* [PATCH AUTOSEL 4.20 207/304] mac80211: Properly handle SKB with radiotap only
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Ilan Peer, Luca Coelho, Johannes Berg, Sasha Levin,
linux-wireless, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Ilan Peer <ilan.peer@intel.com>
[ Upstream commit 8020919a9b99d6c990dc6a50e8215e291fbbe5a6 ]
The monitor interface Rx handling of SKBs that contain only
radiotap information was buggy as it tried to access the
SKB assuming it contains a frame.
To fix this, check the RX_FLAG_NO_PSDU flag in the Rx status
(indicting that the SKB contains only radiotap information),
and do not perform data path specific processing when the flag
is set.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
net/mac80211/rx.c | 26 +++++++++++++++++---------
1 file changed, 17 insertions(+), 9 deletions(-)
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 428f7ad5f9b5..77d996a60f12 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -753,6 +753,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
struct ieee80211_sub_if_data *monitor_sdata =
rcu_dereference(local->monitor_sdata);
bool only_monitor = false;
+ unsigned int min_head_len;
if (status->flag & RX_FLAG_RADIOTAP_HE)
rtap_space += sizeof(struct ieee80211_radiotap_he);
@@ -766,6 +767,8 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
rtap_space += sizeof(*rtap) + rtap->len + rtap->pad;
}
+ min_head_len = rtap_space;
+
/*
* First, we may need to make a copy of the skb because
* (1) we need to modify it for radiotap (if not present), and
@@ -775,18 +778,23 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
* the SKB because it has a bad FCS/PLCP checksum.
*/
- if (ieee80211_hw_check(&local->hw, RX_INCLUDES_FCS)) {
- if (unlikely(origskb->len <= FCS_LEN)) {
- /* driver bug */
- WARN_ON(1);
- dev_kfree_skb(origskb);
- return NULL;
+ if (!(status->flag & RX_FLAG_NO_PSDU)) {
+ if (ieee80211_hw_check(&local->hw, RX_INCLUDES_FCS)) {
+ if (unlikely(origskb->len <= FCS_LEN + rtap_space)) {
+ /* driver bug */
+ WARN_ON(1);
+ dev_kfree_skb(origskb);
+ return NULL;
+ }
+ present_fcs_len = FCS_LEN;
}
- present_fcs_len = FCS_LEN;
+
+ /* also consider the hdr->frame_control */
+ min_head_len += 2;
}
- /* ensure hdr->frame_control and vendor radiotap data are in skb head */
- if (!pskb_may_pull(origskb, 2 + rtap_space)) {
+ /* ensure that the expected data elements are in skb head */
+ if (!pskb_may_pull(origskb, min_head_len)) {
dev_kfree_skb(origskb);
return NULL;
}
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 216/304] net: hns3: fix error handling int the hns3_get_vector_ring_chain
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Huazhong Tan, Peng Li, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Huazhong Tan <tanhuazhong@huawei.com>
[ Upstream commit cda69d244585bc4497d3bb878c22fe2b6ad647c1 ]
When hns3_get_vector_ring_chain() failed in the
hns3_nic_init_vector_data(), it should do the error handling instead
of return directly.
Also, cur_chain should be freed instead of chain and head->next should
be set to NULL in error handling of hns3_get_vector_ring_chain.
This patch fixes them.
Fixes: 73b907a083b8 ("net: hns3: bugfix for buffer not free problem during resetting")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 20fcf0d1c2ce..4aea3f57586e 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2783,9 +2783,10 @@ err_free_chain:
cur_chain = head->next;
while (cur_chain) {
chain = cur_chain->next;
- devm_kfree(&pdev->dev, chain);
+ devm_kfree(&pdev->dev, cur_chain);
cur_chain = chain;
}
+ head->next = NULL;
return -ENOMEM;
}
@@ -2876,7 +2877,7 @@ static int hns3_nic_init_vector_data(struct hns3_nic_priv *priv)
ret = hns3_get_vector_ring_chain(tqp_vector,
&vector_ring_chain);
if (ret)
- return ret;
+ goto map_ring_fail;
ret = h->ae_algo->ops->map_ring_to_vector(h,
tqp_vector->vector_irq, &vector_ring_chain);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 224/304] xfrm6_tunnel: Fix spi check in __xfrm6_tunnel_alloc_spi
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: YueHaibing, Steffen Klassert, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: YueHaibing <yuehaibing@huawei.com>
[ Upstream commit fa89a4593b927b3f59c3b69379f31d3b22272e4e ]
gcc warn this:
net/ipv6/xfrm6_tunnel.c:143 __xfrm6_tunnel_alloc_spi() warn:
always true condition '(spi <= 4294967295) => (0-u32max <= u32max)'
'spi' is u32, which always not greater than XFRM6_TUNNEL_SPI_MAX
because of wrap around. So the second forloop will never reach.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
net/ipv6/xfrm6_tunnel.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index 4a46df8441c9..f5b4febeaa25 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -144,6 +144,9 @@ static u32 __xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr)
index = __xfrm6_tunnel_spi_check(net, spi);
if (index >= 0)
goto alloc_spi;
+
+ if (spi == XFRM6_TUNNEL_SPI_MAX)
+ break;
}
for (spi = XFRM6_TUNNEL_SPI_MIN; spi < xfrm6_tn->spi; spi++) {
index = __xfrm6_tunnel_spi_check(net, spi);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 225/304] mlxsw: spectrum: Properly cleanup LAG uppers when removing port from LAG
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Ido Schimmel, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Ido Schimmel <idosch@mellanox.com>
[ Upstream commit be2d6f421f680e01d58f7cd452646e0d8586d49b ]
When a LAG device or a VLAN device on top of it is enslaved to a bridge,
the driver propagates the CHANGEUPPER event to the LAG's slaves.
This causes each physical port to increase the reference count of the
internal representation of the bridge port by calling
mlxsw_sp_port_bridge_join().
However, when a port is removed from a LAG, the corresponding leave()
function is not called and the reference count is not decremented. This
leads to ugly hacks such as mlxsw_sp_bridge_port_should_destroy() that
try to understand if the bridge port should be destroyed even when its
reference count is not 0.
Instead, make sure that when a port is unlinked from a LAG it would see
the same events as if the LAG (or its uppers) were unlinked from a
bridge.
The above is achieved by walking the LAG's uppers when a port is
unlinked and calling mlxsw_sp_port_bridge_leave() for each upper that is
enslaved to a bridge.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
.../net/ethernet/mellanox/mlxsw/spectrum.c | 23 ++++++++++++++++
.../mellanox/mlxsw/spectrum_switchdev.c | 27 +------------------
2 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 124aee09e953..8f1180fff955 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -4298,6 +4298,25 @@ void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port)
dev_put(mlxsw_sp_port->dev);
}
+static void
+mlxsw_sp_port_lag_uppers_cleanup(struct mlxsw_sp_port *mlxsw_sp_port,
+ struct net_device *lag_dev)
+{
+ struct net_device *br_dev = netdev_master_upper_dev_get(lag_dev);
+ struct net_device *upper_dev;
+ struct list_head *iter;
+
+ if (netif_is_bridge_port(lag_dev))
+ mlxsw_sp_port_bridge_leave(mlxsw_sp_port, lag_dev, br_dev);
+
+ netdev_for_each_upper_dev_rcu(lag_dev, upper_dev, iter) {
+ if (!netif_is_bridge_port(upper_dev))
+ continue;
+ br_dev = netdev_master_upper_dev_get(upper_dev);
+ mlxsw_sp_port_bridge_leave(mlxsw_sp_port, upper_dev, br_dev);
+ }
+}
+
static int mlxsw_sp_lag_create(struct mlxsw_sp *mlxsw_sp, u16 lag_id)
{
char sldr_pl[MLXSW_REG_SLDR_LEN];
@@ -4490,6 +4509,10 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port,
/* Any VLANs configured on the port are no longer valid */
mlxsw_sp_port_vlan_flush(mlxsw_sp_port);
+ /* Make the LAG and its directly linked uppers leave bridges they
+ * are memeber in
+ */
+ mlxsw_sp_port_lag_uppers_cleanup(mlxsw_sp_port, lag_dev);
if (lag->ref_count == 1)
mlxsw_sp_lag_destroy(mlxsw_sp, lag_id);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 69f556ddb934..c514af438fc2 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -292,30 +292,6 @@ mlxsw_sp_bridge_port_destroy(struct mlxsw_sp_bridge_port *bridge_port)
kfree(bridge_port);
}
-static bool
-mlxsw_sp_bridge_port_should_destroy(const struct mlxsw_sp_bridge_port *
- bridge_port)
-{
- struct net_device *dev = bridge_port->dev;
- struct mlxsw_sp *mlxsw_sp;
-
- if (is_vlan_dev(dev))
- mlxsw_sp = mlxsw_sp_lower_get(vlan_dev_real_dev(dev));
- else
- mlxsw_sp = mlxsw_sp_lower_get(dev);
-
- /* In case ports were pulled from out of a bridged LAG, then
- * it's possible the reference count isn't zero, yet the bridge
- * port should be destroyed, as it's no longer an upper of ours.
- */
- if (!mlxsw_sp && list_empty(&bridge_port->vlans_list))
- return true;
- else if (bridge_port->ref_count == 0)
- return true;
- else
- return false;
-}
-
static struct mlxsw_sp_bridge_port *
mlxsw_sp_bridge_port_get(struct mlxsw_sp_bridge *bridge,
struct net_device *brport_dev)
@@ -353,8 +329,7 @@ static void mlxsw_sp_bridge_port_put(struct mlxsw_sp_bridge *bridge,
{
struct mlxsw_sp_bridge_device *bridge_device;
- bridge_port->ref_count--;
- if (!mlxsw_sp_bridge_port_should_destroy(bridge_port))
+ if (--bridge_port->ref_count != 0)
return;
bridge_device = bridge_port->bridge_device;
mlxsw_sp_bridge_port_destroy(bridge_port);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 223/304] mac80211: fix radiotap vendor presence bitmap handling
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Johannes Berg, Luca Coelho, Sasha Levin, linux-wireless, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Johannes Berg <johannes.berg@intel.com>
[ Upstream commit efc38dd7d5fa5c8cdd0c917c5d00947aa0539443 ]
Due to the alignment handling, it actually matters where in the code
we add the 4 bytes for the presence bitmap to the length; the first
field is the timestamp with 8 byte alignment so we need to add the
space for the extra vendor namespace presence bitmap *before* we do
any alignment for the fields.
Move the presence bitmap length accounting to the right place to fix
the alignment for the data properly.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
net/mac80211/rx.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 77d996a60f12..85c365fc7a0c 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -143,6 +143,9 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local,
/* allocate extra bitmaps */
if (status->chains)
len += 4 * hweight8(status->chains);
+ /* vendor presence bitmap */
+ if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA)
+ len += 4;
if (ieee80211_have_rx_timestamp(status)) {
len = ALIGN(len, 8);
@@ -207,8 +210,6 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local,
if (status->flag & RX_FLAG_RADIOTAP_VENDOR_DATA) {
struct ieee80211_vendor_radiotap *rtap = (void *)skb->data;
- /* vendor presence bitmap */
- len += 4;
/* alignment for fixed 6-byte vendor data header */
len = ALIGN(len, 2);
/* vendor data header */
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 229/304] cw1200: Fix concurrency use-after-free bugs in cw1200_hw_scan()
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Jia-Ju Bai, Kalle Valo, Sasha Levin, linux-wireless, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
[ Upstream commit 4f68ef64cd7feb1220232bd8f501d8aad340a099 ]
The function cw1200_bss_info_changed() and cw1200_hw_scan() can be
concurrently executed.
The two functions both access a possible shared variable "frame.skb".
This shared variable is freed by dev_kfree_skb() in cw1200_upload_beacon(),
which is called by cw1200_bss_info_changed(). The free operation is
protected by a mutex lock "priv->conf_mutex" in cw1200_bss_info_changed().
In cw1200_hw_scan(), this shared variable is accessed without the
protection of the mutex lock "priv->conf_mutex".
Thus, concurrency use-after-free bugs may occur.
