[PATCH] mac80211: fix sw crypto

[PATCH] mac80211: fix sw crypto

[Johannes Berg]

What a stupid mistake. In

    commit 813d76694043d00b59475baa1fbfaf54a2eb7fad
    Author: Johannes Berg <johannes@sipsolutions.net>
    Date:   Sun Jan 17 01:47:58 2010 +0100

        mac80211: move control.hw_key assignment

I inserted code testing the wrong flags field,
which means that the test is almost always true
(it's really testing for the peer's WMM support)
and thus the later parts of the stack assume hw
crypto will be done even if that's not true.

Obviously, that broke software crypto. Maxim
said so specifically, and Jochen probably uses
some cipher that iwl3945 doesn't support in
hardware, which might also explain that Maxim
reports that even hw crypto is broken.

Fix this to test the right flags field.

Reported-by: Maxim Levitsky <maximlevitsky@gmail.com>
Reported-by: Jochen Friedrich <jochen@scram.de>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
 net/mac80211/tx.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- wireless-testing.orig/net/mac80211/tx.c	2010-01-25 11:13:59.000000000 +0100
+++ wireless-testing/net/mac80211/tx.c	2010-01-25 11:29:24.000000000 +0100
@@ -559,7 +559,7 @@ ieee80211_tx_h_select_key(struct ieee802
 		}
 
 		if (!skip_hw && tx->key &&
-		    tx->key->conf.flags & KEY_FLAG_UPLOADED_TO_HARDWARE)
+		    tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE)
 			info->control.hw_key = &tx->key->conf;
 	}
 

Re: [PATCH] mac80211: fix sw crypto

[Maxim Levitsky]

On Mon, 2010-01-25 at 11:36 +0100, Johannes Berg wrote: 
> What a stupid mistake. In
> 
>     commit 813d76694043d00b59475baa1fbfaf54a2eb7fad
>     Author: Johannes Berg <johannes@sipsolutions.net>
>     Date:   Sun Jan 17 01:47:58 2010 +0100
> 
>         mac80211: move control.hw_key assignment
> 
> I inserted code testing the wrong flags field,
> which means that the test is almost always true
> (it's really testing for the peer's WMM support)
> and thus the later parts of the stack assume hw
> crypto will be done even if that's not true.
> 
> Obviously, that broke software crypto. Maxim
> said so specifically, and Jochen probably uses
> some cipher that iwl3945 doesn't support in
> hardware, which might also explain that Maxim
> reports that even hw crypto is broken.
This patch works for me. Thanks.

What chipers does iwl3945 support?
I use CCMP, and I also know that software encryption is used by default.

Best regards,
Maxim Levitsky 
> 
> Fix this to test the right flags field.
> 
> Reported-by: Maxim Levitsky <maximlevitsky@gmail.com>
> Reported-by: Jochen Friedrich <jochen@scram.de>
> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> ---
>  net/mac80211/tx.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> --- wireless-testing.orig/net/mac80211/tx.c	2010-01-25 11:13:59.000000000 +0100
> +++ wireless-testing/net/mac80211/tx.c	2010-01-25 11:29:24.000000000 +0100
> @@ -559,7 +559,7 @@ ieee80211_tx_h_select_key(struct ieee802
>  		}
>  
>  		if (!skip_hw && tx->key &&
> -		    tx->key->conf.flags & KEY_FLAG_UPLOADED_TO_HARDWARE)
> +		    tx->key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE)
>  			info->control.hw_key = &tx->key->conf;
>  	}
>  
> 
> 

Re: [PATCH] mac80211: fix sw crypto

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 592 bytes --]

On Mon, 2010-01-25 at 17:21 +0200, Maxim Levitsky wrote:

> > Obviously, that broke software crypto. Maxim
> > said so specifically, and Jochen probably uses
> > some cipher that iwl3945 doesn't support in
> > hardware, which might also explain that Maxim
> > reports that even hw crypto is broken.
> This patch works for me. Thanks.

Thanks for testing.

> What chipers does iwl3945 support?
> I use CCMP, and I also know that software encryption is used by default.

I don't know off-hand, but I'm pretty sure it doesn't do TKIP, or at
least the driver doesn't.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 801 bytes --]

Re: [PATCH] mac80211: fix sw crypto

[Fabio Rossi]

On Monday 25 January 2010 11:36:16 Johannes Berg wrote:

> What a stupid mistake. In
> 
>     commit 813d76694043d00b59475baa1fbfaf54a2eb7fad
>     Author: Johannes Berg <johannes@sipsolutions.net>
>     Date:   Sun Jan 17 01:47:58 2010 +0100
> 
>         mac80211: move control.hw_key assignment
> 
> I inserted code testing the wrong flags field,
> which means that the test is almost always true
> (it's really testing for the peer's WMM support)
> and thus the later parts of the stack assume hw
> crypto will be done even if that's not true.
> 
> Obviously, that broke software crypto. Maxim
> said so specifically, and Jochen probably uses
> some cipher that iwl3945 doesn't support in
> hardware, which might also explain that Maxim
> reports that even hw crypto is broken.
> 
> Fix this to test the right flags field.
> 

Tested-by: Fabio Rossi <rossi.f@inwind.it>

Re: [PATCH] mac80211: fix sw crypto

[Jochen Friedrich]

Fabio Rossi wrote:
> On Monday 25 January 2010 11:36:16 Johannes Berg wrote:
> 
>> What a stupid mistake. In
>>
>>     commit 813d76694043d00b59475baa1fbfaf54a2eb7fad
>>     Author: Johannes Berg <johannes@sipsolutions.net>
>>     Date:   Sun Jan 17 01:47:58 2010 +0100
>>
>>         mac80211: move control.hw_key assignment
>>
>> I inserted code testing the wrong flags field,
>> which means that the test is almost always true
>> (it's really testing for the peer's WMM support)
>> and thus the later parts of the stack assume hw
>> crypto will be done even if that's not true.
>>
>> Obviously, that broke software crypto. Maxim
>> said so specifically, and Jochen probably uses
>> some cipher that iwl3945 doesn't support in
>> hardware, which might also explain that Maxim
>> reports that even hw crypto is broken.
>>
>> Fix this to test the right flags field.
>>
> 
> Tested-by: Fabio Rossi <rossi.f@inwind.it>

Tested-by: Jochen Friedrich <jochen@scram.de>

Thanks,
Jochen