Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Cyril Brulebois]

Jonathan Wiltshire  (2017-02-26):
> Needs approval from d-i RM. Debdiff:
> 
> > diff -Nru wpa-2.5-2+v2.4/debian/changelog wpa-2.4/debian/changelog
> > --- wpa-2.5-2+v2.4/debian/changelog 2016-09-14 11:11:05.0 +0200
> > +++ wpa-2.4/debian/changelog2017-02-20 11:55:12.0 +0100
> > @@ -1,3 +1,21 @@
> > +wpa (2:2.4-1) unstable; urgency=medium
> > +
> > +  [ Vincent Danjean ]
> > +  * Build with libssl1.0-dev (Closes: #828601).
> > +  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).
> > +
> > +  [ Andrew Shadura ]
> > +  * Don't install debian/system-sleep/wpasupplicant (originally introduced
> > +to fix LP: #1422143), it doesn't improve the state of the things,
> > +introduces regressions in some cases, and at all isn't supposed to
> > +work with how wpa-supplicant is started these days (Closes: #835648).
> > +  * Bump the epoch to 2:, so that we can set the upstream version to
> > +what we really mean. It also has to be higher than 2.6 in unstable
> > +and 1:2.6 (what hostapd binary package in unstable has).
> > +  * Drop the binary package epoch override.
> > +
> > + -- Andrew Shadura   Mon, 20 Feb 2017 11:55:11 +0100
> > +
> >  wpa (2.5-2+v2.4-3) unstable; urgency=medium
> >  
> >[ Helmut Grohne ]
[…]

No objections, thanks.


KiBi.


signature.asc
Description: Digital signature

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Jonathan Wiltshire]

On Mon, Feb 20, 2017 at 12:08:51PM +0100, Andrew Shadura wrote:
> On 20/02/17 08:18, Andrew Shadura wrote:
> >> Please upload a targeted fix to unstable.
> 
> I have thought about this a bit more. Even though I don't really like
> the epoch idea, it's an opportunity to clean up the package version a
> bit (1:2.4-1 instead of 2.5-2+v2.4-4stretch1, or even 2:2.4-1 to
> accommodate an existing epoch in hostapd package). How about this debdiff?

Needs approval from d-i RM. Debdiff:

