Date: Sun, 18 Mar 2012 18:25:25 +0100 From: Bernhard Schmidt <bschmidt@freebsd.org> To: Mitsuru IWASAKI <iwasaki@jp.freebsd.org> Cc: freebsd-wireless@freebsd.org Subject: Re: [patch] iwi(4) suspend/resume broken Message-ID: <201203181825.25124.bschmidt@freebsd.org> In-Reply-To: <20120319.010352.124006046.iwasaki@jp.FreeBSD.org> References: <CAAgh0_YaqjXH=yA5Aq7kTcZSik4YE8JTKSSnFRtRsBtMXnPoSQ@mail.gmail.com> <201203171232.42515.bschmidt@freebsd.org> <20120319.010352.124006046.iwasaki@jp.FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--Boundary-00=_FqhZPFMxtAg2aQz
Content-Type: Text/Plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
On Sunday 18 March 2012 17:03:52 Mitsuru IWASAKI wrote:
> Hi,
>
> > I'd really prefer calling ieee80211_stop/init() in the suspend/resume
> > functions, while the driver is not holding the lock (net80211 might
> > call into the driver again).
> >
> > I roughly checked a few things and I even think that calling
> > ieee80211_stop_all() might be enough. Still playing around though.
>
> OK, then I wouldn't commit my patches, I would wait for your patches to
> test here.
> Or may I try to hack ieee80211 layer to fix suspend/resume problem?
>
> Thanks!
Well, I came up with attached diff. It works fine on iwn and wpi
too basically, wpi has a bit of an issue with the scan after resume,
I see probe resquest/response on air but the device doesn't pick
em up sometimes.. still debugging.
Anyways, I'm pretty sure that if you are doing the same for ipw/iwi
it will just work fine. The ieee80211_resume_all/suspend_all calls
will ensure that the appropriate stop/init driver functions are called.
--
Bernhard
--Boundary-00=_FqhZPFMxtAg2aQz
Content-Type: text/x-patch;
charset="ISO-8859-1";
name="suspend.diff"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="suspend.diff"
Index: sys/dev/iwn/if_iwn.c
===================================================================
--- sys/dev/iwn/if_iwn.c (revision 233092)
+++ sys/dev/iwn/if_iwn.c (working copy)
@@ -945,13 +945,9 @@ static int
iwn_suspend(device_t dev)
{
struct iwn_softc *sc = device_get_softc(dev);
- struct ifnet *ifp = sc->sc_ifp;
- struct ieee80211com *ic = ifp->if_l2com;
- struct ieee80211vap *vap = TAILQ_FIRST(&ic->ic_vaps);
+ struct ieee80211com *ic = sc->sc_ifp->if_l2com;
- iwn_stop(sc);
- if (vap != NULL)
- ieee80211_stop(vap);
+ ieee80211_suspend_all(ic);
return 0;
}
@@ -959,20 +955,12 @@ static int
iwn_resume(device_t dev)
{
struct iwn_softc *sc = device_get_softc(dev);
- struct ifnet *ifp = sc->sc_ifp;
- struct ieee80211com *ic = ifp->if_l2com;
- struct ieee80211vap *vap = TAILQ_FIRST(&ic->ic_vaps);
+ struct ieee80211com *ic = sc->sc_ifp->if_l2com;
/* Clear device-specific "PCI retry timeout" register (41h). */
pci_write_config(dev, 0x41, 0, 1);
- if (ifp->if_flags & IFF_UP) {
- iwn_init(sc);
- if (vap != NULL)
- ieee80211_init(vap);
- if (ifp->if_drv_flags & IFF_DRV_RUNNING)
- iwn_start(ifp);
- }
+ ieee80211_resume_all(ic);
return 0;
}
Index: sys/dev/wpi/if_wpi.c
===================================================================
--- sys/dev/wpi/if_wpi.c (revision 233092)
+++ sys/dev/wpi/if_wpi.c (working copy)
@@ -1218,8 +1218,9 @@ static int
wpi_suspend(device_t dev)
{
struct wpi_softc *sc = device_get_softc(dev);
+ struct ieee80211com *ic = sc->sc_ifp->if_l2com;
- wpi_stop(sc);
+ ieee80211_suspend_all(ic);
return 0;
}
@@ -1227,15 +1228,11 @@ static int
wpi_resume(device_t dev)
{
struct wpi_softc *sc = device_get_softc(dev);
- struct ifnet *ifp = sc->sc_ifp;
+ struct ieee80211com *ic = sc->sc_ifp->if_l2com;
pci_write_config(dev, 0x41, 0, 1);
- if (ifp->if_flags & IFF_UP) {
- wpi_init(ifp->if_softc);
- if (ifp->if_drv_flags & IFF_DRV_RUNNING)
- wpi_start(ifp);
- }
+ ieee80211_resume_all(ic);
return 0;
}
--Boundary-00=_FqhZPFMxtAg2aQz--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201203181825.25124.bschmidt>