Protected flag is clear in Monitor Mode

Protected flag is clear in Monitor Mode

[Yeoh Chun-Yeow]

Hi, all

I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
monitor mode. I assume that monitor mode will allows the frame to pass
to upper layer without any modification.

But it seems that the protected flag in frame control is clear while I
capture with wireshark. I has verified that the ath9k has not such
issue and the protected flag is set.

So is this a known issue in monitor mode for ath10k?

-----
Chun-Yeow

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Kalle Valo]

Yeoh Chun-Yeow <yeohchunyeow@gmail.com> writes:

> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
> monitor mode. I assume that monitor mode will allows the frame to pass
> to upper layer without any modification.
>
> But it seems that the protected flag in frame control is clear while I
> capture with wireshark. I has verified that the ath9k has not such
> issue and the protected flag is set.
>
> So is this a known issue in monitor mode for ath10k?

I don't recall anyone mentioning anything like this so I would say this
is not a known issue. But on the other hand, I'm notorious in forgetting
stuff :)

What we really need is a public bug tracker for ath10k.

-- 
Kalle Valo

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Yeoh Chun-Yeow]

Do we have the public bug tracker?

Anyone has used the ath10k as monitor interface and facing the same problem?

----
Chun-Yeow

On Fri, May 16, 2014 at 8:45 PM, Kalle Valo <kvalo@qca.qualcomm.com> wrote:
> Yeoh Chun-Yeow <yeohchunyeow@gmail.com> writes:
>
>> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
>> monitor mode. I assume that monitor mode will allows the frame to pass
>> to upper layer without any modification.
>>
>> But it seems that the protected flag in frame control is clear while I
>> capture with wireshark. I has verified that the ath9k has not such
>> issue and the protected flag is set.
>>
>> So is this a known issue in monitor mode for ath10k?
>
> I don't recall anyone mentioning anything like this so I would say this
> is not a known issue. But on the other hand, I'm notorious in forgetting
> stuff :)
>
> What we really need is a public bug tracker for ath10k.
>
> --
> Kalle Valo

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Michal Kazior]

On 16 May 2014 06:38, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
> Hi, all
>
> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
> monitor mode. I assume that monitor mode will allows the frame to pass
> to upper layer without any modification.
>
> But it seems that the protected flag in frame control is clear while I
> capture with wireshark. I has verified that the ath9k has not such
> issue and the protected flag is set.
>
> So is this a known issue in monitor mode for ath10k?

Good catch!

ath10k doesn't really check if frame was decrypted by firmware or not
(there's no explicit descriptor flag for that).

I think we need to rely on the raw frame decap mode as an implicit
indication a frame must not have it's protected flag stripped.


Michał

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Yeoh Chun-Yeow]

On Mon, May 19, 2014 at 5:25 PM, Michal Kazior <michal.kazior@tieto.com> wrote:
> On 16 May 2014 06:38, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
>> Hi, all
>>
>> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
>> monitor mode. I assume that monitor mode will allows the frame to pass
>> to upper layer without any modification.
>>
>> But it seems that the protected flag in frame control is clear while I
>> capture with wireshark. I has verified that the ath9k has not such
>> issue and the protected flag is set.
>>
>> So is this a known issue in monitor mode for ath10k?
>
> Good catch!
>
> ath10k doesn't really check if frame was decrypted by firmware or not
> (there's no explicit descriptor flag for that).
>
> I think we need to rely on the raw frame decap mode as an implicit
> indication a frame must not have it's protected flag stripped.
>

I try to set TARGET_10X_RX_DECAP_MODE to ATH10K_HW_TXRX_RAW. But still
not working.

So we need firmware change for this?

