- Linux-stable-mirror - lists.linaro.org

[Linux-stable-mirror] Patch "net: mpls: Fix nexthop alive tracking on down events" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: mpls: Fix nexthop alive tracking on down events to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-mpls-fix-nexthop-alive-tracking-on-down-events.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: David Ahern <dsa(a)cumulusnetworks.com> Date: Mon, 13 Mar 2017 16:49:10 -0700 Subject: net: mpls: Fix nexthop alive tracking on down events From: David Ahern <dsa(a)cumulusnetworks.com> [ Upstream commit 61733c91c454a61be0ffc93fe46a5d5f2f048c1c ] Alive tracking of nexthops can account for a link twice if the carrier goes down followed by an admin down of the same link rendering multipath routes useless. This is similar to 79099aab38c8 for UNREGISTER events and DOWN events. Fix by tracking number of alive nexthops in mpls_ifdown similar to the logic in mpls_ifup. Checking the flags per nexthop once after all events have been processed is simpler than trying to maintian a running count through all event combinations. Also, WRITE_ONCE is used instead of ACCESS_ONCE to set rt_nhn_alive per a comment from checkpatch: WARNING: Prefer WRITE_ONCE(<FOO>, <BAR>) over ACCESS_ONCE(<FOO>) = <BAR> Fixes: c89359a42e2a4 ("mpls: support for dead routes") Signed-off-by: David Ahern <dsa(a)cumulusnetworks.com> Acked-by: Robert Shearman <rshearma(a)brocade.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- net/mpls/af_mpls.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -937,6 +937,8 @@ static void mpls_ifdown(struct net_devic { struct mpls_route __rcu **platform_label; struct net *net = dev_net(dev); + unsigned int nh_flags = RTNH_F_DEAD | RTNH_F_LINKDOWN; + unsigned int alive; unsigned index; platform_label = rtnl_dereference(net->mpls.platform_label); @@ -946,9 +948,11 @@ static void mpls_ifdown(struct net_devic if (!rt) continue; + alive = 0; change_nexthops(rt) { if (rtnl_dereference(nh->nh_dev) != dev) - continue; + goto next; + switch (event) { case NETDEV_DOWN: case NETDEV_UNREGISTER: @@ -956,13 +960,16 @@ static void mpls_ifdown(struct net_devic /* fall through */ case NETDEV_CHANGE: nh->nh_flags |= RTNH_F_LINKDOWN; - if (event != NETDEV_UNREGISTER) - ACCESS_ONCE(rt->rt_nhn_alive) = rt->rt_nhn_alive - 1; break; } if (event == NETDEV_UNREGISTER) RCU_INIT_POINTER(nh->nh_dev, NULL); +next: + if (!(nh->nh_flags & nh_flags)) + alive++; } endfor_nexthops(rt); + + WRITE_ONCE(rt->rt_nhn_alive, alive); } } Patches currently in stable-queue which might be from dsa(a)cumulusnetworks.com are queue-4.9/net-mpls-fix-nexthop-alive-tracking-on-down-events.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net/mlx5: Fix create autogroup prev initializer" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net/mlx5: Fix create autogroup prev initializer to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-mlx5-fix-create-autogroup-prev-initializer.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Paul Blakey <paulb(a)mellanox.com> Date: Fri, 10 Mar 2017 14:33:01 +0200 Subject: net/mlx5: Fix create autogroup prev initializer From: Paul Blakey <paulb(a)mellanox.com> [ Upstream commit af36370569eb37420e1e78a2e60c277b781fcd00 ] The autogroups list is a list of non overlapping group boundaries sorted by their start index. If the autogroups list wasn't empty and an empty group slot was found at the start of the list, the new group was added to the end of the list instead of the beginning, as the prev initializer was incorrect. When this was repeated, it caused multiple groups to have overlapping boundaries. Fixed that by correctly initializing the prev pointer to the start of the list. Fixes: eccec8da3b4e ('net/mlx5: Keep autogroups list ordered') Signed-off-by: Paul Blakey <paulb(a)mellanox.com> Reviewed-by: Mark Bloch <markb(a)mellanox.com> Signed-off-by: Saeed Mahameed <saeedm(a)mellanox.