Windows networking bottleneck - all socket notifications coming from 1 thread inside winsock?

understood

I was a little worried if I played with it I might go into a spiral of
BSOD doom.

But it didn’t.

Seems to work fine, but not much difference. Setting the value to
anything other than 0 (need to verify 0) seems to allow interrupts to
use core 3 as well as core 1. It’s just a single hex core with HT, so a
lot of the NUMA stuff / multiprocessor stuff doesn’t apply.

But it’s certainly behaving differently now, and I’m not really pegging
core 0 any more. It definitely seems to prefer it, but seems more
willing to offload work to core 3 when it’s getting busy, and overall
tps has gone up.

Maybe time to buy a real network card and plug it in for testing.

Thanks

Adrien

------ Original Message ------
From: “Tim Roberts”
To: “Windows System Software Devs Interest List”
Sent: 9/05/2017 1:43:42 PM
Subject: Re: [ntdev] Windows networking bottleneck - all socket
notifications coming from 1 thread inside winsock?

>Adrien de Croy wrote:
>>
>> I looked in the registry under the Enum\PCI key for the device, and
>> there are keys like “interrupt management” and “affinity policy”… I
>> wonder if a bit of playing with some of those attributes may help.
>
>It is possible that dinking with the “affinity policy” might force the
>system to spread the interrupts to other processors.
>
>HOWEVER, the authors of the driver would not have forced an affinity
>policy without a good reason. To be more specific, it may be that all
>of their interrupts are forced to CPU 0 because their driver cannot
>handle multiple simultaneous interrupts in multiple processors.
>
>–
>Tim Roberts, xxxxx@probo.com
>Providenza & Boekelheide, Inc.
>
>
>—
>NTDEV is sponsored by OSR
>
>Visit the list online at:
>http:
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:</http:></http:></http:>

sorry that should read core 0 and 2

Adrien

------ Original Message ------
From: “Adrien de Croy”
To: “Windows System Software Devs Interest List”
Sent: 9/05/2017 4:42:20 PM
Subject: Re: [ntdev] Windows networking bottleneck - all socket
notifications coming from 1 thread inside winsock?

>understood
>
>I was a little worried if I played with it I might go into a spiral of
>BSOD doom.
>
>But it didn’t.
>
>Seems to work fine, but not much difference. Setting the value to
>anything other than 0 (need to verify 0) seems to allow interrupts to
>use core 3 as well as core 1. It’s just a single hex core with HT, so a
>lot of the NUMA stuff / multiprocessor stuff doesn’t apply.
>
>But it’s certainly behaving differently now, and I’m not really pegging
>core 0 any more. It definitely seems to prefer it, but seems more
>willing to offload work to core 3 when it’s getting busy, and overall
>tps has gone up.
>
>Maybe time to buy a real network card and plug it in for testing.
>
>Thanks
>
>Adrien
>
>------ Original Message ------
>From: “Tim Roberts”
>To: “Windows System Software Devs Interest List”
>Sent: 9/05/2017 1:43:42 PM
>Subject: Re: [ntdev] Windows networking bottleneck - all socket
>notifications coming from 1 thread inside winsock?
>
>>Adrien de Croy wrote:
>>>
>>> I looked in the registry under the Enum\PCI key for the device, and
>>> there are keys like “interrupt management” and “affinity policy”…
>>>I
>>> wonder if a bit of playing with some of those attributes may help.
>>
>>It is possible that dinking with the “affinity policy” might force the
>>system to spread the interrupts to other processors.
>>
>>HOWEVER, the authors of the driver would not have forced an affinity
>>policy without a good reason. To be more specific, it may be that all
>>of their interrupts are forced to CPU 0 because their driver cannot
>>handle multiple simultaneous interrupts in multiple processors.
>>
>>–
>>Tim Roberts, xxxxx@probo.com
>>Providenza & Boekelheide, Inc.
>>
>>
>>—
>>NTDEV is sponsored by OSR
>>
>>Visit the list online at:
>>http:
>>
>>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>>software drivers!
>>Details at http:
>>
>>To unsubscribe, visit the List Server section of OSR Online at
>>http:
>
>
>—
>NTDEV is sponsored by OSR
>
>Visit the list online at:
>http:
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:</http:></http:></http:></http:></http:></http:>

