LinuxLists.cc - [PATCH v2] d80211: Add software sequence support

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Fri, 16 Feb 2007 13:25:47 +0100, Ivo van Doorn wrote:
> On Friday 16 February 2007 13:03, Jiri Benc wrote:
> > On Thu, 15 Feb 2007 21:57:22 +0100, Jiri Benc wrote:
> > > Thinking about it more, sdata should be the correct place. It's more
> > > per-interface thing than per-BSS (although the difference is small in
> > > this particular case).
> >
> > ...and thinking about it even more, that won't work with WDS.
>
> You mean storing the sequence counter per-interface won't work,
> or per-bss?

Both.

When you have a WDS link and some other wireless interface at the same
time, their source addresses are (in most cases) the same. You want to
have the sequence control counter shared by them. Actually, it will
probably work most of time even when they have different counters. The
situation where it won't work is if you have two virtual interfaces,
one in STA and second in WDS mode, both connected to the same AP. I
can't figure out how this particular setup could be useful but it's
generally not disallowed.

Anyway, using different counter for a WDS interface probably violates
802.11 (although it's not much specific in this case). I believe it
will work in most cases, though.

Any ideas how to solve that?

Thanks,

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 19:09:32

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, 15 Feb 2007 17:42:04 +0100, Johannes Berg wrote:
> Can this be a library function instead? Our TX path is already cluttered
> enough...

The advantage of non-library approach in this case is that you don't
need to search for a BSS the frame belongs to. Ivo's patch searches for
it but I think it's possible to remember it from the past.

If this feature is implemented as a library function you need to search
for a BSS in any case. That's way more overhead than one condition in
the tx path.

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 20:48:49

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Wednesday 14 February 2007 08:23, Ivo van Doorn wrote:
> diff --git a/net/d80211/ieee80211_i.h b/net/d80211/ieee80211_i.h
> index 5ab9eb5..1891e05 100644
> --- a/net/d80211/ieee80211_i.h
> +++ b/net/d80211/ieee80211_i.h
> @@ -80,6 +80,7 @@ struct ieee80211_sta_bss {
> u8 ssid[IEEE80211_MAX_SSID_LEN];
> size_t ssid_len;
> u16 capability; /* host byte order */
> + u16 sequence;
> int hw_mode;
> int channel;
> int freq;
I am not so sure that storing this information in struct ieee80211_sta_bss is
correct. As far as I can tell, that is strictly for storing information scan
results and beacon information. Nothing outside of ieee80211_sta.c uses it.
struct sta_info or struct ieee80211_if_ap (sta interfaces have a pointer to
the struct ieee80211_if_ap on the master device) may be a better choice. I
think struct sta_info is the right one though there is a problem - it isn't
allocated until a successful association.

-Michael Wu

Attachments:

(No filename) (955.00 B)
(No filename) (189.00 B)
Download all attachments

2007-02-16 12:26:21

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Friday 16 February 2007 13:03, Jiri Benc wrote:
> On Thu, 15 Feb 2007 21:57:22 +0100, Jiri Benc wrote:
> > Thinking about it more, sdata should be the correct place. It's more
> > per-interface thing than per-BSS (although the difference is small in
> > this particular case).
>
> ...and thinking about it even more, that won't work with WDS.

You mean storing the sequence counter per-interface won't work,
or per-bss?

Ivo

2007-02-15 22:18:05

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

> > I am not so sure that storing this information in struct ieee80211_sta_bss
> is
> > correct. As far as I can tell, that is strictly for storing information
> scan
> > results and beacon information. Nothing outside of ieee80211_sta.c uses
> it.
> > struct sta_info or struct ieee80211_if_ap (sta interfaces have a pointer
> to
> > the struct ieee80211_if_ap on the master device) may be a better choice. I
> > think struct sta_info is the right one though there is a problem - it
> isn't
> > allocated until a successful association.
>
> Thinking about it more, sdata should be the correct place. It's more
> per-interface thing than per-BSS (although the difference is small in
> this particular case).

That would also optimize the sequence handler since the BSS
does not have to be looked up anymore. I'll create the patch to do this asap.

Ivo

2007-02-16 22:07:51

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Friday 16 February 2007 22:14, Johannes Berg wrote:
> On Fri, 2007-02-16 at 13:58 +0100, Jiri Benc wrote:
>
> > Both.
> >
> > When you have a WDS link and some other wireless interface at the same
> > time, their source addresses are (in most cases) the same. You want to
> > have the sequence control counter shared by them. Actually, it will
> > probably work most of time even when they have different counters. The
> > situation where it won't work is if you have two virtual interfaces,
> > one in STA and second in WDS mode, both connected to the same AP. I
> > can't figure out how this particular setup could be useful but it's
> > generally not disallowed.
> >
> > Anyway, using different counter for a WDS interface probably violates
> > 802.11 (although it's not much specific in this case). I believe it
> > will work in most cases, though.
> >
> > Any ideas how to solve that?
>
> Make the counter per-mac address?

Wouldn't that be the same as per-interface?

Ivo

2007-02-16 12:03:44

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, 15 Feb 2007 21:57:22 +0100, Jiri Benc wrote:
> Thinking about it more, sdata should be the correct place. It's more
> per-interface thing than per-BSS (although the difference is small in
> this particular case).

...and thinking about it even more, that won't work with WDS.

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 19:22:35

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thursday 15 February 2007 14:02, Michael Wu wrote:
> If we're going to switch to a library approach, let's do it all at once. I
> don't want some features requiring flags to toggle and others requiring a
> function to call.
>
To be specific, I don't like the library approach when we can simply set a
flag and let the stack handle it. It may clutter the TX path, but it reduces
complexity on the driver end. Also, this particular test can be moved from
the TX path to device registration with some changes to the way the TX
handlers are handled. I will need to check if there are other TX handlers
that would benefit from this to make it worthwhile. For things like
encryption, I think library functions or callbacks will really be needed to
do things right, but for software sequence numbers, I'd rather stick with a
simple flag to enable/disable it. It also allows us to report if a particular
device supports software sequence numbers which the use of a library function
would not specify so directly.

-Michael Wu

Attachments:

(No filename) (1.01 kB)
(No filename) (189.00 B)
Download all attachments

2007-02-15 18:54:28

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

Hi,

> > Then the driver can just call it whenever it needs the sequence number
> > for a frame. It'll need to be a bit smart about fragmentation but I
> > think that's ok.
> >
> > I may be radical, but I think that a lot of things hardware can do
> > belong into library functions the driver can use instead of cluttering
> > the stack with it.
>
> I'm inclined to agree, FWIW...

Ok, I'll change it. New patch will arrive in a few days.

Ivo

2007-02-16 14:29:56

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Friday 16 February 2007 13:58, Jiri Benc wrote:
> On Fri, 16 Feb 2007 13:25:47 +0100, Ivo van Doorn wrote:
> > On Friday 16 February 2007 13:03, Jiri Benc wrote:
> > > On Thu, 15 Feb 2007 21:57:22 +0100, Jiri Benc wrote:
> > > > Thinking about it more, sdata should be the correct place. It's more
> > > > per-interface thing than per-BSS (although the difference is small in
> > > > this particular case).
> > >
> > > ...and thinking about it even more, that won't work with WDS.
> >
> > You mean storing the sequence counter per-interface won't work,
> > or per-bss?
>
> Both.
>
> When you have a WDS link and some other wireless interface at the same
> time, their source addresses are (in most cases) the same. You want to
> have the sequence control counter shared by them. Actually, it will
> probably work most of time even when they have different counters. The
> situation where it won't work is if you have two virtual interfaces,
> one in STA and second in WDS mode, both connected to the same AP. I
> can't figure out how this particular setup could be useful but it's
> generally not disallowed.
>
> Anyway, using different counter for a WDS interface probably violates
> 802.11 (although it's not much specific in this case). I believe it
> will work in most cases, though.
>
> Any ideas how to solve that?

Are there more situations where it might be important to know
if the WDS interface has to keep in mind that there is a virtual STA interface
on the same network? Because it might be worthwhile to make a check for that
in the creation of the WDS interface (and in the STA interface).

Ivo

2007-02-19 19:40:10

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Fri, 16 Feb 2007 23:12:06 +0100, Johannes Berg wrote:
> On Fri, 2007-02-16 at 23:07 +0100, Ivo van Doorn wrote:
> > On Friday 16 February 2007 22:14, Johannes Berg wrote:
> > > Make the counter per-mac address?
> >
> > Wouldn't that be the same as per-interface?
>
> Don't WDS interfaces have the same MAC as the corresponding AP
> interface?

They do.

Per-MAC address counters would work but it's another overhead we need
to add (allocate a new reference-counted structure, put a pointer to it
into sdata, correctly deal with it when MAC address changes).

I think we can live with the counter in sdata. It's not perfect but it
should work for all useful cases. The only situation where it won't
work I have been able to imagine is the case I described (STA and WDS
with the same address) and this particular setup is currently being
denied by d80211 (and it's not allowed by 802.11, btw).

