* Packet injection with ath9k
@ 2010-05-07 16:13 Roberto Riggio
2010-05-07 16:35 ` Gábor Stefanik
0 siblings, 1 reply; 6+ messages in thread
From: Roberto Riggio @ 2010-05-07 16:13 UTC (permalink / raw)
To: linux-wireless
Hi,
I'm writing an application to inject traffic over a wireless interface.
This app
is working fine on an x86 machine. However if i compile the same app for
an arm platform, no frame are sent over the wireless interface (ath9k).
I'm guessing that this is because of some alignment issues but i cannot
track
the piece of code that is actually parsing the frame. I've found the
__ieee80211_parse_tx_radiotap in net/mac80211/tx.c function, but it is
not called when i try to inject some traffic, so the frame are dropped
before that.
Any hints?
Thanks
R.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Packet injection with ath9k
2010-05-07 16:13 Packet injection with ath9k Roberto Riggio
@ 2010-05-07 16:35 ` Gábor Stefanik
2010-05-07 19:09 ` Roberto Riggio
0 siblings, 1 reply; 6+ messages in thread
From: Gábor Stefanik @ 2010-05-07 16:35 UTC (permalink / raw)
To: Roberto Riggio; +Cc: linux-wireless
Hi!
Are you sure it is not your injector that is having alignment issues?
AFAIK the radiotap parser explicitly uses endianness-aware function
everywhere.
On Fri, May 7, 2010 at 6:13 PM, Roberto Riggio
<roberto.riggio@create-net.org> wrote:
> Hi,
>
> I'm writing an application to inject traffic over a wireless interface. This
> app
> is working fine on an x86 machine. However if i compile the same app for
> an arm platform, no frame are sent over the wireless interface (ath9k).
>
> I'm guessing that this is because of some alignment issues but i cannot
> track
> the piece of code that is actually parsing the frame. I've found the
> __ieee80211_parse_tx_radiotap in net/mac80211/tx.c function, but it is
> not called when i try to inject some traffic, so the frame are dropped
> before that.
>
> Any hints?
>
> Thanks
> R.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
Vista: [V]iruses, [I]ntruders, [S]pyware, [T]rojans and [A]dware. :-)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Packet injection with ath9k
2010-05-07 16:35 ` Gábor Stefanik
@ 2010-05-07 19:09 ` Roberto Riggio
2010-05-09 20:24 ` Gábor Stefanik
0 siblings, 1 reply; 6+ messages in thread
From: Roberto Riggio @ 2010-05-07 19:09 UTC (permalink / raw)
To: Gábor Stefanik; +Cc: linux-wireless
Hi,
well, that is what i wanted to understand. These are the struct
that I'm using to compose the rediotap header:
struct ieee80211_radiotap_header {
u_int8_t it_version;
u_int8_t it_pad;
u_int16_t it_len;
u_int32_t it_present;
} __attribute__((__packed__));
struct click_radiotap_header {
struct ieee80211_radiotap_header wt_ihdr;
u_int8_t wt_rate;
u_int8_t wt_txpower;
u_int8_t wt_rts_retries;
u_int8_t wt_data_retries;
};
The flags are set in order to take into account the fields that I
specify. But i do not know at which point the frame are
dropped.
R.
On 05/07/2010 06:35 PM, Gábor Stefanik wrote:
> Hi!
> Are you sure it is not your injector that is having alignment issues?
> AFAIK the radiotap parser explicitly uses endianness-aware function
> everywhere.
>
> On Fri, May 7, 2010 at 6:13 PM, Roberto Riggio
> <roberto.riggio@create-net.org> wrote:
>
>> Hi,
>>
>> I'm writing an application to inject traffic over a wireless interface. This
>> app
>> is working fine on an x86 machine. However if i compile the same app for
>> an arm platform, no frame are sent over the wireless interface (ath9k).
>>
>> I'm guessing that this is because of some alignment issues but i cannot
>> track
>> the piece of code that is actually parsing the frame. I've found the
>> __ieee80211_parse_tx_radiotap in net/mac80211/tx.c function, but it is
>> not called when i try to inject some traffic, so the frame are dropped
>> before that.
>>
>> Any hints?
>>
>> Thanks
>> R.
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>
>>
>
>
>
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: Packet injection with ath9k
2010-05-07 19:09 ` Roberto Riggio
@ 2010-05-09 20:24 ` Gábor Stefanik
2010-05-12 16:42 ` Roberto Riggio
0 siblings, 1 reply; 6+ messages in thread
From: Gábor Stefanik @ 2010-05-09 20:24 UTC (permalink / raw)
To: Roberto Riggio; +Cc: linux-wireless
2010/5/7 Roberto Riggio <roberto.riggio@create-net.org>:
> Hi,
>
> well, that is what i wanted to understand. These are the struct
> that I'm using to compose the rediotap header:
>
> struct ieee80211_radiotap_header {
> u_int8_t it_version;
> u_int8_t it_pad;
> u_int16_t it_len;
> u_int32_t it_present;
AFAIK these 2 fields need to be little-endian even on big-endian
machines. If your system is big-endian, this can cause problems.
