[PATCH] cfg80211: mark all WEXT handlers _GPL

[PATCH] cfg80211: mark all WEXT handlers _GPL

[Johannes Berg]

The fact that these are exported is a technical detail
of the conversion period -- we don't want anybody to
start relying on these. Ultimately we want things to
use cfg80211 only, and once everything that is in wext
is converted to cfg80211 drivers will not need to touch
wext _at all_.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
 net/wireless/scan.c        |    4 ++--
 net/wireless/wext-compat.c |   24 ++++++++++++------------
 2 files changed, 14 insertions(+), 14 deletions(-)

--- wireless-testing.orig/net/wireless/scan.c	2009-04-20 18:47:21.000000000 +0200
+++ wireless-testing/net/wireless/scan.c	2009-04-20 18:47:45.000000000 +0200
@@ -653,7 +653,7 @@ int cfg80211_wext_siwscan(struct net_dev
 	cfg80211_put_dev(rdev);
 	return err;
 }
-EXPORT_SYMBOL(cfg80211_wext_siwscan);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwscan);
 
 static void ieee80211_scan_add_ies(struct iw_request_info *info,
 				   struct cfg80211_bss *bss,
@@ -962,5 +962,5 @@ int cfg80211_wext_giwscan(struct net_dev
 	cfg80211_put_dev(rdev);
 	return res;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwscan);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwscan);
 #endif
--- wireless-testing.orig/net/wireless/wext-compat.c	2009-04-20 18:47:21.000000000 +0200
+++ wireless-testing/net/wireless/wext-compat.c	2009-04-20 18:47:33.000000000 +0200
@@ -57,7 +57,7 @@ int cfg80211_wext_giwname(struct net_dev
 
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwname);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwname);
 
 int cfg80211_wext_siwmode(struct net_device *dev, struct iw_request_info *info,
 			  u32 *mode, char *extra)
@@ -108,7 +108,7 @@ int cfg80211_wext_siwmode(struct net_dev
 
 	return ret;
 }
-EXPORT_SYMBOL(cfg80211_wext_siwmode);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwmode);
 
 int cfg80211_wext_giwmode(struct net_device *dev, struct iw_request_info *info,
 			  u32 *mode, char *extra)
@@ -143,7 +143,7 @@ int cfg80211_wext_giwmode(struct net_dev
 	}
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwmode);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwmode);
 
 
 int cfg80211_wext_giwrange(struct net_device *dev,
@@ -239,7 +239,7 @@ int cfg80211_wext_giwrange(struct net_de
 
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwrange);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwrange);
 
 int cfg80211_wext_siwmlme(struct net_device *dev,
 			  struct iw_request_info *info,
@@ -283,7 +283,7 @@ int cfg80211_wext_siwmlme(struct net_dev
 		return -EOPNOTSUPP;
 	}
 }
-EXPORT_SYMBOL(cfg80211_wext_siwmlme);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwmlme);
 
 
 /**
@@ -313,7 +313,7 @@ struct ieee80211_channel *cfg80211_wext_
 	}
 
 }
-EXPORT_SYMBOL(cfg80211_wext_freq);
+EXPORT_SYMBOL_GPL(cfg80211_wext_freq);
 
 int cfg80211_wext_siwrts(struct net_device *dev,
 			 struct iw_request_info *info,
@@ -338,7 +338,7 @@ int cfg80211_wext_siwrts(struct net_devi
 
 	return err;
 }
-EXPORT_SYMBOL(cfg80211_wext_siwrts);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwrts);
 
 int cfg80211_wext_giwrts(struct net_device *dev,
 			 struct iw_request_info *info,
@@ -352,7 +352,7 @@ int cfg80211_wext_giwrts(struct net_devi
 
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwrts);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwrts);
 
 int cfg80211_wext_siwfrag(struct net_device *dev,
 			  struct iw_request_info *info,
@@ -379,7 +379,7 @@ int cfg80211_wext_siwfrag(struct net_dev
 
 	return err;
 }
-EXPORT_SYMBOL(cfg80211_wext_siwfrag);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwfrag);
 
 int cfg80211_wext_giwfrag(struct net_device *dev,
 			  struct iw_request_info *info,
@@ -393,7 +393,7 @@ int cfg80211_wext_giwfrag(struct net_dev
 
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwfrag);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwfrag);
 
 int cfg80211_wext_siwretry(struct net_device *dev,
 			   struct iw_request_info *info,
@@ -434,7 +434,7 @@ int cfg80211_wext_siwretry(struct net_de
 
 	return err;
 }
-EXPORT_SYMBOL(cfg80211_wext_siwretry);
+EXPORT_SYMBOL_GPL(cfg80211_wext_siwretry);
 
 int cfg80211_wext_giwretry(struct net_device *dev,
 			   struct iw_request_info *info,
@@ -464,4 +464,4 @@ int cfg80211_wext_giwretry(struct net_de
 
 	return 0;
 }
-EXPORT_SYMBOL(cfg80211_wext_giwretry);
+EXPORT_SYMBOL_GPL(cfg80211_wext_giwretry);

Re: [PATCH] cfg80211: mark all WEXT handlers _GPL

[Zhu Yi]

On Tue, 2009-04-21 at 00:49 +0800, Johannes Berg wrote:
> The fact that these are exported is a technical detail
> of the conversion period -- we don't want anybody to
> start relying on these. Ultimately we want things to
> use cfg80211 only, and once everything that is in wext
> is converted to cfg80211 drivers will not need to touch
> wext _at all_.

Johannes, why don't you want them to be used by others, say some fullmac
drivers? These functions can be used as glue from wext to cfg80211 in
the conversion period. After cfg80211 is ultimately used, they can be
removed all together of course.

Thanks,
-yi
> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> ---
>  net/wireless/scan.c        |    4 ++--
>  net/wireless/wext-compat.c |   24 ++++++++++++------------
>  2 files changed, 14 insertions(+), 14 deletions(-)
> 
> --- wireless-testing.orig/net/wireless/scan.c	2009-04-20 18:47:21.000000000 +0200
> +++ wireless-testing/net/wireless/scan.c	2009-04-20 18:47:45.000000000 +0200
> @@ -653,7 +653,7 @@ int cfg80211_wext_siwscan(struct net_dev
>  	cfg80211_put_dev(rdev);
>  	return err;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwscan);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwscan);
>  
>  static void ieee80211_scan_add_ies(struct iw_request_info *info,
>  				   struct cfg80211_bss *bss,
> @@ -962,5 +962,5 @@ int cfg80211_wext_giwscan(struct net_dev
>  	cfg80211_put_dev(rdev);
>  	return res;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwscan);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwscan);
>  #endif
> --- wireless-testing.orig/net/wireless/wext-compat.c	2009-04-20 18:47:21.000000000 +0200
> +++ wireless-testing/net/wireless/wext-compat.c	2009-04-20 18:47:33.000000000 +0200
> @@ -57,7 +57,7 @@ int cfg80211_wext_giwname(struct net_dev
>  
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwname);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwname);
>  
>  int cfg80211_wext_siwmode(struct net_device *dev, struct iw_request_info *info,
>  			  u32 *mode, char *extra)
> @@ -108,7 +108,7 @@ int cfg80211_wext_siwmode(struct net_dev
>  
>  	return ret;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwmode);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwmode);
>  
>  int cfg80211_wext_giwmode(struct net_device *dev, struct iw_request_info *info,
>  			  u32 *mode, char *extra)
> @@ -143,7 +143,7 @@ int cfg80211_wext_giwmode(struct net_dev
>  	}
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwmode);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwmode);
>  
> 
>  int cfg80211_wext_giwrange(struct net_device *dev,
> @@ -239,7 +239,7 @@ int cfg80211_wext_giwrange(struct net_de
>  
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwrange);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwrange);
>  
>  int cfg80211_wext_siwmlme(struct net_device *dev,
>  			  struct iw_request_info *info,
> @@ -283,7 +283,7 @@ int cfg80211_wext_siwmlme(struct net_dev
>  		return -EOPNOTSUPP;
>  	}
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwmlme);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwmlme);
>  
> 
>  /**
> @@ -313,7 +313,7 @@ struct ieee80211_channel *cfg80211_wext_
>  	}
>  
>  }
> -EXPORT_SYMBOL(cfg80211_wext_freq);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_freq);
>  
>  int cfg80211_wext_siwrts(struct net_device *dev,
>  			 struct iw_request_info *info,
> @@ -338,7 +338,7 @@ int cfg80211_wext_siwrts(struct net_devi
>  
>  	return err;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwrts);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwrts);
>  
>  int cfg80211_wext_giwrts(struct net_device *dev,
>  			 struct iw_request_info *info,
> @@ -352,7 +352,7 @@ int cfg80211_wext_giwrts(struct net_devi
>  
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwrts);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwrts);
>  
>  int cfg80211_wext_siwfrag(struct net_device *dev,
>  			  struct iw_request_info *info,
> @@ -379,7 +379,7 @@ int cfg80211_wext_siwfrag(struct net_dev
>  
>  	return err;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwfrag);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwfrag);
>  
>  int cfg80211_wext_giwfrag(struct net_device *dev,
>  			  struct iw_request_info *info,
> @@ -393,7 +393,7 @@ int cfg80211_wext_giwfrag(struct net_dev
>  
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwfrag);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwfrag);
>  
>  int cfg80211_wext_siwretry(struct net_device *dev,
>  			   struct iw_request_info *info,
> @@ -434,7 +434,7 @@ int cfg80211_wext_siwretry(struct net_de
>  
>  	return err;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_siwretry);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_siwretry);
>  
>  int cfg80211_wext_giwretry(struct net_device *dev,
>  			   struct iw_request_info *info,
> @@ -464,4 +464,4 @@ int cfg80211_wext_giwretry(struct net_de
>  
>  	return 0;
>  }
> -EXPORT_SYMBOL(cfg80211_wext_giwretry);
> +EXPORT_SYMBOL_GPL(cfg80211_wext_giwretry);
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] cfg80211: mark all WEXT handlers _GPL

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 1020 bytes --]

On Tue, 2009-04-21 at 09:08 +0800, Zhu Yi wrote:
> On Tue, 2009-04-21 at 00:49 +0800, Johannes Berg wrote:
> > The fact that these are exported is a technical detail
> > of the conversion period -- we don't want anybody to
> > start relying on these. Ultimately we want things to
> > use cfg80211 only, and once everything that is in wext
> > is converted to cfg80211 drivers will not need to touch
> > wext _at all_.
> 
> Johannes, why don't you want them to be used by others, say some fullmac
> drivers? These functions can be used as glue from wext to cfg80211 in
> the conversion period. After cfg80211 is ultimately used, they can be
> removed all together of course.

I don't expect any binary drivers to start using them anyway, but on the
odd chance that somebody was considering that ... I just don't want any
arguments at the conversion point that "we still need these symbols
because we released a driver that uses them" which would mean we
couldn't actually convert properly.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] cfg80211: mark all WEXT handlers _GPL

[Zhu Yi]

On Tue, 2009-04-21 at 16:27 +0800, Johannes Berg wrote:
> On Tue, 2009-04-21 at 09:08 +0800, Zhu Yi wrote:
> > On Tue, 2009-04-21 at 00:49 +0800, Johannes Berg wrote:
> > > The fact that these are exported is a technical detail
> > > of the conversion period -- we don't want anybody to
> > > start relying on these. Ultimately we want things to
> > > use cfg80211 only, and once everything that is in wext
> > > is converted to cfg80211 drivers will not need to touch
> > > wext _at all_.
> > 
> > Johannes, why don't you want them to be used by others, say some fullmac
> > drivers? These functions can be used as glue from wext to cfg80211 in
> > the conversion period. After cfg80211 is ultimately used, they can be
> > removed all together of course.
> 
> I don't expect any binary drivers to start using them anyway, but on the
> odd chance that somebody was considering that ... I just don't want any
> arguments at the conversion point that "we still need these symbols
> because we released a driver that uses them" which would mean we
> couldn't actually convert properly.

Fair enough, no binary drivers. I don't have any problem with your
patch. Just the comment "we don't want anybody to start relying on
these" makes me confuse.

Thanks,
-yi

Re: [PATCH] cfg80211: mark all WEXT handlers _GPL

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 1623 bytes --]

On Tue, 2009-04-21 at 16:31 +0800, Zhu Yi wrote:
> On Tue, 2009-04-21 at 16:27 +0800, Johannes Berg wrote:
> > On Tue, 2009-04-21 at 09:08 +0800, Zhu Yi wrote:
> > > On Tue, 2009-04-21 at 00:49 +0800, Johannes Berg wrote:
> > > > The fact that these are exported is a technical detail
> > > > of the conversion period -- we don't want anybody to
> > > > start relying on these. Ultimately we want things to
> > > > use cfg80211 only, and once everything that is in wext
> > > > is converted to cfg80211 drivers will not need to touch
> > > > wext _at all_.
> > > 
> > > Johannes, why don't you want them to be used by others, say some fullmac
> > > drivers? These functions can be used as glue from wext to cfg80211 in
> > > the conversion period. After cfg80211 is ultimately used, they can be
> > > removed all together of course.
> > 
> > I don't expect any binary drivers to start using them anyway, but on the
> > odd chance that somebody was considering that ... I just don't want any
> > arguments at the conversion point that "we still need these symbols
> > because we released a driver that uses them" which would mean we
> > couldn't actually convert properly.
> 
> Fair enough, no binary drivers. I don't have any problem with your
> patch. Just the comment "we don't want anybody to start relying on
> these" makes me confuse.

Sorry, my mistake, that should have read "don't want any binary
drivers". Everyone is free to use them of course, but I want the option
to remove them easily and without Ubuntu (...) complaining that it broke
their latest wireless drivers.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] cfg80211: mark all WEXT handlers _GPL

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 862 bytes --]

On Tue, 2009-04-21 at 10:27 +0200, Johannes Berg wrote:

> > Johannes, why don't you want them to be used by others, say some fullmac
> > drivers? These functions can be used as glue from wext to cfg80211 in
> > the conversion period. After cfg80211 is ultimately used, they can be
> > removed all together of course.
> 
> I don't expect any binary drivers to start using them anyway, but on the
> odd chance that somebody was considering that ... I just don't want any
> arguments at the conversion point that "we still need these symbols
> because we released a driver that uses them" which would mean we
> couldn't actually convert properly.

s/released a driver/released a binary driver/

After all, I'm saying they're ok for _GPL -- since that means in the
case this actually happens we can then easily convert that driver too.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]