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1015,7 +1015,7 @@ static struct mlx5_flow_group *create_au u32 *match_criteria) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct list_head *prev = ft->node.children.prev; + struct list_head *prev = &ft->node.children; unsigned int candidate_index = 0; struct mlx5_flow_group *fg; void *match_criteria_addr; Patches currently in stable-queue which might be from paulb(a)mellanox.com are queue-4.9/net-mlx5-fix-create-autogroup-prev-initializer.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net/mlx5: Don't save PCI state when PCI error is detected" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net/mlx5: Don't save PCI state when PCI error is detected to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-mlx5-don-t-save-pci-state-when-pci-error-is-detected.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Daniel Jurgens <danielj(a)mellanox.com> Date: Fri, 10 Mar 2017 14:33:02 +0200 Subject: net/mlx5: Don't save PCI state when PCI error is detected From: Daniel Jurgens <danielj(a)mellanox.com> [ Upstream commit 5d47f6c89d568ab61712d8c40676fbb020b68752 ] When a PCI error is detected the PCI state could be corrupt, don't save it in that flow. Save the state after initialization. After restoring the PCI state during slot reset save it again, restoring the state destroys the previously saved state info. Fixes: 05ac2c0b7438 ('net/mlx5: Fix race between PCI error handlers and health work') Signed-off-by: Daniel Jurgens <danielj(a)mellanox.com> Signed-off-by: Saeed Mahameed <saeedm(a)mellanox.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1283,6 +1283,7 @@ static int init_one(struct pci_dev *pdev if (err) goto clean_load; + pci_save_state(pdev); return 0; clean_load: @@ -1331,9 +1332,8 @@ static pci_ers_result_t mlx5_pci_err_det mlx5_enter_error_state(dev); mlx5_unload_one(dev, priv, false); - /* In case of kernel call save the pci state and drain the health wq */ + /* In case of kernel call drain the health wq */ if (state) { - pci_save_state(pdev); mlx5_drain_health_wq(dev); mlx5_pci_disable_device(dev); } @@ -1385,6 +1385,7 @@ static pci_ers_result_t mlx5_pci_slot_re pci_set_master(pdev); pci_restore_state(pdev); + pci_save_state(pdev); if (wait_vital(pdev)) { dev_err(&pdev->dev, "%s: wait_vital timed out\n", __func__); Patches currently in stable-queue which might be from danielj(a)mellanox.com are queue-4.9/ib-core-fix-calculation-of-maximum-roce-mtu.patch queue-4.9/net-mlx5-don-t-save-pci-state-when-pci-error-is-detected.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net/mlx4_core: Avoid delays during VF driver device shutdown" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net/mlx4_core: Avoid delays during VF driver device shutdown to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-mlx4_core-avoid-delays-during-vf-driver-device-shutdown.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Jack Morgenstein <jackm(a)dev.mellanox.co.il> Date: Mon, 13 Mar 2017 19:29:08 +0200 Subject: net/mlx4_core: Avoid delays during VF driver device shutdown From: Jack Morgenstein <jackm(a)dev.mellanox.co.il> [ Upstream commit 4cbe4dac82e423ecc9a0ba46af24a860853259f4 ] Some Hypervisors detach VFs from VMs by instantly causing an FLR event to be generated for a VF. In the mlx4 case, this will cause that VF's comm channel to be disabled before the VM has an opportunity to invoke the VF device's "shutdown" method. For such Hypervisors, there is a race condition between the VF's shutdown method and its internal-error detection/reset thread. The internal-error detection/reset thread (which runs every 5 seconds) also detects a disabled comm channel. If the internal-error detection/reset flow wins the race, we still get delays (while that flow tries repeatedly to detect comm-channel recovery). The cited commit fixed the command timeout problem when the internal-error detection/reset flow loses the race. This commit avoids the unneeded delays when the internal-error detection/reset flow wins. Fixes: d585df1c5ccf ("net/mlx4_core: Avoid command timeouts during VF driver device shutdown") Signed-off-by: Jack Morgenstein <jackm(a)dev.mellanox.co.il> Reported-by: Simon Xiao <sixiao(a)microsoft.com> Signed-off-by: Tariq Toukan <tariqt(a)mellanox.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/mellanox/mlx4/cmd.c | 11 +++++++++++ drivers/net/ethernet/mellanox/mlx4/main.c | 11 +++++++++++ include/linux/mlx4/device.h | 1 + 3 files changed, 23 insertions(+) --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -2304,6 +2304,17 @@ static int sync_toggles(struct mlx4_dev rd_toggle = swab32(readl(&priv->mfunc.comm->slave_read)); if (wr_toggle == 0xffffffff || rd_toggle == 0xffffffff) { /* PCI might be offline */ + + /* If device removal has been requested, + * do not continue retrying. + */ + if (dev->persist->interface_state & + MLX4_INTERFACE_STATE_NOWAIT) { + mlx4_warn(dev, + "communication channel is offline\n"); + return -EIO; + } + msleep(100); wr_toggle = swab32(readl(&priv->mfunc.comm-> slave_write)); --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -1940,6 +1940,14 @@ static int mlx4_comm_check_offline(struc (u32)(1 << COMM_CHAN_OFFLINE_OFFSET)); if (!offline_bit) return 0; + + /* If device removal has been requested, + * do not continue retrying. + */ + if (dev->persist->interface_state & + MLX4_INTERFACE_STATE_NOWAIT) + break; + /* There are cases as part of AER/Reset flow that PF needs * around 100 msec to load. We therefore sleep for 100 msec * to allow other tasks to make use of that CPU during this @@ -3954,6 +3962,9 @@ static void mlx4_remove_one(struct pci_d struct devlink *devlink = priv_to_devlink(priv); int active_vfs = 0; + if (mlx4_is_slave(dev)) + persist->interface_state |= MLX4_INTERFACE_STATE_NOWAIT; + mutex_lock(&persist->interface_state_mutex); persist->interface_state |= MLX4_INTERFACE_STATE_DELETION; mutex_unlock(&persist->interface_state_mutex); --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -476,6 +476,7 @@ enum { enum { MLX4_INTERFACE_STATE_UP = 1 << 0, MLX4_INTERFACE_STATE_DELETION = 1 << 1, + MLX4_INTERFACE_STATE_NOWAIT = 1 << 2, }; #define MSTR_SM_CHANGE_MASK (MLX4_EQ_PORT_INFO_MSTR_SM_SL_CHANGE_MASK | \ Patches currently in stable-queue which might be from jackm(a)dev.mellanox.co.il are queue-4.9/net-mlx4_core-avoid-delays-during-vf-driver-device-shutdown.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: initialize msg.msg_flags in recvfrom" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: initialize msg.msg_flags in recvfrom to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-initialize-msg.msg_flags-in-recvfrom.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Alexander Potapenko <glider(a)google.com> Date: Wed, 8 Mar 2017 18:08:16 +0100 Subject: net: initialize msg.msg_flags in recvfrom From: Alexander Potapenko <glider(a)google.com> [ Upstream commit 9f138fa609c47403374a862a08a41394be53d461 ] KMSAN reports a use of uninitialized memory in put_cmsg() because msg.msg_flags in recvfrom haven't been initialized properly. The flag values don't affect the result on this path, but it's still a good idea to initialize them explicitly. Signed-off-by: Alexander Potapenko <glider(a)google.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- net/socket.c | 1 + 1 file changed, 1 insertion(+) --- a/net/socket.c +++ b/net/socket.c @@ -1702,6 +1702,7 @@ SYSCALL_DEFINE6(recvfrom, int, fd, void /* We assume all kernel code knows the size of sockaddr_storage */ msg.msg_namelen = 0; msg.msg_iocb = NULL; + msg.msg_flags = 0; if (sock->file->f_flags & O_NONBLOCK) flags |= MSG_DONTWAIT; err = sock_recvmsg(sock, &msg, flags); Patches currently in stable-queue which might be from glider(a)google.com are queue-4.9/net-initialize-msg.msg_flags-in-recvfrom.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: bcmgenet: synchronize irq0 status between the isr and task" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: bcmgenet: synchronize irq0 status between the isr and task to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Doug Berger <opendmb(a)gmail.com> Date: Thu, 9 Mar 2017 16:58:47 -0800 Subject: net: bcmgenet: synchronize irq0 status between the isr and task From: Doug Berger <opendmb(a)gmail.com> [ Upstream commit 07c52d6a0b955a8a28834f9354793cfc4b81d0e9 ] Add a spinlock to ensure that irq0_stat is not unintentionally altered as the result of preemption. Also removed unserviced irq0 interrupts and removed irq1_stat since there is no bottom half service for those interrupts. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger <opendmb(a)gmail.com> Reviewed-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 73 +++++++++++++------------ drivers/net/ethernet/broadcom/genet/bcmgenet.h | 6 +- 2 files changed, 44 insertions(+), 35 deletions(-) --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2513,24 +2513,28 @@ static int bcmgenet_init_dma(struct bcmg /* Interrupt bottom half */ static void bcmgenet_irq_task(struct work_struct *work) { + unsigned long flags; + unsigned int status; struct bcmgenet_priv *priv = container_of( work, struct bcmgenet_priv, bcmgenet_irq_work); netif_dbg(priv, intr, priv->dev, "%s\n", __func__); - if (priv->irq0_stat & UMAC_IRQ_MPD_R) { - priv->irq0_stat &= ~UMAC_IRQ_MPD_R; + spin_lock_irqsave(&priv->lock, flags); + status = priv->irq0_stat; + priv->irq0_stat = 0; + spin_unlock_irqrestore(&priv->lock, flags); + + if (status & UMAC_IRQ_MPD_R) { netif_dbg(priv, wol, priv->dev, "magic packet detected, waking up\n"); bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC); } /* Link UP/DOWN event */ - if (priv->irq0_stat & UMAC_IRQ_LINK_EVENT) { + if (status & UMAC_IRQ_LINK_EVENT) phy_mac_interrupt(priv->phydev, - !!(priv->irq0_stat & UMAC_IRQ_LINK_UP)); - priv->irq0_stat &= ~UMAC_IRQ_LINK_EVENT; - } + !!(status & UMAC_IRQ_LINK_UP)); } /* bcmgenet_isr1: handle Rx and Tx priority queues */ @@ -2539,22 +2543,21 @@ static irqreturn_t bcmgenet_isr1(int irq struct bcmgenet_priv *priv = dev_id; struct bcmgenet_rx_ring *rx_ring; struct bcmgenet_tx_ring *tx_ring; - unsigned int index; + unsigned int index, status; - /* Save irq status for bottom-half processing. */ - priv->irq1_stat = - bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_STAT) & + /* Read irq status */ + status = bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_STAT) & ~bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_MASK_STATUS); /* clear interrupts */ - bcmgenet_intrl2_1_writel(priv, priv->irq1_stat, INTRL2_CPU_CLEAR); + bcmgenet_intrl2_1_writel(priv, status, INTRL2_CPU_CLEAR); netif_dbg(priv, intr, priv->dev, - "%s: IRQ=0x%x\n", __func__, priv->irq1_stat); + "%s: IRQ=0x%x\n", __func__, status); /* Check Rx priority queue interrupts */ for (index = 0; index < priv->hw_params->rx_queues; index++) { - if (!(priv->irq1_stat & BIT(UMAC_IRQ1_RX_INTR_SHIFT + index))) + if (!