Re: [OE-core] [PATCH v2] wpa-supplicant: add patches to support RSNE/RSNXE overriding

[Paul Barker <paul@pbarker.dev>]

[-- Attachment #1: Type: text/plain, Size: 4244 bytes --]

On Fri, 2026-04-24 at 18:40 +0800, Jiacheng Shi (Consultant) via
lists.openembedded.org wrote:
> On 4/24/2026 5:54 PM, Paul Barker wrote:
> > On Fri, 2026-04-24 at 17:36 +0800, Jiacheng Shi via
> > lists.openembedded.org wrote:
> > > Add RSN/RSNXE override support in wpa_supplicant to pass Wi-Fi
> > > certification test FC 5.2.3.3.
> > > 
> > > During the 2EAPMSG exchange, the STA did not apply the SNonce_Cookie
> > > provided by the AP as required by the test specification, resulting
> > > in sniffer validation failure.
> > > To address this, allow RSN/RSNX overriding in STA mode and explicitly
> > > signal SNonce_Cookie–based override capability.
> > > 
> > > Signed-off-by: Jiacheng Shi <jiacheng.shi@oss.qualcomm.com>
> > > ---
> > > v2:
> > >   * Change commit message
> > > 
> > >   ...r-specific-element-types-for-RSNE-RS.patch |  34 +
> > >   ...-overriding-elements-into-IE-parsing.patch |  58 ++
> > >   ...ide-element-to-override-RSNE-content.patch |  36 +
> > >   ...ride-element-to-override-RSNXE-conte.patch |  36 +
> > >   ...eature-flags-to-indicate-RSN-overrid.patch |  67 ++
> > >   ...a-capability-flag-for-RSN-overriding.patch |  65 ++
> > >   ...ions-to-access-RSNE-RSNXE-from-BSS-e.patch | 448 +++++++++++
> > >   ...bilities-for-AKM-suites-available-wi.patch |  66 ++
> > >   .../0013-RSNE-RSNXE-overriding-for-STA.patch  | 739 ++++++++++++++++++
> > >   ...-Selection-element-to-indicate-which.patch | 415 ++++++++++
> > >   ...cookie-to-indicate-support-for-RSN-o.patch |  84 ++
> > >   .../wpa-supplicant/wpa-supplicant_2.11.bb     |  11 +
> > >   12 files changed, 2059 insertions(+)
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0005-Define-WFA-vendor-specific-element-types-for-RSNE-RS.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0006-Add-RSN-overriding-elements-into-IE-parsing.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0007-Allow-RSNE-Override-element-to-override-RSNE-content.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0008-Allow-RSNXE-Override-element-to-override-RSNXE-conte.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0009-Add-QCA-vendor-feature-flags-to-indicate-RSN-overrid.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0010-nl80211-Add-a-capability-flag-for-RSN-overriding.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0011-Use-helper-functions-to-access-RSNE-RSNXE-from-BSS-e.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0012-Make-driver-capabilities-for-AKM-suites-available-wi.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0013-RSNE-RSNXE-overriding-for-STA.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0014-RSNO-Use-the-RSN-Selection-element-to-indicate-which.patch
> > >   create mode 100644 meta/recipes-connectivity/wpa-supplicant/wpa-supplicant/0015-RSNO-Use-SNonce-cookie-to-indicate-support-for-RSN-o.patch
> > Hi,
> > 
> > This is a lot to backport for a new feature. Have you tried asking
> > upstream for a new release? That would be a better path forward if they
> > have important new features in their development branch.
> > 
> > Thanks,
> 
> Hi,
> 
> Since we don't have control over the upstream release schedule,
> would it be possible to merge these certification-related patches first?
> Once wpa_suoolicant version 2.12 is ready, we can then proceed to remove them accordingly.

I think we need to at least ask upstream to make a release. hostap v2.11
was released back in July 2024, and there have been a lot of new
features added since then.

If upstream have been asked for a new release, and it doesn't look like
one is forthcoming, then the question is - how stable and well tested is
upstream's main branch? If it's in a good state, we could consider an
update to a more recent git commit hash instead of backporting patches.

Best regards,

-- 
Paul Barker


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 252 bytes --]