If we consider this as a problem later, we can always change to per-MAC
address counters or something as it's completely internal to the stack
and no driver changes would be required.

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 20:57:25

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, 15 Feb 2007 15:48:20 -0500, Michael Wu wrote:
> I am not so sure that storing this information in struct ieee80211_sta_bss is
> correct. As far as I can tell, that is strictly for storing information scan
> results and beacon information. Nothing outside of ieee80211_sta.c uses it.
> struct sta_info or struct ieee80211_if_ap (sta interfaces have a pointer to
> the struct ieee80211_if_ap on the master device) may be a better choice. I
> think struct sta_info is the right one though there is a problem - it isn't
> allocated until a successful association.

Thinking about it more, sdata should be the correct place. It's more
per-interface thing than per-BSS (although the difference is small in
this particular case).

Thanks,

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 18:08:10

by John W. Linville

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, Feb 15, 2007 at 05:42:04PM +0100, Johannes Berg wrote:

> Then the driver can just call it whenever it needs the sequence number
> for a frame. It'll need to be a bit smart about fragmentation but I
> think that's ok.
>
> I may be radical, but I think that a lot of things hardware can do
> belong into library functions the driver can use instead of cluttering
> the stack with it.

I'm inclined to agree, FWIW...

--
John W. Linville
[email protected]

2007-02-16 22:12:17

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Fri, 2007-02-16 at 23:07 +0100, Ivo van Doorn wrote:
> On Friday 16 February 2007 22:14, Johannes Berg wrote:
> > On Fri, 2007-02-16 at 13:58 +0100, Jiri Benc wrote:
> >
> > > Both.
> > >
> > > When you have a WDS link and some other wireless interface at the same
> > > time, their source addresses are (in most cases) the same. You want to
> > > have the sequence control counter shared by them. Actually, it will
> > > probably work most of time even when they have different counters. The
> > > situation where it won't work is if you have two virtual interfaces,
> > > one in STA and second in WDS mode, both connected to the same AP. I
> > > can't figure out how this particular setup could be useful but it's
> > > generally not disallowed.
> > >
> > > Anyway, using different counter for a WDS interface probably violates
> > > 802.11 (although it's not much specific in this case). I believe it
> > > will work in most cases, though.
> > >
> > > Any ideas how to solve that?
> >
> > Make the counter per-mac address?
>
> Wouldn't that be the same as per-interface?

Don't WDS interfaces have the same MAC as the corresponding AP
interface?

johannes

Attachments:

signature.asc (190.00 B)
This is a digitally signed message part

2007-02-16 21:14:52

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Fri, 2007-02-16 at 13:58 +0100, Jiri Benc wrote:

> Both.
>
> When you have a WDS link and some other wireless interface at the same
> time, their source addresses are (in most cases) the same. You want to
> have the sequence control counter shared by them. Actually, it will
> probably work most of time even when they have different counters. The
> situation where it won't work is if you have two virtual interfaces,
> one in STA and second in WDS mode, both connected to the same AP. I
> can't figure out how this particular setup could be useful but it's
> generally not disallowed.
>
> Anyway, using different counter for a WDS interface probably violates
> 802.11 (although it's not much specific in this case). I believe it
> will work in most cases, though.
>
> Any ideas how to solve that?

Make the counter per-mac address?

johannes

Attachments:

signature.asc (190.00 B)
This is a digitally signed message part

2007-02-15 16:42:13

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Wed, 2007-02-14 at 14:23 +0100, Ivo van Doorn wrote:
> Most hardware can keep track of sequence numbers themselves,
> unfortunately *most* doesn't cover all devices. ;)
> This patch will keep track of the (per-bss) sequence number
> if the flag IEEE80211_HW_SOFTWARE_SEQUENCE has been set.

Can this be a library function instead? Our TX path is already cluttered
enough...

> +static void ieee80211_include_sequence(struct net_device *dev,
> + struct ieee80211_hdr *hdr)
> +{
> + struct ieee80211_sta_bss *bss;
> + u8 *bssid = ieee80211_get_bssid(hdr, sizeof(*hdr));
> +
> + bss = ieee80211_rx_bss_get(dev, bssid);
> + if (!bss)
> + return;
> +
> + /*
> + * Set the sequence number for this frame.
> + */
> + hdr->seq_ctrl = cpu_to_le16(bss->sequence & IEEE80211_SCTL_SEQ);
> +
> + /*
> + * Increase the sequence number.
> + */
> + bss->sequence = (bss->sequence + 0x10) & IEEE80211_SCTL_SEQ;
> +
> + ieee80211_rx_bss_put(dev, bss);
> +}

This could be exported getting an skb instead of the ieee80211_hdr.

