| Message ID | 20171024153521.GA112921@beast (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, 24 Oct 2017 17:35:21 +0200, Kees Cook wrote: > > In preparation for unconditionally passing the struct timer_list pointer to > all timer callbacks, switch to using the new timer_setup() and from_timer() > to pass the timer pointer explicitly. > > Cc: Jaroslav Kysela <perex@perex.cz> > Cc: Takashi Iwai <tiwai@suse.com> > Cc: David Howells <dhowells@redhat.com> > Cc: alsa-devel@alsa-project.org > Signed-off-by: Kees Cook <keescook@chromium.org> I'll skip this as sound/oss/* is already disabled in recent kernels in kbuild, and we're going to remove the whole code soon later. thanks, Takashi
On Tue, Oct 24, 2017 at 12:25 PM, Takashi Iwai <tiwai@suse.de> wrote: > On Tue, 24 Oct 2017 17:35:21 +0200, > Kees Cook wrote: >> >> In preparation for unconditionally passing the struct timer_list pointer to >> all timer callbacks, switch to using the new timer_setup() and from_timer() >> to pass the timer pointer explicitly. >> >> Cc: Jaroslav Kysela <perex@perex.cz> >> Cc: Takashi Iwai <tiwai@suse.com> >> Cc: David Howells <dhowells@redhat.com> >> Cc: alsa-devel@alsa-project.org >> Signed-off-by: Kees Cook <keescook@chromium.org> > > I'll skip this as sound/oss/* is already disabled in recent kernels in > kbuild, and we're going to remove the whole code soon later. At the end of these conversions, the old init_timer() and setup_timer() APIs will be removed, which would break anything still trying to use it (obviously). But if sound/oss/* is _already_ unbuildable, then sure, this is fine by me to leave out. :) Thanks! -Kees
On Tue, 24 Oct 2017 22:07:34 +0200, Kees Cook wrote: > > On Tue, Oct 24, 2017 at 12:25 PM, Takashi Iwai <tiwai@suse.de> wrote: > > On Tue, 24 Oct 2017 17:35:21 +0200, > > Kees Cook wrote: > >> > >> In preparation for unconditionally passing the struct timer_list pointer to > >> all timer callbacks, switch to using the new timer_setup() and from_timer() > >> to pass the timer pointer explicitly. > >> > >> Cc: Jaroslav Kysela <perex@perex.cz> > >> Cc: Takashi Iwai <tiwai@suse.com> > >> Cc: David Howells <dhowells@redhat.com> > >> Cc: alsa-devel@alsa-project.org > >> Signed-off-by: Kees Cook <keescook@chromium.org> > > > > I'll skip this as sound/oss/* is already disabled in recent kernels in > > kbuild, and we're going to remove the whole code soon later. > > At the end of these conversions, the old init_timer() and > setup_timer() APIs will be removed, which would break anything still > trying to use it (obviously). But if sound/oss/* is _already_ > unbuildable, then sure, this is fine by me to leave out. :) I see a few still remaining users of setup_timer() in sound/*. Do you already have conversion patches? They are all trivial, so I can cook up in minutes, too. thanks, Takashi
On Wed, Oct 25, 2017 at 3:06 AM, Takashi Iwai <tiwai@suse.de> wrote: > On Tue, 24 Oct 2017 22:07:34 +0200, > Kees Cook wrote: >> >> On Tue, Oct 24, 2017 at 12:25 PM, Takashi Iwai <tiwai@suse.de> wrote: >> > On Tue, 24 Oct 2017 17:35:21 +0200, >> > Kees Cook wrote: >> >> >> >> In preparation for unconditionally passing the struct timer_list pointer to >> >> all timer callbacks, switch to using the new timer_setup() and from_timer() >> >> to pass the timer pointer explicitly. >> >> >> >> Cc: Jaroslav Kysela <perex@perex.cz> >> >> Cc: Takashi Iwai <tiwai@suse.com> >> >> Cc: David Howells <dhowells@redhat.com> >> >> Cc: alsa-devel@alsa-project.org >> >> Signed-off-by: Kees Cook <keescook@chromium.org> >> > >> > I'll skip this as sound/oss/* is already disabled in recent kernels in >> > kbuild, and we're going to remove the whole code soon later. >> >> At