BMC/Intel-BMC/linux.git/net/ncsi, branch dev-4.3

net/ncsi: Fix possible deadlock caught with lockdep

2016-02-29T01:22:41+00:00

Bringing the interfaces down with ifdown -a directly after boot causes a lockdep warning. It appears that a network interrupt would result in an attempt to acquire that lock again. The solution is to soft disable interrupts. ================================= [ INFO: inconsistent lock state ] 4.3.6 #1 Not tainted --------------------------------- inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage. ip/934 [HC0[0]:SC0[0]:HE1:SE1] takes: (&(&ndp->ndp_package_lock)->rlock){+.?...}, at: [] ncsi_stop_dev+0x14/0x50 {IN-SOFTIRQ-W} state was registered at: [] _raw_spin_lock+0x28/0x38 [] ncsi_add_package+0x64/0xf0 [] ncsi_rsp_handler_sp+0x80/0xe0 [] ncsi_rcv_rsp+0xd4/0x104 [] __netif_receive_skb_core+0x6c4/0x808 [] netif_receive_skb_internal+0xb4/0x138 [] napi_gro_receive+0x48/0x9c [] ftgmac100_poll+0x360/0x59c [] net_rx_action+0xe8/0x2a0 [] __do_softirq+0x108/0x26c [] do_softirq+0x48/0x70 [] __local_bh_enable_ip+0xc8/0x104 [] __dev_queue_xmit+0x654/0x6c4 [] ncsi_xmit_cmd+0x1d4/0x208 [] ncsi_dev_start+0xd0/0x3c0 [] ncsi_dev_work+0x1b8/0x1fc [] process_one_work+0x228/0x3cc [] worker_thread+0x2a4/0x3d8 [] kthread+0xc4/0xd8 [] ret_from_fork+0x14/0x28 irq event stamp: 2009 hardirqs last enabled at (2009): [] __local_bh_enable_ip+0xe4/0x104 hardirqs last disabled at (2007): [] __local_bh_enable_ip+0x64/0x104 softirqs last enabled at (2008): [] dev_deactivate_many+0x270/0x2ac softirqs last disabled at (2006): [] dev_deactivate_many+0x254/0x2ac other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&(&ndp->ndp_package_lock)->rlock); lock(&(&ndp->ndp_package_lock)->rlock); *** DEADLOCK *** 1 lock held by ip/934: #0: (rtnl_mutex){+.+.+.}, at: [] devinet_ioctl+0x15c/0x6c8 stack backtrace: CPU: 0 PID: 934 Comm: ip Not tainted 4.3.6 #1 Hardware name: ASpeed SoC [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [] (show_stack) from [] (print_usage_bug.part.11+0x220/0x288) [] (print_usage_bug.part.11) from [] (mark_lock+0x400/0x678) [] (mark_lock) from [] (__lock_acquire+0xa0c/0x1a9c) [] (__lock_acquire) from [] (lock_acquire+0x9c/0xbc) [] (lock_acquire) from [] (_raw_spin_lock+0x28/0x38) [] (_raw_spin_lock) from [] (ncsi_stop_dev+0x14/0x50) [] (ncsi_stop_dev) from [] (ftgmac100_stop+0x1c/0x28) [] (ftgmac100_stop) from [] (__dev_close_many+0xa0/0xc8) [] (__dev_close_many) from [] (__dev_close+0x20/0x34) [] (__dev_close) from [] (__dev_change_flags+0x8c/0x138) [] (__dev_change_flags) from [] (dev_change_flags+0x18/0x48) [] (dev_change_flags) from [] (devinet_ioctl+0x32c/0x6c8) [] (devinet_ioctl) from [] (sock_ioctl+0x26c/0x2d0) [] (sock_ioctl) from [] (do_vfs_ioctl+0x588/0x67c) [] (do_vfs_ioctl) from [] (SyS_ioctl+0x34/0x5c) [] (SyS_ioctl) from [] (ret_fast_syscall+0x0/0x1c) Signed-off-by: Cyril Bur Signed-off-by: Joel Stanley

ncsi: Fix not releasing lock in ncsi_rcv_rsp()

2016-02-02T02:57:40+00:00

Signed-off-by: Joel Stanley

net/nsci: Remove use of dev_queue_xmit_sk

2015-12-10T03:00:31+00:00

We do not use both arguments, so the code path does not change. Upstream removes the _sk variant in 4.4, so this helps with upstreaming. Signed-off-by: Joel Stanley

net/ncsi: Introduce ncsi_stop_dev()

2015-11-25T04:25:50+00:00

This introduces ncsi_stop_dev() which is to be called when network interrupt is going to be brought down. The NCSI interface won't be disabled and we rely on the driver to stop Rx and Tx path. Signed-off-by: Gavin Shan Signed-off-by: Joel Stanley

net/ncsi: Rename ncsi_stop_dev() to ncsi_suspend_dev()

2015-11-25T04:25:49+00:00

This renames ncsi_stop_dev() to ncsi_suspend_dev(). The NCSI device state ncsi_dev_state_stop_* are renamed to ncsi_dev_state_suspend_* accordingly. No logical changes introduced. Signed-off-by: Gavin Shan Signed-off-by: Joel Stanley

net/ncsi: Disable interrupts when taking ndp->ndp_req_lock

2015-11-25T04:25:47+00:00

The spinlock ndp->ndp_req_lock can be requested in NCSI packet reception path that is done in softIRQ handler. So the external interrupt should be disabled when taking the spinlock. Otherwise, dead-lock could happen: softIRQ handler tries to acquire the lock that has been taken by kernel thread (e.g. workqueue). Signed-off-by: Gavin Shan Signed-off-by: Joel Stanley

net/ncsi: Netlink support

2015-11-18T04:14:19+00:00

This adds netlink support so that userland can retrieve the NCSI topology on the specified interface. Also, more netlink commands are introduced to retrieve information from the active channel or configure the active channel with provided options. Signed-off-by: Gavin Shan

net/ncsi: Manage NCSI device

2015-11-18T04:14:19+00:00

There are multiple packages and channels connecting to NCSI enabled interface. The available packages and channels are probed with NCSI packets when NCSI device is started. Among the available packages and channels, only one channel should be selected to be active to provide service. When the active channel enters to failure state, we need choose another channel as active one to do failover. This implements the state machine to support above tasks. Also, the public interfaces are exposed for NIC driver to manage NCSI device. Signed-off-by: Gavin Shan

net/ncsi: Packet handler

2015-11-18T04:14:18+00:00

NCSI packet encapsulated in ethernet frame is recognized by the ethernet protocol field (0x88F8). According to NCSI spec, the destination MAC field should be 0xFF's, but the source MAC field could be arbitrary one since channel field in NCSI header indicates the source of the packet. There are 3 types of packets depending on the type field in the NCSI packet header: command, response, Asynchronous Event Notification (AEN). The command packets are sent from NCSI enabled interace to the NCSI package or channel for purpose of information retrival or configuration. The response packets, as responses to the received command packets, originate from NCSI package or channel and terminate at NCSI enabled interface. AEN packets are sent from NCSI channel for asynchronous events that might indicate failure detected on the active channel. This implements the functions to handle NCSI packets. Signed-off-by: Gavin Shan

net/ncsi: Resource management

2015-11-18T04:14:17+00:00

According to NCSI spec (DSP0222), the NCSI enabled interface can connected to multiple packages simultaneously, up to 8 packages. Each package includes multiple channels, up to 32 channels. At one moment, one channel is enabled to provide service to the NCSI enabled interface. Besides, each channel comprises capabilities, modes, filters, version and statistics etc. All of them are resources to NCSI protocol stack. At the meanwhile, the NCSI device seen from NIC driver is represented by "struct ncsi_dev", which is expected to populated and started by NIC driver before the NIC can work. All possible NCSI packages and NCSI channels connected to the NCSI interface are tracked from the NCSI device. Also, the NCSI device recognizes active channel that is currently providing service to NCSI enabled interface. Also, the NCSI requests (pairs of command and response) are embedded in NCSI device. This introduces the data structs to represents the NCSI resources mentioned as above. Also, functions used by NCSI stack internally are implemented. Besides, this introduces kernel config option CONFIG_NET_NCSI to enable NCSI stack. Signed-off-by: Gavin Shan