Then the driver can just call it whenever it needs the sequence number
for a frame. It'll need to be a bit smart about fragmentation but I
think that's ok.

I may be radical, but I think that a lot of things hardware can do
belong into library functions the driver can use instead of cluttering
the stack with it.

johannes

Attachments:

signature.asc (190.00 B)
This is a digitally signed message part

2007-02-15 19:03:19

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thursday 15 February 2007 11:42, Johannes Berg wrote:
> On Wed, 2007-02-14 at 14:23 +0100, Ivo van Doorn wrote:
> > Most hardware can keep track of sequence numbers themselves,
> > unfortunately *most* doesn't cover all devices. ;)
> > This patch will keep track of the (per-bss) sequence number
> > if the flag IEEE80211_HW_SOFTWARE_SEQUENCE has been set.
>
> Can this be a library function instead? Our TX path is already cluttered
> enough...
>
If we're going to switch to a library approach, let's do it all at once. I
don't want some features requiring flags to toggle and others requiring a
function to call.

-Michael Wu

Attachments:

(No filename) (633.00 B)
(No filename) (189.00 B)
Download all attachments

2007-02-19 19:53:07

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Mon, 2007-02-19 at 20:40 +0100, Jiri Benc wrote:

> Per-MAC address counters would work but it's another overhead we need
> to add (allocate a new reference-counted structure, put a pointer to it
> into sdata, correctly deal with it when MAC address changes).

Yup, I agree. Put it into drivers :P

> I think we can live with the counter in sdata. It's not perfect but it
> should work for all useful cases. The only situation where it won't
> work I have been able to imagine is the case I described (STA and WDS
> with the same address) and this particular setup is currently being
> denied by d80211 (and it's not allowed by 802.11, btw).
>
> If we consider this as a problem later, we can always change to per-MAC
> address counters or something as it's completely internal to the stack
> and no driver changes would be required.

Ok. I'm fine with putting it into the stack. I'll take Michael by the
word though that he'll do dynamically registered tx handlers later ;)

johannes

Attachments:

signature.asc (190.00 B)
This is a digitally signed message part

2007-02-15 19:11:09

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, 15 Feb 2007 14:02:43 -0500, Michael Wu wrote:
> If we're going to switch to a library approach, let's do it all at once. I
> don't want some features requiring flags to toggle and others requiring a
> function to call.

Entirely library approach is not possible. So we're going to end up
with mixed environment in any case.

Jiri

--
Jiri Benc
SUSE Labs

2007-02-15 20:04:57

by Jouni Malinen

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Thu, Feb 15, 2007 at 02:22:07PM -0500, Michael Wu wrote:

> To be specific, I don't like the library approach when we can simply set a
> flag and let the stack handle it.

In general, I would prefer not to use library approach for things like
this sequence number support. In this case, there may not even be need
for that flag if the operation itself is fast enough. In other words, I
don't think it would cause problems for other drivers if the seq# were
set regardless of whether it is neede or not. Just doing this without
the flag may be the cleanest way and not likely to use much more CPU
than verification of a flag before doing this..

--
Jouni Malinen PGP id EFC895FA

2007-03-10 19:33:48

by Michael Büsch

[permalink] [raw]

Subject: Re: [PATCH v3] d80211: Add software sequence support

On Saturday 10 March 2007 20:21, Ivo van Doorn wrote:
> Most hardware can keep track of sequence numbers themselves,
> unfortunately *most* doesn't cover all devices. ;)
> This patch will keep track of the (per-bss) sequence number
> if the flag IEEE80211_HW_SOFTWARE_SEQUENCE has been set.

The question has been asked (by Jouni), but I didn't see
a good answer (maybe I missed it).
Why is this a flag at all? The sequence counter is such a
trivial thing that having an extra conditional is just _more_
overhead. Why not simply calculate the seq all the time and
let hardware override it, if it has the capability to do so?
Calculation of the seq is cheap. No need to bloat code.