> diff -Nru wpa-2.5-2+v2.4/debian/changelog wpa-2.4/debian/changelog
> --- wpa-2.5-2+v2.4/debian/changelog   2016-09-14 11:11:05.0 +0200
> +++ wpa-2.4/debian/changelog  2017-02-20 11:55:12.0 +0100
> @@ -1,3 +1,21 @@
> +wpa (2:2.4-1) unstable; urgency=medium
> +
> +  [ Vincent Danjean ]
> +  * Build with libssl1.0-dev (Closes: #828601).
> +  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).
> +
> +  [ Andrew Shadura ]
> +  * Don't install debian/system-sleep/wpasupplicant (originally introduced
> +to fix LP: #1422143), it doesn't improve the state of the things,
> +introduces regressions in some cases, and at all isn't supposed to
> +work with how wpa-supplicant is started these days (Closes: #835648).
> +  * Bump the epoch to 2:, so that we can set the upstream version to
> +what we really mean. It also has to be higher than 2.6 in unstable
> +and 1:2.6 (what hostapd binary package in unstable has).
> +  * Drop the binary package epoch override.
> +
> + -- Andrew Shadura   Mon, 20 Feb 2017 11:55:11 +0100
> +
>  wpa (2.5-2+v2.4-3) unstable; urgency=medium
>  
>[ Helmut Grohne ]
> diff -Nru wpa-2.5-2+v2.4/debian/control wpa-2.4/debian/control
> --- wpa-2.5-2+v2.4/debian/control 2016-08-05 20:43:38.0 +0200
> +++ wpa-2.4/debian/control2017-02-14 17:02:28.0 +0100
> @@ -8,7 +8,7 @@
>  Priority: optional
>  Build-Depends: debhelper (>> 9.20120115),
>   libdbus-1-dev,
> - libssl-dev,
> + libssl1.0-dev,
>   libqt4-dev,
>   libncurses5-dev,
>   libpcsclite-dev,
> diff -Nru 
> wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
> wpa-2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch
> --- wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch   
> 1970-01-01 01:00:00.0 +0100
> +++ wpa-2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch  
> 2017-02-14 17:07:16.0 +0100
> @@ -0,0 +1,56 @@
> +From: Jouni Malinen 
> +Date: Fri Mar 18 16:16:40 2016 +0200
> +Subject: nl80211: Do not add NL80211_ATTR_SMPS_MODE attribute if HT is 
> disabled
> +
> +SMPS mode is applicable only for HT and including an attribute to
> +configure it when HT is disabled could result in the AP start operation
> +failing. Fix this by adding the attribute only in cases where HT is
> +enabled.
> +
> +Upstream commit ee298f1b1f7efd7eb5fd510f36b25ff88208017c
> +--- a/src/drivers/driver_nl80211.c
>  b/src/drivers/driver_nl80211.c
> +@@ -3419,24 +3419,26 @@
> + nla_put_u32(msg, NL80211_ATTR_CIPHER_SUITE_GROUP, suite))
> + goto fail;
> + 
> +-switch (params->smps_mode) {
> +-case HT_CAP_INFO_SMPS_DYNAMIC:
> +-wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
> +-smps_mode = NL80211_SMPS_DYNAMIC;
> +-break;
> +-case HT_CAP_INFO_SMPS_STATIC:
> +-wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
> +-smps_mode = NL80211_SMPS_STATIC;
> +-break;
> +-default:
> +-/* invalid - fallback to smps off */
> +-case HT_CAP_INFO_SMPS_DISABLED:
> +-wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
> +-smps_mode = NL80211_SMPS_OFF;
> +-break;
> ++if (params->ht_opmode != -1) {
> ++switch (params->smps_mode) {
> ++case HT_CAP_INFO_SMPS_DYNAMIC:
> ++wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
> ++smps_mode = NL80211_SMPS_DYNAMIC;
> ++break;
> ++case HT_CAP_INFO_SMPS_STATIC:
> ++wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
> ++smps_mode = NL80211_SMPS_STATIC;
> ++break;
> ++default:
> ++/* invalid - fallback to smps off */
> ++case HT_CAP_INFO_SMPS_DISABLED:
> ++wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
> ++smps_mode = NL80211_SMPS_OFF;
> ++break;
> ++}
> ++if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
> ++goto fail;
> + }
> +-if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
> +-goto fail;
> + 
> + if (params->beacon_ies) {
> + wpa_hexdump_buf(MSG_DEBUG, "nl80211: beacon_ies",
> diff -Nru wpa-2.5-2+v2.4/debian/patches/series 

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Ivo De Decker]

Control: tags -1 confirmed

Hi Andrew,

On Mon, Feb 20, 2017 at 12:08:51PM +0100, Andrew Shadura wrote:
> On 20/02/17 08:18, Andrew Shadura wrote:
> >> Please upload a targeted fix to unstable.
> 
> I have thought about this a bit more. Even though I don't really like
> the epoch idea, it's an opportunity to clean up the package version a
> bit (1:2.4-1 instead of 2.5-2+v2.4-4stretch1, or even 2:2.4-1 to
> accommodate an existing epoch in hostapd package).

OK, thanks!

> How about this debdiff?

You don't need pre-approval for uploads to unstable, but the diff looks ok to
me, so I'm happy to give it.

Please note that this will need a d-i ack (for the udeb), but we can ask for
that after it has been in unstable for a few days.

Cheers,

Ivo

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Andrew Shadura]

On 20/02/17 08:18, Andrew Shadura wrote:
>> Please upload a targeted fix to unstable.

I have thought about this a bit more. Even though I don't really like
the epoch idea, it's an opportunity to clean up the package version a
bit (1:2.4-1 instead of 2.5-2+v2.4-4stretch1, or even 2:2.4-1 to
accommodate an existing epoch in hostapd package). How about this debdiff?

