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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/gpu/drm/gma500/psb_lid.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/gma500/psb_lid.c b/drivers/gpu/drm/gma500/psb_lid.c index 1d2ebb5e530f..be6dda58fcae 100644 --- a/drivers/gpu/drm/gma500/psb_lid.c +++ b/drivers/gpu/drm/gma500/psb_lid.c @@ -23,9 +23,9 @@ #include "psb_intel_reg.h" #include <linux/spinlock.h>
-static void psb_lid_timer_func(unsigned long data) +static void psb_lid_timer_func(struct timer_list *t) { - struct drm_psb_private * dev_priv = (struct drm_psb_private *)data; + struct drm_psb_private *dev_priv = from_timer(dev_priv, t, lid_timer); struct drm_device *dev = (struct drm_device *)dev_priv->dev; struct timer_list *lid_timer = &dev_priv->lid_timer; unsigned long irq_flags; @@ -77,10 +77,8 @@ void psb_lid_timer_init(struct drm_psb_private *dev_priv) spin_lock_init(&dev_priv->lid_lock); spin_lock_irqsave(&dev_priv->lid_lock, irq_flags);
- init_timer(lid_timer); + timer_setup(lid_timer, psb_lid_timer_func, 0);
- lid_timer->data = (unsigned long)dev_priv; - lid_timer->function = psb_lid_timer_func; lid_timer->expires = jiffies + PSB_LID_DELAY;
add_timer(lid_timer);
On Tue, Oct 24, 2017 at 08:16:09AM -0700, 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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org
Do you expect drm folks to apply this, or is this part of a larger refactoring?
A notch more context in the commit message would help ... -Daniel
drivers/gpu/drm/gma500/psb_lid.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/gma500/psb_lid.c b/drivers/gpu/drm/gma500/psb_lid.c index 1d2ebb5e530f..be6dda58fcae 100644 --- a/drivers/gpu/drm/gma500/psb_lid.c +++ b/drivers/gpu/drm/gma500/psb_lid.c @@ -23,9 +23,9 @@ #include "psb_intel_reg.h" #include <linux/spinlock.h>
-static void psb_lid_timer_func(unsigned long data) +static void psb_lid_timer_func(struct timer_list *t) {
- struct drm_psb_private * dev_priv = (struct drm_psb_private *)data;
- struct drm_psb_private *dev_priv = from_timer(dev_priv, t, lid_timer); struct drm_device *dev = (struct drm_device *)dev_priv->dev; struct timer_list *lid_timer = &dev_priv->lid_timer; unsigned long irq_flags;
@@ -77,10 +77,8 @@ void psb_lid_timer_init(struct drm_psb_private *dev_priv) spin_lock_init(&dev_priv->lid_lock); spin_lock_irqsave(&dev_priv->lid_lock, irq_flags);
- init_timer(lid_timer);
- timer_setup(lid_timer, psb_lid_timer_func, 0);
lid_timer->data = (unsigned long)dev_priv;
lid_timer->function = psb_lid_timer_func; lid_timer->expires = jiffies + PSB_LID_DELAY;
add_timer(lid_timer);
-- 2.7.4
-- Kees Cook Pixel Security _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
On Mon, Oct 30, 2017 at 3:08 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Tue, Oct 24, 2017 at 08:16:09AM -0700, 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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org
Do you expect drm folks to apply this, or is this part of a larger refactoring?
If the drm tree includes -rc3, you can carry these. If you don't want to carry these and want the timer tree to carry them, we can do that too.
A notch more context in the commit message would help ...
Sorry about that, my added context for this go lost in later conversion patches.
-Kees
On Mon, Oct 30, 2017 at 03:05:29PM -0700, Kees Cook wrote:
On Mon, Oct 30, 2017 at 3:08 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Tue, Oct 24, 2017 at 08:16:09AM -0700, 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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org
Do you expect drm folks to apply this, or is this part of a larger refactoring?
If the drm tree includes -rc3, you can carry these. If you don't want to carry these and want the timer tree to carry them, we can do that too.
Applied to drm-misc-next for 4.16 (we're way past freeze for 4.15 already).
Thanks, Daniel
A notch more context in the commit message would help ...
Sorry about that, my added context for this go lost in later conversion patches.
-Kees
-- Kees Cook Pixel Security _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
On Tue, Oct 31, 2017 at 3:18 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Mon, Oct 30, 2017 at 03:05:29PM -0700, Kees Cook wrote:
On Mon, Oct 30, 2017 at 3:08 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Tue, Oct 24, 2017 at 08:16:09AM -0700, 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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org
Do you expect drm folks to apply this, or is this part of a larger refactoring?
If the drm tree includes -rc3, you can carry these. If you don't want to carry these and want the timer tree to carry them, we can do that too.
Applied to drm-misc-next for 4.16 (we're way past freeze for 4.15 already).
Since this is one of the few remaining "non-trivial" users of the ancient init_timer() API, would you mind if the timers tree carried this for 4.15? I'm trying to entirely remove the init_timer() API (and if I can, remove the old setup_*timer() API too).
Thanks!
-Kees
On Tue, Oct 31, 2017 at 08:08:14AM -0700, Kees Cook wrote:
On Tue, Oct 31, 2017 at 3:18 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Mon, Oct 30, 2017 at 03:05:29PM -0700, Kees Cook wrote:
On Mon, Oct 30, 2017 at 3:08 AM, Daniel Vetter daniel@ffwll.ch wrote:
On Tue, Oct 24, 2017 at 08:16:09AM -0700, 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: Patrik Jakobsson patrik.r.jakobsson@gmail.com Cc: David Airlie airlied@linux.ie Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org
Do you expect drm folks to apply this, or is this part of a larger refactoring?
If the drm tree includes -rc3, you can carry these. If you don't want to carry these and want the timer tree to carry them, we can do that too.
Applied to drm-misc-next for 4.16 (we're way past freeze for 4.15 already).
Since this is one of the few remaining "non-trivial" users of the ancient init_timer() API, would you mind if the timers tree carried this for 4.15? I'm trying to entirely remove the init_timer() API (and if I can, remove the old setup_*timer() API too).
I was contemplating before applying it whether I should ask ...
Oh well, problem is that drm-misc is non-rebasing, but you can just apply it twice. git usually figures it out.
Acked-by: Daniel Vetter daniel.vetter@ffwll.ch
in case you do so. -Daniel
dri-devel@lists.freedesktop.org
-
Daniel Vetter -
Kees Cook