> This 3rd version of the patch has the sequence number stored
> in the ieee80211_sub_if_data structure.
>
> Signed-off-by: Ivo van Doorn <[email protected]>
>
> ---
>
> diff --git a/include/net/mac80211.h b/include/net/mac80211.h
> index 82ea43f..e32e28a 100644
> --- a/include/net/mac80211.h
> +++ b/include/net/mac80211.h
> @@ -515,6 +515,9 @@ struct ieee80211_hw {
> * normal operation. */
> #define IEEE80211_HW_MONITOR_DURING_OPER (1<<9)
>
> + /* Does the HOST need to insert the frame sequence counter */
> +#define IEEE80211_HW_SOFTWARE_SEQUENCE (1<<10)
> +
> /* please fill this gap when adding new flags */
>
> /* calculate Michael MIC for an MSDU when doing hwcrypto */
> diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
> index 577dbe3..79c2a60 100644
> --- a/net/mac80211/ieee80211.c
> +++ b/net/mac80211/ieee80211.c
> @@ -54,6 +54,20 @@ static u8 * ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len);
> static int ieee80211_mgmt_start_xmit(struct sk_buff *skb,
> struct net_device *dev);
>
> +static void ieee80211_include_sequence(struct ieee80211_sub_if_data *sdata,
> + struct ieee80211_hdr *hdr)
> +{
> + /*
> + * Set the sequence number for this frame.
> + */
> + hdr->seq_ctrl = cpu_to_le16(sdata->sequence & IEEE80211_SCTL_SEQ);
> +
> + /*
> + * Increase the sequence number.
> + */
> + sdata->sequence = (sdata->sequence + 0x10) & IEEE80211_SCTL_SEQ;
> +}
> +
> struct ieee80211_key_conf *
> ieee80211_key_data2conf(struct ieee80211_local *local,
> const struct ieee80211_key *data)
> @@ -445,6 +459,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> size_t hdrlen, per_fragm, num_fragm, payload_len, left;
> struct sk_buff **frags, *first, *frag;
> int i;
> + u16 seq;
> u8 *pos;
> int frag_threshold = tx->local->fragmentation_threshold;
>
> @@ -463,6 +478,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> goto fail;
>
> hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREFRAGS);
> + seq = le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ;
> pos = first->data + hdrlen + per_fragm;
> left = payload_len - per_fragm;
> for (i = 0; i < num_fragm - 1; i++) {
> @@ -490,7 +506,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> memcpy(fhdr, first->data, hdrlen);
> if (i == num_fragm - 2)
> fhdr->frame_control &= cpu_to_le16(~IEEE80211_FCTL_MOREFRAGS);
> - fhdr->seq_ctrl = cpu_to_le16(i + 1);
> + fhdr->seq_ctrl = cpu_to_le16(seq + ((i + 1) & IEEE80211_SCTL_FRAG));
> copylen = left > per_fragm ? per_fragm : left;
> memcpy(skb_put(frag, copylen), pos, copylen);
>
> @@ -900,6 +916,17 @@ ieee80211_tx_h_check_assoc(struct ieee80211_txrx_data *tx)
> return TXRX_CONTINUE;
> }
>
> +static ieee80211_txrx_result
> +ieee80211_tx_h_sequence(struct ieee80211_txrx_data *tx)
> +{
> + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
> +
> + if ((tx->local->hw.flags & IEEE80211_HW_SOFTWARE_SEQUENCE) &&
> + ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_control)) >= 24)
> + ieee80211_include_sequence(tx->sdata, hdr);
> +
> + return TXRX_CONTINUE;
> +}
>
> /* This function is called whenever the AP is about to exceed the maximum limit
> * of buffered frames for power saving STAs. This situation should not really
> @@ -1770,6 +1797,10 @@ struct sk_buff * ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
> skb_reserve(skb, local->hw.extra_tx_headroom);
> memcpy(skb_put(skb, bh_len), b_head, bh_len);
>
> + if (hw->flags & IEEE80211_HW_SOFTWARE_SEQUENCE)
> + ieee80211_include_sequence(sdata,
> + (struct ieee80211_hdr *)skb->data);
> +
> ieee80211_beacon_add_tim(local, ap, skb);
>
> if (b_tail) {
> @@ -4381,6 +4412,7 @@ static ieee80211_rx_handler ieee80211_rx_handlers[] =
> static ieee80211_tx_handler ieee80211_tx_handlers[] =
> {
> ieee80211_tx_h_check_assoc,
> + ieee80211_tx_h_sequence,
> ieee80211_tx_h_ps_buf,
> ieee80211_tx_h_select_key,
> ieee80211_tx_h_michael_mic_add,
> diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
> index 9df8ef0..4b7c427 100644
> --- a/net/mac80211/ieee80211_i.h
> +++ b/net/mac80211/ieee80211_i.h
> @@ -317,6 +317,8 @@ struct ieee80211_sub_if_data {
> int ieee802_1x; /* IEEE 802.1X PAE - drop packet to/from unauthorized
> * port */
>
> + u16 sequence;
> +
> /* Fragment table for host-based reassembly */
> struct ieee80211_fragment_entry fragments[IEEE80211_FRAGMENT_MAX];
> unsigned int fragment_next;
> -
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
>

--
Greetings Michael.

