> On Sep 24, 2018, at 8:30 AM, Eric Dumazet wrote:
>
> On Sun, Sep 23, 2018 at 10:04 PM David Miller wrote:
>>
>> Series applied, thanks Eric.
>
> Thanks David.
>
> Song, would you please this additional patch ?
>
> diff --git a/net/core/netpoll.c b/net/core/netpoll.c
> index
> 3219a2932
On Tue, Sep 25, 2018 at 11:25 AM Song Liu wrote:
>
> Hi Michael,
>
> This may not be related. But I am looking at this:
>
> bnxt_poll_work() {
>
> while (1) {
>
> if (rx_pkts == budget)
> return
> }
> }
>
> With budget o
> On Sep 25, 2018, at 7:43 AM, Michael Chan wrote:
>
> On Tue, Sep 25, 2018 at 7:20 AM Eric Dumazet wrote:
>>
>> On Tue, Sep 25, 2018 at 7:02 AM Michael Chan
>> wrote:
>>>
>>> On Mon, Sep 24, 2018 at 2:18 PM Song Liu wrote:
> On Sep 24, 2018, at 2:05 PM, Eric Dumaze
On Tue, Sep 25, 2018 at 7:20 AM Eric Dumazet wrote:
>
> On Tue, Sep 25, 2018 at 7:02 AM Michael Chan
> wrote:
> >
> > On Mon, Sep 24, 2018 at 2:18 PM Song Liu wrote:
> > >
> > >
> > >
> > > > On Sep 24, 2018, at 2:05 PM, Eric Dumazet wrote:
> > > >
> > > >>
> > > >> Interesting, maybe a bnxt s
On Tue, Sep 25, 2018 at 7:02 AM Michael Chan wrote:
>
> On Mon, Sep 24, 2018 at 2:18 PM Song Liu wrote:
> >
> >
> >
> > > On Sep 24, 2018, at 2:05 PM, Eric Dumazet wrote:
> > >
> > >>
> > >> Interesting, maybe a bnxt specific issue.
> > >>
> > >> It seems their model is to process TX/RX notifica
On Mon, Sep 24, 2018 at 2:18 PM Song Liu wrote:
>
>
>
> > On Sep 24, 2018, at 2:05 PM, Eric Dumazet wrote:
> >
> >>
> >> Interesting, maybe a bnxt specific issue.
> >>
> >> It seems their model is to process TX/RX notification in the same queue,
> >> they throw away RX events if budget == 0
> >>
> On Sep 24, 2018, at 2:05 PM, Eric Dumazet wrote:
>
>>
>> Interesting, maybe a bnxt specific issue.
>>
>> It seems their model is to process TX/RX notification in the same queue,
>> they throw away RX events if budget == 0
>>
>> It means commit e7b9569102995ebc26821789628eef45bd9840d8 is w
>
> Interesting, maybe a bnxt specific issue.
>
> It seems their model is to process TX/RX notification in the same queue,
> they throw away RX events if budget == 0
>
> It means commit e7b9569102995ebc26821789628eef45bd9840d8 is wrong and
> must be reverted.
>
> Otherwise, we have a possibility of
On Mon, Sep 24, 2018 at 1:00 PM Song Liu wrote:
>
>
>
> > On Sep 24, 2018, at 12:41 PM, Eric Dumazet wrote:
> >
> > On Mon, Sep 24, 2018 at 12:31 PM Song Liu wrote:
> >
> >> This triggers dev_watchdog() on a simple netperf TCP_RR on bnxt (I haven't
> >> tested other drivers yet).
> >>
> >> I gue
> On Sep 24, 2018, at 12:41 PM, Eric Dumazet wrote:
>
> On Mon, Sep 24, 2018 at 12:31 PM Song Liu wrote:
>
>> This triggers dev_watchdog() on a simple netperf TCP_RR on bnxt (I haven't
>> tested other drivers yet).
>>
>> I guess this is because NAPI_STATE_SCHED is set when poll_one_napi() c
On Mon, Sep 24, 2018 at 12:31 PM Song Liu wrote:
> This triggers dev_watchdog() on a simple netperf TCP_RR on bnxt (I haven't
> tested other drivers yet).
>
> I guess this is because NAPI_STATE_SCHED is set when poll_one_napi() calls
> napi->poll(). And then cleared by napi->poll().
Which part o
> On Sep 24, 2018, at 8:30 AM, Eric Dumazet wrote:
>
> On Sun, Sep 23, 2018 at 10:04 PM David Miller wrote:
>>
>> Series applied, thanks Eric.
>
> Thanks David.
>
> Song, would you please this additional patch ?
>
> diff --git a/net/core/netpoll.c b/net/core/netpoll.c
> index
> 3219a2932
On Sun, Sep 23, 2018 at 10:04 PM David Miller wrote:
>
> Series applied, thanks Eric.
Thanks David.
Song, would you please this additional patch ?
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index
3219a2932463096566ce8ff336ecdf699422dd65..2ad45babe621b2c979ad5496b7df4342e4efbaa6
10064
From: Eric Dumazet
Date: Fri, 21 Sep 2018 15:27:37 -0700
> As diagnosed by Song Liu, ndo_poll_controller() can
> be very dangerous on loaded hosts, since the cpu
> calling ndo_poll_controller() might steal all NAPI
> contexts (for all RX/TX queues of the NIC).
>
> This capture, showing one ksoft
On Sun, Sep 23, 2018 at 12:29 PM David Miller wrote:
>
> From: Eric Dumazet
> Date: Fri, 21 Sep 2018 15:27:37 -0700
>
> > As diagnosed by Song Liu, ndo_poll_controller() can
> > be very dangerous on loaded hosts, since the cpu
> > calling ndo_poll_controller() might steal all NAPI
> > contexts (f
From: Eric Dumazet
Date: Fri, 21 Sep 2018 15:27:37 -0700
> As diagnosed by Song Liu, ndo_poll_controller() can
> be very dangerous on loaded hosts, since the cpu
> calling ndo_poll_controller() might steal all NAPI
> contexts (for all RX/TX queues of the NIC).
>
> This capture, showing one ksoft
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC).
This capture, showing one ksoftirqd eating all cycles
can last for unlimited amount of time, since one
17 matches
Mail list logo
