linux/drivers/hv
Dexuan Cui 37c2578c0c Drivers: hv: vmbus: Offload the handling of channels to two workqueues
vmbus_process_offer() mustn't call channel->sc_creation_callback()
directly for sub-channels, because sc_creation_callback() ->
vmbus_open() may never get the host's response to the
OPEN_CHANNEL message (the host may rescind a channel at any time,
e.g. in the case of hot removing a NIC), and vmbus_onoffer_rescind()
may not wake up the vmbus_open() as it's blocked due to a non-zero
vmbus_connection.offer_in_progress, and finally we have a deadlock.

The above is also true for primary channels, if the related device
drivers use sync probing mode by default.

And, usually the handling of primary channels and sub-channels can
depend on each other, so we should offload them to different
workqueues to avoid possible deadlock, e.g. in sync-probing mode,
NIC1's netvsc_subchan_work() can race with NIC2's netvsc_probe() ->
rtnl_lock(), and causes deadlock: the former gets the rtnl_lock
and waits for all the sub-channels to appear, but the latter
can't get the rtnl_lock and this blocks the handling of sub-channels.

The patch can fix the multiple-NIC deadlock described above for
v3.x kernels (e.g. RHEL 7.x) which don't support async-probing
of devices, and v4.4, v4.9, v4.14 and v4.18 which support async-probing
but don't enable async-probing for Hyper-V drivers (yet).

The patch can also fix the hang issue in sub-channel's handling described
above for all versions of kernels, including v4.19 and v4.20-rc4.

So actually the patch should be applied to all the existing kernels,
not only the kernels that have 8195b1396e.

Fixes: 8195b1396e ("hv_netvsc: fix deadlock on hotplug")
Cc: stable@vger.kernel.org
Cc: Stephen Hemminger <sthemmin@microsoft.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-03 08:01:01 +01:00
..
channel_mgmt.c Drivers: hv: vmbus: Offload the handling of channels to two workqueues 2018-12-03 08:01:01 +01:00
channel.c Drivers: hv: vmbus: check the creation_status in vmbus_establish_gpadl() 2018-11-26 20:33:56 +01:00
connection.c Drivers: hv: vmbus: Offload the handling of channels to two workqueues 2018-12-03 08:01:01 +01:00
hv_balloon.c hv_balloon: Replace spin_is_locked() with lockdep 2018-10-15 20:54:17 +02:00
hv_fcopy.c Drivers: hv: fcopy: restore correct transfer length 2017-09-22 10:29:54 +02:00
hv_kvp.c Drivers: hv: kvp: Fix the recent regression caused by incorrect clean-up 2018-11-11 09:00:20 -08:00
hv_snapshot.c Merge 4.11-rc4 into char-misc-next 2017-03-27 09:13:04 +02:00
hv_trace_balloon.h hv_balloon: trace post_status 2018-03-06 09:57:17 -08:00
hv_trace.c hv: add SPDX license to trace 2018-03-28 13:24:56 +02:00
hv_trace.h hv: add SPDX license to trace 2018-03-28 13:24:56 +02:00
hv_util.c use the new async probing feature for the hyperv drivers 2018-07-03 13:02:28 +02:00
hv_utils_transport.c vfs: do bulk POLL* -> EPOLL* replacement 2018-02-11 14:34:03 -08:00
hv_utils_transport.h Drivers: hv: util: move waiting for release to hv_utils_transport itself 2017-03-16 16:42:00 +09:00
hv.c Drivers: hv: vmbus: Fix synic per-cpu context initialization 2018-09-12 09:46:46 +02:00
hyperv_vmbus.h Drivers: hv: vmbus: Offload the handling of channels to two workqueues 2018-12-03 08:01:01 +01:00
Kconfig hv: add SPDX license id to Kconfig 2018-03-28 13:24:56 +02:00
Makefile hv_balloon: trace post_status 2018-03-06 09:57:17 -08:00
ring_buffer.c vmbus: split ring buffer allocation from open 2018-09-25 20:33:24 +02:00
vmbus_drv.c vmbus: pass channel to hv_process_channel_removal 2018-09-25 20:33:24 +02:00