-- 
Cheers,
  Andrew
diff -Nru wpa-2.5-2+v2.4/debian/changelog wpa-2.4/debian/changelog
--- wpa-2.5-2+v2.4/debian/changelog 2016-09-14 11:11:05.0 +0200
+++ wpa-2.4/debian/changelog2017-02-20 11:55:12.0 +0100
@@ -1,3 +1,21 @@
+wpa (2:2.4-1) unstable; urgency=medium
+
+  [ Vincent Danjean ]
+  * Build with libssl1.0-dev (Closes: #828601).
+  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).
+
+  [ Andrew Shadura ]
+  * Don't install debian/system-sleep/wpasupplicant (originally introduced
+to fix LP: #1422143), it doesn't improve the state of the things,
+introduces regressions in some cases, and at all isn't supposed to
+work with how wpa-supplicant is started these days (Closes: #835648).
+  * Bump the epoch to 2:, so that we can set the upstream version to
+what we really mean. It also has to be higher than 2.6 in unstable
+and 1:2.6 (what hostapd binary package in unstable has).
+  * Drop the binary package epoch override.
+
+ -- Andrew Shadura   Mon, 20 Feb 2017 11:55:11 +0100
+
 wpa (2.5-2+v2.4-3) unstable; urgency=medium
 
   [ Helmut Grohne ]
diff -Nru wpa-2.5-2+v2.4/debian/control wpa-2.4/debian/control
--- wpa-2.5-2+v2.4/debian/control   2016-08-05 20:43:38.0 +0200
+++ wpa-2.4/debian/control  2017-02-14 17:02:28.0 +0100
@@ -8,7 +8,7 @@
 Priority: optional
 Build-Depends: debhelper (>> 9.20120115),
  libdbus-1-dev,
- libssl-dev,
+ libssl1.0-dev,
  libqt4-dev,
  libncurses5-dev,
  libpcsclite-dev,
diff -Nru 
wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
wpa-2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch
--- wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
1970-01-01 01:00:00.0 +0100
+++ wpa-2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch
2017-02-14 17:07:16.0 +0100
@@ -0,0 +1,56 @@
+From: Jouni Malinen 
+Date: Fri Mar 18 16:16:40 2016 +0200
+Subject: nl80211: Do not add NL80211_ATTR_SMPS_MODE attribute if HT is disabled
+
+SMPS mode is applicable only for HT and including an attribute to
+configure it when HT is disabled could result in the AP start operation
+failing. Fix this by adding the attribute only in cases where HT is
+enabled.
+
+Upstream commit ee298f1b1f7efd7eb5fd510f36b25ff88208017c
+--- a/src/drivers/driver_nl80211.c
 b/src/drivers/driver_nl80211.c
+@@ -3419,24 +3419,26 @@
+   nla_put_u32(msg, NL80211_ATTR_CIPHER_SUITE_GROUP, suite))
+   goto fail;
+ 
+-  switch (params->smps_mode) {
+-  case HT_CAP_INFO_SMPS_DYNAMIC:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
+-  smps_mode = NL80211_SMPS_DYNAMIC;
+-  break;
+-  case HT_CAP_INFO_SMPS_STATIC:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
+-  smps_mode = NL80211_SMPS_STATIC;
+-  break;
+-  default:
+-  /* invalid - fallback to smps off */
+-  case HT_CAP_INFO_SMPS_DISABLED:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
+-  smps_mode = NL80211_SMPS_OFF;
+-  break;
++  if (params->ht_opmode != -1) {
++  switch (params->smps_mode) {
++  case HT_CAP_INFO_SMPS_DYNAMIC:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
++  smps_mode = NL80211_SMPS_DYNAMIC;
++  break;
++  case HT_CAP_INFO_SMPS_STATIC:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
++  smps_mode = NL80211_SMPS_STATIC;
++  break;
++  default:
++  /* invalid - fallback to smps off */
++  case HT_CAP_INFO_SMPS_DISABLED:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
++  smps_mode = NL80211_SMPS_OFF;
++  break;
++  }
++  if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
++  goto fail;
+   }
+-  if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
+-  goto fail;
+ 
+   if (params->beacon_ies) {
+   wpa_hexdump_buf(MSG_DEBUG, "nl80211: beacon_ies",
diff -Nru wpa-2.5-2+v2.4/debian/patches/series wpa-2.4/debian/patches/series
--- wpa-2.5-2+v2.4/debian/patches/series2016-08-09 20:48:10.0 
+0200
+++ wpa-2.4/debian/patches/series   2017-02-14 17:02:28.0 +0100
@@ -31,3 +31,4 @@
 