the end of these conversions, the old init_timer() and >> setup_timer() APIs will be removed, which would break anything still >> trying to use it (obviously). But if sound/oss/* is _already_ >> unbuildable, then sure, this is fine by me to leave out. :) > > I see a few still remaining users of setup_timer() in sound/*. > Do you already have conversion patches? They are all trivial, so I > can cook up in minutes, too. I haven't sent the trivial ones (since I am expecting to do a treewide mechanical change for those). I can send you the rest as a patch instead, if you want? -Kees
On Wed, 25 Oct 2017 12:17:25 +0200, Kees Cook wrote: > > On Wed, Oct 25, 2017 at 3:06 AM, Takashi Iwai <tiwai@suse.de> wrote: > > On Tue, 24 Oct 2017 22:07:34 +0200, > > Kees Cook wrote: > >> > >> On Tue, Oct 24, 2017 at 12:25 PM, Takashi Iwai <tiwai@suse.de> wrote: > >> > On Tue, 24 Oct 2017 17:35:21 +0200, > >> > Kees Cook wrote: > >> >> > >> >> In preparation for unconditionally passing the struct timer_list pointer to > >> >> all timer callbacks, switch to using the new timer_setup() and from_timer() > >> >> to pass the timer pointer explicitly. > >> >> > >> >> Cc: Jaroslav Kysela <perex@perex.cz> > >> >> Cc: Takashi Iwai <tiwai@suse.com> > >> >> Cc: David Howells <dhowells@redhat.com> > >> >> Cc: alsa-devel@alsa-project.org > >> >> Signed-off-by: Kees Cook <keescook@chromium.org> > >> > > >> > I'll skip this as sound/oss/* is already disabled in recent kernels in > >> > kbuild, and we're going to remove the whole code soon later. > >> > >> At the end of these conversions, the old init_timer() and > >> setup_timer() APIs will be removed, which would break anything still > >> trying to use it (obviously). But if sound/oss/* is _already_ > >> unbuildable, then sure, this is fine by me to leave out. :) > > > > I see a few still remaining users of setup_timer() in sound/*. > > Do you already have conversion patches? They are all trivial, so I > > can cook up in minutes, too. > > I haven't sent the trivial ones (since I am expecting to do a treewide > mechanical change for those). I can send you the rest as a patch > instead, if you want? Yeah, sure, earlier is easier for me :) Takashi
diff --git a/sound/oss/waveartist.c b/sound/oss/waveartist.c index 4f0c3a232e41..c051c8e6023e 100644 --- a/sound/oss/waveartist.c +++ b/sound/oss/waveartist.c @@ -151,6 +151,7 @@ static DEFINE_SPINLOCK(waveartist_lock); #define machine_is_netwinder() 0 #else static struct timer_list vnc_timer; +static int timer_nr_waveartist_devs; static void vnc_configure_mixer(struct wavnc_info *devc, unsigned int input_mask); static int vnc_private_ioctl(int dev, unsigned int cmd, int __user *arg); @@ -1414,8 +1415,8 @@ attach_waveartist(struct address_info *hw, const struct waveartist_mixer_info *m else { #ifdef CONFIG_ARCH_NETWINDER if (machine_is_netwinder()) { - setup_timer(&vnc_timer, vnc_slider_tick, - nr_waveartist_devs); + timer_nr_waveartist_devs = nr_waveartist_devs; + timer_setup(&vnc_timer, vnc_slider_tick, 0); mod_timer(&vnc_timer, jiffies); vnc_configure_mixer(devc, 0); @@ -1799,11 +1800,11 @@ vnc_slider(struct wavnc_info *devc) } static void -vnc_slider_tick(unsigned long data) +vnc_slider_tick(struct timer_list *unused) { int next_timeout; - if (vnc_slider(adev_info + data)) + if (vnc_slider(adev_info + timer_nr_waveartist_devs)) next_timeout = 5; // mixer reported change else next_timeout = VNC_TIMER_PERIOD;
In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: David Howells <dhowells@redhat.com> Cc: alsa-devel@alsa-project.org Signed-off-by: Kees Cook <keescook@chromium.org> --- sound/oss/waveartist.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)