2007-03-23 17:52:50

[permalink] [raw]

Subject: Re: [PATCH v3] d80211: Add software sequence support

On Sun, 11 Mar 2007 00:25:30 +0100, Ivo van Doorn wrote:
> +static void ieee80211_include_sequence(struct ieee80211_sub_if_data *sdata,
> + struct ieee80211_hdr *hdr)

I'd prefer specifying this as inline function.

> +{
> + /*
> + * Set the sequence number for this frame.
> + */
> + hdr->seq_ctrl = cpu_to_le16(sdata->sequence & IEEE80211_SCTL_SEQ);

No need to do "& IEEE80211_SCTL_SEQ" here.

> +
> + /*
> + * Increase the sequence number.
> + */

Please reduce this and previous comment to one line each.

> + sdata->sequence = (sdata->sequence + 0x10) & IEEE80211_SCTL_SEQ;
> +}
> +
> struct ieee80211_key_conf *
> ieee80211_key_data2conf(struct ieee80211_local *local,
> const struct ieee80211_key *data)
> @@ -445,6 +459,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> size_t hdrlen, per_fragm, num_fragm, payload_len, left;
> struct sk_buff **frags, *first, *frag;
> int i;
> + u16 seq;
> u8 *pos;
> int frag_threshold = tx->local->fragmentation_threshold;
>
> @@ -463,6 +478,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> goto fail;
>
> hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREFRAGS);
> + seq = le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ;
> pos = first->data + hdrlen + per_fragm;
> left = payload_len - per_fragm;
> for (i = 0; i < num_fragm - 1; i++) {
> @@ -490,7 +506,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
> memcpy(fhdr, first->data, hdrlen);
> if (i == num_fragm - 2)
> fhdr->frame_control &= cpu_to_le16(~IEEE80211_FCTL_MOREFRAGS);
> - fhdr->seq_ctrl = cpu_to_le16(i + 1);
> + fhdr->seq_ctrl = cpu_to_le16(seq + ((i + 1) & IEEE80211_SCTL_FRAG));

I'd prefer | instead of + here.

Thanks,

Jiri

--
Jiri Benc
SUSE Labs

2007-03-10 14:39:12

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

> > If we consider this as a problem later, we can always change to per-MAC
> > address counters or something as it's completely internal to the stack
> > and no driver changes would be required.
>
> Ok. I'm fine with putting it into the stack. I'll take Michael by the
> word though that he'll do dynamically registered tx handlers later ;)

So just to be clear, this patch does not need any tweaking anymore right?
Latest patch series for rt2x00 accidently had software sequence handling
removed out of rt2400pci and rt2500pci so it is actually broken and at the moment.
So I need to either restore it back into the rt2x00 drivers or the mac80211 stack needs
to get the support so rt2x00 can use it. :)

Ivo

2007-03-10 23:25:46

[permalink] [raw]

Subject: Re: [PATCH v3] d80211: Add software sequence support

On Saturday 10 March 2007 20:32, Michael Buesch wrote:
> On Saturday 10 March 2007 20:21, Ivo van Doorn wrote:
> > Most hardware can keep track of sequence numbers themselves,
> > unfortunately *most* doesn't cover all devices. ;)
> > This patch will keep track of the (per-bss) sequence number
> > if the flag IEEE80211_HW_SOFTWARE_SEQUENCE has been set.
>
> The question has been asked (by Jouni), but I didn't see
> a good answer (maybe I missed it).
> Why is this a flag at all? The sequence counter is such a
> trivial thing that having an extra conditional is just _more_
> overhead. Why not simply calculate the seq all the time and
> let hardware override it, if it has the capability to do so?
> Calculation of the seq is cheap. No need to bloat code.

