Wireless Daemon for Linux
 help / color / mirror / Atom feed
From: James Prestwood <prestwoj@gmail.com>
To: John Brandt <brandtwjohn@gmail.com>, iwd@lists.linux.dev
Subject: Re: [PATCH 05/11] sae: make sae_process_commit callable in AP mode
Date: Wed, 24 Apr 2024 05:08:41 -0700	[thread overview]
Message-ID: <34794f6d-211b-4f91-91f7-4de8ab063841@gmail.com> (raw)
In-Reply-To: <20240421125050.6649-6-brandtwjohn@gmail.com>

Hi John

On 4/21/24 5:50 AM, John Brandt wrote:
> As an AP, the function sae_process_commit will pick the group offered by
> the client. In a subsuquent commit the offered group will first be
> verified before calling sae_process_commit. The AP will reply with a
> Commit frame, calculate current keys, and move to the COMMITTED state.
> ---
>   src/sae.c | 32 ++++++++++++++++++++------------
>   1 file changed, 20 insertions(+), 12 deletions(-)
>
> diff --git a/src/sae.c b/src/sae.c
> index 314fc28f..2c97d94c 100644
> --- a/src/sae.c
> +++ b/src/sae.c
> @@ -48,6 +48,8 @@ static bool debug;
>   #define SAE_SYNC_MAX		3
>   #define SAE_MAX_ASSOC_RETRY	3
>   
> +static bool sae_send_commit(struct sae_sm *sm, bool retry);
> +
>   #define sae_debug(fmat, ...) \
>   ({	\
>   	if (debug) \
> @@ -797,8 +799,12 @@ static int sae_process_commit(struct sae_sm *sm, const uint8_t *from,
>   					const uint8_t *frame, size_t len)
>   {
>   	uint8_t *ptr = (uint8_t *) frame;
> -	unsigned int nbytes = l_ecc_curve_get_scalar_bytes(sm->curve);
> +	unsigned int nbytes;
> +
> +	if (sm->handshake->authenticator && sae_set_group(sm, l_get_le16(frame)) < 0)
> +		return -1;
>   
> +	nbytes = l_ecc_curve_get_scalar_bytes(sm->curve);
>   	ptr += 2;
>   
>   	sm->p_scalar = l_ecc_scalar_new(sm->curve, ptr, nbytes);
> @@ -824,20 +830,22 @@ static int sae_process_commit(struct sae_sm *sm, const uint8_t *from,
>   	 * it is evidence of a reflection attack) and the t0 (retransmission)
>   	 * timer shall be set.
>   	 */
> -	if (l_ecc_scalars_are_equal(sm->p_scalar, sm->scalar) ||
> -			l_ecc_points_are_equal(sm->p_element, sm->element)) {
> -		l_warn("peer scalar or element matched own, discarding frame");
> +	if ((sm->scalar && l_ecc_scalars_are_equal(sm->p_scalar, sm->scalar)) ||
> +			(sm->element && l_ecc_points_are_equal(sm->p_element, sm->element)))
So we isolated this check to only the supplicant side now, but do we 
also need to do the same check after the keys are calculated for AP mode?
>   		return -ENOMSG;
> -	}
>   
>   	sm->sc++;
>   
> -	sae_calculate_keys(sm);
> -
> -	if (!sae_send_confirm(sm))
> -		return -EPROTO;
> -
> -	sm->state = SAE_STATE_CONFIRMED;
> +	if (sm->handshake->authenticator) {
> +		sae_send_commit(sm, false);
> +		sae_calculate_keys(sm);
> +		sm->state = SAE_STATE_COMMITTED;
> +	} else {
> +		sae_calculate_keys(sm);
> +		if (!sae_send_confirm(sm))
> +			return -EPROTO;
> +		sm->state = SAE_STATE_CONFIRMED;
> +	}
>   
>   	return 0;
>   }
> @@ -1008,7 +1016,7 @@ static int sae_verify_nothing(struct sae_sm *sm, uint16_t transaction,
>   	/*
>   	 * TODO: This does not handle the transition from NOTHING -> CONFIRMED
>   	 * as this is only relevant to the AP or in Mesh mode which is not
> -	 * yet supported.
> +	 * yet fully supported.
>   	 */
>   	if (transaction != SAE_STATE_COMMITTED)
>   		return -EBADMSG;

  reply	other threads:[~2024-04-24 12:08 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-21 12:50 [PATCH 00/11] Basic SAE support for AP mode John Brandt
2024-04-21 12:50 ` [PATCH 01/11] ap: ability to advertise PSK and SAE John Brandt
2024-04-21 12:50 ` [PATCH 02/11] ap: accept PSK/SAE in auth depending on config John Brandt
2024-04-24 12:05   ` James Prestwood
2024-04-21 12:50 ` [PATCH 03/11] sae: add function sae_set_group John Brandt
2024-04-24 12:05   ` James Prestwood
2024-04-21 12:50 ` [PATCH 04/11] sae: refactor and add function sae_calculate_keys John Brandt
2024-04-24 12:06   ` James Prestwood
2024-04-21 12:50 ` [PATCH 05/11] sae: make sae_process_commit callable in AP mode John Brandt
2024-04-24 12:08   ` James Prestwood [this message]
2024-04-21 12:50 ` [PATCH 06/11] sae: verify offered group " John Brandt
2024-04-21 12:50 ` [PATCH 07/11] sae: support reception of Confirm frame by AP John Brandt
2024-04-24 12:08   ` James Prestwood
2024-04-21 12:50 ` [PATCH 08/11] ap: add support to handle SAE authentication John Brandt
2024-04-24 12:06   ` James Prestwood
2024-04-21 12:50 ` [PATCH 09/11] ap: enable start of 4-way HS after SAE John Brandt
2024-04-21 12:50 ` [PATCH 10/11] eapol: support PTK derivation with SHA256 John Brandt
2024-04-21 12:50 ` [PATCH 11/11] eapol: encrypt key data for AKM-defined ciphers John Brandt
2024-04-22 13:52 ` [PATCH 00/11] Basic SAE support for AP mode James Prestwood
2024-04-24 12:07   ` James Prestwood
2024-04-29  0:04     ` John Brandt
2024-04-29 12:00       ` James Prestwood
2024-04-30 23:27         ` KeithG

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=34794f6d-211b-4f91-91f7-4de8ab063841@gmail.com \
    --to=prestwoj@gmail.com \
    --cc=brandtwjohn@gmail.com \
    --cc=iwd@lists.linux.dev \
    /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