[linux-azure]Pull Request: Fixes for lp:1807757

[Joseph Salisbury]

On 1/10/19 6:52 AM, Kleber Souza wrote:
> On 12/19/18 4:54 PM, Joseph Salisbury wrote:
>> BugLink: https://bugs.launchpad.net/bugs/1807757
>>
>> == SRU Justification ==
>> Microsoft has identified a multiple-NIC-deadlock issue on Hyper-V VMs.
>> This bug is fixed by the seven commits in this pull request.
>>
>> These commits fix a regression introduced by mainline commit 8195b1396ec8.
>>
>> These commits are also needed in Cosmic and Bionic.  Cosmic only requires
>> two of these commits and Bionic requres one addition commit.  Due to this,
>> their SRU requests will be sent separate.
>>
>> == Regression Potential ==
>> Low, limited to hyperv.
>>
>> == Test Case ==
>> Microsoft ran tests with these commits are confirm they resolve the bug
>> and do not introduce any regressions.
>>
>>
>> The following changes since commit 71779dccb6cf812ba546ffead73d6adcf676d9c9:
>>
>>   UBUNTU: Ubuntu-azure-4.15.0-1032.33~16.04.1 (2018-11-09 18:59:04 -0200)
>>
>> are available in the Git repository at:
>>
>>   kernel.ubuntu.com:/srv/kernel.ubuntu.com/git/jsalisbury/bugs/lp1807757/linux-azure/linux-azure.git 
>>
>> for you to fetch changes up to 3eff5f1124b788f7bf096164d57fdd5401a5db24:
>>
>>   Drivers: hv: vmbus: Offload the handling of channels to two workqueues (2018-12-19 10:29:16 -0500)
> Hi Joe,
>
> This PR was sent with the patches on top of xenial/linux-azure, which is
> a backport of bionic/linux-azure. We generally apply the patches on top
> of the bionic tree and those patches will be included on the xenial tree
> once we rebase it. Unless there's a reason not to apply the changes on
> the bionic one.
>
> Could you please clarify if those changes are supposed to be applied to
> all linux-azure based on 4.15 or if there's a reason to keep them only
> on xenial?
There is no reason to only apply them to Xenial.  The changes should be
applied to linux-azure based on 4.15.
>
>
> Thanks,
>
> Kleber
>
>
>> ----------------------------------------------------------------
>> Dexuan Cui (4):
>>       Drivers: hv: vmbus: Fix the offer_in_progress in vmbus_process_offer()
>>       hv_netvsc: Fix a deadlock by getting rtnl lock earlier in netvsc_probe()
>>       Drivers: hv: vmbus: check the creation_status in vmbus_establish_gpadl()
>>       Drivers: hv: vmbus: Offload the handling of channels to two workqueues
>>
>> Stephen Hemminger (2):
>>       hv_netvsc: fix network namespace issues with VF support
>>       hv_netvsc: split sub-channel setup into async and sync
>>
>>  drivers/hv/channel.c              |   8 ++
>>  drivers/hv/channel_mgmt.c         | 194 +++++++++++++++++++++++++-------------
>>  drivers/hv/connection.c           |  24 ++++-
>>  drivers/hv/hyperv_vmbus.h         |   7 ++
>>  drivers/net/hyperv/hyperv_net.h   |   4 +-
>>  drivers/net/hyperv/netvsc.c       |  37 +++++++-
>>  drivers/net/hyperv/netvsc_drv.c   |  69 +++++++++-----
>>  drivers/net/hyperv/rndis_filter.c |  61 +++---------
>>  include/linux/hyperv.h            |   7 ++
>>  9 files changed, 267 insertions(+), 144 deletions(-)
>>
>