It would also safe me a patch for rt2400pci and rt2500pci as well. ;)

Well if nobody objects against setting the sequence number
always, then below patch would be sufficient. :)

Signed-off-by: Ivo van Doorn <[email protected]>

---

diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 577dbe3..05a0929 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -54,6 +54,20 @@ static u8 * ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len);
static int ieee80211_mgmt_start_xmit(struct sk_buff *skb,
struct net_device *dev);

+static void ieee80211_include_sequence(struct ieee80211_sub_if_data *sdata,
+ struct ieee80211_hdr *hdr)
+{
+ /*
+ * Set the sequence number for this frame.
+ */
+ hdr->seq_ctrl = cpu_to_le16(sdata->sequence & IEEE80211_SCTL_SEQ);
+
+ /*
+ * Increase the sequence number.
+ */
+ sdata->sequence = (sdata->sequence + 0x10) & IEEE80211_SCTL_SEQ;
+}
+
struct ieee80211_key_conf *
ieee80211_key_data2conf(struct ieee80211_local *local,
const struct ieee80211_key *data)
@@ -445,6 +459,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
size_t hdrlen, per_fragm, num_fragm, payload_len, left;
struct sk_buff **frags, *first, *frag;
int i;
+ u16 seq;
u8 *pos;
int frag_threshold = tx->local->fragmentation_threshold;

@@ -463,6 +478,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
goto fail;

hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREFRAGS);
+ seq = le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ;
pos = first->data + hdrlen + per_fragm;
left = payload_len - per_fragm;
for (i = 0; i < num_fragm - 1; i++) {
@@ -490,7 +506,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
memcpy(fhdr, first->data, hdrlen);
if (i == num_fragm - 2)
fhdr->frame_control &= cpu_to_le16(~IEEE80211_FCTL_MOREFRAGS);
- fhdr->seq_ctrl = cpu_to_le16(i + 1);
+ fhdr->seq_ctrl = cpu_to_le16(seq + ((i + 1) & IEEE80211_SCTL_FRAG));
copylen = left > per_fragm ? per_fragm : left;
memcpy(skb_put(frag, copylen), pos, copylen);

@@ -900,6 +916,16 @@ ieee80211_tx_h_check_assoc(struct ieee80211_txrx_data *tx)
return TXRX_CONTINUE;
}

+static ieee80211_txrx_result
+ieee80211_tx_h_sequence(struct ieee80211_txrx_data *tx)
+{
+ struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
+
+ if (ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_control)) >= 24)
+ ieee80211_include_sequence(tx->sdata, hdr);
+
+ return TXRX_CONTINUE;
+}

/* This function is called whenever the AP is about to exceed the maximum limit
* of buffered frames for power saving STAs. This situation should not really
@@ -1770,6 +1796,8 @@ struct sk_buff * ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
skb_reserve(skb, local->hw.extra_tx_headroom);
memcpy(skb_put(skb, bh_len), b_head, bh_len);

+ ieee80211_include_sequence(sdata, (struct ieee80211_hdr *)skb->data);
+
ieee80211_beacon_add_tim(local, ap, skb);

if (b_tail) {
@@ -4381,6 +4409,7 @@ static ieee80211_rx_handler ieee80211_rx_handlers[] =
static ieee80211_tx_handler ieee80211_tx_handlers[] =
{
ieee80211_tx_h_check_assoc,
+ ieee80211_tx_h_sequence,
ieee80211_tx_h_ps_buf,
ieee80211_tx_h_select_key,
ieee80211_tx_h_michael_mic_add,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 9df8ef0..4b7c427 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -317,6 +317,8 @@ struct ieee80211_sub_if_data {
int ieee802_1x; /* IEEE 802.1X PAE - drop packet to/from unauthorized
* port */

+ u16 sequence;
+
/* Fragment table for host-based reassembly */
struct ieee80211_fragment_entry fragments[IEEE80211_FRAGMENT_MAX];
unsigned int fragment_next;

2007-03-10 19:21:57

[permalink] [raw]

Subject: [PATCH v3] d80211: Add software sequence support

Most hardware can keep track of sequence numbers themselves,
unfortunately *most* doesn't cover all devices. ;)
This patch will keep track of the (per-bss) sequence number
if the flag IEEE80211_HW_SOFTWARE_SEQUENCE has been set.

This 3rd version of the patch has the sequence number stored
in the ieee80211_sub_if_data structure.

Signed-off-by: Ivo van Doorn <[email protected]>

---

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 82ea43f..e32e28a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -515,6 +515,9 @@ struct ieee80211_hw {
* normal operation. */
#define IEEE80211_HW_MONITOR_DURING_OPER (1<<9)

+ /* Does the HOST need to insert the frame sequence counter */
+#define IEEE80211_HW_SOFTWARE_SEQUENCE (1<<10)
+
/* please fill this gap when adding new flags */

/* calculate Michael MIC for an MSDU when doing hwcrypto */
diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c
index 577dbe3..79c2a60 100644
--- a/net/mac80211/ieee80211.c
+++ b/net/mac80211/ieee80211.c
@@ -54,6 +54,20 @@ static u8 * ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len);
static int ieee80211_mgmt_start_xmit(struct sk_buff *skb,
struct net_device *dev);

+static void ieee80211_include_sequence(struct ieee80211_sub_if_data *sdata,
+ struct ieee80211_hdr *hdr)
+{
+ /*
+ * Set the sequence number for this frame.
+ */
+ hdr->seq_ctrl = cpu_to_le16(sdata->sequence & IEEE80211_SCTL_SEQ);
+
+ /*
+ * Increase the sequence number.
+ */
+ sdata->sequence = (sdata->sequence + 0x10) & IEEE80211_SCTL_SEQ;
+}
+
struct ieee80211_key_conf *
ieee80211_key_data2conf(struct ieee80211_local *local,
const struct ieee80211_key *data)
@@ -445,6 +459,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
size_t hdrlen, per_fragm, num_fragm, payload_len, left;
struct sk_buff **frags, *first, *frag;
int i;
+ u16 seq;
u8 *pos;
int frag_threshold = tx->local->fragmentation_threshold;

@@ -463,6 +478,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
goto fail;

hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_MOREFRAGS);
+ seq = le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ;
pos = first->data + hdrlen + per_fragm;
left = payload_len - per_fragm;
for (i = 0; i < num_fragm - 1; i++) {
@@ -490,7 +506,7 @@ ieee80211_tx_h_fragment(struct ieee80211_txrx_data *tx)
memcpy(fhdr, first->data, hdrlen);
if (i == num_fragm - 2)
fhdr->frame_control &= cpu_to_le16(~IEEE80211_FCTL_MOREFRAGS);
- fhdr->seq_ctrl = cpu_to_le16(i + 1);
+ fhdr->seq_ctrl = cpu_to_le16(seq + ((i + 1) & IEEE80211_SCTL_FRAG));
copylen = left > per_fragm ? per_fragm : left;
memcpy(skb_put(frag, copylen), pos, copylen);

@@ -900,6 +916,17 @@ ieee80211_tx_h_check_assoc(struct ieee80211_txrx_data *tx)
return TXRX_CONTINUE;
}

+static ieee80211_txrx_result
+ieee80211_tx_h_sequence(struct ieee80211_txrx_data *tx)
+{
+ struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx->skb->data;
+
+ if ((tx->local->hw.flags & IEEE80211_HW_SOFTWARE_SEQUENCE) &&
+ ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_control)) >= 24)
+ ieee80211_include_sequence(tx->sdata, hdr);
+
+ return TXRX_CONTINUE;
+}

/* This function is called whenever the AP is about to exceed the maximum limit
* of buffered frames for power saving STAs. This situation should not really
@@ -1770,6 +1797,10 @@ struct sk_buff * ieee80211_beacon_get(struct ieee80211_hw *hw, int if_id,
skb_reserve(skb, local->hw.extra_tx_headroom);
memcpy(skb_put(skb, bh_len), b_head, bh_len);

+ if (hw->flags & IEEE80211_HW_SOFTWARE_SEQUENCE)
+ ieee80211_include_sequence(sdata,
+ (struct ieee80211_hdr *)skb->data);
+
ieee80211_beacon_add_tim(local, ap, skb);

if (b_tail) {
@@ -4381,6 +4412,7 @@ static ieee80211_rx_handler ieee80211_rx_handlers[] =
static ieee80211_tx_handler ieee80211_tx_handlers[] =
{
ieee80211_tx_h_check_assoc,
+ ieee80211_tx_h_sequence,
ieee80211_tx_h_ps_buf,
ieee80211_tx_h_select_key,
ieee80211_tx_h_michael_mic_add,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 9df8ef0..4b7c427 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -317,6 +317,8 @@ struct ieee80211_sub_if_data {
int ieee802_1x; /* IEEE 802.1X PAE - drop packet to/from unauthorized
* port */

+ u16 sequence;
+
/* Fragment table for host-based reassembly */
struct ieee80211_fragment_entry fragments[IEEE80211_FRAGMENT_MAX];
unsigned int fragment_next;

2007-03-10 17:51:17

[permalink] [raw]

Subject: Re: [PATCH v2] d80211: Add software sequence support

On Saturday 10 March 2007 09:38, Ivo van Doorn wrote:
> So just to be clear, this patch does not need any tweaking anymore right?
Just stick the counter in sdata if you haven't done so already. Should be fine
once that's done.

Thanks,
-Michael Wu

Attachments:

(No filename) (249.00 B)
(No filename) (189.00 B)
Download all attachments

2007-03-10 19:01:57