> } __attribute__((__packed__));
>
> struct click_radiotap_header {
> struct ieee80211_radiotap_header wt_ihdr;
> u_int8_t wt_rate;
> u_int8_t wt_txpower;
> u_int8_t wt_rts_retries;
> u_int8_t wt_data_retries;
> };
>
> The flags are set in order to take into account the fields that I
> specify. But i do not know at which point the frame are
> dropped.
>
> R.
>
> On 05/07/2010 06:35 PM, Gábor Stefanik wrote:
>>
>> Hi!
>> Are you sure it is not your injector that is having alignment issues?
>> AFAIK the radiotap parser explicitly uses endianness-aware function
>> everywhere.
>>
>> On Fri, May 7, 2010 at 6:13 PM, Roberto Riggio
>> <roberto.riggio@create-net.org> wrote:
>>
>>>
>>> Hi,
>>>
>>> I'm writing an application to inject traffic over a wireless interface.
>>> This
>>> app
>>> is working fine on an x86 machine. However if i compile the same app for
>>> an arm platform, no frame are sent over the wireless interface (ath9k).
>>>
>>> I'm guessing that this is because of some alignment issues but i cannot
>>> track
>>> the piece of code that is actually parsing the frame. I've found the
>>> __ieee80211_parse_tx_radiotap in net/mac80211/tx.c function, but it is
>>> not called when i try to inject some traffic, so the frame are dropped
>>> before that.
>>>
>>> Any hints?
>>>
>>> Thanks
>>> R.
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-wireless"
>>> in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>
>>>
>>
>>
>>
>
>
--
Vista: [V]iruses, [I]ntruders, [S]pyware, [T]rojans and [A]dware. :-)
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: Packet injection with ath9k
2010-05-09 20:24 ` Gábor Stefanik
@ 2010-05-12 16:42 ` Roberto Riggio
2010-05-12 16:46 ` Johannes Berg
0 siblings, 1 reply; 6+ messages in thread
From: Roberto Riggio @ 2010-05-12 16:42 UTC (permalink / raw)
To: Gábor Stefanik; +Cc: linux-wireless
On 05/09/2010 10:24 PM, Gábor Stefanik wrote:
>> u_int16_t it_len;
>> u_int32_t it_present;
>>
> AFAIK these 2 fields need to be little-endian even on big-endian
> machines. If your system is big-endian, this can cause problems.
>
Thanks, that was indeed the issue. I totally missed this info
in the radiotap website. Everything is working fine now.
R.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Packet injection with ath9k
2010-05-12 16:42 ` Roberto Riggio
@ 2010-05-12 16:46 ` Johannes Berg
0 siblings, 0 replies; 6+ messages in thread
From: Johannes Berg @ 2010-05-12 16:46 UTC (permalink / raw)
To: Roberto Riggio; +Cc: Gábor Stefanik, linux-wireless
On Wed, 2010-05-12 at 18:42 +0200, Roberto Riggio wrote:
> On 05/09/2010 10:24 PM, Gábor Stefanik wrote:
> >> u_int16_t it_len;
> >> u_int32_t it_present;
> >>
> > AFAIK these 2 fields need to be little-endian even on big-endian
> > machines. If your system is big-endian, this can cause problems.
> >
>
> Thanks, that was indeed the issue. I totally missed this info
> in the radiotap website. Everything is working fine now.
I'd make it more prominent, but ... this is what it says now:
Important Radiotap Characteristics
* Fields are strictly ordered; The developer can specify any
combination of fields, but the data must appear following the
radiotap header in the order they are specified in the
it_present bitmask (or more accurately, in the order the bit
numbers for the it_present bitmask are defined).
* Data is specified in little endian byte-order, all data fields
including the it_version, it_len and it_present fields in the
radiotap header are to be specified in little endian byte-order.
This wiki has adopted the Linux convention of using __le64,
__le32 and __le16 for 64-, 32- and 16-bit little endian
quantities.
johannes
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2010-05-12 16:46 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-05-07 16:13 Packet injection with ath9k Roberto Riggio
2010-05-07 16:35 ` Gábor Stefanik
2010-05-07 19:09 ` Roberto Riggio
2010-05-09 20:24 ` Gábor Stefanik
2010-05-12 16:42 ` Roberto Riggio
2010-05-12 16:46 ` Johannes Berg
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox