid this problem, added check to set HCI_QUIRK_USE_BDADDR_PROPERTY based on local-bd-address node entry. When this quirk is not set, the public Bluetooth address read by host from controller though HCI Read BD Address command is considered as valid. Fixes: e668eb1e1578 ("Bluetooth: hci_core: Don't stop BT if the BD address missing in dts") Signed-off-by: Janaki Ramaiah Thota Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_qca.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index 94b8c406f0c0..06193546ebb6 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -7,6 +7,7 @@ * * Copyright (C) 2007 Texas Instruments, Inc. * Copyright (c) 2010, 2012, 2018 The Linux Foundation. All rights reserved. + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved. * * Acknowledgements: * This file is based on hci_ll.c, which was... @@ -1904,7 +1905,17 @@ static int qca_setup(struct hci_uart *hu) case QCA_WCN6750: case QCA_WCN6855: case QCA_WCN7850: - set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); + + /* Set BDA quirk bit for reading BDA value from fwnode property + * only if that property exist in DT. + */ + if (fwnode_property_present(dev_fwnode(hdev->dev.parent), "local-bd-address")) { + set_bit(HCI_QUIRK_USE_BDADDR_PROPERTY, &hdev->quirks); + bt_dev_info(hdev, "setting quirk bit to read BDA from fwnode later"); + } else { + bt_dev_dbg(hdev, "local-bd-address` is not present in the devicetree so not setting quirk bit for BDA"); + } + hci_set_aosp_capable(hdev); ret = qca_read_soc_version(hdev, &ver, soc_type); From 6abf9dd26bb1699c17d601b9a292577d01827c0e Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Fri, 26 Jan 2024 17:00:24 +0800 Subject: [PATCH 59/73] Bluetooth: qca: Fix triggering coredump implementation hci_coredump_qca() uses __hci_cmd_sync() to send a vendor-specific command to trigger firmware coredump, but the command does not have any event as its sync response, so it is not suitable to use __hci_cmd_sync(), fixed by using __hci_cmd_send(). Fixes: 06d3fdfcdf5c ("Bluetooth: hci_qca: Add qcom devcoredump support") Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_qca.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index 06193546ebb6..edd2a81b4d5e 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -1807,13 +1807,12 @@ static int qca_power_on(struct hci_dev *hdev) static void hci_coredump_qca(struct hci_dev *hdev) { + int err; static const u8 param[] = { 0x26 }; - struct sk_buff *skb; - skb = __hci_cmd_sync(hdev, 0xfc0c, 1, param, HCI_CMD_TIMEOUT); - if (IS_ERR(skb)) - bt_dev_err(hdev, "%s: trigger crash failed (%ld)", __func__, PTR_ERR(skb)); - kfree_skb(skb); + err = __hci_cmd_send(hdev, 0xfc0c, 1, param); + if (err < 0) + bt_dev_err(hdev, "%s: trigger crash failed (%d)", __func__, err); } static int qca_get_data_path_id(struct hci_dev *hdev, __u8 *data_path_id) From 7e0f122c65912740327e4c54472acaa5f85868cb Mon Sep 17 00:00:00 2001 From: Ignat Korchagin Date: Thu, 22 Feb 2024 10:33:08 +0000 Subject: [PATCH 60/73] netfilter: nf_tables: allow NFPROTO_INET in nft_(match/target)_validate() Commit d0009effa886 ("netfilter: nf_tables: validate NFPROTO_* family") added some validation of NFPROTO_* families in the nft_compat module, but it broke the ability to use legacy iptables modules in dual-stack nftables. While with legacy iptables one had to independently manage IPv4 and IPv6 tables, with nftables it is possible to have dual-stack tables sharing the rules. Moreover, it was possible to use rules based on legacy iptables match/target modules in dual-stack nftables. As an example, the program from [2] creates an INET dual-stack family table using an xt_bpf based rule, which looks like the following (the actual output was generated with a patched nft tool as the current nft tool does not parse dual stack tables with legacy match rules, so consider it for illustrative purposes only): table inet testfw { chain input { type filter hook prerouting priority filter; policy accept; bytecode counter packets 0 bytes 0 accept } } After d0009effa886 ("netfilter: nf_tables: validate NFPROTO_* family") we get EOPNOTSUPP for the above program. Fix this by allowing NFPROTO_INET for nft_(match/target)_validate(), but also restrict the functions to classic iptables hooks. Changes in v3: * clarify that upstream nft will not display such configuration properly and that the output was generated with a patched nft tool * remove example program from commit description and link to it instead * no code changes otherwise Changes in v2: * restrict nft_(match/target)_validate() to classic iptables hooks * rewrite example program to use unmodified libnftnl Fixes: d0009effa886 ("netfilter: nf_tables: validate NFPROTO_* family") Link: https://lore.kernel.org/all/Zc1PfoWN38UuFJRI@calendula/T/#mc947262582c90fec044c7a3398cc92fac7afea72 [1] Link: https://lore.kernel.org/all/20240220145509.53357-1-ignat@cloudflare.com/ [2] Reported-by: Jordan Griege Signed-off-by: Ignat Korchagin Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_compat.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 1f9474fefe84..d3d11dede545 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -359,10 +359,20 @@ static int nft_target_validate(const struct nft_ctx *ctx, if (ctx->family != NFPROTO_IPV4 && ctx->family != NFPROTO_IPV6 && + ctx->family != NFPROTO_INET && ctx->family != NFPROTO_BRIDGE && ctx->family != NFPROTO_ARP) return -EOPNOTSUPP; + ret = nft_chain_validate_hooks(ctx->chain, + (1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING)); + if (ret) + return ret; + if (nft_is_base_chain(ctx->chain)) { const struct nft_base_chain *basechain = nft_base_chain(ctx->chain); @@ -610,10 +620,20 @@ static int nft_match_validate(const struct nft_ctx *ctx, if (ctx->family != NFPROTO_IPV4 && ctx->family != NFPROTO_IPV6 && + ctx->family != NFPROTO_INET && ctx->family != NFPROTO_BRIDGE && ctx->family != NFPROTO_ARP) return -EOPNOTSUPP; + ret = nft_chain_validate_hooks(ctx->chain, + (1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING)); + if (ret) + return ret; + if (nft_is_base_chain(ctx->chain)) { const struct nft_base_chain *basechain = nft_base_chain(ctx->chain); From 62e7151ae3eb465e0ab52a20c941ff33bb6332e9 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 27 Feb 2024 16:17:51 +0100 Subject: [PATCH 61/73] netfilter: bridge: confirm multicast packets before passing them up the stack conntrack nf_confirm logic cannot handle cloned skbs referencing the same nf_conn entry, which will happen for multicast (broadcast) frames on bridges. Example: macvlan0 | br0 / \ ethX ethY ethX (or Y) receives a L2 multicast or broadcast packet containing an IP packet, flow is not yet in conntrack table. 