---
Chun-Yeow

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Michal Kazior]

On 19 May 2014 12:43, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
> On Mon, May 19, 2014 at 5:25 PM, Michal Kazior <michal.kazior@tieto.com> wrote:
>> On 16 May 2014 06:38, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
>>> Hi, all
>>>
>>> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
>>> monitor mode. I assume that monitor mode will allows the frame to pass
>>> to upper layer without any modification.
>>>
>>> But it seems that the protected flag in frame control is clear while I
>>> capture with wireshark. I has verified that the ath9k has not such
>>> issue and the protected flag is set.
>>>
>>> So is this a known issue in monitor mode for ath10k?
>>
>> Good catch!
>>
>> ath10k doesn't really check if frame was decrypted by firmware or not
>> (there's no explicit descriptor flag for that).
>>
>> I think we need to rely on the raw frame decap mode as an implicit
>> indication a frame must not have it's protected flag stripped.
>>
>
> I try to set TARGET_10X_RX_DECAP_MODE to ATH10K_HW_TXRX_RAW. But still
> not working.
>
> So we need firmware change for this?

There's absolutely no need to *set* the rx decap mode to raw.

What I meant was to look at decap mode in htt rx descriptor of each
frame when deciding whether to strip the protected flag or not. I've
made a patch but I need to test a few things before sending it out.


Michał

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Adrian Chadd]

Hi,

Just create an ath10k project in github under the qca group so you
have a place to stick issues and things like firmware documentation.



-a


On 16 May 2014 05:45, Kalle Valo <kvalo@qca.qualcomm.com> wrote:
> Yeoh Chun-Yeow <yeohchunyeow@gmail.com> writes:
>
>> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
>> monitor mode. I assume that monitor mode will allows the frame to pass
>> to upper layer without any modification.
>>
>> But it seems that the protected flag in frame control is clear while I
>> capture with wireshark. I has verified that the ath9k has not such
>> issue and the protected flag is set.
>>
>> So is this a known issue in monitor mode for ath10k?
>
> I don't recall anyone mentioning anything like this so I would say this
> is not a known issue. But on the other hand, I'm notorious in forgetting
> stuff :)
>
> What we really need is a public bug tracker for ath10k.
>
> --
> Kalle Valo
>
> _______________________________________________
> ath10k mailing list
> ath10k@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/ath10k

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Protected flag is clear in Monitor Mode

[Yeoh Chun-Yeow]

Ok got it.

----
Chun-Yeow

On Mon, May 19, 2014 at 7:06 PM, Michal Kazior <michal.kazior@tieto.com> wrote:
> On 19 May 2014 12:43, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
>> On Mon, May 19, 2014 at 5:25 PM, Michal Kazior <michal.kazior@tieto.com> wrote:
>>> On 16 May 2014 06:38, Yeoh Chun-Yeow <yeohchunyeow@gmail.com> wrote:
>>>> Hi, all
>>>>
>>>> I try to use ath10k (firmware version 10.1.467.2-1_ as standalone
>>>> monitor mode. I assume that monitor mode will allows the frame to pass
>>>> to upper layer without any modification.
>>>>
>>>> But it seems that the protected flag in frame control is clear while I
>>>> capture with wireshark. I has verified that the ath9k has not such
>>>> issue and the protected flag is set.
>>>>
>>>> So is this a known issue in monitor mode for ath10k?
>>>
>>> Good catch!
>>>
>>> ath10k doesn't really check if frame was decrypted by firmware or not
>>> (there's no explicit descriptor flag for that).
>>>
>>> I think we need to rely on the raw frame decap mode as an implicit
>>> indication a frame must not have it's protected flag stripped.
>>>
>>
>> I try to set TARGET_10X_RX_DECAP_MODE to ATH10K_HW_TXRX_RAW. But still
>> not working.
>>
>> So we need firmware change for this?
>
> There's absolutely no need to *set* the rx decap mode to raw.
>
> What I meant was to look at decap mode in htt rx descriptor of each
> frame when deciding whether to strip the protected flag or not. I've
> made a patch but I need to test a few things before sending it out.
>
>
> Michał

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k