Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-11-11 Thread Brian King
On 10/27/2016 10:26 AM, Eric Dumazet wrote: > On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: >> We recently encountered a bug where a few customers using ibmveth on the >> same LPAR hit an issue where a TCP session hung when large receive was >> enabled. Closer analysis revealed that the

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-11-09 Thread Brian King
On 11/06/2016 03:22 PM, Jonathan Maxwell wrote: > On Thu, Nov 3, 2016 at 8:40 AM, Brian King wrote: >> On 10/27/2016 10:26 AM, Eric Dumazet wrote: >>> On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: We recently encountered a bug where a few customers using

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-11-06 Thread Jonathan Maxwell
On Thu, Nov 3, 2016 at 8:40 AM, Brian King wrote: > On 10/27/2016 10:26 AM, Eric Dumazet wrote: >> On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: >>> We recently encountered a bug where a few customers using ibmveth on the >>> same LPAR hit an issue where a TCP

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-11-02 Thread Brian King
On 10/27/2016 10:26 AM, Eric Dumazet wrote: > On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: >> We recently encountered a bug where a few customers using ibmveth on the >> same LPAR hit an issue where a TCP session hung when large receive was >> enabled. Closer analysis revealed that the

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-29 Thread Jonathan Maxwell
On Fri, Oct 28, 2016 at 5:08 AM, Eric Dumazet wrote: > On Thu, 2016-10-27 at 12:54 -0500, Thomas Falcon wrote: >> On 10/27/2016 10:26 AM, Eric Dumazet wrote: >> > On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: >> >> We recently encountered a bug where a few

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-27 Thread Eric Dumazet
On Thu, 2016-10-27 at 12:54 -0500, Thomas Falcon wrote: > On 10/27/2016 10:26 AM, Eric Dumazet wrote: > > On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: > >> We recently encountered a bug where a few customers using ibmveth on the > >> same LPAR hit an issue where a TCP session hung when

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-27 Thread Thomas Falcon
On 10/27/2016 10:26 AM, Eric Dumazet wrote: > On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: >> We recently encountered a bug where a few customers using ibmveth on the >> same LPAR hit an issue where a TCP session hung when large receive was >> enabled. Closer analysis revealed that the

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-27 Thread Eric Dumazet
On Wed, 2016-10-26 at 11:09 +1100, Jon Maxwell wrote: > We recently encountered a bug where a few customers using ibmveth on the > same LPAR hit an issue where a TCP session hung when large receive was > enabled. Closer analysis revealed that the session was stuck because the > one side was

Re: [PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-27 Thread Thomas Falcon
On 10/25/2016 07:09 PM, Jon Maxwell wrote: > We recently encountered a bug where a few customers using ibmveth on the > same LPAR hit an issue where a TCP session hung when large receive was > enabled. Closer analysis revealed that the session was stuck because the > one side was advertising a

[PATCH net-next] ibmveth: v1 calculate correct gso_size and set gso_type

2016-10-25 Thread Jon Maxwell
We recently encountered a bug where a few customers using ibmveth on the same LPAR hit an issue where a TCP session hung when large receive was enabled. Closer analysis revealed that the session was stuck because the one side was advertising a zero window repeatedly. We narrowed this down to