To fix these bugs, the original calls to mutex_lock(&priv->conf_mutex) and
mutex_unlock(&priv->conf_mutex) are moved to the places, which can
protect the accesses to the shared variable.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/st/cw1200/scan.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/net/wireless/st/cw1200/scan.c b/drivers/net/wireless/st/cw1200/scan.c
index 67213f11acbd..0a9eac93dd01 100644
--- a/drivers/net/wireless/st/cw1200/scan.c
+++ b/drivers/net/wireless/st/cw1200/scan.c
@@ -78,6 +78,10 @@ int cw1200_hw_scan(struct ieee80211_hw *hw,
if (req->n_ssids > WSM_SCAN_MAX_NUM_OF_SSIDS)
return -EINVAL;
+ /* will be unlocked in cw1200_scan_work() */
+ down(&priv->scan.lock);
+ mutex_lock(&priv->conf_mutex);
+
frame.skb = ieee80211_probereq_get(hw, priv->vif->addr, NULL, 0,
req->ie_len);
if (!frame.skb)
@@ -86,19 +90,15 @@ int cw1200_hw_scan(struct ieee80211_hw *hw,
if (req->ie_len)
skb_put_data(frame.skb, req->ie, req->ie_len);
- /* will be unlocked in cw1200_scan_work() */
- down(&priv->scan.lock);
- mutex_lock(&priv->conf_mutex);
-
ret = wsm_set_template_frame(priv, &frame);
if (!ret) {
/* Host want to be the probe responder. */
ret = wsm_set_probe_responder(priv, true);
}
if (ret) {
+ dev_kfree_skb(frame.skb);
mutex_unlock(&priv->conf_mutex);
up(&priv->scan.lock);
- dev_kfree_skb(frame.skb);
return ret;
}
@@ -120,10 +120,9 @@ int cw1200_hw_scan(struct ieee80211_hw *hw,
++priv->scan.n_ssids;
}
- mutex_unlock(&priv->conf_mutex);
-
if (frame.skb)
dev_kfree_skb(frame.skb);
+ mutex_unlock(&priv->conf_mutex);
queue_work(priv->workqueue, &priv->scan.work);
return 0;
}
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 230/304] net: hns3: add max vector number check for pf
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Jian Shen, Peng Li, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Jian Shen <shenjian15@huawei.com>
[ Upstream commit 75edb610860fda65ceedb017fc69afabd2806b8b ]
Each pf supports max 64 vectors and 128 tqps. For 2p/4p core scenario,
there may be more than 64 cpus online. So the result of min_t(u16,
num_Online_cpus(), tqp_num) may be more than 64. This patch adds check
for the vector number.
Fixes: dd38c72604dc ("net: hns3: fix for coalesce configuration lost during reset")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 85d534a24f2c..beb5dbd5e98d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2902,6 +2902,8 @@ map_ring_fail:
static int hns3_nic_alloc_vector_data(struct hns3_nic_priv *priv)
{
+#define HNS3_VECTOR_PF_MAX_NUM 64
+
struct hnae3_handle *h = priv->ae_handle;
struct hns3_enet_tqp_vector *tqp_vector;
struct hnae3_vector_info *vector;
@@ -2914,6 +2916,8 @@ static int hns3_nic_alloc_vector_data(struct hns3_nic_priv *priv)
/* RSS size, cpu online and vector_num should be the same */
/* Should consider 2p/4p later */
vector_num = min_t(u16, num_online_cpus(), tqp_num);
+ vector_num = min_t(u16, vector_num, HNS3_VECTOR_PF_MAX_NUM);
+
vector = devm_kcalloc(&pdev->dev, vector_num, sizeof(*vector),
GFP_KERNEL);
if (!vector)
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 232/304] net: hns3: don't restore rules when flow director is disabled
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Jian Shen, Peng Li, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Jian Shen <shenjian15@huawei.com>
[ Upstream commit 8edc2285b7285c5fbffe438dc8113260d3e76656 ]
When user disables flow director, all the rules will be disabled. But
when reset happens, it will restore all the rules again. It's not
reasonable. This patch fixes it by add flow director status check before
restore fules.
Fixes: 6871af29b3ab ("net: hns3: Add reset handle for flow director")
Fixes: c17852a8932f ("net: hns3: Add support for enable/disable flow director")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index ffdd96020860..a7895aefe291 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -4339,6 +4339,10 @@ static int hclge_restore_fd_entries(struct hnae3_handle *handle)
if (!hnae3_dev_fd_supported(hdev))
return -EOPNOTSUPP;
+ /* if fd is disabled, should not restore it when reset */
+ if (!hdev->fd_cfg.fd_en)
+ return 0;
+
hlist_for_each_entry_safe(rule, node, &hdev->fd_rule_list, rule_node) {
ret = hclge_config_action(hdev, HCLGE_FD_STAGE_1, rule);
if (!ret)
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 234/304] iwlwifi: mvm: fix setting HE ppe FW config
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Naftali Goldstein, Luca Coelho, Kalle Valo, Sasha Levin,
linux-wireless, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Naftali Goldstein <naftali.goldstein@intel.com>
[ Upstream commit 189b8d441b0f7825f0b4278851c52afaa0515ed2 ]
The FW expects to get the ppe value for each NSS-BW pair in the same
format as in the he phy capabilities IE, which means that a value of 0
implies ppe should be used for BPSK (mcs 0). If there are no PPE
thresholds in the IE, or if for some NSS-RU pair there's no threshold
set for it (this could happen because it's a variable-sized field), it
means no PPE should not be used for that pair, so the value sent to FW
should be 7 which corresponds to "none".
Fixes: 514c30696fbc ("iwlwifi: add support for IEEE802.11ax")
Signed-off-by: Naftali Goldstein <naftali.goldstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/intel/iwlwifi/fw/api/mac.h | 2 +-
drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 8 +++++++-
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
index 1dd23f846fb9..f3ccd7948340 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
@@ -442,7 +442,7 @@ struct iwl_he_backoff_conf {
* Support for Nss x BW (or RU) matrix:
* (0=SISO, 1=MIMO2) x (0-20MHz, 1-40MHz, 2-80MHz, 3-160MHz)
* Each entry contains 2 QAM thresholds for 8us and 16us:
- * 0=BPSK, 1=QPSK, 2=16QAM, 3=64QAM, 4=256QAM, 5=1024QAM, 6/7=RES
+ * 0=BPSK, 1=QPSK, 2=16QAM, 3=64QAM, 4=256QAM, 5=1024QAM, 6=RES, 7=NONE
* i.e. QAM_th1 < QAM_th2 such if TX uses QAM_tx:
* QAM_tx < QAM_th1 --> PPE=0us
* QAM_th1 <= QAM_tx < QAM_th2 --> PPE=8us
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 00f831d88366..5a42c617c54c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2005,7 +2005,13 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm,
if (sta->he_cap.he_cap_elem.mac_cap_info[4] & IEEE80211_HE_MAC_CAP4_BQR)
sta_ctxt_cmd.htc_flags |= cpu_to_le32(IWL_HE_HTC_BQR_SUPP);
- /* If PPE Thresholds exist, parse them into a FW-familiar format */
+ /*
+ * Initialize the PPE thresholds to "None" (7), as described in Table
+ * 9-262ac of 80211.ax/D3.0.
+ */
+ memset(&sta_ctxt_cmd.pkt_ext, 7, sizeof(sta_ctxt_cmd.pkt_ext));
+
+ /* If PPE Thresholds exist, parse them into a FW-familiar format. */
if (sta->he_cap.he_cap_elem.phy_cap_info[6] &
IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) {
u8 nss = (sta->he_cap.ppe_thres[0] &
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 236/304] mlx5: update timecounter at least twice per counter overflow
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Miroslav Lichvar, Richard Cochran, Ariel Levkovich,
Saeed Mahameed, Sasha Levin, netdev, linux-rdma
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Miroslav Lichvar <mlichvar@redhat.com>
[ Upstream commit 5d8678365c90b9ce1fd2243ff5ea562609f6cec1 ]
The timecounter needs to be updated at least once in half of the
cyclecounter interval to prevent timecounter_cyc2time() interpreting a
new timestamp as an old value and causing a backward jump.
This would be an issue if the timecounter multiplier was so small that
the update interval would not be limited by the 64-bit overflow in
multiplication.
Shorten the calculated interval to make sure the timecounter is updated
in time even when the system clock is slowed down by up to 10%, the
multiplier is increased by up to 10%, and the scheduled overflow check
is late by 15%.
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Ariel Levkovich <lariel@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
index 0d90b1b4a3d3..2d6168ee99e8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
@@ -511,14 +511,14 @@ void mlx5_init_clock(struct mlx5_core_dev *mdev)
ktime_to_ns(ktime_get_real()));
/* Calculate period in seconds to call the overflow watchdog - to make
- * sure counter is checked at least once every wrap around.
+ * sure counter is checked at least twice every wrap around.
* The period is calculated as the minimum between max HW cycles count
* (The clock source mask) and max amount of cycles that can be
* multiplied by clock multiplier where the result doesn't exceed
* 64bits.
*/
overflow_cycles = div64_u64(~0ULL >> 1, clock->cycles.mult);
- overflow_cycles = min(overflow_cycles, clock->cycles.mask >> 1);
+ overflow_cycles = min(overflow_cycles, div_u64(clock->cycles.mask, 3));
ns = cyclecounter_cyc2ns(&clock->cycles, overflow_cycles,
frac, &frac);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 242/304] ath10k: fix tx_stats memory leak
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Zhi Chen, Kalle Valo, Sasha Levin, ath10k, linux-wireless, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Zhi Chen <zhichen@codeaurora.org>
[ Upstream commit 386f97e3b201d18578abb0e7037b85a1ae50c0a3 ]
Memory of tx_stats was allocated when a STA was added. But it's not freed
if the STA failed to be added to driver. This issue could be seen in MDK3
attack case when STA number reached the limit.
Tested: QCA9984 with firmware ver 10.4-3.9.0.1-00005
Signed-off-by: Zhi Chen <zhichen@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/ath/ath10k/mac.c | 22 +++++++++++++---------
1 file changed, 13 insertions(+), 9 deletions(-)
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 6436dc229be5..400495858e4e 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -6293,15 +6293,6 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
ar->num_stations + 1, ar->max_num_stations,
ar->num_peers + 1, ar->max_num_peers);
- if (ath10k_debug_is_extd_tx_stats_enabled(ar)) {
- arsta->tx_stats = kzalloc(sizeof(*arsta->tx_stats),
- GFP_KERNEL);
- if (!arsta->tx_stats) {
- ret = -ENOMEM;
- goto exit;
- }
- }
-
num_tdls_stations = ath10k_mac_tdls_vif_stations_count(hw, vif);
num_tdls_vifs = ath10k_mac_tdls_vifs_count(hw);
@@ -6323,12 +6314,22 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
goto exit;
}
+ if (ath10k_debug_is_extd_tx_stats_enabled(ar)) {
+ arsta->tx_stats = kzalloc(sizeof(*arsta->tx_stats),
+ GFP_KERNEL);
+ if (!arsta->tx_stats) {
+ ret = -ENOMEM;
+ goto exit;
+ }
+ }
+
ret = ath10k_peer_create(ar, vif, sta, arvif->vdev_id,
sta->addr, peer_type);
if (ret) {
ath10k_warn(ar, "failed to add peer %pM for vdev %d when adding a new sta: %i\n",
sta->addr, arvif->vdev_id, ret);
ath10k_mac_dec_num_stations(arvif, sta);
+ kfree(arsta->tx_stats);
goto exit;
}
@@ -6341,6 +6342,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
spin_unlock_bh(&ar->data_lock);
ath10k_peer_delete(ar, arvif->vdev_id, sta->addr);
ath10k_mac_dec_num_stations(arvif, sta);
+ kfree(arsta->tx_stats);
ret = -ENOENT;
goto exit;
}
@@ -6361,6 +6363,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
ath10k_peer_delete(ar, arvif->vdev_id,
sta->addr);
ath10k_mac_dec_num_stations(arvif, sta);
+ kfree(arsta->tx_stats);
goto exit;
}
@@ -6372,6 +6375,7 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
sta->addr, arvif->vdev_id, ret);
ath10k_peer_delete(ar, arvif->vdev_id, sta->addr);
ath10k_mac_dec_num_stations(arvif, sta);
+ kfree(arsta->tx_stats);
if (num_tdls_stations != 0)
goto exit;
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 247/304] i40e: define proper net_device::neigh_priv_len
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Konstantin Khorenko, Jeff Kirsher, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Konstantin Khorenko <khorenko@virtuozzo.com>
[ Upstream commit 31389b53b3e0b535867af9090a5d19ec64768d55 ]
Out of bound read reported by KASan.
i40iw_net_event() reads unconditionally 16 bytes from
neigh->primary_key while the memory allocated for
"neighbour" struct is evaluated in neigh_alloc() as
tbl->entry_size + dev->neigh_priv_len
where "dev" is a net_device.
But the driver does not setup dev->neigh_priv_len and
we read beyond the neigh entry allocated memory,
so the patch in the next mail fixes this.
Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 419cf3faada6..f97c3d5ab884 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -12339,6 +12339,9 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
ether_addr_copy(netdev->dev_addr, mac_addr);
ether_addr_copy(netdev->perm_addr, mac_addr);
+ /* i40iw_net_event() reads 16 bytes from neigh->primary_key */
+ netdev->neigh_priv_len = sizeof(u32) * 4;
+
netdev->priv_flags |= IFF_UNICAST_FLT;
netdev->priv_flags |= IFF_SUPP_NOFCS;
/* Setup netdev TC information */
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 248/304] ice: Do not enable NAPI on q_vectors that have no rings
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Young Xiao, Jeff Kirsher, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Young Xiao <YangX92@hotmail.com>
[ Upstream commit eec903769b4ea476591ffff73bb7359f14f38c51 ]
If ice driver has q_vectors w/ active NAPI that has no rings,
then this will result in a divide by zero error. To correct it
I am updating the driver code so that we only support NAPI on
q_vectors that have 1 or more rings allocated to them.
See commit 13a8cd191a2b ("i40e: Do not enable NAPI on q_vectors
that have no rings") for detail.
Signed-off-by: Young Xiao <YangX92@hotmail.com>
Acked-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/intel/ice/ice_main.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 333312a1d595..945000449279 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -2563,8 +2563,12 @@ static void ice_napi_enable_all(struct ice_vsi *vsi)
if (!vsi->netdev)
return;
- for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
- napi_enable(&vsi->q_vectors[q_idx]->napi);
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
+ struct ice_q_vector *q_vector = vsi->q_vectors[q_idx];
+
+ if (q_vector->rx.ring || q_vector->tx.ring)
+ napi_enable(&q_vector->napi);
+ }
}
/**
@@ -2931,8 +2935,12 @@ static void ice_napi_disable_all(struct ice_vsi *vsi)
if (!vsi->netdev)
return;
- for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
- napi_disable(&vsi->q_vectors[q_idx]->napi);
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
+ struct ice_q_vector *q_vector = vsi->q_vectors[q_idx];
+
+ if (q_vector->rx.ring || q_vector->tx.ring)
+ napi_disable(&q_vector->napi);
+ }
}
/**
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 249/304] igb: Fix an issue that PME is not enabled during runtime suspend
From: Sasha Levin @ 2019-01-28 15:42 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Kai-Heng Feng, Jeff Kirsher, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Kai-Heng Feng <kai.heng.feng@canonical.com>
[ Upstream commit 1fb3a7a75e2efcc83ef21f2434069cddd6fae6f5 ]
I210 ethernet card doesn't wakeup when a cable gets plugged. It's
because its PME is not set.
Since commit 42eca2302146 ("PCI: Don't touch card regs after runtime
suspend D3"), if the PCI state is saved, pci_pm_runtime_suspend() stops
calling pci_finish_runtime_suspend(), which enables the PCI PME.
To fix the issue, let's not to save PCI states when it's runtime
suspend, to let the PCI subsystem enables PME.
Fixes: 42eca2302146 ("PCI: Don't touch card regs after runtime suspend D3")
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/intel/igb/igb_main.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 5df88ad8ac81..93f150784cfc 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -8770,9 +8770,11 @@ static int __igb_shutdown(struct pci_dev *pdev, bool *enable_wake,
rtnl_unlock();
#ifdef CONFIG_PM
- retval = pci_save_state(pdev);
- if (retval)
- return retval;
+ if (!runtime) {
+ retval = pci_save_state(pdev);
+ if (retval)
+ return retval;
+ }
#endif
status = rd32(E1000_STATUS);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 272/304] niu: fix missing checks of niu_pci_eeprom_read
From: Sasha Levin @ 2019-01-28 15:43 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Kangjie Lu, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Kangjie Lu <kjlu@umn.edu>
[ Upstream commit 26fd962bde0b15e54234fe762d86bc0349df1de4 ]
niu_pci_eeprom_read() may fail, so we should check its return value
before using the read data.
Signed-off-by: Kangjie Lu <kjlu@umn.edu>
Acked-by: Shannon Nelson <shannon.lee.nelson@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/ethernet/sun/niu.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c
index 9319d84bf49f..d84501441edd 100644
--- a/drivers/net/ethernet/sun/niu.c
+++ b/drivers/net/ethernet/sun/niu.c
@@ -8100,6 +8100,8 @@ static int niu_pci_vpd_scan_props(struct niu *np, u32 start, u32 end)
start += 3;
prop_len = niu_pci_eeprom_read(np, start + 4);
+ if (prop_len < 0)
+ return prop_len;
err = niu_pci_vpd_get_propname(np, start + 5, namebuf, 64);
if (err < 0)
return err;
@@ -8144,8 +8146,12 @@ static int niu_pci_vpd_scan_props(struct niu *np, u32 start, u32 end)
netif_printk(np, probe, KERN_DEBUG, np->dev,
"VPD_SCAN: Reading in property [%s] len[%d]\n",
namebuf, prop_len);
- for (i = 0; i < prop_len; i++)
- *prop_buf++ = niu_pci_eeprom_read(np, off + i);
+ for (i = 0; i < prop_len; i++) {
+ err = niu_pci_eeprom_read(np, off + i);
+ if (err >= 0)
+ *prop_buf = err;
+ ++prop_buf;
+ }
}
start += len;
--
2.19.1
^ permalink raw reply related
* Re: [PATCH net-next v2 1/2] net: dsa: mv88e6xxx: Save switch rules
From: Miquel Raynal @ 2019-01-28 15:57 UTC (permalink / raw)
To: Andrew Lunn
Cc: Florian Fainelli, Vivien Didelot, David S. Miller, netdev,
linux-kernel, Thomas Petazzoni, Gregory Clement, Antoine Tenart,
Maxime Chevallier, Nadav Haklai
In-Reply-To: <20190128144417.GG4765@lunn.ch>
Hi Andrew,
Thanks for helping!
Andrew Lunn <andrew@lunn.ch> wrote on Mon, 28 Jan 2019 15:44:17 +0100:
> > I don't see where VLAN and bridge information are cached, can you point
> > me to the relevant locations?
>
> Miquèl
>
> The bridge should have all that information. You need to ask it to
> enumerate the current configuration and replay it to the switch.
>
> There might be something in the Mellanox driver you can copy? But i've
> not looked, i'm just guessing.
I am still searching but so far I did not find a mechanism reading the
configuration of the bridge out of a 'net' object. Indeed there are
multiple lists with the configuration but they are all 'mellanox'
objects, they do not belong to the core.
Maybe I don't find this configuration because I don't know what it is.
I imagine this configuration being one (or multiple) list(s), stored
somewhere in a net_device being a bridge. Am I on the wrong path?
Otherwise I might just save my own structures in net/dsa/switch.c like
I did for the mv88e6xx driver, and once this works, net-folks might
want to optimize the memory consumption and re-use the bridge
configuration directly?
> We also need to think about how we are going to test this. There is a
> lot of state information in a switch. So we are going to need some
> pretty good tests to show we have recreated all of it.
My understanding of all this is rather short, until know I used what
you proposed in the v1 of this series but I am all ears if I need to
add anything to my test list.
Thanks,
Miquèl
^ permalink raw reply
* [PATCH AUTOSEL 4.20 283/304] isdn: hisax: hfc_pci: Fix a possible concurrency use-after-free bug in HFCPCI_l1hw()
From: Sasha Levin @ 2019-01-28 15:43 UTC (permalink / raw)
To: linux-kernel, stable; +Cc: Jia-Ju Bai, David S . Miller, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
[ Upstream commit 7418e6520f22a2e35815122fa5a53d5bbfa2c10f ]
In drivers/isdn/hisax/hfc_pci.c, the functions hfcpci_interrupt() and
HFCPCI_l1hw() may be concurrently executed.
HFCPCI_l1hw()
line 1173: if (!cs->tx_skb)
hfcpci_interrupt()
line 942: spin_lock_irqsave();
line 1066: dev_kfree_skb_irq(cs->tx_skb);
Thus, a possible concurrency use-after-free bug may occur
in HFCPCI_l1hw().
To fix these bugs, the calls to spin_lock_irqsave() and
spin_unlock_irqrestore() are added in HFCPCI_l1hw(), to protect the
access to cs->tx_skb.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/isdn/hisax/hfc_pci.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/isdn/hisax/hfc_pci.c b/drivers/isdn/hisax/hfc_pci.c
index ea0e4c6de3fb..0109e0e8bcb6 100644
--- a/drivers/isdn/hisax/hfc_pci.c
+++ b/drivers/isdn/hisax/hfc_pci.c
@@ -1170,11 +1170,13 @@ HFCPCI_l1hw(struct PStack *st, int pr, void *arg)
if (cs->debug & L1_DEB_LAPD)
debugl1(cs, "-> PH_REQUEST_PULL");
#endif
+ spin_lock_irqsave(&cs->lock, flags);
if (!cs->tx_skb) {
test_and_clear_bit(FLG_L1_PULL_REQ, &st->l1.Flags);
st->l1.l1l2(st, PH_PULL | CONFIRM, NULL);
} else
test_and_set_bit(FLG_L1_PULL_REQ, &st->l1.Flags);
+ spin_unlock_irqrestore(&cs->lock, flags);
break;
case (HW_RESET | REQUEST):
spin_lock_irqsave(&cs->lock, flags);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.20 293/304] bpf: fix check_map_access smin_value test when pointer contains offset
From: Sasha Levin @ 2019-01-28 15:43 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Daniel Borkmann, Alexei Starovoitov, Sasha Levin, netdev
In-Reply-To: <20190128154341.47195-1-sashal@kernel.org>
From: Daniel Borkmann <daniel@iogearbox.net>
[ Upstream commit b7137c4eab85c1cf3d46acdde90ce1163b28c873 ]
In check_map_access() we probe actual bounds through __check_map_access()
with offset of reg->smin_value + off for lower bound and offset of
reg->umax_value + off for the upper bound. However, even though the
reg->smin_value could have a negative value, the final result of the
sum with off could be positive when pointer arithmetic with known and
unknown scalars is combined. In this case we reject the program with
an error such as "R<x> min value is negative, either use unsigned index
or do a if (index >=0) check." even though the access itself would be
fine. Therefore extend the check to probe whether the actual resulting
reg->smin_value + off is less than zero.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
kernel/bpf/verifier.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index eedc7bd4185d..8ae4c7f8eaf6 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1345,13 +1345,17 @@ static int check_map_access(struct bpf_verifier_env *env, u32 regno,
*/
if (env->log.level)
print_verifier_state(env, state);
+
/* The minimum value is only important with signed
* comparisons where we can't assume the floor of a
* value is 0. If we are using signed variables for our
* index'es we need to make sure that whatever we use
* will have a set floor within our range.
*/
- if (reg->smin_value < 0) {
+ if (reg->smin_value < 0 &&
+ (reg->smin_value == S64_MIN ||
+ (off + reg->smin_value != (s64)(s32)(off + reg->smin_value)) ||
+ reg->smin_value + off < 0)) {
verbose(env, "R%d min value is negative, either use unsigned index or do a if (index >=0) check.\n",
regno);
return -EACCES;
--
2.19.1
^ permalink raw reply related
* [PATCH v2 net-next 3/3] qede: Error recovery process
From: Michal Kalderon @ 2019-01-28 17:27 UTC (permalink / raw)
To: michal.kalderon, ariel.elior, davem; +Cc: netdev, natechancellor, Tomer Tayar
In-Reply-To: <20190128172756.27361-1-michal.kalderon@cavium.com>
From: Tomer Tayar <tomer.tayar@cavium.com>
This patch adds the error recovery process in the qede driver.
The process includes a partial/customized driver unload and load, which
allows it to look like a short suspend period to the kernel while
preserving the net devices' state.
Signed-off-by: Tomer Tayar <tomer.tayar@cavium.com>
Signed-off-by: Ariel Elior <ariel.elior@cavium.com>
Signed-off-by: Michal Kalderon <michal.kalderon@cavium.com>
---
drivers/net/ethernet/qlogic/qede/qede.h | 3 +
drivers/net/ethernet/qlogic/qede/qede_main.c | 292 ++++++++++++++++++++++-----
drivers/net/ethernet/qlogic/qede/qede_rdma.c | 63 ++++--
include/linux/qed/qede_rdma.h | 10 +-
4 files changed, 294 insertions(+), 74 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index 613249d..8434164 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -162,6 +162,7 @@ struct qede_rdma_dev {
struct list_head entry;
struct list_head rdma_event_list;
struct workqueue_struct *rdma_wq;
+ bool exp_recovery;
};
struct qede_ptp;
@@ -264,6 +265,7 @@ struct qede_dev {
enum QEDE_STATE {
QEDE_STATE_CLOSED,
QEDE_STATE_OPEN,
+ QEDE_STATE_RECOVERY,
};
#define HILO_U64(hi, lo) ((((u64)(hi)) << 32) + (lo))
@@ -462,6 +464,7 @@ struct qede_fastpath {
#define QEDE_CSUM_UNNECESSARY BIT(1)
#define QEDE_TUNN_CSUM_UNNECESSARY BIT(2)
+#define QEDE_SP_RECOVERY 0
#define QEDE_SP_RX_MODE 1
#ifdef CONFIG_RFS_ACCEL
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 5a74fcb..6b4d966 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -133,23 +133,12 @@ enum qede_pci_private {
static void qede_remove(struct pci_dev *pdev);
static void qede_shutdown(struct pci_dev *pdev);
static void qede_link_update(void *dev, struct qed_link_output *link);
+static void qede_schedule_recovery_handler(void *dev);
+static void qede_recovery_handler(struct qede_dev *edev);
static void qede_get_eth_tlv_data(void *edev, void *data);
static void qede_get_generic_tlv_data(void *edev,
struct qed_generic_tlvs *data);
-/* The qede lock is used to protect driver state change and driver flows that
- * are not reentrant.
- */
-void __qede_lock(struct qede_dev *edev)
-{
- mutex_lock(&edev->qede_lock);
-}
-
-void __qede_unlock(struct qede_dev *edev)
-{
- mutex_unlock(&edev->qede_lock);
-}
-
#ifdef CONFIG_QED_SRIOV
static int qede_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan, u8 qos,
__be16 vlan_proto)
@@ -231,6 +220,7 @@ static int qede_sriov_configure(struct pci_dev *pdev, int num_vfs_param)
.arfs_filter_op = qede_arfs_filter_op,
#endif
.link_update = qede_link_update,
+ .schedule_recovery_handler = qede_schedule_recovery_handler,
.get_generic_tlv_data = qede_get_generic_tlv_data,
.get_protocol_tlv_data = qede_get_eth_tlv_data,
},
@@ -950,11 +940,57 @@ static int qede_alloc_fp_array(struct qede_dev *edev)
return -ENOMEM;
}
+/* The qede lock is used to protect driver state change and driver flows that
+ * are not reentrant.
+ */
+void __qede_lock(struct qede_dev *edev)
+{
+ mutex_lock(&edev->qede_lock);
+}
+
+void __qede_unlock(struct qede_dev *edev)
+{
+ mutex_unlock(&edev->qede_lock);
+}
+
+/* This version of the lock should be used when acquiring the RTNL lock is also
+ * needed in addition to the internal qede lock.
+ */
+void qede_lock(struct qede_dev *edev)
+{
+ rtnl_lock();
+ __qede_lock(edev);
+}
+
+void qede_unlock(struct qede_dev *edev)
+{
+ __qede_unlock(edev);
+ rtnl_unlock();
+}
+
static void qede_sp_task(struct work_struct *work)
{
struct qede_dev *edev = container_of(work, struct qede_dev,
sp_task.work);
+ /* The locking scheme depends on the specific flag:
+ * In case of QEDE_SP_RECOVERY, acquiring the RTNL lock is required to
+ * ensure that ongoing flows are ended and new ones are not started.
+ * In other cases - only the internal qede lock should be acquired.
+ */
+
+ if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) {
+#ifdef CONFIG_QED_SRIOV
+ /* SRIOV must be disabled outside the lock to avoid a deadlock.
+ * The recovery of the active VFs is currently not supported.
+ */
+ qede_sriov_configure(edev->pdev, 0);
+#endif
+ qede_lock(edev);
+ qede_recovery_handler(edev);
+ qede_unlock(edev);
+ }
+
__qede_lock(edev);
if (test_and_clear_bit(QEDE_SP_RX_MODE, &edev->sp_flags))
@@ -1031,6 +1067,7 @@ static void qede_log_probe(struct qede_dev *edev)
enum qede_probe_mode {
QEDE_PROBE_NORMAL,
+ QEDE_PROBE_RECOVERY,
};
static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level,
@@ -1051,6 +1088,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level,
probe_params.dp_module = dp_module;
probe_params.dp_level = dp_level;
probe_params.is_vf = is_vf;
+ probe_params.recov_in_prog = (mode == QEDE_PROBE_RECOVERY);
cdev = qed_ops->common->probe(pdev, &probe_params);
if (!cdev) {
rc = -ENODEV;
@@ -1078,11 +1116,20 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level,
if (rc)
goto err2;
- edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module,
- dp_level);
- if (!edev) {
- rc = -ENOMEM;
- goto err2;
+ if (mode != QEDE_PROBE_RECOVERY) {
+ edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module,
+ dp_level);
+ if (!edev) {
+ rc = -ENOMEM;
+ goto err2;
+ }
+ } else {
+ struct net_device *ndev = pci_get_drvdata(pdev);
+
+ edev = netdev_priv(ndev);
+ edev->cdev = cdev;
+ memset(&edev->stats, 0, sizeof(edev->stats));
+ memcpy(&edev->dev_info, &dev_info, sizeof(dev_info));
}
if (is_vf)
@@ -1090,28 +1137,31 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level,
qede_init_ndev(edev);
- rc = qede_rdma_dev_add(edev);
+ rc = qede_rdma_dev_add(edev, (mode == QEDE_PROBE_RECOVERY));
if (rc)
goto err3;
- /* Prepare the lock prior to the registration of the netdev,
- * as once it's registered we might reach flows requiring it
- * [it's even possible to reach a flow needing it directly
- * from there, although it's unlikely].
- */
- INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task);
- mutex_init(&edev->qede_lock);
- rc = register_netdev(edev->ndev);
- if (rc) {
- DP_NOTICE(edev, "Cannot register net-device\n");
- goto err4;
+ if (mode != QEDE_PROBE_RECOVERY) {
+ /* Prepare the lock prior to the registration of the netdev,
+ * as once it's registered we might reach flows requiring it
+ * [it's even possible to reach a flow needing it directly
+ * from there, although it's unlikely].
+ */
+ INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task);
+ mutex_init(&edev->qede_lock);
+
+ rc = register_netdev(edev->ndev);
+ if (rc) {
+ DP_NOTICE(edev, "Cannot register net-device\n");
+ goto err4;
+ }
}
edev->ops->common->set_name(cdev, edev->ndev->name);
/* PTP not supported on VFs */
if (!is_vf)
- qede_ptp_enable(edev, true);
+ qede_ptp_enable(edev, (mode == QEDE_PROBE_NORMAL));
edev->ops->register_ops(cdev, &qede_ll_ops, edev);
@@ -1126,7 +1176,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level,
return 0;
err4:
- qede_rdma_dev_remove(edev);
+ qede_rdma_dev_remove(edev, (mode == QEDE_PROBE_RECOVERY));
err3:
free_netdev(edev->ndev);
err2:
@@ -1162,6 +1212,7 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id)
enum qede_remove_mode {
QEDE_REMOVE_NORMAL,
+ QEDE_REMOVE_RECOVERY,
};
static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode)
@@ -1172,15 +1223,19 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode)
DP_INFO(edev, "Starting qede_remove\n");
- qede_rdma_dev_remove(edev);
- unregister_netdev(ndev);
- cancel_delayed_work_sync(&edev->sp_task);
+ qede_rdma_dev_remove(edev, (mode == QEDE_REMOVE_RECOVERY));
- qede_ptp_disable(edev);
+ if (mode != QEDE_REMOVE_RECOVERY) {
+ unregister_netdev(ndev);
- edev->ops->common->set_power_state(cdev, PCI_D0);
+ cancel_delayed_work_sync(&edev->sp_task);
- pci_set_drvdata(pdev, NULL);
+ edev->ops->common->set_power_state(cdev, PCI_D0);
+
+ pci_set_drvdata(pdev, NULL);
+ }
+
+ qede_ptp_disable(edev);
/* Use global ops since we've freed edev */
qed_ops->common->slowpath_stop(cdev);
@@ -1194,7 +1249,8 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode)
* [e.g., QED register callbacks] won't break anything when
* accessing the netdevice.
*/
- free_netdev(ndev);
+ if (mode != QEDE_REMOVE_RECOVERY)
+ free_netdev(ndev);
dev_info(&pdev->dev, "Ending qede_remove successfully\n");
}
@@ -1539,6 +1595,58 @@ static int qede_alloc_mem_load(struct qede_dev *edev)
return 0;
}
+static void qede_empty_tx_queue(struct qede_dev *edev,
+ struct qede_tx_queue *txq)
+{
+ unsigned int pkts_compl = 0, bytes_compl = 0;
+ struct netdev_queue *netdev_txq;
+ int rc, len = 0;
+
+ netdev_txq = netdev_get_tx_queue(edev->ndev, txq->ndev_txq_id);
+
+ while (qed_chain_get_cons_idx(&txq->tx_pbl) !=
+ qed_chain_get_prod_idx(&txq->tx_pbl)) {
+ DP_VERBOSE(edev, NETIF_MSG_IFDOWN,
+ "Freeing a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n",
+ txq->index, qed_chain_get_cons_idx(&txq->tx_pbl),
+ qed_chain_get_prod_idx(&txq->tx_pbl));
+
+ rc = qede_free_tx_pkt(edev, txq, &len);
+ if (rc) {
+ DP_NOTICE(edev,
+ "Failed to free a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n",
+ txq->index,
+ qed_chain_get_cons_idx(&txq->tx_pbl),
+ qed_chain_get_prod_idx(&txq->tx_pbl));
+ break;
+ }
+
+ bytes_compl += len;
+ pkts_compl++;
+ txq->sw_tx_cons++;
+ }
+
+ netdev_tx_completed_queue(netdev_txq, pkts_compl, bytes_compl);
+}
+
+static void qede_empty_tx_queues(struct qede_dev *edev)
+{
+ int i;
+
+ for_each_queue(i)
+ if (edev->fp_array[i].type & QEDE_FASTPATH_TX) {
+ int cos;
+
+ for_each_cos_in_txq(edev, cos) {
+ struct qede_fastpath *fp;
+
+ fp = &edev->fp_array[i];
+ qede_empty_tx_queue(edev,
+ &fp->txq[cos]);
+ }
+ }
+}
+
/* This function inits fp content and resets the SB, RXQ and TXQ structures */
static void qede_init_fp(struct qede_dev *edev)
{
@@ -2053,6 +2161,7 @@ static int qede_start_queues(struct qede_dev *edev, bool clear_stats)
enum qede_unload_mode {
QEDE_UNLOAD_NORMAL,
+ QEDE_UNLOAD_RECOVERY,
};
static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode,
@@ -2068,7 +2177,8 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode,
clear_bit(QEDE_FLAGS_LINK_REQUESTED, &edev->flags);
- edev->state = QEDE_STATE_CLOSED;
+ if (mode != QEDE_UNLOAD_RECOVERY)
+ edev->state = QEDE_STATE_CLOSED;
qede_rdma_dev_event_close(edev);
@@ -2076,17 +2186,20 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode,
netif_tx_disable(edev->ndev);
netif_carrier_off(edev->ndev);
- /* Reset the link */
- memset(&link_params, 0, sizeof(link_params));
- link_params.link_up = false;
- edev->ops->common->set_link(edev->cdev, &link_params);
- rc = qede_stop_queues(edev);
- if (rc) {
- qede_sync_free_irqs(edev);
- goto out;
- }
+ if (mode != QEDE_UNLOAD_RECOVERY) {
+ /* Reset the link */
+ memset(&link_params, 0, sizeof(link_params));
+ link_params.link_up = false;
+ edev->ops->common->set_link(edev->cdev, &link_params);
- DP_INFO(edev, "Stopped Queues\n");
+ rc = qede_stop_queues(edev);
+ if (rc) {
+ qede_sync_free_irqs(edev);
+ goto out;
+ }
+
+ DP_INFO(edev, "Stopped Queues\n");
+ }
qede_vlan_mark_nonconfigured(edev);
edev->ops->fastpath_stop(edev->cdev);
@@ -2102,18 +2215,26 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode,
qede_napi_disable_remove(edev);
+ if (mode == QEDE_UNLOAD_RECOVERY)
+ qede_empty_tx_queues(edev);
+
qede_free_mem_load(edev);
qede_free_fp_array(edev);
out:
if (!is_locked)
__qede_unlock(edev);
+
+ if (mode != QEDE_UNLOAD_RECOVERY)
+ DP_NOTICE(edev, "Link is down\n");
+
DP_INFO(edev, "Ending qede unload\n");
}
enum qede_load_mode {
QEDE_LOAD_NORMAL,
QEDE_LOAD_RELOAD,
+ QEDE_LOAD_RECOVERY,
};
static int qede_load(struct qede_dev *edev, enum qede_load_mode mode,
@@ -2293,6 +2414,77 @@ static void qede_link_update(void *dev, struct qed_link_output *link)
}
}
+static void qede_schedule_recovery_handler(void *dev)
+{
+ struct qede_dev *edev = dev;
+
+ if (edev->state == QEDE_STATE_RECOVERY) {
+ DP_NOTICE(edev,
+ "Avoid scheduling a recovery handling since already in recovery state\n");
+ return;
+ }
+
+ set_bit(QEDE_SP_RECOVERY, &edev->sp_flags);
+ schedule_delayed_work(&edev->sp_task, 0);
+
+ DP_INFO(edev, "Scheduled a recovery handler\n");
+}
+
+static void qede_recovery_failed(struct qede_dev *edev)
+{
+ netdev_err(edev->ndev, "Recovery handling has failed. Power cycle is needed.\n");
+
+ netif_device_detach(edev->ndev);
+
+ if (edev->cdev)
+ edev->ops->common->set_power_state(edev->cdev, PCI_D3hot);
+}
+
+static void qede_recovery_handler(struct qede_dev *edev)
+{
+ u32 curr_state = edev->state;
+ int rc;
+
+ DP_NOTICE(edev, "Starting a recovery process\n");
+
+ /* No need to acquire first the qede_lock since is done by qede_sp_task
+ * before calling this function.