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Andrew Shadura]

On 20/02/17 00:08, Ivo De Decker wrote:
> We really, really want to avoid going through t-p-u if at all possible. In
> this case, there doesn't seem to be a real issue with doing a revert in
> unstable. There isn't really any point in having a version in unstable that
> isn't meant for testing. The 2.6 version can go to experimental.

It was meant for testing, but unfortunately the issues with it turned
out to be more serious than I thought, which is why I decided to stick
with an older version for now.

> As for the confusion, this can be avoided by using an epoch instead
> of the version you're using now (but that is really your call).

It was already in experimental, and it went to unstable at the point
when it seemed it had enough of testing, but it turned out it didn't. As
for the epoch, I'm not sure it's a good idea, as the package already
uses it in a bit inconsistent way (for one of the binary packages only).

On the other hand, all fixes I'm including are backports of fixes
included in the version currently in unstable.

> Please upload a targeted fix to unstable.

-- 
Cheers,
  Andrew

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Ivo De Decker]

Hi Andrew,

On Sun, Feb 19, 2017 at 11:33:32PM +0100, Andrew Shadura wrote:
>   On Tue, Feb 14, 2017 at 05:25:42PM +0100, Andrew Shadura wrote:
>   > Please unblock package wpa.
>   > This fix has to go through testing-proposed-updates.
> 
>   Why? Is there any reason you can't just revert the version in unstable? That
>   way we don't have to push a new untested upload to stretch. If this upload
>   goes through unstable, there is a higher change that issues are detected 
> (and
>   fixed).
> 
> Because the version in unstable is a new upstream version, and the version in
> testing is already a revert of a newer upstream version. I don't want to 
> confuse
> things even more.
> The changes I propose are quite minimal, and the version we have in testing 
> has
> been there for quite some time already. I don't think there's a high risk of
> introducing a regression with a bug fix.

We really, really want to avoid going through t-p-u if at all possible. In
this case, there doesn't seem to be a real issue with doing a revert in
unstable. There isn't really any point in having a version in unstable that
isn't meant for testing. The 2.6 version can go to experimental. As for the
confusion, this can be avoided by using an epoch instead of the version you're
using now (but that is really your call).

Please upload a targeted fix to unstable.

Cheers,

Ivo

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Andrew Shadura]

On 19 Feb 2017 23:15, "Ivo De Decker"  wrote:

Control: tags -1 moreinfo

Hi Andrew,

On Tue, Feb 14, 2017 at 05:25:42PM +0100, Andrew Shadura wrote:
> Please unblock package wpa.

> This fix has to go through testing-proposed-updates.

Why? Is there any reason you can't just revert the version in unstable? That
way we don't have to push a new untested upload to stretch. If this upload
goes through unstable, there is a higher change that issues are detected
(and
fixed).


Because the version in unstable is a new upstream version, and the version
in testing is already a revert of a newer upstream version. I don't want to
confuse things even more.

The changes I propose are quite minimal, and the version we have in testing
has been there for quite some time already. I don't think there's a high
risk of introducing a regression with a bug fix.

-- 
Cheers,
Andrew

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Ivo De Decker]

Control: tags -1 moreinfo

Hi Andrew,

On Tue, Feb 14, 2017 at 05:25:42PM +0100, Andrew Shadura wrote:
> Please unblock package wpa.

> This fix has to go through testing-proposed-updates.

Why? Is there any reason you can't just revert the version in unstable? That
way we don't have to push a new untested upload to stretch. If this upload
goes through unstable, there is a higher change that issues are detected (and
fixed).

Cheers,

Ivo

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Andrew Shadura]

On 14/02/17 17:25, Andrew Shadura wrote:
> This fix has to go through testing-proposed-updates.
> 
> Thanks for considering.
> 
> unblock wpa/2.5-2+v2.4-4stretch1

Sorry, the debdiff I attached was made against a never released version
of the package (differing only in the version number). Attaching the
correct one.

-- 
Cheers,
  Andrew
diff -Nru wpa-2.5-2+v2.4/debian/changelog wpa-2.5-2+v2.4/debian/changelog
--- wpa-2.5-2+v2.4/debian/changelog 2016-09-14 11:11:05.0 +0200
+++ wpa-2.5-2+v2.4/debian/changelog 2017-02-14 17:10:53.0 +0100
@@ -1,3 +1,17 @@
+wpa (2.5-2+v2.4-4stretch1) testing; urgency=medium
+
+  [ Vincent Danjean ]
+  * Build with libssl1.0-dev (Closes: #828601).
+  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).
+
+  [ Andrew Shadura ]
+  * Don't install debian/system-sleep/wpasupplicant (originally introduced
+to fix LP: #1422143), it doesn't improve the state of the things,
+introduces regressions in some cases, and at all isn't supposed to
+work with how wpa-supplicant is started these days (Closes: #835648).
+
+ -- Andrew Shadura   Tue, 14 Feb 2017 17:10:53 +0100
+
 wpa (2.5-2+v2.4-3) unstable; urgency=medium
 
   [ Helmut Grohne ]
diff -Nru wpa-2.5-2+v2.4/debian/control wpa-2.5-2+v2.4/debian/control
--- wpa-2.5-2+v2.4/debian/control   2016-08-05 20:43:38.0 +0200
+++ wpa-2.5-2+v2.4/debian/control   2017-02-14 17:02:28.0 +0100
@@ -8,7 +8,7 @@
 Priority: optional
 Build-Depends: debhelper (>> 9.20120115),
  libdbus-1-dev,
- libssl-dev,
+ libssl1.0-dev,
  libqt4-dev,
  libncurses5-dev,
  libpcsclite-dev,
diff -Nru 
wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch
--- wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
1970-01-01 01:00:00.0 +0100
+++ wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch 
2017-02-14 17:07:16.0 +0100
@@ -0,0 +1,56 @@
+From: Jouni Malinen 
+Date: Fri Mar 18 16:16:40 2016 +0200
+Subject: nl80211: Do not add NL80211_ATTR_SMPS_MODE attribute if HT is disabled
+
+SMPS mode is applicable only for HT and including an attribute to
+configure it when HT is disabled could result in the AP start operation
+failing. Fix this by adding the attribute only in cases where HT is
+enabled.
+
+Upstream commit ee298f1b1f7efd7eb5fd510f36b25ff88208017c
+--- a/src/drivers/driver_nl80211.c
 b/src/drivers/driver_nl80211.c
+@@ -3419,24 +3419,26 @@
+   nla_put_u32(msg, NL80211_ATTR_CIPHER_SUITE_GROUP, suite))
+   goto fail;
+ 
+-  switch (params->smps_mode) {
+-  case HT_CAP_INFO_SMPS_DYNAMIC:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
+-  smps_mode = NL80211_SMPS_DYNAMIC;
+-  break;
+-  case HT_CAP_INFO_SMPS_STATIC:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
+-  smps_mode = NL80211_SMPS_STATIC;
+-  break;
+-  default:
+-  /* invalid - fallback to smps off */
+-  case HT_CAP_INFO_SMPS_DISABLED:
+-  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
+-  smps_mode = NL80211_SMPS_OFF;
+-  break;
++  if (params->ht_opmode != -1) {
++  switch (params->smps_mode) {
++  case HT_CAP_INFO_SMPS_DYNAMIC:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
++  smps_mode = NL80211_SMPS_DYNAMIC;
++  break;
++  case HT_CAP_INFO_SMPS_STATIC:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
++  smps_mode = NL80211_SMPS_STATIC;
++  break;
++  default:
++  /* invalid - fallback to smps off */
++  case HT_CAP_INFO_SMPS_DISABLED:
++  wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - off");
++  smps_mode = NL80211_SMPS_OFF;
++  break;
++  }
++  if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
++  goto fail;
+   }
+-  if (nla_put_u32(msg, NL80211_ATTR_SMPS_MODE, smps_mode))
+-  goto fail;
+ 
+   if (params->beacon_ies) {
+   wpa_hexdump_buf(MSG_DEBUG, "nl80211: beacon_ies",
diff -Nru wpa-2.5-2+v2.4/debian/patches/lowercase-mschapv2.patch 
wpa-2.5-2+v2.4/debian/patches/lowercase-mschapv2.patch
--- wpa-2.5-2+v2.4/debian/patches/lowercase-mschapv2.patch  1970-01-01 
01:00:00.0 +0100
+++ wpa-2.5-2+v2.4/debian/patches/lowercase-mschapv2.patch  2017-01-20 
21:32:38.0 +0100
@@ -0,0 +1,19 @@
+From: Andrew Shadura 
+Subject: Temporarily accept never supported MSCHAPv2 syntax and issue a 
warning.
+
+--- a/src/eap_peer/eap_ttls.c
 