(status & BIT(UMAC_IRQ1_RX_INTR_SHIFT + index))) continue; rx_ring = &priv->rx_rings[index]; @@ -2567,7 +2570,7 @@ static irqreturn_t bcmgenet_isr1(int irq /* Check Tx priority queue interrupts */ for (index = 0; index < priv->hw_params->tx_queues; index++) { - if (!(priv->irq1_stat & BIT(index))) + if (!(status & BIT(index))) continue; tx_ring = &priv->tx_rings[index]; @@ -2587,19 +2590,20 @@ static irqreturn_t bcmgenet_isr0(int irq struct bcmgenet_priv *priv = dev_id; struct bcmgenet_rx_ring *rx_ring; struct bcmgenet_tx_ring *tx_ring; + unsigned int status; + unsigned long flags; - /* Save irq status for bottom-half processing. */ - priv->irq0_stat = - bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT) & + /* Read irq status */ + status = bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT) & ~bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_MASK_STATUS); /* clear interrupts */ - bcmgenet_intrl2_0_writel(priv, priv->irq0_stat, INTRL2_CPU_CLEAR); + bcmgenet_intrl2_0_writel(priv, status, INTRL2_CPU_CLEAR); netif_dbg(priv, intr, priv->dev, - "IRQ=0x%x\n", priv->irq0_stat); + "IRQ=0x%x\n", status); - if (priv->irq0_stat & UMAC_IRQ_RXDMA_DONE) { + if (status & UMAC_IRQ_RXDMA_DONE) { rx_ring = &priv->rx_rings[DESC_INDEX]; if (likely(napi_schedule_prep(&rx_ring->napi))) { @@ -2608,7 +2612,7 @@ static irqreturn_t bcmgenet_isr0(int irq } } - if (priv->irq0_stat & UMAC_IRQ_TXDMA_DONE) { + if (status & UMAC_IRQ_TXDMA_DONE) { tx_ring = &priv->tx_rings[DESC_INDEX]; if (likely(napi_schedule_prep(&tx_ring->napi))) { @@ -2617,22 +2621,23 @@ static irqreturn_t bcmgenet_isr0(int irq } } - if (priv->irq0_stat & (UMAC_IRQ_PHY_DET_R | - UMAC_IRQ_PHY_DET_F | - UMAC_IRQ_LINK_EVENT | - UMAC_IRQ_HFB_SM | - UMAC_IRQ_HFB_MM | - UMAC_IRQ_MPD_R)) { - /* all other interested interrupts handled in bottom half */ - schedule_work(&priv->bcmgenet_irq_work); - } - if ((priv->hw_params->flags & GENET_HAS_MDIO_INTR) && - priv->irq0_stat & (UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR)) { - priv->irq0_stat &= ~(UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR); + status & (UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR)) { wake_up(&priv->wq); } + /* all other interested interrupts handled in bottom half */ + status &= (UMAC_IRQ_LINK_EVENT | + UMAC_IRQ_MPD_R); + if (status) { + /* Save irq status for bottom-half processing. */ + spin_lock_irqsave(&priv->lock, flags); + priv->irq0_stat |= status; + spin_unlock_irqrestore(&priv->lock, flags); + + schedule_work(&priv->bcmgenet_irq_work); + } + return IRQ_HANDLED; } @@ -3334,6 +3339,8 @@ static int bcmgenet_probe(struct platfor goto err; } + spin_lock_init(&priv->lock); + SET_NETDEV_DEV(dev, &pdev->dev); dev_set_drvdata(&pdev->dev, dev); ether_addr_copy(dev->dev_addr, macaddr); --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -623,11 +623,13 @@ struct bcmgenet_priv { struct work_struct bcmgenet_irq_work; int irq0; int irq1; - unsigned int irq0_stat; - unsigned int irq1_stat; int wol_irq; bool wol_irq_disabled; + /* shared status */ + spinlock_t lock; + unsigned int irq0_stat; + /* HW descriptors/checksum variables */ bool desc_64b_en; bool desc_rxchk_en; Patches currently in stable-queue which might be from opendmb(a)gmail.com are queue-4.9/net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch queue-4.9/net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch queue-4.9/net-bcmgenet-correct-mib-access-of-unimac-runt-counters.patch queue-4.9/net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch queue-4.9/net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch queue-4.9/net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: bcmgenet: reserved phy revisions must be checked first" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: bcmgenet: reserved phy revisions must be checked first to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Doug Berger <opendmb(a)gmail.com> Date: Thu, 9 Mar 2017 16:58:45 -0800 Subject: net: bcmgenet: reserved phy revisions must be checked first From: Doug Berger <opendmb(a)gmail.com> [ Upstream commit eca4bad73409aedc6ff22f823c18b67a4f08c851 ] The reserved gphy_rev value of 0x01ff must be tested before the old or new scheme for GPHY major versioning are tested, otherwise it will be treated as 0xff00 according to the old scheme. Fixes: b04a2f5b9ff5 ("net: bcmgenet: add support for new GENET PHY revision scheme") Signed-off-by: Doug Berger <opendmb(a)gmail.com> Reviewed-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -3233,6 +3233,12 @@ static void bcmgenet_set_hw_params(struc */ gphy_rev = reg & 0xffff; + /* This is reserved so should require special treatment */ + if (gphy_rev == 0 || gphy_rev == 0x01ff) { + pr_warn("Invalid GPHY revision detected: 0x%04x\n", gphy_rev); + return; + } + /* This is the good old scheme, just GPHY major, no minor nor patch */ if ((gphy_rev & 0xf0) != 0) priv->gphy_rev = gphy_rev << 8; @@ -3241,12 +3247,6 @@ static void bcmgenet_set_hw_params(struc else if ((gphy_rev & 0xff00) != 0) priv->gphy_rev = gphy_rev; - /* This is reserved so should require special treatment */ - else if (gphy_rev == 0 || gphy_rev == 0x01ff) { - pr_warn("Invalid GPHY revision detected: 0x%04x\n", gphy_rev); - return; - } - #ifdef CONFIG_PHYS_ADDR_T_64BIT if (!(params->flags & GENET_HAS_40BITS)) pr_warn("GENET does not support 40-bits PA\n"); Patches currently in stable-queue which might be from opendmb(a)gmail.com are queue-4.9/net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch queue-4.9/net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch queue-4.9/net-bcmgenet-correct-mib-access-of-unimac-runt-counters.patch queue-4.9/net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch queue-4.9/net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch queue-4.9/net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: bcmgenet: Power up the internal PHY before probing the MII" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: bcmgenet: Power up the internal PHY before probing the MII to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Doug Berger <opendmb(a)gmail.com> Date: Thu, 9 Mar 2017 16:58:48 -0800 Subject: net: bcmgenet: Power up the internal PHY before probing the MII From: Doug Berger <opendmb(a)gmail.com> [ Upstream commit 6be371b053dc86f11465cc1abce2e99bda0a0574 ] When using the internal PHY it must be powered up when the MII is probed or the PHY will not be detected. Since the PHY is powered up at reset this has not been a problem. However, when the kernel is restarted with kexec the PHY will likely be powered down when the kernel starts so it will not be detected and the Ethernet link will not be established. This commit explicitly powers up the internal PHY when the GENET driver is probed to correct this behavior. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger <opendmb(a)gmail.com> Reviewed-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 8 ++++++++ 1 file changed, 8 insertions(+) --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -3296,6 +3296,7 @@ static int bcmgenet_probe(struct platfor const void *macaddr; struct resource *r; int err = -EIO; + const char *phy_mode_str; /* Up to GENET_MAX_MQ_CNT + 1 TX queues and RX queues */ dev = alloc_etherdev_mqs(sizeof(*priv), GENET_MAX_MQ_CNT + 1, @@ -3403,6 +3404,13 @@ static int bcmgenet_probe(struct platfor priv->clk_eee = NULL; } + /* If this is an internal GPHY, power it on now, before UniMAC is + * brought out of reset as absolutely no UniMAC activity is allowed + */ + if (dn && !of_property_read_string(dn, "phy-mode", &phy_mode_str) && + !strcasecmp(phy_mode_str, "internal")) + bcmgenet_power_up(priv, GENET_POWER_PASSIVE); + err = reset_umac(priv); if (err) goto err_clk_disable; Patches currently in stable-queue which might be from opendmb(a)gmail.com are queue-4.9/net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch queue-4.9/net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch queue-4.9/net-bcmgenet-correct-mib-access-of-unimac-runt-counters.patch queue-4.9/net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch queue-4.9/net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch queue-4.9/net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: bcmgenet: power down internal phy if open or resume fails" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: bcmgenet: power down internal phy if open or resume fails to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Doug Berger <opendmb(a)gmail.com> Date: Thu, 9 Mar 2017 16:58:46 -0800 Subject: net: bcmgenet: power down internal phy if open or resume fails From: Doug Berger <opendmb(a)gmail.com> [ Upstream commit 7627409cc4970e8c8b9de6945ad86a575290a94e ] Since the internal PHY is powered up during the open and resume functions it should be powered back down if the functions fail. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger <opendmb(a)gmail.com> Reviewed-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 4 ++++ 1 file changed, 4 insertions(+) --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2857,6 +2857,8 @@ err_irq0: err_fini_dma: bcmgenet_fini_dma(priv); err_clk_disable: + if (priv->internal_phy) + bcmgenet_power_down(priv, GENET_POWER_PASSIVE); clk_disable_unprepare(priv->clk); return ret; } @@ -3560,6 +3562,8 @@ static int bcmgenet_resume(struct device return 0; out_clk_disable: + if (priv->internal_phy) + bcmgenet_power_down(priv, GENET_POWER_PASSIVE); clk_disable_unprepare(priv->clk); return ret; } Patches currently in stable-queue which might be from opendmb(a)gmail.com are queue-4.9/net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch queue-4.9/net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch queue-4.9/net-bcmgenet-correct-mib-access-of-unimac-runt-counters.patch queue-4.9/net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch queue-4.9/net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch queue-4.9/net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch

7 years, 4 months

1
0
0 0

[Linux-stable-mirror] Patch "net: bcmgenet: correct the RBUF_OVFL_CNT and RBUF_ERR_CNT MIB values" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled net: bcmgenet: correct the RBUF_OVFL_CNT and RBUF_ERR_CNT MIB values to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Dec 18 14:12:34 CET 2017 From: Doug Berger <opendmb(a)gmail.com> Date: Thu, 9 Mar 2017 16:58:43 -0800 Subject: net: bcmgenet: correct the RBUF_OVFL_CNT and RBUF_ERR_CNT MIB values From: Doug Berger <opendmb(a)gmail.com> [ Upstream commit ffff71328a3c321f7c14cc1edd33577717037744 ] The location of the RBUF overflow and error counters has moved between different version of the GENET MAC. This commit corrects the driver to read from the correct locations depending on the version of the GENET MAC. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger <opendmb(a)gmail.com> Reviewed-by: Florian Fainelli <f.fainelli(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)verizon.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 60 ++++++++++++++++++++++--- drivers/net/ethernet/broadcom/genet/bcmgenet.h | 10 ++-- 2 files changed, 60 insertions(+), 10 deletions(-) --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -1,7 +1,7 @@ /* * Broadcom GENET (Gigabit Ethernet) controller driver * - * Copyright (c) 2014 Broadcom Corporation + * Copyright (c) 2014-2017 Broadcom * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -778,8 +778,9 @@ static const struct bcmgenet_stats bcmge STAT_GENET_RUNT("rx_runt_bytes", mib.rx_runt_bytes), /* Misc UniMAC counters */ STAT_GENET_MISC("rbuf_ovflow_cnt", mib.rbuf_ovflow_cnt, - UMAC_RBUF_OVFL_CNT), - STAT_GENET_MISC("rbuf_err_cnt", mib.rbuf_err_cnt, UMAC_RBUF_ERR_CNT), + UMAC_RBUF_OVFL_CNT_V1), + STAT_GENET_MISC("rbuf_err_cnt", mib.rbuf_err_cnt, + UMAC_RBUF_ERR_CNT_V1), STAT_GENET_MISC("mdf_err_cnt", mib.mdf_err_cnt, UMAC_MDF_ERR_CNT), STAT_GENET_SOFT_MIB("alloc_rx_buff_failed", mib.alloc_rx_buff_failed), STAT_GENET_SOFT_MIB("rx_dma_failed", mib.rx_dma_failed), @@ -821,6 +822,45 @@ static void bcmgenet_get_strings(struct } } +static u32 bcmgenet_update_stat_misc(struct bcmgenet_priv *priv, u16 offset) +{ + u16 new_offset; + u32 val; + + switch (offset) { + case UMAC_RBUF_OVFL_CNT_V1: + if (GENET_IS_V2(priv)) + new_offset = RBUF_OVFL_CNT_V2; + else + new_offset = RBUF_OVFL_CNT_V3PLUS; + + val = bcmgenet_rbuf_readl(priv, new_offset); + /* clear if overflowed */ + if (val == ~0) + bcmgenet_rbuf_writel(priv, 0, new_offset); + break; + case UMAC_RBUF_ERR_CNT_V1: + if (GENET_IS_V2(priv)) + new_offset = RBUF_ERR_CNT_V2; + else + new_offset = RBUF_ERR_CNT_V3PLUS; + + val = bcmgenet_rbuf_readl(priv, new_offset); + /* clear if overflowed */ + if (val == ~0) + bcmgenet_rbuf_writel(priv, 0, new_offset); + break; + default: + val = bcmgenet_umac_readl(priv, offset); + /* clear if overflowed */ + if (val == ~0) + bcmgenet_umac_writel(priv, 0, offset); + break; + } + + return val; +} + static void bcmgenet_update_mib_counters(struct bcmgenet_priv *priv) { int i, j = 0; @@ -845,10 +885,16 @@ static void bcmgenet_update_mib_counters UMAC_MIB_START + j + offset); break; case BCMGENET_STAT_MISC: - val = bcmgenet_umac_readl(priv, s->reg_offset); - /* clear if overflowed */ - if (val == ~0) - bcmgenet_umac_writel(priv, 0, s->reg_offset); + if (GENET_IS_V1(priv)) { + val = bcmgenet_umac_readl(priv, s->reg_offset); + /* clear if overflowed */ + if (val == ~0) + bcmgenet_umac_writel(priv, 0, + s->reg_offset); + } else { + val = bcmgenet_update_stat_misc(priv, + s->reg_offset); + } break; } --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014 Broadcom Corporation + * Copyright (c) 2014-2017 Broadcom * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -214,7 +214,9 @@ struct bcmgenet_mib_counters { #define MDIO_REG_SHIFT 16 #define MDIO_REG_MASK 0x1F -#define UMAC_RBUF_OVFL_CNT 0x61C +#define UMAC_RBUF_OVFL_CNT_V1 0x61C +#define RBUF_OVFL_CNT_V2 0x80 +#define RBUF_OVFL_CNT_V3PLUS 0x94 #define UMAC_MPD_CTRL 0x620 #define MPD_EN (1 << 0) @@ -224,7 +226,9 @@ struct bcmgenet_mib_counters { #define UMAC_MPD_PW_MS 0x624 #define UMAC_MPD_PW_LS 0x628 -#define UMAC_RBUF_ERR_CNT 0x634 +#define UMAC_RBUF_ERR_CNT_V1 0x634 +#define RBUF_ERR_CNT_V2 0x84 +#define RBUF_ERR_CNT_V3PLUS 0x98 #define UMAC_MDF_ERR_CNT 0x638 #define UMAC_MDF_CTRL 0x650 #define UMAC_MDF_ADDR 0x654 Patches currently in stable-queue which might be from opendmb(a)gmail.com are queue-4.9/net-bcmgenet-synchronize-irq0-status-between-the-isr-and-task.patch queue-4.9/net-bcmgenet-correct-the-rbuf_ovfl_cnt-and-rbuf_err_cnt-mib-values.patch queue-4.9/net-bcmgenet-correct-mib-access-of-unimac-runt-counters.patch queue-4.9/net-bcmgenet-power-down-internal-phy-if-open-or-resume-fails.patch queue-4.9/net-bcmgenet-reserved-phy-revisions-must-be-checked-first.patch queue-4.9/net-bcmgenet-power-up-the-internal-phy-before-probing-the-mii.patch

7 years, 4 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror