[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a16f2e4d-45fe-47b7-23fc-af4f506ac059@cogentembedded.com>
Date: Fri, 18 May 2018 20:28:05 +0300
From: Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
To: Pavel Tatashin <pasha.tatashin@...cle.com>,
Jeff Kirsher <jeffrey.t.kirsher@...el.com>, davem@...emloft.net
Cc: netdev@...r.kernel.org, nhorman@...hat.com, sassmann@...hat.com,
jogreene@...hat.com
Subject: Re: [net-next 3/6] ixgbe: release lock for the duration of
ixgbe_suspend_close()
On 05/18/2018 02:37 PM, Pavel Tatashin wrote:
> * parallelized this function, so drop lock for the
>>
>> Parallelizing? Else the sentence doesn't parse for me. :-)
My comment hardly makes sense when you removed all the context...
> Hi Sergei,
>
> In a separate series I parallelized device_shutdown(), see:
> http://lkml.kernel.org/r/20180516024004.28977-1-pasha.tatashin@oracle.com
>
> But, this particular patch should be dropped, as discussed in this thread:
> http://lkml.kernel.org/r/20180503035931.22439-2-pasha.tatashin@oracle.com
>
>
> Alexander Duyck, made a point that a generic RTNL scalability fix should be done. This particular patch might introduce a race, since it relies on assumption that RTNL is not needed in this place because ixgbe_close() does not have it, but Alexander Duyck, says that the callers of ixgbe_close() are assumed to own this lock.
My comment was about the English grammar only. :-)
> Thank you,
> Pavel
MBR, Sergei
Powered by blists - more mailing lists