Bug#855143: unblock: wpa/2.5-2+v2.4-4stretch1

[Andrew Shadura]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: unblock

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Please unblock package wpa.

wpa-supplicant 2.6 brought some regressions making it unfit for stretch,
so I decided to keep 2.4 in testing. However, two minor fixes didn't
make it into 2.4 package, and later a grave bug was found in 2.4
compared to 2.3 jessie ships. Because of this I'm asking to unblock a
version that fixes these bugs:

  * Build with libssl1.0-dev (Closes: #828601).

Without this fix, it wouldn't be possible to build the following
fix.

  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).

This is a grave bug which breaks hostapd presumably for many users.

  * Don't install debian/system-sleep/wpasupplicant (originally introduced
to fix LP: #1422143), it doesn't improve the state of the things,
introduces regressions in some cases, and at all isn't supposed to
work with how wpa-supplicant is started these days (Closes: #835648).

This bug slows down shutdown for many users, and as the fix is just
not installing a single file, I don't see a reason to not ship it.

This fix has to go through testing-proposed-updates.

Thanks for considering.

unblock wpa/2.5-2+v2.4-4stretch1

- -- 
Cheers,
  Andrew

-BEGIN PGP SIGNATURE-

iQExBAEBCAAbBQJYoy+GFBxhbmRyZXdzaEBkZWJpYW4ub3JnAAoJEJ1bI/kYT6UU
1tYH/3h8eaMFB386jUQtIXvAZbruzSm29FlDlXslMgaSX1SaBqbIYS2frxJ6RfvT
lU5A5gtaNMqA+TjFeump0sHPjLiwDhNnyDJFgZx9XNqDdedBpaEDRQf6ffv02j1P
WyY7AZFT+Kw7gijwujx4Bb35oeFUgDUWQ7MvDB10gbhPmRVM2kT2YryVNIwkyGhR
mflwqXw7AsK3dcVo0BCGPrxIMy9Jl4oKxTUVLeQwtOQr5jy2HlOSyt6Vz396bkFq
aerMYFkk0m2SEITddDbwMybYD+Xv0tiLyAfOeL4HFswmCiMW26BEk2HzI5OREOdK
vtDHQ7P1mSS5BTUKbn8LNwqn2lc=
=CZE0
-END PGP SIGNATURE-
diff -Nru wpa-2.5-2+v2.4/debian/changelog wpa-2.5-2+v2.4/debian/changelog
--- wpa-2.5-2+v2.4/debian/changelog	2016-09-14 11:08:29.0 +0200
+++ wpa-2.5-2+v2.4/debian/changelog	2017-02-14 17:10:53.0 +0100
@@ -1,4 +1,18 @@
-wpa (2.5-2+v2.4-4) unstable; urgency=medium
+wpa (2.5-2+v2.4-4stretch1) testing; urgency=medium
+
+  [ Vincent Danjean ]
+  * Build with libssl1.0-dev (Closes: #828601).
+  * Add an upstream patch to fix hostapd in SMPS mode (Closes: #854719).
+
+  [ Andrew Shadura ]
+  * Don't install debian/system-sleep/wpasupplicant (originally introduced
+to fix LP: #1422143), it doesn't improve the state of the things,
+introduces regressions in some cases, and at all isn't supposed to
+work with how wpa-supplicant is started these days (Closes: #835648).
+
+ -- Andrew Shadura   Tue, 14 Feb 2017 17:10:53 +0100
+
+wpa (2.5-2+v2.4-3) unstable; urgency=medium
 
