[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e49ae0e2-abe5-73f5-93e7-1f2b6386ded2@solarflare.com>
Date: Wed, 15 Jun 2016 17:48:49 +0100
From: Edward Cree <ecree@...arflare.com>
To: <linux-net-drivers@...arflare.com>, <netdev@...r.kernel.org>,
<davem@...emloft.net>
Subject: [PATCH net-next 11/17] sfc: Take mac_lock before calling
efx_ef10_filter_table_probe
From: Martin Habets <mhabets@...arflare.com>
When trying to enslave an SFC interface to a bond the following BUG_ON was
hit:
kernel BUG [in ef10.c]!
CPU: 0 PID: 4383 Comm: ifenslave Tainted: G
...
Call Trace:
efx_ef10_filter_add_vlan+0x121/0x180 [sfc]
efx_ef10_filter_table_probe+0x2a2/0x4f0 [sfc]
efx_ef10_set_mac_address+0x370/0x6d0 [sfc]
efx_set_mac_address+0x7d/0x120 [sfc]
dev_set_mac_address+0x43/0xa0
bond_enslave+0x337/0xea0 [bonding]
This comes from function efx_ef10_filter_vlan_sync_rx_mode.
To solve the bug we ensure the mac_lock is taken before calling
efx_ef10_filter_add_vlan. But to avoid a priority inversion mac_lock must
be taken before filter_sem.
To satisfy these requirements we end up taking mac_lock in
efx_ef10_vport_set_mac_address, efx_ef10_set_mac_address,
efx_ef10_sriov_set_vf_vlan and efx_probe_filters.
Signed-off-by: Edward Cree <ecree@...arflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 4 ++++
drivers/net/ethernet/sfc/ef10_sriov.c | 7 +++++--
drivers/net/ethernet/sfc/efx.c | 2 ++
3 files changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index a3c00ff..add9f46 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -4666,6 +4666,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx)
efx_device_detach_sync(efx);
efx_net_stop(efx->net_dev);
+
+ mutex_lock(&efx->mac_lock);
down_write(&efx->filter_sem);
efx_ef10_filter_table_remove(efx);
@@ -4678,6 +4680,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx)
efx_ef10_filter_table_probe(efx);
up_write(&efx->filter_sem);
+ mutex_unlock(&efx->mac_lock);
+
if (was_enabled)
efx_net_open(efx->net_dev);
netif_device_attach(efx->net_dev);
diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c
index 3c17f27..a76610a 100644
--- a/drivers/net/ethernet/sfc/ef10_sriov.c
+++ b/drivers/net/ethernet/sfc/ef10_sriov.c
@@ -554,6 +554,7 @@ int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, u16 vlan,
efx_device_detach_sync(vf->efx);
efx_net_stop(vf->efx->net_dev);
+ mutex_lock(&vf->efx->mac_lock);
down_write(&vf->efx->filter_sem);
vf->efx->type->filter_table_remove(vf->efx);
@@ -630,6 +631,7 @@ restore_filters:
goto reset_nic_up_write;
up_write(&vf->efx->filter_sem);
+ mutex_unlock(&vf->efx->mac_lock);
up_write(&vf->efx->filter_sem);
@@ -642,9 +644,10 @@ restore_filters:
return rc;
reset_nic_up_write:
- if (vf->efx)
+ if (vf->efx) {
up_write(&vf->efx->filter_sem);
-
+ mutex_unlock(&vf->efx->mac_lock);
+ }
reset_nic:
if (vf->efx) {
netif_err(efx, drv, efx->net_dev,
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 902bcf2..130ee17 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -1729,6 +1729,7 @@ static int efx_probe_filters(struct efx_nic *efx)
spin_lock_init(&efx->filter_lock);
init_rwsem(&efx->filter_sem);
+ mutex_lock(&efx->mac_lock);
down_write(&efx->filter_sem);
rc = efx->type->filter_table_probe(efx);
if (rc)
@@ -1767,6 +1768,7 @@ static int efx_probe_filters(struct efx_nic *efx)
#endif
out_unlock:
up_write(&efx->filter_sem);
+ mutex_unlock(&efx->mac_lock);
return rc;
}
Powered by blists - more mailing lists