This patch adds an low-level driver independent entry to read the TSF v=
alue into the debugfs of mac80211. This makes debugging the IBSS handli=
ng of wifi drivers easier.
Signed-off-by: Alina Friedrichsen <[email protected]>
---
diff -urN compat-wireless-2009-01-22.orig/net/mac80211/debugfs.c compat=
-wireless-2009-01-22/net/mac80211/debugfs.c
--- compat-wireless-2009-01-22.orig/net/mac80211/debugfs.c 2009-01-22 2=
0:51:46.000000000 +0100
+++ compat-wireless-2009-01-22/net/mac80211/debugfs.c 2009-01-23 03:49:=
45.000000000 +0100
@@ -61,6 +61,8 @@
local->wep_iv & 0xffffff);
DEBUGFS_READONLY_FILE(rate_ctrl_alg, 100, "%s",
local->rate_ctrl ? local->rate_ctrl->ops->name : "<unset>");
+DEBUGFS_READONLY_FILE(tsf, 20, "%#018llx",
+ (unsigned long long) (local->ops->get_tsf ? loca=
l->ops->get_tsf(local_to_hw(local)) : 0));
=20
/* statistics stuff */
=20
@@ -202,6 +204,7 @@
DEBUGFS_ADD(long_retry_limit);
DEBUGFS_ADD(total_ps_buffered);
DEBUGFS_ADD(wep_iv);
+ DEBUGFS_ADD(tsf);
=20
statsd =3D debugfs_create_dir("statistics", phyd);
local->debugfs.statistics =3D statsd;
@@ -255,6 +258,7 @@
DEBUGFS_DEL(long_retry_limit);
DEBUGFS_DEL(total_ps_buffered);
DEBUGFS_DEL(wep_iv);
+ DEBUGFS_DEL(tsf);
=20
DEBUGFS_STATS_DEL(transmitted_fragment_count);
DEBUGFS_STATS_DEL(multicast_transmitted_frame_count);
diff -urN compat-wireless-2009-01-22.orig/net/mac80211/ieee80211_i.h co=
mpat-wireless-2009-01-22/net/mac80211/ieee80211_i.h
--- compat-wireless-2009-01-22.orig/net/mac80211/ieee80211_i.h 2009-01-=
22 20:51:46.000000000 +0100
+++ compat-wireless-2009-01-22/net/mac80211/ieee80211_i.h 2009-01-23 03=
:55:44.000000000 +0100
@@ -744,6 +744,7 @@
struct dentry *long_retry_limit;
struct dentry *total_ps_buffered;
struct dentry *wep_iv;
+ struct dentry *tsf;
struct dentry *statistics;
struct local_debugfsdentries_statsdentries {
struct dentry *transmitted_fragment_count;
--=20
Psssst! Schon vom neuen GMX MultiMessenger geh=F6rt? Der kann`s mit all=
en: http://www.gmx.net/de/go/multimessenger
Hello Luis!
> Alina, a few notes on this series. I see you have used the mac80211
> ops get_tsf via debugfs, and then added one for ath5k and ath9k. It
> would seem a driver-specific get tsf from debugfs is not needed then.
> The set_tsf is the only one that seems to be driver specific and coul=
d
> probably go into their own debugfs. Anyway how about instead of all
> this just adding a new mac80211 ops for set_tsf for mac80211 as well
> and then just add nl80211 commands to be able to get/set/reset them?
> This way if the driver implements them it'll be through a generic
> interface and also we can start using it via iw.
If the others agree I would very happy to implement it. :)
I only don't want to change to much in the infrastructure, but this wou=
ld be the nicest way to implement it.
Regards
Alina
--=20
NUR NOCH BIS 31.01.! GMX FreeDSL - Telefonanschluss + DSL=20
f=FCr nur 16,37 EURO/mtl.!* http://dsl.gmx.de/?ac=3DOM.AD.PD003K11308T4=
569a
On Thu, Jan 22, 2009 at 8:33 PM, Alina Friedrichsen <[email protected]> wrote:
> This patch adds an low-level driver independent entry to read the TSF value into the debugfs of mac80211. This makes debugging the IBSS handling of wifi drivers easier.
>
> Signed-off-by: Alina Friedrichsen <[email protected]>
Alina, a few notes on this series. I see you have used the mac80211
ops get_tsf via debugfs, and then added one for ath5k and ath9k. It
would seem a driver-specific get tsf from debugfs is not needed then.
The set_tsf is the only one that seems to be driver specific and could
probably go into their own debugfs. Anyway how about instead of all
this just adding a new mac80211 ops for set_tsf for mac80211 as well
and then just add nl80211 commands to be able to get/set/reset them?
This way if the driver implements them it'll be through a generic
interface and also we can start using it via iw.
Luis
On Fri, 2009-01-23 at 05:47 -0800, Luis R. Rodriguez wrote:
> Alina, a few notes on this series. I see you have used the mac80211
> ops get_tsf via debugfs, and then added one for ath5k and ath9k. It
> would seem a driver-specific get tsf from debugfs is not needed then.
> The set_tsf is the only one that seems to be driver specific and could
> probably go into their own debugfs.
agree
> Anyway how about instead of all
> this just adding a new mac80211 ops for set_tsf for mac80211 as well
> and then just add nl80211 commands to be able to get/set/reset them?
> This way if the driver implements them it'll be through a generic
> interface and also we can start using it via iw.
disagree. who the hell needs that, except for debugging? Why allow
setting it anyway?
johannes