A quick look at the Intel documentation for this chip suggests that it might not support RSS fully. Try testing with a known IP performance tool and see what results you get

Sent from Mailhttps: for Windows 10

From: Adrien de Croymailto:xxxxx
Sent: May 9, 2017 12:45 AM
To: Windows System Software Devs Interest Listmailto:xxxxx
Subject: Re: [ntdev] Windows networking bottleneck - all socket notifications coming from 1 thread inside winsock?

sorry that should read core 0 and 2

Adrien

------ Original Message ------
From: “Adrien de Croy”
To: “Windows System Software Devs Interest List”
Sent: 9/05/2017 4:42:20 PM
Subject: Re: [ntdev] Windows networking bottleneck - all socket
notifications coming from 1 thread inside winsock?

>understood
>
>I was a little worried if I played with it I might go into a spiral of
>BSOD doom.
>
>But it didn’t.
>
>Seems to work fine, but not much difference. Setting the value to
>anything other than 0 (need to verify 0) seems to allow interrupts to
>use core 3 as well as core 1. It’s just a single hex core with HT, so a
>lot of the NUMA stuff / multiprocessor stuff doesn’t apply.
>
>But it’s certainly behaving differently now, and I’m not really pegging
>core 0 any more. It definitely seems to prefer it, but seems more
>willing to offload work to core 3 when it’s getting busy, and overall
>tps has gone up.
>
>Maybe time to buy a real network card and plug it in for testing.
>
>Thanks
>
>Adrien
>
>------ Original Message ------
>From: “Tim Roberts”
>To: “Windows System Software Devs Interest List”
>Sent: 9/05/2017 1:43:42 PM
>Subject: Re: [ntdev] Windows networking bottleneck - all socket
>notifications coming from 1 thread inside winsock?
>
>>Adrien de Croy wrote:
>>>
>>> I looked in the registry under the Enum\PCI key for the device, and
>>> there are keys like “interrupt management” and “affinity policy”…
>>>I
>>> wonder if a bit of playing with some of those attributes may help.
>>
>>It is possible that dinking with the “affinity policy” might force the
>>system to spread the interrupts to other processors.
>>
>>HOWEVER, the authors of the driver would not have forced an affinity
>>policy without a good reason. To be more specific, it may be that all
>>of their interrupts are forced to CPU 0 because their driver cannot
>>handle multiple simultaneous interrupts in multiple processors.
>>
>>–
>>Tim Roberts, xxxxx@probo.com
>>Providenza & Boekelheide, Inc.
>>
>>
>>—
>>NTDEV is sponsored by OSR
>>
>>Visit the list online at:
>>http:
>>
>>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>>software drivers!
>>Details at http:
>>
>>To unsubscribe, visit the List Server section of OSR Online at
>>http:
>
>
>—
>NTDEV is sponsored by OSR
>
>Visit the list online at:
>http:
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:


NTDEV is sponsored by OSR

Visit the list online at: http:

MONTHLY seminars on crash dump analysis, WDF, Windows internals and software drivers!
Details at http:

To unsubscribe, visit the List Server section of OSR Online at http:</http:></http:></http:></http:></http:></http:></http:></http:></http:></mailto:xxxxx></mailto:xxxxx></https:>

thanks

yeah good call, might try with some other network hardware so different
drivers.

Adrien

------ Original Message ------
From: “Marion Bond”
To: “Windows System Software Devs Interest List”
Sent: 10/05/2017 10:24:50 AM
Subject: RE: [ntdev] Windows networking bottleneck - all socket
notifications coming from 1 thread inside winsock?

>A quick look at the Intel documentation for this chip suggests that it
>might not support RSS fully. Try testing with a known IP performance
>tool and see what results you get
>
>
>
>Sent from Mail https: for
>Windows 10
>
>
>
>From: Adrien de Croy mailto:xxxxx
>Sent: May 9, 2017 12:45 AM
>To: Windows System Software Devs Interest List
>mailto:xxxxx
>Subject: Re: [ntdev] Windows networking bottleneck - all socket
>notifications coming from 1 thread inside winsock?
>
>
>
>
>sorry that should read core 0 and 2
>
>Adrien
>
>
>------ Original Message ------
>From: “Adrien de Croy”
>To: “Windows System Software Devs Interest List”
>Sent: 9/05/2017 4:42:20 PM
>Subject: Re: [ntdev] Windows networking bottleneck - all socket
>notifications coming from 1 thread inside winsock?
>
> >understood
> >
> >I was a little worried if I played with it I might go into a spiral of
> >BSOD doom.
> >
> >But it didn’t.
> >
> >Seems to work fine, but not much difference. Setting the value to
> >anything other than 0 (need to verify 0) seems to allow interrupts to
> >use core 3 as well as core 1. It’s just a single hex core with HT, so
>a
> >lot of the NUMA stuff / multiprocessor stuff doesn’t apply.
> >
> >But it’s certainly behaving differently now, and I’m not really
>pegging
> >core 0 any more. It definitely seems to prefer it, but seems more
> >willing to offload work to core 3 when it’s getting busy, and overall
> >tps has gone up.
> >
> >Maybe time to buy a real network card and plug it in for testing.
> >
> >Thanks
> >
> >Adrien
> >
> >------ Original Message ------
> >From: “Tim Roberts”
> >To: “Windows System Software Devs Interest List”
> >Sent: 9/05/2017 1:43:42 PM
> >Subject: Re: [ntdev] Windows networking bottleneck - all socket
> >notifications coming from 1 thread inside winsock?
> >
> >>Adrien de Croy wrote:
> >>>
> >>> I looked in the registry under the Enum\PCI key for the device,
>and
> >>> there are keys like “interrupt management” and “affinity policy”…
> >>>I
> >>> wonder if a bit of playing with some of those attributes may help.
> >>
> >>It is possible that dinking with the “affinity policy” might force
>the
> >>system to spread the interrupts to other processors.
> >>
> >>HOWEVER, the authors of the driver would not have forced an affinity
> >>policy without a good reason. To be more specific, it may be that
>all
> >>of their interrupts are forced to CPU 0 because their driver cannot
> >>handle multiple simultaneous interrupts in multiple processors.
> >>
> >>–
> >>Tim Roberts, xxxxx@probo.com
> >>Providenza & Boekelheide, Inc.
> >>
> >>
> >>—
> >>NTDEV is sponsored by OSR
> >>
> >>Visit the list online at:
> >>http:
> >>
> >>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
> >>software drivers!
> >>Details at http:
> >>
> >>To unsubscribe, visit the List Server section of OSR Online at
> >>http:
> >
> >
> >—
> >NTDEV is sponsored by OSR
> >
> >Visit the list online at:
> >http:
> >
> >MONTHLY seminars on crash dump analysis, WDF, Windows internals and
> >software drivers!
> >Details at http:
> >
> >To unsubscribe, visit the List Server section of OSR Online at
> >http:
>
>
>—
>NTDEV is sponsored by OSR
>
>Visit the list online at:
>http:
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:
>
>
>
>
>—
>NTDEV is sponsored by OSR
>
>Visit the list online at:
>http:
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:</http:></http:></http:></http:></http:></http:></http:></http:></http:></http:></http:></http:></mailto:xxxxx></mailto:xxxxx></https:>