+ */
+ edev->state = QEDE_STATE_RECOVERY;
+
+ edev->ops->common->recovery_prolog(edev->cdev);
+
+ if (curr_state == QEDE_STATE_OPEN)
+ qede_unload(edev, QEDE_UNLOAD_RECOVERY, true);
+
+ __qede_remove(edev->pdev, QEDE_REMOVE_RECOVERY);
+
+ rc = __qede_probe(edev->pdev, edev->dp_module, edev->dp_level,
+ IS_VF(edev), QEDE_PROBE_RECOVERY);
+ if (rc) {
+ edev->cdev = NULL;
+ goto err;
+ }
+
+ if (curr_state == QEDE_STATE_OPEN) {
+ rc = qede_load(edev, QEDE_LOAD_RECOVERY, true);
+ if (rc)
+ goto err;
+
+ qede_config_rx_mode(edev->ndev);
+ udp_tunnel_get_rx_info(edev->ndev);
+ }
+
+ edev->state = curr_state;
+
+ DP_NOTICE(edev, "Recovery handling is done\n");
+
+ return;
+
+err:
+ qede_recovery_failed(edev);
+}
+
static bool qede_is_txq_full(struct qede_dev *edev, struct qede_tx_queue *txq)
{
struct netdev_queue *netdev_txq;
diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c
index 1900bf7..ffabc2d 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c
@@ -50,6 +50,8 @@ static void _qede_rdma_dev_add(struct qede_dev *edev)
if (!qedr_drv)
return;
+ /* Leftovers from previous error recovery */
+ edev->rdma_info.exp_recovery = false;
edev->rdma_info.qedr_dev = qedr_drv->add(edev->cdev, edev->pdev,
edev->ndev);
}
@@ -87,21 +89,26 @@ static void qede_rdma_destroy_wq(struct qede_dev *edev)
destroy_workqueue(edev->rdma_info.rdma_wq);
}
-int qede_rdma_dev_add(struct qede_dev *edev)
+int qede_rdma_dev_add(struct qede_dev *edev, bool recovery)
{
- int rc = 0;
+ int rc;
- if (qede_rdma_supported(edev)) {
- rc = qede_rdma_create_wq(edev);
- if (rc)
- return rc;
+ if (!qede_rdma_supported(edev))
+ return 0;
- INIT_LIST_HEAD(&edev->rdma_info.entry);
- mutex_lock(&qedr_dev_list_lock);
- list_add_tail(&edev->rdma_info.entry, &qedr_dev_list);
- _qede_rdma_dev_add(edev);
- mutex_unlock(&qedr_dev_list_lock);
- }
+ /* Cannot start qedr while recovering since it wasn't fully stopped */
+ if (recovery)
+ return 0;
+
+ rc = qede_rdma_create_wq(edev);
+ if (rc)
+ return rc;
+
+ INIT_LIST_HEAD(&edev->rdma_info.entry);
+ mutex_lock(&qedr_dev_list_lock);
+ list_add_tail(&edev->rdma_info.entry, &qedr_dev_list);
+ _qede_rdma_dev_add(edev);
+ mutex_unlock(&qedr_dev_list_lock);
return rc;
}
@@ -110,19 +117,30 @@ static void _qede_rdma_dev_remove(struct qede_dev *edev)
{
if (qedr_drv && qedr_drv->remove && edev->rdma_info.qedr_dev)
qedr_drv->remove(edev->rdma_info.qedr_dev);
- edev->rdma_info.qedr_dev = NULL;
}
-void qede_rdma_dev_remove(struct qede_dev *edev)
+void qede_rdma_dev_remove(struct qede_dev *edev, bool recovery)
{
if (!qede_rdma_supported(edev))
return;
- qede_rdma_destroy_wq(edev);
- mutex_lock(&qedr_dev_list_lock);
- _qede_rdma_dev_remove(edev);
- list_del(&edev->rdma_info.entry);
- mutex_unlock(&qedr_dev_list_lock);
+ /* Cannot remove qedr while recovering since it wasn't fully stopped */
+ if (!recovery) {
+ qede_rdma_destroy_wq(edev);
+ mutex_lock(&qedr_dev_list_lock);
+ if (!edev->rdma_info.exp_recovery)
+ _qede_rdma_dev_remove(edev);
+ edev->rdma_info.qedr_dev = NULL;
+ list_del(&edev->rdma_info.entry);
+ mutex_unlock(&qedr_dev_list_lock);
+ } else {
+ if (!edev->rdma_info.exp_recovery) {
+ mutex_lock(&qedr_dev_list_lock);
+ _qede_rdma_dev_remove(edev);
+ mutex_unlock(&qedr_dev_list_lock);
+ }
+ edev->rdma_info.exp_recovery = true;
+ }
}
static void _qede_rdma_dev_open(struct qede_dev *edev)
@@ -204,7 +222,8 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv)
mutex_lock(&qedr_dev_list_lock);
list_for_each_entry(edev, &qedr_dev_list, rdma_info.entry) {
- if (edev->rdma_info.qedr_dev)
+ /* If device has experienced recovery it was already removed */
+ if (edev->rdma_info.qedr_dev && !edev->rdma_info.exp_recovery)
_qede_rdma_dev_remove(edev);
}
qedr_drv = NULL;
@@ -284,6 +303,10 @@ static void qede_rdma_add_event(struct qede_dev *edev,
{
struct qede_rdma_event_work *event_node;
+ /* If a recovery was experienced avoid adding the event */
+ if (edev->rdma_info.exp_recovery)
+ return;
+
if (!edev->rdma_info.qedr_dev)
return;
diff --git a/include/linux/qed/qede_rdma.h b/include/linux/qed/qede_rdma.h
index 9904617..5a00c7a 100644
--- a/include/linux/qed/qede_rdma.h
+++ b/include/linux/qed/qede_rdma.h
@@ -74,21 +74,23 @@ struct qedr_driver {
bool qede_rdma_supported(struct qede_dev *dev);
#if IS_ENABLED(CONFIG_QED_RDMA)
-int qede_rdma_dev_add(struct qede_dev *dev);
+int qede_rdma_dev_add(struct qede_dev *dev, bool recovery);
void qede_rdma_dev_event_open(struct qede_dev *dev);
void qede_rdma_dev_event_close(struct qede_dev *dev);
-void qede_rdma_dev_remove(struct qede_dev *dev);
+void qede_rdma_dev_remove(struct qede_dev *dev, bool recovery);
void qede_rdma_event_changeaddr(struct qede_dev *edr);
#else
-static inline int qede_rdma_dev_add(struct qede_dev *dev)
+static inline int qede_rdma_dev_add(struct qede_dev *dev,
+ bool recovery)
{
return 0;
}
static inline void qede_rdma_dev_event_open(struct qede_dev *dev) {}
static inline void qede_rdma_dev_event_close(struct qede_dev *dev) {}
-static inline void qede_rdma_dev_remove(struct qede_dev *dev) {}
+static inline void qede_rdma_dev_remove(struct qede_dev *dev,
+ bool recovery) {}
static inline void qede_rdma_event_changeaddr(struct qede_dev *edr) {}
#endif
#endif
--
1.8.3.1
^ permalink raw reply related
* [PATCH v2 net-next 2/3] qed: Add infrastructure for error detection and recovery
From: Michal Kalderon @ 2019-01-28 17:27 UTC (permalink / raw)
To: michal.kalderon, ariel.elior, davem; +Cc: netdev, natechancellor, Tomer Tayar
In-Reply-To: <20190128172756.27361-1-michal.kalderon@cavium.com>
From: Tomer Tayar <tomer.tayar@cavium.com>
This patch adds the detection and handling of a parity error ("process kill
event"), including the update of the protocol drivers, and the prevention
of any HW access that will lead to device access towards the host while
recovery is in progress.
It also provides the means for the protocol drivers to trigger a recovery
process on their decision.
Signed-off-by: Tomer Tayar <tomer.tayar@cavium.com>
Signed-off-by: Ariel Elior <ariel.elior@cavium.com>
Signed-off-by: Michal Kalderon <michal.kalderon@cavium.com>
---
drivers/net/ethernet/qlogic/qed/qed.h | 4 ++
drivers/net/ethernet/qlogic/qed/qed_dev.c | 41 +++++++----
drivers/net/ethernet/qlogic/qed/qed_hsi.h | 2 +-
drivers/net/ethernet/qlogic/qed/qed_hw.c | 11 +++
drivers/net/ethernet/qlogic/qed/qed_main.c | 30 ++++++++
drivers/net/ethernet/qlogic/qed/qed_mcp.c | 94 ++++++++++++++++++++++++++
drivers/net/ethernet/qlogic/qed/qed_mcp.h | 32 +++++++++
drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 +
drivers/net/ethernet/qlogic/qed/qed_spq.c | 22 ++++++
drivers/net/ethernet/qlogic/qed/qed_sriov.c | 9 ++-
include/linux/qed/qed_if.h | 20 ++++++
11 files changed, 251 insertions(+), 16 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index b352e31..3b0955d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -804,6 +804,9 @@ struct qed_dev {
u32 mcp_nvm_resp;
+ /* Recovery */
+ bool recov_in_prog;
+
/* Linux specific here */
struct qede_dev *edev;
struct pci_dev *pdev;
@@ -943,6 +946,7 @@ int qed_fill_dev_info(struct qed_dev *cdev,
u32 qed_unzip_data(struct qed_hwfn *p_hwfn,
u32 input_len, u8 *input_buf,
u32 max_size, u8 *unzip_buf);
+void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn);
void qed_get_protocol_stats(struct qed_dev *cdev,
enum qed_mcp_protocol_type type,
union qed_mcp_protocol_stats *stats);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index fa5f07e..b17003d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -2140,6 +2140,11 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
"Load request was sent. Load code: 0x%x\n",
load_code);
+ /* Only relevant for recovery:
+ * Clear the indication after LOAD_REQ is responded by the MFW.
+ */
+ cdev->recov_in_prog = false;
+
qed_mcp_set_capabilities(p_hwfn, p_hwfn->p_main_ptt);
qed_reset_mb_shadow(p_hwfn, p_hwfn->p_main_ptt);
@@ -2291,6 +2296,9 @@ static void qed_hw_timers_stop(struct qed_dev *cdev,
qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0);
qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0);
+ if (cdev->recov_in_prog)
+ return;
+
for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) {
if ((!qed_rd(p_hwfn, p_ptt,
TM_REG_PF_SCAN_ACTIVE_CONN)) &&
@@ -2353,12 +2361,14 @@ int qed_hw_stop(struct qed_dev *cdev)
p_hwfn->hw_init_done = false;
/* Send unload command to MCP */
- rc = qed_mcp_unload_req(p_hwfn, p_ptt);
- if (rc) {
- DP_NOTICE(p_hwfn,
- "Failed sending a UNLOAD_REQ command. rc = %d.\n",
- rc);
- rc2 = -EINVAL;
+ if (!cdev->recov_in_prog) {
+ rc = qed_mcp_unload_req(p_hwfn, p_ptt);
+ if (rc) {
+ DP_NOTICE(p_hwfn,
+ "Failed sending a UNLOAD_REQ command. rc = %d.\n",
+ rc);
+ rc2 = -EINVAL;
+ }
}
qed_slowpath_irq_sync(p_hwfn);
@@ -2400,16 +2410,18 @@ int qed_hw_stop(struct qed_dev *cdev)
qed_wr(p_hwfn, p_ptt, DORQ_REG_PF_DB_ENABLE, 0);
qed_wr(p_hwfn, p_ptt, QM_REG_PF_EN, 0);
- qed_mcp_unload_done(p_hwfn, p_ptt);
- if (rc) {
- DP_NOTICE(p_hwfn,
- "Failed sending a UNLOAD_DONE command. rc = %d.\n",
- rc);
- rc2 = -EINVAL;
+ if (!cdev->recov_in_prog) {
+ rc = qed_mcp_unload_done(p_hwfn, p_ptt);
+ if (rc) {
+ DP_NOTICE(p_hwfn,
+ "Failed sending a UNLOAD_DONE command. rc = %d.\n",
+ rc);
+ rc2 = -EINVAL;
+ }
}
}
- if (IS_PF(cdev)) {
+ if (IS_PF(cdev) && !cdev->recov_in_prog) {
p_hwfn = QED_LEADING_HWFN(cdev);
p_ptt = QED_LEADING_HWFN(cdev)->p_main_ptt;
@@ -3459,6 +3471,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn,
void __iomem *p_doorbells,
enum qed_pci_personality personality)
{
+ struct qed_dev *cdev = p_hwfn->cdev;
int rc = 0;
/* Split PCI bars evenly between hwfns */
@@ -3511,7 +3524,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn,
/* Sending a mailbox to the MFW should be done after qed_get_hw_info()
* is called as it sets the ports number in an engine.
*/
- if (IS_LEAD_HWFN(p_hwfn)) {
+ if (IS_LEAD_HWFN(p_hwfn) && !cdev->recov_in_prog) {
rc = qed_mcp_initiate_pf_flr(p_hwfn, p_hwfn->p_main_ptt);
if (rc)
DP_NOTICE(p_hwfn, "Failed to initiate PF FLR\n");
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index b13cfb4..417121e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -12827,7 +12827,7 @@ enum MFW_DRV_MSG_TYPE {
MFW_DRV_MSG_LLDP_DATA_UPDATED,
MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED,
MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED,
- MFW_DRV_MSG_RESERVED4,
+ MFW_DRV_MSG_ERROR_RECOVERY,
MFW_DRV_MSG_BW_UPDATE,
MFW_DRV_MSG_S_TAG_UPDATE,
MFW_DRV_MSG_GET_LAN_STATS,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c
index 70504dc..72ec1c6b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hw.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c
@@ -703,6 +703,17 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn,
int qed_status = 0;
u32 offset = 0;
+ if (p_hwfn->cdev->recov_in_prog) {
+ DP_VERBOSE(p_hwfn,
+ NETIF_MSG_HW,
+ "Recovery is in progress. Avoid DMAE transaction [{src: addr 0x%llx, type %d}, {dst: addr 0x%llx, type %d}, size %d].\n",
+ src_addr, src_type, dst_addr, dst_type,
+ size_in_dwords);
+
+ /* Let the flow complete w/o any error handling */
+ return 0;
+ }
+
qed_dmae_opcode(p_hwfn,
(src_type == QED_DMAE_ADDRESS_GRC),
(dst_type == QED_DMAE_ADDRESS_GRC),
diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 6adf5bd..b473526 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -359,6 +359,8 @@ static struct qed_dev *qed_probe(struct pci_dev *pdev,
qed_init_dp(cdev, params->dp_module, params->dp_level);
+ cdev->recov_in_prog = params->recov_in_prog;
+
rc = qed_init_pci(cdev, pdev);
if (rc) {
DP_ERR(cdev, "init pci failed\n");
@@ -2203,6 +2205,15 @@ static int qed_nvm_get_image(struct qed_dev *cdev, enum qed_nvm_images type,
return qed_mcp_get_nvm_image(hwfn, type, buf, len);
}
+void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn)
+{
+ struct qed_common_cb_ops *ops = p_hwfn->cdev->protocol_ops.common;
+ void *cookie = p_hwfn->cdev->ops_cookie;
+
+ if (ops && ops->schedule_recovery_handler)
+ ops->schedule_recovery_handler(cookie);
+}
+
static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal,
void *handle)
{
@@ -2226,6 +2237,23 @@ static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode)
return status;
}
+static int qed_recovery_process(struct qed_dev *cdev)
+{
+ struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev);
+ struct qed_ptt *p_ptt;
+ int rc = 0;
+
+ p_ptt = qed_ptt_acquire(p_hwfn);
+ if (!p_ptt)
+ return -EAGAIN;
+
+ rc = qed_start_recovery_process(p_hwfn, p_ptt);
+
+ qed_ptt_release(p_hwfn, p_ptt);
+
+ return rc;
+}
+
static int qed_update_wol(struct qed_dev *cdev, bool enabled)
{
struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev);
@@ -2380,6 +2408,8 @@ static int qed_read_module_eeprom(struct qed_dev *cdev, char *buf,
.nvm_get_image = &qed_nvm_get_image,
.set_coalesce = &qed_set_coalesce,
.set_led = &qed_set_led,
+ .recovery_process = &qed_recovery_process,
+ .recovery_prolog = &qed_recovery_prolog,
.update_drv_state = &qed_update_drv_state,
.update_mac = &qed_update_mac,
.update_mtu = &qed_update_mtu,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 1024484..bb85418 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -1549,6 +1549,60 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool b_up)
return 0;
}
+u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt)
+{
+ u32 path_offsize_addr, path_offsize, path_addr, proc_kill_cnt;
+
+ if (IS_VF(p_hwfn->cdev))
+ return -EINVAL;
+
+ path_offsize_addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base,
+ PUBLIC_PATH);
+ path_offsize = qed_rd(p_hwfn, p_ptt, path_offsize_addr);
+ path_addr = SECTION_ADDR(path_offsize, QED_PATH_ID(p_hwfn));
+
+ proc_kill_cnt = qed_rd(p_hwfn, p_ptt,
+ path_addr +
+ offsetof(struct public_path, process_kill)) &
+ PROCESS_KILL_COUNTER_MASK;
+
+ return proc_kill_cnt;
+}
+
+static void qed_mcp_handle_process_kill(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt)
+{
+ struct qed_dev *cdev = p_hwfn->cdev;
+ u32 proc_kill_cnt;
+
+ /* Prevent possible attentions/interrupts during the recovery handling
+ * and till its load phase, during which they will be re-enabled.
+ */
+ qed_int_igu_disable_int(p_hwfn, p_ptt);
+
+ DP_NOTICE(p_hwfn, "Received a process kill indication\n");
+
+ /* The following operations should be done once, and thus in CMT mode
+ * are carried out by only the first HW function.
+ */
+ if (p_hwfn != QED_LEADING_HWFN(cdev))
+ return;
+
+ if (cdev->recov_in_prog) {
+ DP_NOTICE(p_hwfn,
+ "Ignoring the indication since a recovery process is already in progress\n");
+ return;
+ }
+
+ cdev->recov_in_prog = true;
+
+ proc_kill_cnt = qed_get_process_kill_counter(p_hwfn, p_ptt);
+ DP_NOTICE(p_hwfn, "Process kill counter: %d\n", proc_kill_cnt);
+
+ qed_schedule_recovery_handler(p_hwfn);
+}
+
static void qed_mcp_send_protocol_stats(struct qed_hwfn *p_hwfn,
struct qed_ptt *p_ptt,
enum MFW_DRV_MSG_TYPE type)
@@ -1779,6 +1833,9 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn,
case MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE:
qed_mcp_handle_transceiver_change(p_hwfn, p_ptt);
break;
+ case MFW_DRV_MSG_ERROR_RECOVERY:
+ qed_mcp_handle_process_kill(p_hwfn, p_ptt);
+ break;
case MFW_DRV_MSG_GET_LAN_STATS:
case MFW_DRV_MSG_GET_FCOE_STATS:
case MFW_DRV_MSG_GET_ISCSI_STATS:
@@ -2324,6 +2381,43 @@ int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn,
return 0;
}
+int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
+{
+ struct qed_dev *cdev = p_hwfn->cdev;
+
+ if (cdev->recov_in_prog) {
+ DP_NOTICE(p_hwfn,
+ "Avoid triggering a recovery since such a process is already in progress\n");
+ return -EAGAIN;
+ }
+
+ DP_NOTICE(p_hwfn, "Triggering a recovery process\n");
+ qed_wr(p_hwfn, p_ptt, MISC_REG_AEU_GENERAL_ATTN_35, 0x1);
+
+ return 0;
+}
+
+#define QED_RECOVERY_PROLOG_SLEEP_MS 100
+
+int qed_recovery_prolog(struct qed_dev *cdev)
+{
+ struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev);
+ struct qed_ptt *p_ptt = p_hwfn->p_main_ptt;
+ int rc;
+
+ /* Allow ongoing PCIe transactions to complete */
+ msleep(QED_RECOVERY_PROLOG_SLEEP_MS);
+
+ /* Clear the PF's internal FID_enable in the PXP */
+ rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false);
+ if (rc)
+ DP_NOTICE(p_hwfn,
+ "qed_pglueb_set_pfid_enable() failed. rc = %d.\n",
+ rc);
+
+ return rc;
+}
+
static int
qed_mcp_config_vf_msix_bb(struct qed_hwfn *p_hwfn,
struct qed_ptt *p_ptt, u8 vf_id, u8 num)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
index 387c5e6..6e1d72a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
@@ -441,6 +441,38 @@ int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn,
struct qed_mcp_drv_version *p_ver);
/**
+ * @brief Read the MFW process kill counter
+ *
+ * @param p_hwfn
+ * @param p_ptt
+ *
+ * @return u32
+ */
+u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt);
+
+/**
+ * @brief Trigger a recovery process
+ *
+ * @param p_hwfn
+ * @param p_ptt
+ *
+ * @return int
+ */
+int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt);
+
+/**
+ * @brief A recovery handler must call this function as its first step.
+ * It is assumed that the handler is not run from an interrupt context.
+ *
+ * @param cdev
+ * @param p_ptt
+ *
+ * @return int
+ */
+int qed_recovery_prolog(struct qed_dev *cdev);
+
+/**
* @brief Notify MFW about the change in base device properties
*
* @param p_hwfn
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index 8939ed6..5ce825c 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -518,6 +518,8 @@
0x180824UL
#define MISC_REG_AEU_GENERAL_ATTN_0 \
0x008400UL
+#define MISC_REG_AEU_GENERAL_ATTN_35 \
+ 0x00848cUL
#define CAU_REG_SB_ADDR_MEMORY \
0x1c8000UL
#define CAU_REG_SB_VAR_MEMORY \
diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c
index eb88bbc..3e0f7c4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_spq.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c
@@ -790,6 +790,17 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn)
SPQ_HIGH_PRI_RESERVE_DEFAULT);
}
+static void qed_spq_recov_set_ret_code(struct qed_spq_entry *p_ent,
+ u8 *fw_return_code)
+{
+ if (!fw_return_code)
+ return;
+
+ if (p_ent->elem.hdr.protocol_id == PROTOCOLID_ROCE ||
+ p_ent->elem.hdr.protocol_id == PROTOCOLID_IWARP)
+ *fw_return_code = RDMA_RETURN_OK;
+}
+
/* Avoid overriding of SPQ entries when getting out-of-order completions, by
* marking the completions in a bitmap and increasing the chain consumer only
* for the first successive completed entries.
@@ -825,6 +836,17 @@ int qed_spq_post(struct qed_hwfn *p_hwfn,
return -EINVAL;
}
+ if (p_hwfn->cdev->recov_in_prog) {
+ DP_VERBOSE(p_hwfn,
+ QED_MSG_SPQ,
+ "Recovery is in progress. Skip spq post [cmd %02x protocol %02x]\n",
+ p_ent->elem.hdr.cmd_id, p_ent->elem.hdr.protocol_id);
+
+ /* Let the flow complete w/o any error handling */
+ qed_spq_recov_set_ret_code(p_ent, fw_return_code);
+ return 0;
+ }
+
/* Complete the entry */
rc = qed_spq_fill_entry(p_hwfn, p_ent);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
index ca6290f..71e28be 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
@@ -4447,6 +4447,13 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled)
if (cdev->p_iov_info && cdev->p_iov_info->num_vfs && pci_enabled)
pci_disable_sriov(cdev->pdev);
+ if (cdev->recov_in_prog) {
+ DP_VERBOSE(cdev,
+ QED_MSG_IOV,
+ "Skip SRIOV disable operations in the device since a recovery is in progress\n");
+ goto out;
+ }
+
for_each_hwfn(cdev, i) {
struct qed_hwfn *hwfn = &cdev->hwfns[i];
struct qed_ptt *ptt = qed_ptt_acquire(hwfn);
@@ -4486,7 +4493,7 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled)
qed_ptt_release(hwfn, ptt);
}
-
+out:
qed_iov_set_vfs_to_disable(cdev, false);
return 0;
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index 5f818fd..35170f7 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -763,6 +763,7 @@ struct qed_probe_params {
u32 dp_module;
u8 dp_level;
bool is_vf;
+ bool recov_in_prog;
};
#define QED_DRV_VER_STR_SIZE 12
@@ -809,6 +810,7 @@ struct qed_common_cb_ops {
void (*arfs_filter_op)(void *dev, void *fltr, u8 fw_rc);
void (*link_update)(void *dev,
struct qed_link_output *link);
+ void (*schedule_recovery_handler)(void *dev);
void (*dcbx_aen)(void *dev, struct qed_dcbx_get *get, u32 mib_type);
void (*get_generic_tlv_data)(void *dev, struct qed_generic_tlvs *data);
void (*get_protocol_tlv_data)(void *dev, void *data);
@@ -1057,6 +1059,24 @@ struct qed_common_ops {
void __iomem *db_addr, void *db_data);
/**
+ * @brief recovery_process - Trigger a recovery process
+ *
+ * @param cdev
+ *
+ * @return 0 on success, error otherwise.