   [ Helmut Grohne ]
   * Address FTCBFS: Set PKG_CONFIG (Closes: #836074).
@@ -7,7 +21,7 @@
   * Don't run wpa_cli suspend/resume if /run/wpa_supplicant isn't around
 (Closes: #835648).
 
- -- Andrew Shadura   Wed, 14 Sep 2016 11:08:27 +0200
+ -- Andrew Shadura   Wed, 14 Sep 2016 11:11:01 +0200
 
 wpa (2.5-2+v2.4-2) unstable; urgency=medium
 
diff -Nru wpa-2.5-2+v2.4/debian/control wpa-2.5-2+v2.4/debian/control
--- wpa-2.5-2+v2.4/debian/control	2016-08-05 20:43:38.0 +0200
+++ wpa-2.5-2+v2.4/debian/control	2017-02-14 17:02:28.0 +0100
@@ -8,7 +8,7 @@
 Priority: optional
 Build-Depends: debhelper (>> 9.20120115),
  libdbus-1-dev,
- libssl-dev,
+ libssl1.0-dev,
  libqt4-dev,
  libncurses5-dev,
  libpcsclite-dev,
diff -Nru wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch
--- wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch	1970-01-01 01:00:00.0 +0100
+++ wpa-2.5-2+v2.4/debian/patches/from-upstream-hostapd-fix-SMPS-mode.patch	2017-02-14 17:07:16.0 +0100
@@ -0,0 +1,56 @@
+From: Jouni Malinen 
+Date: Fri Mar 18 16:16:40 2016 +0200
+Subject: nl80211: Do not add NL80211_ATTR_SMPS_MODE attribute if HT is disabled
+
+SMPS mode is applicable only for HT and including an attribute to
+configure it when HT is disabled could result in the AP start operation
+failing. Fix this by adding the attribute only in cases where HT is
+enabled.
+
+Upstream commit ee298f1b1f7efd7eb5fd510f36b25ff88208017c
+--- a/src/drivers/driver_nl80211.c
 b/src/drivers/driver_nl80211.c
+@@ -3419,24 +3419,26 @@
+ 	nla_put_u32(msg, NL80211_ATTR_CIPHER_SUITE_GROUP, suite))
+ 		goto fail;
+ 
+-	switch (params->smps_mode) {
+-	case HT_CAP_INFO_SMPS_DYNAMIC:
+-		wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - dynamic");
+-		smps_mode = NL80211_SMPS_DYNAMIC;
+-		break;
+-	case HT_CAP_INFO_SMPS_STATIC:
+-		wpa_printf(MSG_DEBUG, "nl80211: SMPS mode - static");
+-		smps_mode = NL80211_SMPS_STATIC;
+-		break;
+-	default:
+-		/* invalid - fallback to smps off */
+-	case HT_CAP_INFO_SMPS_DISABLED:
+-		wpa_printf(MSG_DEBUG, "nl80211: SMPS