1. skb passes through bridge and fake-ip (br_netfilter)Prerouting. -> skb->_nfct now references a unconfirmed entry 2. skb is broad/mcast packet. bridge now passes clones out on each bridge interface. 3. skb gets passed up the stack. 4. In macvlan case, macvlan driver retains clone(s) of the mcast skb and schedules a work queue to send them out on the lower devices. The clone skb->_nfct is not a copy, it is the same entry as the original skb. The macvlan rx handler then returns RX_HANDLER_PASS. 5. Normal conntrack hooks (in NF_INET_LOCAL_IN) confirm the orig skb. The Macvlan broadcast worker and normal confirm path will race. This race will not happen if step 2 already confirmed a clone. In that case later steps perform skb_clone() with skb->_nfct already confirmed (in hash table). This works fine. But such confirmation won't happen when eb/ip/nftables rules dropped the packets before they reached the nf_confirm step in postrouting. Pablo points out that nf_conntrack_bridge doesn't allow use of stateful nat, so we can safely discard the nf_conn entry and let inet call conntrack again. This doesn't work for bridge netfilter: skb could have a nat transformation. Also bridge nf prevents re-invocation of inet prerouting via 'sabotage_in' hook. Work around this problem by explicit confirmation of the entry at LOCAL_IN time, before upper layer has a chance to clone the unconfirmed entry. The downside is that this disables NAT and conntrack helpers. Alternative fix would be to add locking to all code parts that deal with unconfirmed packets, but even if that could be done in a sane way this opens up other problems, for example: -m physdev --physdev-out eth0 -j SNAT --snat-to 1.2.3.4 -m physdev --physdev-out eth1 -j SNAT --snat-to 1.2.3.5 For multicast case, only one of such conflicting mappings will be created, conntrack only handles 1:1 NAT mappings. Users should set create a setup that explicitly marks such traffic NOTRACK (conntrack bypass) to avoid this, but we cannot auto-bypass them, ruleset might have accept rules for untracked traffic already, so user-visible behaviour would change. Suggested-by: Pablo Neira Ayuso Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217777 Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter.h | 1 + net/bridge/br_netfilter_hooks.c | 96 ++++++++++++++++++++++ net/bridge/netfilter/nf_conntrack_bridge.c | 30 +++++++ net/netfilter/nf_conntrack_core.c | 1 + 4 files changed, 128 insertions(+) diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 80900d910992..ce660d51549b 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -474,6 +474,7 @@ struct nf_ct_hook { const struct sk_buff *); void (*attach)(struct sk_buff *nskb, const struct sk_buff *skb); void (*set_closing)(struct nf_conntrack *nfct); + int (*confirm)(struct sk_buff *skb); }; extern const struct nf_ct_hook __rcu *nf_ct_hook; diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index ed1720890757..35e10c5a766d 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -43,6 +43,10 @@ #include #endif +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +#include +#endif + static unsigned int brnf_net_id __read_mostly; struct brnf_net { @@ -553,6 +557,90 @@ static unsigned int br_nf_pre_routing(void *priv, return NF_STOLEN; } +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +/* conntracks' nf_confirm logic cannot handle cloned skbs referencing + * the same nf_conn entry, which will happen for multicast (broadcast) + * Frames on bridges. + * + * Example: + * macvlan0 + * br0 + * ethX ethY + * + * ethX (or Y) receives multicast or broadcast packet containing + * an IP packet, not yet in conntrack table. + * + * 1. skb passes through bridge and fake-ip (br_netfilter)Prerouting. + * -> skb->_nfct now references a unconfirmed entry + * 2. skb is broad/mcast packet. bridge now passes clones out on each bridge + * interface. + * 3. skb gets passed up the stack. + * 4. In macvlan case, macvlan driver retains clone(s) of the mcast skb + * and schedules a work queue to send them out on the lower devices. + * + * The clone skb->_nfct is not a copy, it is the same entry as the + * original skb. The macvlan rx handler then returns RX_HANDLER_PASS. + * 5. Normal conntrack hooks (in NF_INET_LOCAL_IN) confirm the orig skb. + * + * The Macvlan broadcast worker and normal confirm path will race. + * + * This race will not happen if step 2 already confirmed a clone. In that + * case later steps perform skb_clone() with skb->_nfct already confirmed (in + * hash table). This works fine. + * + * But such confirmation won't happen when eb/ip/nftables rules dropped the + * packets before they reached the nf_confirm step in postrouting. + * + * Work around this problem by explicit confirmation of the entry at + * LOCAL_IN time, before upper layer has a chance to clone the unconfirmed + * entry. + * + */ +static unsigned int br_nf_local_in(void *priv, + struct sk_buff *skb, + const struct nf_hook_state *state) +{ + struct nf_conntrack *nfct = skb_nfct(skb); + const struct nf_ct_hook *ct_hook; + struct nf_conn *ct; + int ret; + + if (!nfct || skb->pkt_type == PACKET_HOST) + return NF_ACCEPT; + + ct = container_of(nfct, struct nf_conn, ct_general); + if (likely(nf_ct_is_confirmed(ct))) + return NF_ACCEPT; + + WARN_ON_ONCE(skb_shared(skb)); + WARN_ON_ONCE(refcount_read(&nfct->use) != 1); + + /* We can't call nf_confirm here, it would create a dependency + * on nf_conntrack module. + */ + ct_hook = rcu_dereference(nf_ct_hook); + if (!ct_hook) { + skb->_nfct = 0ul; + nf_conntrack_put(nfct); + return NF_ACCEPT; + } + + nf_bridge_pull_encap_header(skb); + ret = ct_hook->confirm(skb); + switch (ret & NF_VERDICT_MASK) { + case NF_STOLEN: + return NF_STOLEN; + default: + nf_bridge_push_encap_header(skb); + break; + } + + ct = container_of(nfct, struct nf_conn, ct_general); + WARN_ON_ONCE(!nf_ct_is_confirmed(ct)); + + return ret; +} +#endif /* PF_BRIDGE/FORWARD *************************************************/ static int br_nf_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb) @@ -964,6 +1052,14 @@ static const struct nf_hook_ops br_nf_ops[] = { .hooknum = NF_BR_PRE_ROUTING, .priority = NF_BR_PRI_BRNF, }, +#if IS_ENABLED(CONFIG_NF_CONNTRACK) + { + .hook = br_nf_local_in, + .pf = NFPROTO_BRIDGE, + .hooknum = NF_BR_LOCAL_IN, + .priority = NF_BR_PRI_LAST, + }, +#endif { .hook = br_nf_forward, .pf = NFPROTO_BRIDGE, diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c index abb090f94ed2..6f877e31709b 100644 --- a/net/bridge/netfilter/nf_conntrack_bridge.c +++ b/net/bridge/netfilter/nf_conntrack_bridge.c @@ -291,6 +291,30 @@ static unsigned int nf_ct_bridge_pre(void *priv, struct sk_buff *skb, return nf_conntrack_in(skb, &bridge_state); } +static unsigned int nf_ct_bridge_in(void *priv, struct sk_buff *skb, + const struct nf_hook_state *state) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct; + + if (skb->pkt_type == PACKET_HOST) + return NF_ACCEPT; + + /* nf_conntrack_confirm() cannot handle concurrent clones, + * this happens for broad/multicast frames with e.g. macvlan on top + * of the bridge device. + */ + ct = nf_ct_get(skb, &ctinfo); + if (!ct || nf_ct_is_confirmed(ct) || nf_ct_is_template(ct)) + return NF_ACCEPT; + + /* let inet prerouting call conntrack again */ + skb->_nfct = 0; + nf_ct_put(ct); + + return NF_ACCEPT; +} + static void nf_ct_bridge_frag_save(struct sk_buff *skb, struct nf_bridge_frag_data *data) { @@ -385,6 +409,12 @@ static struct nf_hook_ops nf_ct_bridge_hook_ops[] __read_mostly = { .hooknum = NF_BR_PRE_ROUTING, .priority = NF_IP_PRI_CONNTRACK, }, + { + .hook = nf_ct_bridge_in, + .pf = NFPROTO_BRIDGE, + .hooknum = NF_BR_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, + }, { .hook = nf_ct_bridge_post, .pf = NFPROTO_BRIDGE, diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 2e5f3864d353..5b876fa7f9af 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -2756,6 +2756,7 @@ static const struct nf_ct_hook nf_conntrack_hook = { .get_tuple_skb = nf_conntrack_get_tuple_skb, .attach = nf_conntrack_attach, .set_closing = nf_conntrack_set_closing, + .confirm = __nf_conntrack_confirm, }; void nf_conntrack_init_end(void) From 6523cf516c55db164f8f73306027b1caebb5628e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 26 Feb 2024 15:21:48 +0100 Subject: [PATCH 62/73] selftests: netfilter: add bridge conntrack + multicast test case Add test case for multicast packet confirm race. Without preceding patch, this should result in: WARNING: CPU: 0 PID: 38 at net/netfilter/nf_conntrack_core.c:1198 __nf_conntrack_confirm+0x3ed/0x5f0 Workqueue: events_unbound macvlan_process_broadcast RIP: 0010:__nf_conntrack_confirm+0x3ed/0x5f0 ? __nf_conntrack_confirm+0x3ed/0x5f0 nf_confirm+0x2ad/0x2d0 nf_hook_slow+0x36/0xd0 ip_local_deliver+0xce/0x110 __netif_receive_skb_one_core+0x4f/0x70 process_backlog+0x8c/0x130 [..] Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- tools/testing/selftests/netfilter/Makefile | 3 +- .../selftests/netfilter/bridge_netfilter.sh | 188 ++++++++++++++++++ 2 files changed, 190 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/netfilter/bridge_netfilter.sh diff --git a/tools/testing/selftests/netfilter/Makefile b/tools/testing/selftests/netfilter/Makefile index db27153eb4a0..936c3085bb83 100644 --- a/tools/testing/selftests/netfilter/Makefile +++ b/tools/testing/selftests/netfilter/Makefile @@ -7,7 +7,8 @@ TEST_PROGS := nft_trans_stress.sh nft_fib.sh nft_nat.sh bridge_brouter.sh \ nft_queue.sh nft_meta.sh nf_nat_edemux.sh \ ipip-conntrack-mtu.sh conntrack_tcp_unreplied.sh \ conntrack_vrf.sh nft_synproxy.sh rpath.sh nft_audit.sh \ - conntrack_sctp_collision.sh xt_string.sh + conntrack_sctp_collision.sh xt_string.sh \ + bridge_netfilter.sh HOSTPKG_CONFIG := pkg-config diff --git a/tools/testing/selftests/netfilter/bridge_netfilter.sh b/tools/testing/selftests/netfilter/bridge_netfilter.sh new file mode 100644 index 000000000000..659b3ab02c8b --- /dev/null +++ b/tools/testing/selftests/netfilter/bridge_netfilter.sh @@ -0,0 +1,188 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Test bridge netfilter + conntrack, a combination that doesn't really work, +# with multicast/broadcast packets racing for hash table insertion. + +# eth0 br0 eth0 +# setup is: ns1 <->,ns0 <-> ns3 +# ns2 <-' `'-> ns4 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 +ret=0 + +sfx=$(mktemp -u "XXXXXXXX") +ns0="ns0-$sfx" +ns1="ns1-$sfx" +ns2="ns2-$sfx" +ns3="ns3-$sfx" +ns4="ns4-$sfx" + +ebtables -V > /dev/null 2>&1 +if [ $? -ne 0 ];then + echo "SKIP: Could not run test without ebtables" + exit $ksft_skip +fi + +ip -Version > /dev/null 2>&1 +if [ $? -ne 0 ];then + echo "SKIP: Could not run test without ip tool" + exit $ksft_skip +fi + +for i in $(seq 0 4); do + eval ip netns add \$ns$i +done + +cleanup() { + for i in $(seq 0 4); do eval ip netns del \$ns$i;done +} + +trap cleanup EXIT + +do_ping() +{ + fromns="$1" + dstip="$2" + + ip netns exec $fromns ping -c 1 -q $dstip > /dev/null + if [ $? -ne 0 ]; then + echo "ERROR: ping from $fromns to $dstip" + ip netns exec ${ns0} nft list ruleset + ret=1 + fi +} + +bcast_ping() +{ + fromns="$1" + dstip="$2" + + for i in $(seq 1 1000); do + ip netns exec $fromns ping -q -f -b -c 1 -q $dstip > /dev/null 2>&1 + if [ $? -ne 0 ]; then + echo "ERROR: ping -b from $fromns to $dstip" + ip netns exec ${ns0} nft list ruleset + fi + done +} + +ip link add veth1 netns ${ns0} type veth peer name eth0 netns ${ns1} +if [ $? -ne 0 ]; then + echo "SKIP: Can't create veth device" + exit $ksft_skip +fi + +ip link add veth2 netns ${ns0} type veth peer name eth0 netns $ns2 +ip link add veth3 netns ${ns0} type veth peer name eth0 netns $ns3 +ip link add veth4 netns ${ns0} type veth peer name eth0 netns $ns4 + +ip -net ${ns0} link set lo up + +for i in $(seq 1 4); do + ip -net ${ns0} link set veth$i up +done + +ip -net ${ns0} link add br0 type bridge stp_state 0 forward_delay 0 nf_call_iptables 1 nf_call_ip6tables 1 nf_call_arptables 1 +if [ $? -ne 0 ]; then + echo "SKIP: Can't create bridge br0" + exit $ksft_skip +fi + +# make veth0,1,2 part of bridge. +for i in $(seq 1 3); do + ip -net ${ns0} link set veth$i master br0 +done + +# add a macvlan on top of the bridge. +MACVLAN_ADDR=ba:f3:13:37:42:23 +ip -net ${ns0} link add link br0 name macvlan0 type macvlan mode private +ip -net ${ns0} link set macvlan0 address ${MACVLAN_ADDR} +ip -net ${ns0} link set macvlan0 up +ip -net ${ns0} addr add 10.23.0.1/24 dev macvlan0 + +# add a macvlan on top of veth4. +MACVLAN_ADDR=ba:f3:13:37:42:24 +ip -net ${ns0} link add link veth4 name macvlan4 type macvlan mode vepa +ip -net ${ns0} link set macvlan4 address ${MACVLAN_ADDR} +ip -net ${ns0} link set macvlan4 up + +# make the macvlan part of the bridge. +# veth4 is not a bridge port, only the macvlan on top of it. +ip -net ${ns0} link set macvlan4 master br0 + +ip -net ${ns0} link set br0 up +ip -net ${ns0} addr add 10.0.0.1/24 dev br0 +ip netns exec ${ns0} sysctl -q net.bridge.bridge-nf-call-iptables=1 +ret=$? +if [ $ret -ne 0 ] ; then + echo "SKIP: bridge netfilter not available" + ret=$ksft_skip +fi + +# for testing, so namespaces will reply to ping -b probes. +ip netns exec ${ns0} sysctl -q net.ipv4.icmp_echo_ignore_broadcasts=0 + +# enable conntrack in ns0 and drop broadcast packets in forward to +# avoid them from getting confirmed in the postrouting hook before +# the cloned skb is passed up the stack. +ip netns exec ${ns0} nft -f - < Date: Mon, 26 Feb 2024 13:40:18 -0800 Subject: [PATCH 63/73] tools: ynl: fix handling of multiple mcast groups We never increment the group number iterator, so all groups get recorded into index 0 of the mcast_groups[] array. As a result YNL can only handle using the last group. For example using the "netdev" sample on kernel with page pool commands results in: $ ./samples/netdev YNL: Multicast group 'mgmt' not found Most families have only one multicast group, so this hasn't been noticed. Plus perhaps developers usually test the last group which would have worked. Fixes: 86878f14d71a ("tools: ynl: user space helpers") Reviewed-by: Donald Hunter Acked-by: Nicolas Dichtel Link: https://lore.kernel.org/r/20240226214019.1255242-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/lib/ynl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/net/ynl/lib/ynl.c b/tools/net/ynl/lib/ynl.c index 6e6d474c8366..45e49671ae87 100644 --- a/tools/net/ynl/lib/ynl.c +++ b/tools/net/ynl/lib/ynl.c @@ -523,6 +523,7 @@ ynl_get_family_info_mcast(struct ynl_sock *ys, const struct nlattr *mcasts) ys->mcast_groups[i].name[GENL_NAMSIZ - 1] = 0; } } + i++; } return 0; From 743ad091fb46e622f1b690385bb15e3cd3daf874 Mon Sep 17 00:00:00 2001 From: Lin Ma Date: Tue, 27 Feb 2024 20:11:28 +0800 Subject: [PATCH 64/73] rtnetlink: fix error logic of IFLA_BRIDGE_FLAGS writing back In the commit d73ef2d69c0d ("rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length"), an adjustment was made to the old loop logic in the function `rtnl_bridge_setlink` to enable the loop to also check the length of the IFLA_BRIDGE_MODE attribute. However, this adjustment removed the `break` statement and led to an error logic of the flags writing back at the end of this function. if (have_flags) memcpy(nla_data(attr), &flags, sizeof(flags)); // attr should point to IFLA_BRIDGE_FLAGS NLA !!! Before the mentioned commit, the `attr` is granted to be IFLA_BRIDGE_FLAGS. However, this is not necessarily true fow now as the updated loop will let the attr point to the last NLA, even an invalid NLA which could cause overflow writes. This patch introduces a new variable `br_flag` to save the NLA pointer that points to IFLA_BRIDGE_FLAGS and uses it to resolve the mentioned error logic. Fixes: d73ef2d69c0d ("rtnetlink: let rtnl_bridge_setlink checks IFLA_BRIDGE_MODE length") Signed-off-by: Lin Ma Acked-by: Nikolay Aleksandrov Link: https://lore.kernel.org/r/20240227121128.608110-1-linma@zju.edu.cn Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 9c4f427f3a50..ae86f751efc3 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -5169,10 +5169,9 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, struct net *net = sock_net(skb->sk); struct ifinfomsg *ifm; struct net_device *dev; - struct nlattr *br_spec, *attr = NULL; + struct nlattr *br_spec, *attr, *br_flags_attr = NULL; int rem, err = -EOPNOTSUPP; u16 flags = 0; - bool have_flags = false; if (nlmsg_len(nlh) < sizeof(*ifm)) return -EINVAL; @@ -5190,11 +5189,11 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC); if (br_spec) { nla_for_each_nested(attr, br_spec, rem) { - if (nla_type(attr) == IFLA_BRIDGE_FLAGS && !have_flags) { + if (nla_type(attr) == IFLA_BRIDGE_FLAGS && !br_flags_attr) { if (nla_len(attr) < sizeof(flags)) return -EINVAL; - have_flags = true; + br_flags_attr = attr; flags = nla_get_u16(attr); } @@ -5238,8 +5237,8 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, } } - if (have_flags) - memcpy(nla_data(attr), &flags, sizeof(flags)); + if (br_flags_attr) + memcpy(nla_data(br_flags_attr), &flags, sizeof(flags)); out: return err; } From 0bb7b09392eb74b152719ae87b1ba5e4bf910ef0 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Tue, 27 Feb 2024 10:49:41 -0800 Subject: [PATCH 65/73] igb: extend PTP timestamp adjustments to i211 The i211 requires the same PTP timestamp adjustments as the i210, according to its datasheet. To ensure consistent timestamping across different platforms, this change extends the existing adjustments to include the i211. The adjustment result are tested and comparable for i210 and i211 based systems. Fixes: 3f544d2a4d5c ("igb: adjust PTP timestamps for Tx/Rx latency") Signed-off-by: Oleksij Rempel Reviewed-by: Jacob Keller Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Link: https://lore.kernel.org/r/20240227184942.362710-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/igb/igb_ptp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 319c544b9f04..f94570556120 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -957,7 +957,7 @@ static void igb_ptp_tx_hwtstamp(struct igb_adapter *adapter) igb_ptp_systim_to_hwtstamp(adapter, &shhwtstamps, regval); /* adjust timestamp for the TX latency based on link speed */ - if (adapter->hw.mac.type == e1000_i210) { + if (hw->mac.type == e1000_i210 || hw->mac.type == e1000_i211) { switch (adapter->link_speed) { case SPEED_10: adjust = IGB_I210_TX_LATENCY_10; @@ -1003,6 +1003,7 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va, ktime_t *timestamp) { struct igb_adapter *adapter = q_vector->adapter; + struct e1000_hw *hw = &adapter->hw; struct skb_shared_hwtstamps ts; __le64 *regval = (__le64 *)va; int adjust = 0; @@ -1022,7 +1023,7 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va, igb_ptp_systim_to_hwtstamp(adapter, &ts, le64_to_cpu(regval[1])); /* adjust timestamp for the RX latency based on link speed */ - if (adapter->hw.mac.type == e1000_i210) { + if (hw->mac.type == e1000_i210 || hw->mac.type == e1000_i211) { switch (adapter->link_speed) { case SPEED_10: adjust = IGB_I210_RX_LATENCY_10; From 51dd4ee0372228ffb0f7709fa7aa0678d4199d06 Mon Sep 17 00:00:00 2001 From: Lukasz Majewski Date: Wed, 28 Feb 2024 09:56:44 +0100 Subject: [PATCH 66/73] net: hsr: Use correct offset for HSR TLV values in supervisory HSR frames Current HSR implementation uses following supervisory frame (even for HSRv1 the HSR tag is not is not present): 00000000: 01 15 4e 00 01 2d XX YY ZZ 94 77 10 88 fb 00 01 00000010: 7e 1c 17 06 XX YY ZZ 94 77 10 1e 06 XX YY ZZ 94 00000020: 77 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00000030: 00 00 00 00 00 00 00 00 00 00 00 00 The current code adds extra two bytes (i.e. sizeof(struct hsr_sup_tlv)) when offset for skb_pull() is calculated. This is wrong, as both 'struct hsrv1_ethhdr_sp' and 'hsrv0_ethhdr_sp' already have 'struct hsr_sup_tag' defined in them, so there is no need for adding extra two bytes. This code was working correctly as with no RedBox support, the check for HSR_TLV_EOT (0x00) was off by two bytes, which were corresponding to zeroed padded bytes for minimal packet size. Fixes: eafaa88b3eb7 ("net: hsr: Add support for redbox supervision frames") Signed-off-by: Lukasz Majewski Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20240228085644.3618044-1-lukma@denx.de Signed-off-by: Paolo Abeni --- net/hsr/hsr_forward.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 2afe28712a7a..5d68cb181695 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -83,7 +83,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb) return false; /* Get next tlv */ - total_length += sizeof(struct hsr_sup_tlv) + hsr_sup_tag->tlv.HSR_TLV_length; + total_length += hsr_sup_tag->tlv.HSR_TLV_length; if (!pskb_may_pull(skb, total_length)) return false; skb_pull(skb, total_length); From 616d82c3cfa2a2146dd7e3ae47bda7e877ee549e Mon Sep 17 00:00:00 2001 From: Alexander Ofitserov Date: Wed, 28 Feb 2024 14:47:03 +0300 Subject: [PATCH 67/73] gtp: fix use-after-free and null-ptr-deref in gtp_newlink() The gtp_link_ops operations structure for the subsystem must be registered after registering the gtp_net_ops pernet operations structure. Syzkaller hit 'general protection fault in gtp_genl_dump_pdp' bug: [ 1010.702740] gtp: GTP module unloaded [ 1010.715877] general protection fault, probably for non-canonical address 0xdffffc0000000001: 0000 [#1] SMP KASAN NOPTI [ 1010.715888] KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] [ 1010.715895] CPU: 1 PID: 128616 Comm: a.out Not tainted 6.8.0-rc6-std-def-alt1 #1 [ 1010.715899] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.0-alt1 04/01/2014 [ 1010.715908] RIP: 0010:gtp_newlink+0x4d7/0x9c0 [gtp] [ 1010.715915] Code: 80 3c 02 00 0f 85 41 04 00 00 48 8b bb d8 05 00 00 e8 ed f6 ff ff 48 89 c2 48 89 c5 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f 85 4f 04 00 00 4c 89 e2 4c 8b 6d 00 48 b8 00 00 00 [ 1010.715920] RSP: 0018:ffff888020fbf180 EFLAGS: 00010203 [ 1010.715929] RAX: dffffc0000000000 RBX: ffff88800399c000 RCX: 0000000000000000 [ 1010.715933] RDX: 0000000000000001 RSI: ffffffff84805280 RDI: 0000000000000282 [ 1010.715938] RBP: 000000000000000d R08: 0000000000000001 R09: 0000000000000000 [ 1010.715942] R10: 0000000000000001 R11: 0000000000000001 R12: ffff88800399cc80 [ 1010.715947] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000400 [ 1010.715953] FS: 00007fd1509ab5c0(0000) GS:ffff88805b300000(0000) knlGS:0000000000000000 [ 1010.715958] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1010.715962] CR2: 0000000000000000 CR3: 000000001c07a000 CR4: 0000000000750ee0 [ 1010.715968] PKRU: 55555554 [ 1010.715972] Call Trace: [ 1010.715985] ? __die_body.cold+0x1a/0x1f [ 1010.715995] ? die_addr+0x43/0x70 [ 1010.716002] ? exc_general_protection+0x199/0x2f0 [ 1010.716016] ? asm_exc_general_protection+0x1e/0x30 [ 1010.716026] ? gtp_newlink+0x4d7/0x9c0 [gtp] [ 1010.716034] ? gtp_net_exit+0x150/0x150 [gtp] [ 1010.716042] __rtnl_newlink+0x1063/0x1700 [ 1010.716051] ? rtnl_setlink+0x3c0/0x3c0 [ 1010.716063] ? is_bpf_text_address+0xc0/0x1f0 [ 1010.716070] ? kernel_text_address.part.0+0xbb/0xd0 [ 1010.716076] ? __kernel_text_address+0x56/0xa0 [ 1010.716084] ? unwind_get_return_address+0x5a/0xa0 [ 1010.716091] ? create_prof_cpu_mask+0x30/0x30 [ 1010.716098] ? arch_stack_walk+0x9e/0xf0 [ 1010.716106] ? stack_trace_save+0x91/0xd0 [ 1010.716113] ? stack_trace_consume_entry+0x170/0x170 [ 1010.716121] ? __lock_acquire+0x15c5/0x5380 [ 1010.716139] ? mark_held_locks+0x9e/0xe0 [ 1010.716148] ? kmem_cache_alloc_trace+0x35f/0x3c0 [ 1010.716155] ? __rtnl_newlink+0x1700/0x1700 [ 1010.716160] rtnl_newlink+0x69/0xa0 [ 1010.716166] rtnetlink_rcv_msg+0x43b/0xc50 [ 1010.716172] ? rtnl_fdb_dump+0x9f0/0x9f0 [ 1010.716179] ? lock_acquire+0x1fe/0x560 [ 1010.716188] ? netlink_deliver_tap+0x12f/0xd50 [ 1010.716196] netlink_rcv_skb+0x14d/0x440 [ 1010.716202] ? rtnl_fdb_dump+0x9f0/0x9f0 [ 1010.716208] ? netlink_ack+0xab0/0xab0 [ 1010.716213] ? netlink_deliver_tap+0x202/0xd50 [ 1010.716220] ? netlink_deliver_tap+0x218/0xd50 [ 1010.716226] ? __virt_addr_valid+0x30b/0x590 [ 1010.716233] netlink_unicast+0x54b/0x800 [ 1010.716240] ? netlink_attachskb+0x870/0x870 [ 1010.716248] ? __check_object_size+0x2de/0x3b0 [ 1010.716254] netlink_sendmsg+0x938/0xe40 [ 1010.716261] ? netlink_unicast+0x800/0x800 [ 1010.716269] ? __import_iovec+0x292/0x510 [ 1010.716276] ? netlink_unicast+0x800/0x800 [ 1010.716284] __sock_sendmsg+0x159/0x190 [ 1010.716290] ____sys_sendmsg+0x712/0x880 [ 1010.716297] ? sock_write_iter+0x3d0/0x3d0 [ 1010.716304] ? __ia32_sys_recvmmsg+0x270/0x270 [ 1010.716309] ? lock_acquire+0x1fe/0x560 [ 1010.716315] ? drain_array_locked+0x90/0x90 [ 1010.716324] ___sys_sendmsg+0xf8/0x170 [ 1010.716331] ? sendmsg_copy_msghdr+0x170/0x170 [ 1010.716337] ? lockdep_init_map_type+0x2c7/0x860 [ 1010.716343] ? lockdep_hardirqs_on_prepare+0x430/0x430 [ 1010.716350] ? debug_mutex_init+0x33/0x70 [ 1010.716360] ? percpu_counter_add_batch+0x8b/0x140 [ 1010.716367] ? lock_acquire+0x1fe/0x560 [ 1010.716373] ? find_held_lock+0x2c/0x110 [ 1010.716384] ? __fd_install+0x1b6/0x6f0 [ 1010.716389] ? lock_downgrade+0x810/0x810 [ 1010.716396] ? __fget_light+0x222/0x290 [ 1010.716403] __sys_sendmsg+0xea/0x1b0 [ 1010.716409] ? __sys_sendmsg_sock+0x40/0x40 [ 1010.716419] ? lockdep_hardirqs_on_prepare+0x2b3/0x430 [ 1010.716425] ? syscall_enter_from_user_mode+0x1d/0x60 [ 1010.716432] do_syscall_64+0x30/0x40 [ 1010.716438] entry_SYSCALL_64_after_hwframe+0x62/0xc7 [ 1010.716444] RIP: 0033:0x7fd1508cbd49 [ 1010.716452] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 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 73 01 c3 48 8b 0d ef 70 0d 00 f7 d8 64 89 01 48 [ 1010.716456] RSP: 002b:00007fff18872348 EFLAGS: 00000202 ORIG_RAX: 000000000000002e [ 1010.716463] RAX: ffffffffffffffda RBX: 000055f72bf0eac0 RCX: 00007fd1508cbd49 [ 1010.716468] RDX: 0000000000000000 RSI: 0000000020000280 RDI: 0000000000000006 [ 1010.716473] RBP: 00007fff18872360 R08: 00007fff18872360 R09: 00007fff18872360 [ 1010.716478] R10: 00007fff18872360 R11: 0000000000000202 R12: 000055f72bf0e1b0 [ 1010.716482] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 [ 1010.716491] Modules linked in: gtp(+) udp_tunnel ib_core uinput af_packet rfkill qrtr joydev hid_generic usbhid hid kvm_intel iTCO_wdt intel_pmc_bxt iTCO_vendor_support kvm snd_hda_codec_generic ledtrig_audio irqbypass crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel snd_hda_intel nls_utf8 snd_intel_dspcfg nls_cp866 psmouse aesni_intel vfat crypto_simd fat cryptd glue_helper snd_hda_codec pcspkr snd_hda_core i2c_i801 snd_hwdep i2c_smbus xhci_pci snd_pcm lpc_ich xhci_pci_renesas xhci_hcd qemu_fw_cfg tiny_power_button button sch_fq_codel vboxvideo drm_vram_helper drm_ttm_helper ttm vboxsf vboxguest snd_seq_midi snd_seq_midi_event snd_seq snd_rawmidi snd_seq_device snd_timer snd soundcore msr fuse efi_pstore dm_mod ip_tables x_tables autofs4 virtio_gpu virtio_dma_buf drm_kms_helper cec rc_core drm virtio_rng virtio_scsi rng_core virtio_balloon virtio_blk virtio_net virtio_console net_failover failover ahci libahci libata evdev scsi_mod input_leds serio_raw virtio_pci intel_agp [ 1010.716674] virtio_ring intel_gtt virtio [last unloaded: gtp] [ 1010.716693] ---[ end trace 04990a4ce61e174b ]--- Cc: stable@vger.kernel.org Signed-off-by: Alexander Ofitserov Fixes: 459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)") Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20240228114703.465107-1-oficerovas@altlinux.org Signed-off-by: Paolo Abeni --- drivers/net/gtp.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 2129ae42c703..2b5357d94ff5 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -1903,26 +1903,26 @@ static int __init gtp_init(void) get_random_bytes(>p_h_initval, sizeof(gtp_h_initval)); - err = rtnl_link_register(>p_link_ops); + err = register_pernet_subsys(>p_net_ops); if (err < 0) goto error_out; - err = register_pernet_subsys(>p_net_ops); + err = rtnl_link_register(>p_link_ops); if (err < 0) - goto unreg_rtnl_link; + goto unreg_pernet_subsys; err = genl_register_family(>p_genl_family); if (err < 0) - goto unreg_pernet_subsys; + goto unreg_rtnl_link; pr_info("GTP module loaded (pdp ctx size %zd bytes)\n", sizeof(struct pdp_ctx)); return 0; -unreg_pernet_subsys: - unregister_pernet_subsys(>p_net_ops); unreg_rtnl_link: rtnl_link_unregister(>p_link_ops); +unreg_pernet_subsys: + unregister_pernet_subsys(>p_net_ops); error_out: pr_err("error loading GTP module loaded\n"); return err; From f7fa16d49837f947ee59492958f9e6f0e51d9a78 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 28 Feb 2024 23:43:57 +0100 Subject: [PATCH 68/73] tls: decrement decrypt_pending if no async completion will be called With mixed sync/async decryption, or failures of crypto_aead_decrypt, we increment decrypt_pending but we never do the corresponding decrement since tls_decrypt_done will not be called. In this case, we should decrement decrypt_pending immediately to avoid getting stuck. For example, the prequeue prequeue test gets stuck with mixed modes (one async decrypt + one sync decrypt). Fixes: 94524d8fc965 ("net/tls: Add support for async decryption of tls records") Signed-off-by: Sabrina Dubroca Link: https://lore.kernel.org/r/c56d5fc35543891d5319f834f25622360e1bfbec.1709132643.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index de96959336c4..9f23ba321efe 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -289,6 +289,8 @@ static int tls_do_decryption(struct sock *sk, return 0; ret = crypto_wait_req(ret, &ctx->async_wait); + } else if (darg->async) { + atomic_dec(&ctx->decrypt_pending); } darg->async = false; From 6caaf104423d809b49a67ee6500191d063b40dc6 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 28 Feb 2024 23:43:58 +0100 Subject: [PATCH 69/73] tls: fix peeking with sync+async decryption If we peek from 2 records with a currently empty rx_list, and the first record is decrypted synchronously but the second record is decrypted async, the following happens: 1. decrypt record 1 (sync) 2. copy from record 1 to the userspace's msg 3. queue the decrypted record to rx_list for future read(!PEEK) 4. decrypt record 2 (async) 5. queue record 2 to rx_list 6. call process_rx_list to copy data from the 2nd record We currently pass copied=0 as skip offset to process_rx_list, so we end up copying once again from the first record. We should skip over the data we've already copied. Seen with selftest tls.12_aes_gcm.recv_peek_large_buf_mult_recs Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records") Signed-off-by: Sabrina Dubroca Link: https://lore.kernel.org/r/1b132d2b2b99296bfde54e8a67672d90d6d16e71.1709132643.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 9f23ba321efe..1394fc44f378 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1950,6 +1950,7 @@ int tls_sw_recvmsg(struct sock *sk, struct strp_msg *rxm; struct tls_msg *tlm; ssize_t copied = 0; + ssize_t peeked = 0; bool async = false; int target, err; bool is_kvec = iov_iter_is_kvec(&msg->msg_iter); @@ -2097,8 +2098,10 @@ int tls_sw_recvmsg(struct sock *sk, if (err < 0) goto put_on_rx_list_err; - if (is_peek) + if (is_peek) { + peeked += chunk; goto put_on_rx_list; + } if (partially_consumed) { rxm->offset += chunk; @@ -2137,8 +2140,8 @@ int tls_sw_recvmsg(struct sock *sk, /* Drain records from the rx_list & copy if required */ if (is_peek || is_kvec) - err = process_rx_list(ctx, msg, &control, copied, - decrypted, is_peek, NULL); + err = process_rx_list(ctx, msg, &control, copied + peeked, + decrypted - peeked, is_peek, NULL); else err = process_rx_list(ctx, msg, &control, 0, async_copy_bytes, is_peek, NULL); From 41532b785e9d79636b3815a64ddf6a096647d011 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 28 Feb 2024 23:43:59 +0100 Subject: [PATCH 70/73] tls: separate no-async decryption request handling from async If we're not doing async, the handling is much simpler. There's no reference counting, we just need to wait for the completion to wake us up and return its result. We should preferably also use a separate crypto_wait. I'm not seeing a UAF as I did in the past, I think aec7961916f3 ("tls: fix race between async notify and socket close") took care of it. This will make the next fix easier. Signed-off-by: Sabrina Dubroca Link: https://lore.kernel.org/r/47bde5f649707610eaef9f0d679519966fc31061.1709132643.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 1394fc44f378..1fd37fe13ffd 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -274,9 +274,15 @@ static int tls_do_decryption(struct sock *sk, DEBUG_NET_WARN_ON_ONCE(atomic_read(&ctx->decrypt_pending) < 1); atomic_inc(&ctx->decrypt_pending); } else { + DECLARE_CRYPTO_WAIT(wait); + aead_request_set_callback(aead_req, CRYPTO_TFM_REQ_MAY_BACKLOG, - crypto_req_done, &ctx->async_wait); + crypto_req_done, &wait); + ret = crypto_aead_decrypt(aead_req); + if (ret == -EINPROGRESS || ret == -EBUSY) + ret = crypto_wait_req(ret, &wait); + return ret; } ret = crypto_aead_decrypt(aead_req); @@ -285,10 +291,7 @@ static int tls_do_decryption(struct sock *sk, ret = ret ?: -EINPROGRESS; } if (ret == -EINPROGRESS) { - if (darg->async) - return 0; - - ret = crypto_wait_req(ret, &ctx->async_wait); + return 0; } else if (darg->async) { atomic_dec(&ctx->decrypt_pending); } From 13114dc5543069f7b97991e3b79937b6da05f5b0 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 28 Feb 2024 23:44:00 +0100 Subject: [PATCH 71/73] tls: fix use-after-free on failed backlog decryption When the decrypt request goes to the backlog and crypto_aead_decrypt returns -EBUSY, tls_do_decryption will wait until all async decryptions have completed. If one of them fails, tls_do_decryption will return -EBADMSG and tls_decrypt_sg jumps to the error path, releasing all the pages. But the pages have been passed to the async callback, and have already been released by tls_decrypt_done. The only true async case is when crypto_aead_decrypt returns -EINPROGRESS. With -EBUSY, we already waited so we can tell tls_sw_recvmsg that the data is available for immediate copy, but we need to notify tls_decrypt_sg (via the new ->async_done flag) that the memory has already been released. Fixes: 859054147318 ("net: tls: handle backlogging of crypto requests") Signed-off-by: Sabrina Dubroca Link: https://lore.kernel.org/r/4755dd8d9bebdefaa19ce1439b833d6199d4364c.1709132643.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 1fd37fe13ffd..211f57164cb6 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -52,6 +52,7 @@ struct tls_decrypt_arg { struct_group(inargs, bool zc; bool async; + bool async_done; u8 tail; ); @@ -286,15 +287,18 @@ static int tls_do_decryption(struct sock *sk, } ret = crypto_aead_decrypt(aead_req); + if (ret == -EINPROGRESS) + return 0; + if (ret == -EBUSY) { ret = tls_decrypt_async_wait(ctx); - ret = ret ?: -EINPROGRESS; - } - if (ret == -EINPROGRESS) { - return 0; - } else if (darg->async) { - atomic_dec(&ctx->decrypt_pending); + darg->async_done = true; + /* all completions have run, we're not doing async anymore */ + darg->async = false; + return ret; } + + atomic_dec(&ctx->decrypt_pending); darg->async = false; return ret; @@ -1593,8 +1597,11 @@ static int tls_decrypt_sg(struct sock *sk, struct iov_iter *out_iov, /* Prepare and submit AEAD request */ err = tls_do_decryption(sk, sgin, sgout, dctx->iv, data_len + prot->tail_size, aead_req, darg); - if (err) + if (err) { + if (darg->async_done) + goto exit_free_skb; goto exit_free_pages; + } darg->skb = clear_skb ?: tls_strp_msg(ctx); clear_skb = NULL; @@ -1606,6 +1613,9 @@ static int tls_decrypt_sg(struct sock *sk, struct iov_iter *out_iov, return err; } + if (unlikely(darg->async_done)) + return 0; + if (prot->tail_size) darg->tail = dctx->tail; From 3d6423ef8d517e8924bec3f22c40285a90d652f3 Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Fri, 23 Feb 2024 11:41:52 +0100 Subject: [PATCH 72/73] kunit: Fix again checksum tests on big endian CPUs Commit b38460bc463c ("kunit: Fix checksum tests on big endian CPUs") fixed endianness issues with kunit checksum tests, but then commit 6f4c45cbcb00 ("kunit: Add tests for csum_ipv6_magic and ip_fast_csum") introduced new issues on big endian CPUs. Those issues are once again reflected by the warnings reported by sparse. So, fix them with the same approach, perform proper conversion in order to support both little and big endian CPUs. Once the conversions are properly done and the right types used, the sparse warnings are cleared as well. Reported-by: Erhard Furtner Fixes: 6f4c45cbcb00 ("kunit: Add tests for csum_ipv6_magic and ip_fast_csum") Signed-off-by: Christophe Leroy Tested-by: Charlie Jenkins Tested-by: Guenter Roeck Acked-by: Paolo Abeni Acked-by: Palmer Dabbelt Link: https://lore.kernel.org/r/73df3a9e95c2179119398ad1b4c84cdacbd8dfb6.1708684443.git.christophe.leroy@csgroup.eu Signed-off-by: Jakub Kicinski --- lib/checksum_kunit.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/lib/checksum_kunit.c b/lib/checksum_kunit.c index 225bb7701460..bf70850035c7 100644 --- a/lib/checksum_kunit.c +++ b/lib/checksum_kunit.c @@ -215,7 +215,7 @@ static const u32 init_sums_no_overflow[] = { 0xffff0000, 0xfffffffb, }; -static const __sum16 expected_csum_ipv6_magic[] = { +static const u16 expected_csum_ipv6_magic[] = { 0x18d4, 0x3085, 0x2e4b, 0xd9f4, 0xbdc8, 0x78f, 0x1034, 0x8422, 0x6fc0, 0xd2f6, 0xbeb5, 0x9d3, 0x7e2a, 0x312e, 0x778e, 0xc1bb, 0x7cf2, 0x9d1e, 0xca21, 0xf3ff, 0x7569, 0xb02e, 0xca86, 0x7e76, 0x4539, 0x45e3, 0xf28d, @@ -241,7 +241,7 @@ static const __sum16 expected_csum_ipv6_magic[] = { 0x3845, 0x1014 }; -static const __sum16 expected_fast_csum[] = { +static const u16 expected_fast_csum[] = { 0xda83, 0x45da, 0x4f46, 0x4e4f, 0x34e, 0xe902, 0xa5e9, 0x87a5, 0x7187, 0x5671, 0xf556, 0x6df5, 0x816d, 0x8f81, 0xbb8f, 0xfbba, 0x5afb, 0xbe5a, 0xedbe, 0xabee, 0x6aac, 0xe6b, 0xea0d, 0x67ea, 0x7e68, 0x8a7e, 0x6f8a, @@ -577,7 +577,8 @@ static void test_csum_no_carry_inputs(struct kunit *test) static void test_ip_fast_csum(struct kunit *test) { - __sum16 csum_result, expected; + __sum16 csum_result; + u16 expected; for (int len = IPv4_MIN_WORDS; len < IPv4_MAX_WORDS; len++) { for (int index = 0; index < NUM_IP_FAST_CSUM_TESTS; index++) { @@ -586,7 +587,7 @@ static void test_ip_fast_csum(struct kunit *test) expected_fast_csum[(len - IPv4_MIN_WORDS) * NUM_IP_FAST_CSUM_TESTS + index]; - CHECK_EQ(expected, csum_result); + CHECK_EQ(to_sum16(expected), csum_result); } } } @@ -598,7 +599,7 @@ static void test_csum_ipv6_magic(struct kunit *test) const struct in6_addr *daddr; unsigned int len; unsigned char proto; - unsigned int csum; + __wsum csum; const int daddr_offset = sizeof(struct in6_addr); const int len_offset = sizeof(struct in6_addr) + sizeof(struct in6_addr); @@ -611,10 +612,10 @@ static void test_csum_ipv6_magic(struct kunit *test) saddr = (const struct in6_addr *)(random_buf + i); daddr = (const struct in6_addr *)(random_buf + i + daddr_offset); - len = *(unsigned int *)(random_buf + i + len_offset); + len = le32_to_cpu(*(__le32 *)(random_buf + i + len_offset)); proto = *(random_buf + i + proto_offset); - csum = *(unsigned int *)(random_buf + i + csum_offset); - CHECK_EQ(expected_csum_ipv6_magic[i], + csum = *(__wsum *)(random_buf + i + csum_offset); + CHECK_EQ(to_sum16(expected_csum_ipv6_magic[i]), csum_ipv6_magic(saddr, daddr, len, proto, csum)); } #endif /* !CONFIG_NET */ From 640f41ed33b5a420e05daf395afae85e6b20c003 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 29 Feb 2024 11:05:15 -0800 Subject: [PATCH 73/73] dpll: fix build failure due to rcu_dereference_check() on unknown type MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Tasmiya reports that their compiler complains that we deref a pointer to unknown type with rcu_dereference_rtnl(): include/linux/rcupdate.h:439:9: error: dereferencing pointer to incomplete type ‘struct dpll_pin’ Unclear what compiler it is, at the moment, and we can't report but since DPLL can't be a module - move the code from the header into the source file. Fixes: 0d60d8df6f49 ("dpll: rely on rcu for netdev_dpll_pin()") Reported-by: Tasmiya Nalatwad Link: https://lore.kernel.org/all/3fcf3a2c-1c1b-42c1-bacb-78fdcd700389@linux.vnet.ibm.com/ Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240229190515.2740221-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/dpll/dpll_core.c | 5 +++++ include/linux/dpll.h | 8 ++++---- 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/dpll/dpll_core.c b/drivers/dpll/dpll_core.c index 4c2bb27c99fe..241db366b2c7 100644 --- a/drivers/dpll/dpll_core.c +++ b/drivers/dpll/dpll_core.c @@ -42,6 +42,11 @@ struct dpll_pin_registration { void *priv; }; +struct dpll_pin *netdev_dpll_pin(const struct net_device *dev) +{ + return rcu_dereference_rtnl(dev->dpll_pin); +} + struct dpll_device *dpll_device_get_by_id(int id) { if (xa_get_mark(&dpll_device_xa, id, DPLL_REGISTERED)) diff --git a/include/linux/dpll.h b/include/linux/dpll.h index 4ec2fe9caf5a..c60591308ae8 100644 --- a/include/linux/dpll.h +++ b/include/linux/dpll.h @@ -169,13 +169,13 @@ int dpll_device_change_ntf(struct dpll_device *dpll); int dpll_pin_change_ntf(struct dpll_pin *pin); +#if !IS_ENABLED(CONFIG_DPLL) static inline struct dpll_pin *netdev_dpll_pin(const struct net_device *dev) { -#if IS_ENABLED(CONFIG_DPLL) - return rcu_dereference_rtnl(dev->dpll_pin); -#else return NULL; -#endif } +#else +struct dpll_pin *netdev_dpll_pin(const struct net_device *dev); +#endif #endif