> -Original Message-
> From: Robert Shearman [mailto:rshea...@brocade.com]
> Sent: Friday, December 2, 2016 5:54 AM
> To: Alexander Duyck
> Cc: David Miller ; Netdev ;
> Duyck, Alexander H
> Subject: Re: [PATCH net] fib_trie: Avoid expensive update of suffix length
On 01/12/16 18:29, Alexander Duyck wrote:
On Wed, Nov 30, 2016 at 4:27 PM, Robert Shearman wrote:
On 29/11/16 23:14, Alexander Duyck wrote:
On Tue, Nov 29, 2016 at 9:46 AM, Robert Shearman
Fixes: 5405afd1a306 ("fib_trie: Add tracking value for suffix length")
Signed-off-by: Robert Shearman
On Wed, Nov 30, 2016 at 4:27 PM, Robert Shearman wrote:
> On 29/11/16 23:14, Alexander Duyck wrote:
>>
>> On Tue, Nov 29, 2016 at 9:46 AM, Robert Shearman
>> wrote:
>>>
>>> With certain distributions of routes it can take a long time to add
>>> and delete further routes at scale. For example, wit
On 29/11/16 23:14, Alexander Duyck wrote:
On Tue, Nov 29, 2016 at 9:46 AM, Robert Shearman wrote:
With certain distributions of routes it can take a long time to add
and delete further routes at scale. For example, with a route for
10.37.96.0/20 present it takes 47s to add ~200k contiguous /24
On Tue, Nov 29, 2016 at 9:46 AM, Robert Shearman wrote:
> With certain distributions of routes it can take a long time to add
> and delete further routes at scale. For example, with a route for
> 10.37.96.0/20 present it takes 47s to add ~200k contiguous /24 routes
> from 8.0.0.0/24 through to 11.
With certain distributions of routes it can take a long time to add
and delete further routes at scale. For example, with a route for
10.37.96.0/20 present it takes 47s to add ~200k contiguous /24 routes
from 8.0.0.0/24 through to 11.138.207.0/24. Perf shows the bottleneck
is update_suffix:
