Linux wireless drivers development
 help / color / mirror / Atom feed
From: Pooventhiran G <pooventhiran.g@oss.qualcomm.com>
To: Johannes Berg <johannes@sipsolutions.net>
Cc: linux-wireless@vger.kernel.org
Subject: Re: [PATCH wireless-next v2 02/16] wifi: nl80211: Add kernel interfaces for Seamless Mobility Domain setup
Date: Mon, 5 Oct 2026 21:34:48 +0530	[thread overview]
Message-ID: <c354da3f-857e-4de2-92f5-e300842c2276@oss.qualcomm.com> (raw)
In-Reply-To: <bb6a1ef2348fd1d43420083b8bfa81b95fbf571c.camel@sipsolutions.net>



On 10/02/2026 12:20 pm, Johannes Berg wrote:
> On Thu, 2026-09-24 at 08:10 +0530, Pooventhiran G wrote:
>>
>> +++ b/include/uapi/linux/nl80211.h
>> @@ -3190,6 +3190,12 @@ enum nl80211_commands {
>>   *	known station to transmit a frame. This is relevant to know whether
>>   *	MLD address translation happened or to disable it when sending a frame.
>>   *
>> + * @NL80211_ATTR_SMD_PARAMS: Nested attribute to indicate support for Seamless
>> + *	Mobility Domain (SMD); it contains parameters to define capabilities of
>> + *	an AP managed by an SMD-ME. Valid if the driver supports
>> + *	%NL80211_EXT_FEATURE_SMD.
>> + *	Used with %NL80211_CMD_START_AP. See &enum nl80211_smd_params_attrs.
> 
> "Valid if the driver supports" doesn't make sense - it should clearly be
> arranged to be "valid if present" (which doesn't even need to be
> stated). Then you can get rid of patch 1 entirely, though my request for
> documentation still stands.
> 

Sure, I will fix it in v3.

>> +/**
>> + * enum nl80211_smd_type - Type of the SMD
>> + *
>> + * @NL80211_SMD_TYPE_PER_AP_MLD_MAC_SAP: Separate MAC SAP is used per AP MLD.
>> + * @NL80211_SMD_TYPE_PER_SMD_MAC_SAP: one MAC SAP is used in the SMD.
>> + * @NUM_NL80211_SMD_TYPES: internal.
>> + * @NL80211_SMD_TYPE_MAX: Max type of the SMD.
>> + *
>> + * This value represents the data path models between the non-AP MLD and DS.
>> + * DS mapping for the non-AP MLD is updated if
>> + * %NL80211_SMD_TYPE_PER_AP_MLD_MAC_SAP is used.
>> + *
>> + * All AP MLDs within an SMD shall indicate the same SMD type (refer
>> + * IEEE P802.11bn/D2.0, Aug 2026, subclause 37.16.1.2).
>> + */
>> +enum nl80211_smd_type {
>> +	NL80211_SMD_TYPE_PER_AP_MLD_MAC_SAP,
>> +	NL80211_SMD_TYPE_PER_SMD_MAC_SAP,
>> +
>> +	NUM_NL80211_SMD_TYPES,
>> +	NL80211_SMD_TYPE_MAX = NUM_NL80211_SMD_TYPES - 1,
>> +};
> 
> I don't think this makes sense, any given driver might be able to
> implement both.
> 
>> +/**
>> + * enum nl80211_smd_ptk_mode - PTK Mode used in the SMD
>> + *
>> + * @NL80211_SMD_PTK_MODE_PER_SMD_PTK: one PTK is used by all AP MLDs in the SMD.
>> + * @NL80211_SMD_PTK_MODE_PER_AP_MLD_PTK: Separate PTK is used per AP MLD.
>> + * @NUM_NL80211_SMD_PTK_MODES: internal.
>> + * @NL80211_SMD_PTK_MODE_MAX: Max type of PTK mode.
>> + *
>> + * This value represents the security mode of the SMD that dictates the PTK used
>> + * in current and target AP MLDs to protect communications with the non-AP MLD
>> + * (refer IEEE P802.11bn/D2.0, Aug 2026, subclause 37.16.1.3.2)
>> + */
>> +enum nl80211_smd_ptk_mode {
>> +	NL80211_SMD_PTK_MODE_PER_SMD_PTK,
>> +	NL80211_SMD_PTK_MODE_PER_AP_MLD_PTK,
>> +
>> +	NUM_NL80211_SMD_PTK_MODES,
>> +	NL80211_SMD_PTK_MODE_MAX = NUM_NL80211_SMD_PTK_MODES - 1,
>> +};
> 
> Same here.
> 
>> +/**
>> + * enum nl80211_smd_params_attrs - SMD Domain parameters
>> + *
>> + * @NL80211_SMD_PARAMS_ATTR_UNSPEC: Invalid attribute.
>> + * @NL80211_SMD_PARAMS_ATTR_IDENTIFIER: Required (binary) attribute of ETH_ALEN
>> + *	bytes used to define the unique identifier for the SMD.
> 
> What? Surely you're not asking the driver to advertise the SMD ID vs.
> hostapd just setting it, so I don't see how this makes sense?
> 
>> + * @NL80211_SMD_PARAMS_ATTR_TIMEOUT: Required (u16) attribute to define the SMD
>> + *	Preparation timeout (in 64 TUs, values 0 and 1 are reserved) for
>> + *	the target AP MLD.
> 
> Doesn't make sense either - if anything the driver might need to set a
> minimum or something?
> 
>> + * @NL80211_SMD_PARAMS_ATTR_DL_DATA_FORWARDING: Flag attribute to indicate that
>> + *	the device supports forwarding of buffered DL data packets.
> 
> What am I reading ... please get someone else to review this code first.
> 
>> + * @NL80211_SMD_PARAMS_ATTR_MAX_PEER_APMLDS: Required (u8) attribute to define
>> + *	the maximum number of AP MLDs a non-AP MLD can prepare with. Value is
>> + *	the maximum number of AP MLDs - 1. A value of 0 means that the non-AP
>> + *	MLD can prepare with 1 AP MLD.
> 
> That doesn't make sense at the driver level at all, each individual
> driver should have no notion of how many others a peer prepared with?
> 
>> + * @NL80211_SMD_PARAMS_ATTR_TYPE: Required (u8) attribute to indicate the
>> + *	SMD Type (&enum nl80211_smd_type).
>> + * @NL80211_SMD_PARAMS_ATTR_PTK_MODE: Required (u8) attribute to indicate the
>> + *	PTK mode used for protection (&enum nl80211_smd_ptk_mode).
> 
> See above.
> 
> johannes

I will clean this up in v3 since these parameters are not required in lower
layers and can be contained within hostapd.

  reply	other threads:[~2026-10-05 16:04 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-24  2:40 [PATCH wireless-next v2 00/16] wifi: Add Seamless Mobility Domain (SMD) AP support Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 01/16] wifi: nl80211: Define Seamless Mobility Domain (SMD) device capability Pooventhiran G
2026-10-02  6:41   ` Johannes Berg
2026-10-05 16:01     ` Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 02/16] wifi: nl80211: Add kernel interfaces for Seamless Mobility Domain setup Pooventhiran G
2026-10-02  6:50   ` Johannes Berg
2026-10-05 16:04     ` Pooventhiran G [this message]
2026-09-24  2:40 ` [PATCH wireless-next v2 03/16] wifi: cfg80211/mac80211: Configure AP with SMD capabilities Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 04/16] wifi: cfg80211/mac80211: Parse SMD parameters in STA addition/modification Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 05/16] wifi: nl80211/cfg80211: Indicate STA creation via SMD BSS Transition Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 06/16] wifi: nl80211/mac80211: Add SMD BSS Transition sub-state STA flags Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 07/16] wifi: mac80211: Add driver_op for SMD substate changes Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 08/16] wifi: mac80211: Send BlockAck policy in AMPDU action Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 09/16] wifi: mac80211: Define layouts for SMD BSS Transition context Pooventhiran G
2026-10-01 20:09   ` Johannes Berg
2026-10-01 20:12     ` Johannes Berg
2026-10-07 11:35     ` Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 10/16] wifi: nl80211: Define attributes to pack " Pooventhiran G
2026-10-01 20:19   ` Johannes Berg
2026-10-07 11:42     ` Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 11/16] wifi: cfg80211/mac80211: Handle UHR Link Reconfiguration frame Pooventhiran G
2026-10-04 10:23   ` Johannes Berg
2026-10-05 16:16     ` Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 12/16] wifi: nl80211: Pack SMD dynamic context along with frame Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 13/16] wifi: nl80211/cfg80211: Add support for SMD context programming Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 14/16] wifi: mac80211: Add mac80211 support to handle NL80211_CMD_SET_SMD_CTX Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 15/16] wifi: nl80211/cfg80211: Add support for querying SMD context for target AP MLD Pooventhiran G
2026-09-24  2:40 ` [PATCH wireless-next v2 16/16] wifi: mac80211: Add mac80211 support to handle NL80211_CMD_GET_SMD_CTX Pooventhiran G
2026-10-04 10:49 ` [PATCH wireless-next v2 00/16] wifi: Add Seamless Mobility Domain (SMD) AP support Johannes Berg
2026-10-07 12:09   ` Pooventhiran G

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c354da3f-857e-4de2-92f5-e300842c2276@oss.qualcomm.com \
    --to=pooventhiran.g@oss.qualcomm.com \
    --cc=johannes@sipsolutions.net \
    --cc=linux-wireless@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox