Re: [zd1211-devs] zd1211 vendor driver on 2.6.35.10
Nerijus Baliunas wrote: > What are standard instructions? It works with hostapd here... Perhaps I should say the patch looks incomplete, because I have seen the "official" change I mentioned below, and it is more than just two lines. >> in any case, the "official" change has >> already happened - once something is in the wireless-testing branch of the >> kernel, it is just a matter of time before it reaches Linus himself. The >> process >> usually takes a few months between a change appearing in wireless testing to >> it >> appearing in the official release. > > Great. You wrote the patch was for 2.6.35.10, and my old message was about it entering the wireless-testing branch after 2.6.38 (and before the 2.6.39 merge windows, I think). I just checked that 2.6.39 is in rc7 now, so you probably just need to wait a week or two - and to be sure about it, read the changelog on www.kernel.org. -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Zd1211-devs mailing list - http://zd1211.wiki.sourceforge.net/ Unsubscribe: https://lists.sourceforge.net/lists/listinfo/zd1211-devs
Re: [zd1211-devs] zd1211 vendor driver on 2.6.35.10
On Thu, 12 May 2011 08:58:31 +0100 Hin-Tak Leung wrote: > > There is a patch at http://ankalagon.ru/notes/patch-na-zd1211rw which adds > > AP > > mode in a very simple way, and it actually works! I had to modify the first > > hunk slightly for kernel 2.6.38, modified patch below: > > It looks like a non-standard(ugly?) hack though, and not likely to work with > standard linux AP-setup instruction... What are standard instructions? It works with hostapd here... > in any case, the "official" change has > already happened - once something is in the wireless-testing branch of the > kernel, it is just a matter of time before it reaches Linus himself. The > process > usually takes a few months between a change appearing in wireless testing to > it > appearing in the official release. Great. Regards, Nerijus -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Zd1211-devs mailing list - http://zd1211.wiki.sourceforge.net/ Unsubscribe: https://lists.sourceforge.net/lists/listinfo/zd1211-devs
Re: [zd1211-devs] zd1211 vendor driver on 2.6.35.10
Nerijus Baliunas wrote: > On Wed, 16 Feb 2011 01:28:19 + (GMT) Hin-Tak Leung > wrote: > >> I use it mainly for the AP mode, on an old machine that has just been >> resurrected after over a year, so the last it worked was on 2.6.31.x . >> There are a couple of old and new quirks: a dozen of 'We are trapped in >> zd_SigProcess, Going to release all,id=3,block=1' is new to 2.6.35.x, but >> the driver has always needed if down/up and or iwlist scan before a client >> station can connect, and that's still the case; and under extended use (a >> few days, and/or a few GB traffic) it can still bring the AP machine down. >> >> I see some great person has just had a whole bunch of patches committed to >> wireless-testing in the last few months for enabling AP mode for the >> in-kernel zd1211rw driver... so 2.6.39(?) or thereabout it would be >> generally available...so the use of the vendor driver for AP mode would >> soon not be needed any more. I'd probably give the compat-wireless a try >> soon enough. > > There is a patch at http://ankalagon.ru/notes/patch-na-zd1211rw which adds AP > mode in a very simple way, and it actually works! I had to modify the first > hunk slightly for kernel 2.6.38, modified patch below: It looks like a non-standard(ugly?) hack though, and not likely to work with standard linux AP-setup instruction... in any case, the "official" change has already happened - once something is in the wireless-testing branch of the kernel, it is just a matter of time before it reaches Linus himself. The process usually takes a few months between a change appearing in wireless testing to it appearing in the official release. -- Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ___ Zd1211-devs mailing list - http://zd1211.wiki.sourceforge.net/ Unsubscribe: https://lists.sourceforge.net/lists/listinfo/zd1211-devs
Re: [zd1211-devs] zd1211 vendor driver on 2.6.35.10
On Wed, 16 Feb 2011 01:28:19 + (GMT) Hin-Tak Leung
wrote:
> I use it mainly for the AP mode, on an old machine that has just been
> resurrected after over a year, so the last it worked was on 2.6.31.x . There
> are a couple of old and new quirks: a dozen of 'We are trapped in
> zd_SigProcess, Going to release all,id=3,block=1' is new to 2.6.35.x, but the
> driver has always needed if down/up and or iwlist scan before a client
> station can connect, and that's still the case; and under extended use (a few
> days, and/or a few GB traffic) it can still bring the AP machine down.
>
> I see some great person has just had a whole bunch of patches committed to
> wireless-testing in the last few months for enabling AP mode for the
> in-kernel zd1211rw driver... so 2.6.39(?) or thereabout it would be generally
> available...so the use of the vendor driver for AP mode would soon not be
> needed any more. I'd probably give the compat-wireless a try soon enough.
There is a patch at http://ankalagon.ru/notes/patch-na-zd1211rw which adds AP
mode in a very simple
way, and it actually works! I had to modify the first hunk slightly for kernel
2.6.38, modified patch below:
--- a/drivers/net/wireless/zd1211rw/zd_mac.c2009-08-17 01:19:38.0
+0400
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c2009-08-22 21:05:06.268498867
+0400
@@ -882,6 +882,7 @@
case NL80211_IFTYPE_MESH_POINT:
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_ADHOC:
+ case NL80211_IFTYPE_AP:
mac->type = vif->type;
break;
default:
@@ -1046,6 +1047,7 @@
dev_dbg_f(zd_mac_dev(mac), "changes: %x\n", changes);
if (mac->type == NL80211_IFTYPE_MESH_POINT ||
+ mac->type == NL80211_IFTYPE_AP ||
mac->type == NL80211_IFTYPE_ADHOC) {
associated = true;
if (changes & BSS_CHANGED_BEACON) {
@@ -1143,6 +1145,7 @@
hw->wiphy->interface_modes =
BIT(NL80211_IFTYPE_MESH_POINT) |
BIT(NL80211_IFTYPE_STATION) |
+ BIT(NL80211_IFTYPE_AP) |
BIT(NL80211_IFTYPE_ADHOC);
hw->max_signal = 100;
Regards,
Nerijus
--
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
___
Zd1211-devs mailing list - http://zd1211.wiki.sourceforge.net/
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/zd1211-devs