+ */
+ int (*recovery_process)(struct qed_dev *cdev);
+
+/**
+ * @brief recovery_prolog - Execute the prolog operations of a recovery process
+ *
+ * @param cdev
+ *
+ * @return 0 on success, error otherwise.
+ */
+ int (*recovery_prolog)(struct qed_dev *cdev);
+
+/**
* @brief update_drv_state - API to inform the change in the driver state.
*
* @param cdev
--
1.8.3.1
^ permalink raw reply related
* [PATCH v2 net-next 1/3] qed: Revise load sequence to avoid PCI errors
From: Michal Kalderon @ 2019-01-28 17:27 UTC (permalink / raw)
To: michal.kalderon, ariel.elior, davem; +Cc: netdev, natechancellor, Tomer Tayar
In-Reply-To: <20190128172756.27361-1-michal.kalderon@cavium.com>
From: Tomer Tayar <tomer.tayar@cavium.com>
Initiating final cleanup after an ungraceful driver unload can lead to bad
PCI accesses towards the host.
This patch revises the load sequence so final cleanup is sent while the
internal master enable is cleared, to prevent the host accesses, and clears
the internal error indications just before enabling the internal master
enable.
Signed-off-by: Tomer Tayar <tomer.tayar@cavium.com>
Signed-off-by: Ariel Elior <ariel.elior@cavium.com>
Signed-off-by: Michal Kalderon <michal.kalderon@cavium.com>
---
drivers/net/ethernet/qlogic/qed/qed.h | 1 -
drivers/net/ethernet/qlogic/qed/qed_dev.c | 117 ++++++++++++++----------
drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 12 +++
drivers/net/ethernet/qlogic/qed/qed_int.c | 126 +++++++++++++-------------
drivers/net/ethernet/qlogic/qed/qed_int.h | 3 +
drivers/net/ethernet/qlogic/qed/qed_mcp.c | 21 +++++
drivers/net/ethernet/qlogic/qed/qed_mcp.h | 10 ++
7 files changed, 178 insertions(+), 112 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index 24a9016..b352e31 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -554,7 +554,6 @@ struct qed_hwfn {
u8 dp_level;
char name[NAME_SIZE];
- bool first_on_engine;
bool hw_init_done;
u8 num_funcs_on_engine;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 8f65514..fa5f07e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1959,11 +1959,6 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn,
(p_hwfn->hw_info.personality == QED_PCI_FCOE) ? 1 : 0);
STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_ROCE_RT_OFFSET, 0);
- /* Cleanup chip from previous driver if such remains exist */
- rc = qed_final_cleanup(p_hwfn, p_ptt, rel_pf_id, false);
- if (rc)
- return rc;
-
/* Sanity check before the PF init sequence that uses DMAE */
rc = qed_dmae_sanity(p_hwfn, p_ptt, "pf_phase");
if (rc)
@@ -2007,17 +2002,15 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn,
return rc;
}
-static int qed_change_pci_hwfn(struct qed_hwfn *p_hwfn,
- struct qed_ptt *p_ptt,
- u8 enable)
+int qed_pglueb_set_pfid_enable(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt, bool b_enable)
{
- u32 delay_idx = 0, val, set_val = enable ? 1 : 0;
+ u32 delay_idx = 0, val, set_val = b_enable ? 1 : 0;
- /* Change PF in PXP */
- qed_wr(p_hwfn, p_ptt,
- PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER, set_val);
+ /* Configure the PF's internal FID_enable for master transactions */
+ qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER, set_val);
- /* wait until value is set - try for 1 second every 50us */
+ /* Wait until value is set - try for 1 second every 50us */
for (delay_idx = 0; delay_idx < 20000; delay_idx++) {
val = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER);
@@ -2071,13 +2064,19 @@ static int qed_vf_start(struct qed_hwfn *p_hwfn,
return 0;
}
+static void qed_pglueb_clear_err(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
+{
+ qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR,
+ BIT(p_hwfn->abs_pf_id));
+}
+
int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
{
struct qed_load_req_params load_req_params;
u32 load_code, resp, param, drv_mb_param;
bool b_default_mtu = true;
struct qed_hwfn *p_hwfn;
- int rc = 0, mfw_rc, i;
+ int rc = 0, i;
u16 ether_type;
if ((p_params->int_mode == QED_INT_MODE_MSI) && (cdev->num_hwfns > 1)) {
@@ -2092,7 +2091,7 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
}
for_each_hwfn(cdev, i) {
- struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
+ p_hwfn = &cdev->hwfns[i];
/* If management didn't provide a default, set one of our own */
if (!p_hwfn->hw_info.mtu) {
@@ -2105,9 +2104,6 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
continue;
}
- /* Enable DMAE in PXP */
- rc = qed_change_pci_hwfn(p_hwfn, p_hwfn->p_main_ptt, true);
-
rc = qed_calc_hw_mode(p_hwfn);
if (rc)
return rc;
@@ -2148,8 +2144,34 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
qed_reset_mb_shadow(p_hwfn, p_hwfn->p_main_ptt);
- p_hwfn->first_on_engine = (load_code ==
- FW_MSG_CODE_DRV_LOAD_ENGINE);
+ /* Clean up chip from previous driver if such remains exist.
+ * This is not needed when the PF is the first one on the
+ * engine, since afterwards we are going to init the FW.
+ */
+ if (load_code != FW_MSG_CODE_DRV_LOAD_ENGINE) {
+ rc = qed_final_cleanup(p_hwfn, p_hwfn->p_main_ptt,
+ p_hwfn->rel_pf_id, false);
+ if (rc) {
+ DP_NOTICE(p_hwfn, "Final cleanup failed\n");
+ goto load_err;
+ }
+ }
+
+ /* Log and clear previous pglue_b errors if such exist */
+ qed_pglueb_rbc_attn_handler(p_hwfn, p_hwfn->p_main_ptt);
+
+ /* Enable the PF's internal FID_enable in the PXP */
+ rc = qed_pglueb_set_pfid_enable(p_hwfn, p_hwfn->p_main_ptt,
+ true);
+ if (rc)
+ goto load_err;
+
+ /* Clear the pglue_b was_error indication.
+ * In E4 it must be done after the BME and the internal
+ * FID_enable for the PF are set, since VDMs may cause the
+ * indication to be set again.
+ */
+ qed_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt);
switch (load_code) {
case FW_MSG_CODE_DRV_LOAD_ENGINE:
@@ -2180,39 +2202,29 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
break;
}
- if (rc)
+ if (rc) {
DP_NOTICE(p_hwfn,
"init phase failed for loadcode 0x%x (rc %d)\n",
- load_code, rc);
+ load_code, rc);
+ goto load_err;
+ }
- /* ACK mfw regardless of success or failure of initialization */
- mfw_rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt,
- DRV_MSG_CODE_LOAD_DONE,
- 0, &load_code, ¶m);
+ rc = qed_mcp_load_done(p_hwfn, p_hwfn->p_main_ptt);
if (rc)
return rc;
- if (mfw_rc) {
- DP_NOTICE(p_hwfn, "Failed sending LOAD_DONE command\n");
- return mfw_rc;
- }
-
- /* Check if there is a DID mismatch between nvm-cfg/efuse */
- if (param & FW_MB_PARAM_LOAD_DONE_DID_EFUSE_ERROR)
- DP_NOTICE(p_hwfn,
- "warning: device configuration is not supported on this board type. The device may not function as expected.\n");
/* send DCBX attention request command */
DP_VERBOSE(p_hwfn,
QED_MSG_DCB,
"sending phony dcbx set command to trigger DCBx attention handling\n");
- mfw_rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt,
- DRV_MSG_CODE_SET_DCBX,
- 1 << DRV_MB_PARAM_DCBX_NOTIFY_SHIFT,
- &load_code, ¶m);
- if (mfw_rc) {
+ rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt,
+ DRV_MSG_CODE_SET_DCBX,
+ 1 << DRV_MB_PARAM_DCBX_NOTIFY_SHIFT,
+ &resp, ¶m);
+ if (rc) {
DP_NOTICE(p_hwfn,
"Failed to send DCBX attention request\n");
- return mfw_rc;
+ return rc;
}
p_hwfn->hw_init_done = true;
@@ -2261,6 +2273,12 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params)
}
return 0;
+
+load_err:
+ /* The MFW load lock should be released also when initialization fails.
+ */
+ qed_mcp_load_done(p_hwfn, p_hwfn->p_main_ptt);
+ return rc;
}
#define QED_HW_STOP_RETRY_LIMIT (10)
@@ -2395,14 +2413,16 @@ int qed_hw_stop(struct qed_dev *cdev)
p_hwfn = QED_LEADING_HWFN(cdev);
p_ptt = QED_LEADING_HWFN(cdev)->p_main_ptt;
- /* Disable DMAE in PXP - in CMT, this should only be done for
- * first hw-function, and only after all transactions have
- * stopped for all active hw-functions.
+ /* Clear the PF's internal FID_enable in the PXP.
+ * In CMT this should only be done for first hw-function, and
+ * only after all transactions have stopped for all active
+ * hw-functions.
*/
- rc = qed_change_pci_hwfn(p_hwfn, p_ptt, false);
+ rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false);
if (rc) {
DP_NOTICE(p_hwfn,
- "qed_change_pci_hwfn failed. rc = %d.\n", rc);
+ "qed_pglueb_set_pfid_enable() failed. rc = %d.\n",
+ rc);
rc2 = -EINVAL;
}
}
@@ -2502,9 +2522,8 @@ static void qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn)
PGLUE_B_REG_PGL_ADDR_94_F0_BB, 0);
}
- /* Clean Previous errors if such exist */
- qed_wr(p_hwfn, p_hwfn->p_main_ptt,
- PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR, 1 << p_hwfn->abs_pf_id);
+ /* Clean previous pglue_b errors if such exist */
+ qed_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt);
/* enable internal target-read */
qed_wr(p_hwfn, p_hwfn->p_main_ptt,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
index acccd85..e4b4e3b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
@@ -473,6 +473,18 @@ int qed_final_cleanup(struct qed_hwfn *p_hwfn,
qed_set_queue_coalesce(u16 rx_coal, u16 tx_coal, void *p_handle);
/**
+ * @brief qed_pglueb_set_pfid_enable - Enable or disable PCI BUS MASTER
+ *
+ * @param p_hwfn
+ * @param p_ptt
+ * @param b_enable - true/false
+ *
+ * @return int
+ */
+int qed_pglueb_set_pfid_enable(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt, bool b_enable);
+
+/**
* @brief db_recovery_add - add doorbell information to the doorbell
* recovery mechanism.
*
diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index 9234091..e23980e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -255,112 +255,114 @@ static int qed_grc_attn_cb(struct qed_hwfn *p_hwfn)
#define PGLUE_ATTENTION_ICPL_VALID (1 << 23)
#define PGLUE_ATTENTION_ZLR_VALID (1 << 25)
#define PGLUE_ATTENTION_ILT_VALID (1 << 23)
-static int qed_pglub_rbc_attn_cb(struct qed_hwfn *p_hwfn)
+
+int qed_pglueb_rbc_attn_handler(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt)
{
u32 tmp;
- tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_TX_ERR_WR_DETAILS2);
+ tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_WR_DETAILS2);
if (tmp & PGLUE_ATTENTION_VALID) {
u32 addr_lo, addr_hi, details;
- addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_lo = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_WR_ADD_31_0);
- addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_hi = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_WR_ADD_63_32);
- details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ details = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_WR_DETAILS);
- DP_INFO(p_hwfn,
- "Illegal write by chip to [%08x:%08x] blocked.\n"
- "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
- "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
- addr_hi, addr_lo, details,
- (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
- (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
- GET_FIELD(details,
- PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
- tmp,
- GET_FIELD(tmp,
- PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0,
- GET_FIELD(tmp,
- PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
- GET_FIELD(tmp,
- PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0);
+ DP_NOTICE(p_hwfn,
+ "Illegal write by chip to [%08x:%08x] blocked.\n"
+ "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
+ "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
+ addr_hi, addr_lo, details,
+ (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
+ (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
+ GET_FIELD(details,
+ PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
+ tmp,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0);
}
- tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_TX_ERR_RD_DETAILS2);
+ tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_RD_DETAILS2);
if (tmp & PGLUE_ATTENTION_RD_VALID) {
u32 addr_lo, addr_hi, details;
- addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_lo = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_RD_ADD_31_0);
- addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_hi = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_RD_ADD_63_32);
- details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ details = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_TX_ERR_RD_DETAILS);
- DP_INFO(p_hwfn,
- "Illegal read by chip from [%08x:%08x] blocked.\n"
- " Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
- " Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
- addr_hi, addr_lo, details,
- (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
- (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
- GET_FIELD(details,
- PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
- tmp,
- GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1
- : 0,
- GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
- GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1
- : 0);
+ DP_NOTICE(p_hwfn,
+ "Illegal read by chip from [%08x:%08x] blocked.\n"
+ "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
+ "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
+ addr_hi, addr_lo, details,
+ (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
+ (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
+ GET_FIELD(details,
+ PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
+ tmp,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
+ GET_FIELD(tmp,
+ PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0);
}
- tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL);
+ tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL);
if (tmp & PGLUE_ATTENTION_ICPL_VALID)
- DP_INFO(p_hwfn, "ICPL error - %08x\n", tmp);
+ DP_NOTICE(p_hwfn, "ICPL error - %08x\n", tmp);
- tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS);
+ tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS);
if (tmp & PGLUE_ATTENTION_ZLR_VALID) {
u32 addr_hi, addr_lo;
- addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_lo = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_MASTER_ZLR_ERR_ADD_31_0);
- addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_hi = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_MASTER_ZLR_ERR_ADD_63_32);
- DP_INFO(p_hwfn, "ZLR eror - %08x [Address %08x:%08x]\n",
- tmp, addr_hi, addr_lo);
+ DP_NOTICE(p_hwfn, "ZLR error - %08x [Address %08x:%08x]\n",
+ tmp, addr_hi, addr_lo);
}
- tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_VF_ILT_ERR_DETAILS2);
+ tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_VF_ILT_ERR_DETAILS2);
if (tmp & PGLUE_ATTENTION_ILT_VALID) {
u32 addr_hi, addr_lo, details;
- addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_lo = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_VF_ILT_ERR_ADD_31_0);
- addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ addr_hi = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_VF_ILT_ERR_ADD_63_32);
- details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+ details = qed_rd(p_hwfn, p_ptt,
PGLUE_B_REG_VF_ILT_ERR_DETAILS);
- DP_INFO(p_hwfn,
- "ILT error - Details %08x Details2 %08x [Address %08x:%08x]\n",
- details, tmp, addr_hi, addr_lo);
+ DP_NOTICE(p_hwfn,
+ "ILT error - Details %08x Details2 %08x [Address %08x:%08x]\n",
+ details, tmp, addr_hi, addr_lo);
}
/* Clear the indications */
- qed_wr(p_hwfn, p_hwfn->p_dpc_ptt,
- PGLUE_B_REG_LATCHED_ERRORS_CLR, (1 << 2));
+ qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_LATCHED_ERRORS_CLR, BIT(2));
return 0;
}
+static int qed_pglueb_rbc_attn_cb(struct qed_hwfn *p_hwfn)
+{
+ return qed_pglueb_rbc_attn_handler(p_hwfn, p_hwfn->p_dpc_ptt);
+}
+
#define QED_DORQ_ATTENTION_REASON_MASK (0xfffff)
#define QED_DORQ_ATTENTION_OPAQUE_MASK (0xffff)
#define QED_DORQ_ATTENTION_OPAQUE_SHIFT (0x0)
@@ -540,7 +542,7 @@ enum aeu_invert_reg_special_type {
{"PGLUE misc_flr", ATTENTION_SINGLE,
NULL, MAX_BLOCK_ID},
{"PGLUE B RBC", ATTENTION_PAR_INT,
- qed_pglub_rbc_attn_cb, BLOCK_PGLUE_B},
+ qed_pglueb_rbc_attn_cb, BLOCK_PGLUE_B},
{"PGLUE misc_mctp", ATTENTION_SINGLE,
NULL, MAX_BLOCK_ID},
{"Flash event", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID},
diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h
index d81a62e..1f356ed 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.h
@@ -431,4 +431,7 @@ int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt,
#define QED_MAPPING_MEMORY_SIZE(dev) (NUM_OF_SBS(dev))
+int qed_pglueb_rbc_attn_handler(struct qed_hwfn *p_hwfn,
+ struct qed_ptt *p_ptt);
+
#endif
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index e7f18e3..1024484 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -1070,6 +1070,27 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
return 0;
}
+int qed_mcp_load_done(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
+{
+ u32 resp = 0, param = 0;
+ int rc;
+
+ rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_LOAD_DONE, 0, &resp,
+ ¶m);
+ if (rc) {
+ DP_NOTICE(p_hwfn,
+ "Failed to send a LOAD_DONE command, rc = %d\n", rc);
+ return rc;
+ }
+
+ /* Check if there is a DID mismatch between nvm-cfg/efuse */
+ if (param & FW_MB_PARAM_LOAD_DONE_DID_EFUSE_ERROR)
+ DP_NOTICE(p_hwfn,
+ "warning: device configuration is not supported on this board type. The device may not function as expected.\n");
+
+ return 0;
+}
+
int qed_mcp_unload_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
{
struct qed_mcp_mb_params mb_params;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
index eddf677..387c5e6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
@@ -801,6 +801,16 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
struct qed_load_req_params *p_params);
/**
+ * @brief Sends a LOAD_DONE message to the MFW
+ *
+ * @param p_hwfn
+ * @param p_ptt
+ *
+ * @return int - 0 - Operation was successful.
+ */
+int qed_mcp_load_done(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt);
+
+/**
* @brief Sends a UNLOAD_REQ message to the MFW
*
* @param p_hwfn
--
1.8.3.1
^ permalink raw reply related
* [PATCH v2 net-next 0/3] qed*: Error recovery process
From: Michal Kalderon @ 2019-01-28 17:27 UTC (permalink / raw)
To: michal.kalderon, ariel.elior, davem; +Cc: netdev, natechancellor
Parity errors might happen in the device's memories due to momentary bit
flips which are caused by radiation.
Errors that are not correctable initiate a process kill event, which blocks
the device access towards the host and the network, and a recovery process
is started in the management FW and in the driver.
This series adds the support of this process in the qed core module and in
the qede driver (patches 2 & 3).
Patch 1 in the series revises the load sequence, to avoid PCI errors that
might be observed during a recovery process.
Changes in v2:
- Addressed issue found in https://patchwork.ozlabs.org/patch/1030545/
The change was done be removing the enum and passing a boolean to
the related functions.
Tomer Tayar (3):
qed: Revise load sequence to avoid PCI errors
qed: Add infrastructure for error detection and recovery
qede: Error recovery process
drivers/net/ethernet/qlogic/qed/qed.h | 5 +-
drivers/net/ethernet/qlogic/qed/qed_dev.c | 158 +++++++------
drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 12 +
drivers/net/ethernet/qlogic/qed/qed_hsi.h | 2 +-
drivers/net/ethernet/qlogic/qed/qed_hw.c | 11 +
drivers/net/ethernet/qlogic/qed/qed_int.c | 126 +++++------
drivers/net/ethernet/qlogic/qed/qed_int.h | 3 +
drivers/net/ethernet/qlogic/qed/qed_main.c | 30 +++
drivers/net/ethernet/qlogic/qed/qed_mcp.c | 115 ++++++++++
drivers/net/ethernet/qlogic/qed/qed_mcp.h | 42 ++++
drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 +
drivers/net/ethernet/qlogic/qed/qed_spq.c | 22 ++
drivers/net/ethernet/qlogic/qed/qed_sriov.c | 9 +-
drivers/net/ethernet/qlogic/qede/qede.h | 3 +
drivers/net/ethernet/qlogic/qede/qede_main.c | 292 ++++++++++++++++++++-----
drivers/net/ethernet/qlogic/qede/qede_rdma.c | 63 ++++--
include/linux/qed/qed_if.h | 20 ++
include/linux/qed/qede_rdma.h | 10 +-
18 files changed, 723 insertions(+), 202 deletions(-)
--
1.8.3.1
^ permalink raw reply
* [PATCH AUTOSEL 4.19 012/258] wil6210: fix memory leak in wil_find_tx_bcast_2
From: Sasha Levin @ 2019-01-28 15:55 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Lior David, Maya Erez, Kalle Valo, Sasha Levin, linux-wireless,
wil6210, netdev
In-Reply-To: <20190128155924.51521-1-sashal@kernel.org>
From: Lior David <liord@codeaurora.org>
[ Upstream commit 664497400c89a4d40aee51bcf48bbd2e4dc71104 ]
A successful call to wil_tx_ring takes skb reference so
it will only be freed in wil_tx_complete. Consume the skb
in wil_find_tx_bcast_2 to prevent memory leak.
Signed-off-by: Lior David <liord@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/ath/wil6210/txrx.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index 6a7943e487fb..75c8aa297107 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -1313,6 +1313,8 @@ static struct wil_ring *wil_find_tx_bcast_2(struct wil6210_priv *wil,
wil_dbg_txrx(wil, "BCAST DUP -> ring %d\n", i);
wil_set_da_for_vring(wil, skb2, i);
wil_tx_ring(wil, vif, v2, skb2);
+ /* successful call to wil_tx_ring takes skb2 ref */
+ dev_kfree_skb_any(skb2);
} else {
wil_err(wil, "skb_copy failed\n");
}
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.19 011/258] wil6210: fix reset flow for Talyn-mb
From: Sasha Levin @ 2019-01-28 15:55 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Alexei Avshalom Lazar, Maya Erez, Kalle Valo, Sasha Levin,
linux-wireless, wil6210, netdev
In-Reply-To: <20190128155924.51521-1-sashal@kernel.org>
From: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
[ Upstream commit d083b2e2b7db5cca1791643d036e6597af27f49b ]
With current reset flow, Talyn sometimes get stuck causing PCIe
enumeration to fail. Fix this by removing some reset flow operations
that are not relevant for Talyn.
Setting bit 15 in RGF_HP_CTRL is WBE specific and is not in use for
all wil6210 devices.
For Sparrow, BIT_HPAL_PERST_FROM_PAD and BIT_CAR_PERST_RST were set
as a WA an HW issue.
Signed-off-by: Alexei Avshalom Lazar <ailizaro@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/ath/wil6210/main.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 7debed6bec06..a0fe8cbad104 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -995,10 +995,13 @@ static int wil_target_reset(struct wil6210_priv *wil, int no_flash)
wil_dbg_misc(wil, "Resetting \"%s\"...\n", wil->hw_name);
- /* Clear MAC link up */
- wil_s(wil, RGF_HP_CTRL, BIT(15));
- wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0, BIT_HPAL_PERST_FROM_PAD);
- wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0, BIT_CAR_PERST_RST);
+ if (wil->hw_version < HW_VER_TALYN) {
+ /* Clear MAC link up */
+ wil_s(wil, RGF_HP_CTRL, BIT(15));
+ wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0,
+ BIT_HPAL_PERST_FROM_PAD);
+ wil_s(wil, RGF_USER_CLKS_CTL_SW_RST_MASK_0, BIT_CAR_PERST_RST);
+ }
wil_halt_cpu(wil);
--
2.19.1
^ permalink raw reply related
* [PATCH AUTOSEL 4.19 013/258] ath10k: assign 'n_cipher_suites' for WCN3990
From: Sasha Levin @ 2019-01-28 15:55 UTC (permalink / raw)
To: linux-kernel, stable
Cc: Brian Norris, Rakesh Pillai, Govind Singh, Kalle Valo,
Sasha Levin, ath10k, linux-wireless, netdev
In-Reply-To: <20190128155924.51521-1-sashal@kernel.org>
From: Brian Norris <briannorris@chromium.org>
[ Upstream commit 2bd345cd2bfc0bd44528896313c0b45f087bdf67 ]
Commit 2ea9f12cefe4 ("ath10k: add new cipher suite support") added a new
n_cipher_suites HW param with a fallback value and a warning log. Commit
03a72288c546 ("ath10k: wmi: add hw params entry for wcn3990") later
added WCN3990 HW entries, but it missed the n_cipher_suites.
Rather than seeing this warning every boot
ath10k_snoc 18800000.wifi: invalid hw_params.n_cipher_suites 0
let's provide the appropriate value.
Cc: Rakesh Pillai <pillair@qti.qualcomm.com>
Cc: Govind Singh <govinds@qti.qualcomm.com>
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/net/wireless/ath/ath10k/core.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index c40cd129afe7..5210cffb5344 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -532,6 +532,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
.hw_ops = &wcn3990_ops,
.decap_align_bytes = 1,
.num_peers = TARGET_HL_10_TLV_NUM_PEERS,
+ .n_cipher_suites = 8,
.ast_skid_limit = TARGET_HL_10_TLV_AST_SKID_LIMIT,
.num_wds_entries = TARGET_HL_10_TLV_NUM_WDS_ENTRIES,
.target_64bit = true,
--
2.19.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox