From: Dave Airlie airlied@redhat.com
So we have a few places where the drm drivers would like to sleep to be nice to the system, mainly in the modesetting paths, but we also have two cases were atomic modesetting must take place, panic writing and kernel debugger. So provide a central inline to determine if a sleep or delay should be used and use this in the intel and radeon drivers.
Based on patch from Michel Dänzer michel.daenzer@amd.com
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=43941
Signed-off-by: Dave Airlie airlied@redhat.com --- drivers/gpu/drm/i915/intel_drv.h | 4 ++-- drivers/gpu/drm/radeon/atom.c | 2 ++ include/drm/drmP.h | 8 ++++++++ 3 files changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index a1b4343..cf2326b 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -39,7 +39,7 @@ ret__ = -ETIMEDOUT; \ break; \ } \ - if (W && !(in_atomic() || in_dbg_master())) msleep(W); \ + if (W && drm_can_sleep()) msleep(W); \ } \ ret__; \ }) @@ -48,7 +48,7 @@ #define wait_for_atomic(COND, MS) _wait_for(COND, MS, 0)
#define MSLEEP(x) do { \ - if (in_dbg_master()) \ + if (!drm_can_sleeep()) \ mdelay(x); \ else \ msleep(x); \ diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c index 14cc88a..d1bd239 100644 --- a/drivers/gpu/drm/radeon/atom.c +++ b/drivers/gpu/drm/radeon/atom.c @@ -665,6 +665,8 @@ static void atom_op_delay(atom_exec_context *ctx, int *ptr, int arg) SDEBUG(" count: %d\n", count); if (arg == ATOM_UNIT_MICROSEC) udelay(count); + else if (!drm_can_sleep()) + mdelay(count); else msleep(count); } diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 1f9e951..f1c01bc 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -1696,5 +1696,13 @@ extern void drm_platform_exit(struct drm_driver *driver, struct platform_device extern int drm_get_platform_dev(struct platform_device *pdev, struct drm_driver *driver);
+/* returns true if currently okay to sleep */ +static __inline__ bool drm_can_sleep(void) +{ + if (in_atomic() || in_dbg_master() || irqs_disabled()) + return false; + return true; +} + #endif /* __KERNEL__ */ #endif
On Don, 2012-01-05 at 09:57 +0000, Dave Airlie wrote:
From: Dave Airlie airlied@redhat.com
So we have a few places where the drm drivers would like to sleep to be nice to the system, mainly in the modesetting paths, but we also have two cases were atomic modesetting must take place, panic writing and kernel debugger. So provide a central inline to determine if a sleep or delay should be used and use this in the intel and radeon drivers.
Based on patch from Michel Dänzer michel.daenzer@amd.com
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=43941
Signed-off-by: Dave Airlie airlied@redhat.com
drivers/gpu/drm/i915/intel_drv.h | 4 ++-- drivers/gpu/drm/radeon/atom.c | 2 ++ include/drm/drmP.h | 8 ++++++++ 3 files changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index a1b4343..cf2326b 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h
[...]
@@ -48,7 +48,7 @@ #define wait_for_atomic(COND, MS) _wait_for(COND, MS, 0)
#define MSLEEP(x) do { \
- if (in_dbg_master()) \
- if (!drm_can_sleeep()) \ mdelay(x); \
So for MSLEEP it's not enough if it can sleep, it has to be able to 'sleeep'? ;) Looks like this macro is unused.
Looks good to me otherwise, thanks.
On Thu, Jan 05, 2012 at 09:57:56AM +0000, Dave Airlie wrote:
From: Dave Airlie airlied@redhat.com
So we have a few places where the drm drivers would like to sleep to be nice to the system, mainly in the modesetting paths, but we also have two cases were atomic modesetting must take place, panic writing and kernel debugger. So provide a central inline to determine if a sleep or delay should be used and use this in the intel and radeon drivers.
Based on patch from Michel Dänzer michel.daenzer@amd.com
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=43941
Signed-off-by: Dave Airlie airlied@redhat.com
If MSLEEP is indeed unused, I think it should just die - we have way too much yelling cruft from dri1 yonder lying around that hides important details like this. Otherwise
Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
On Thu, 5 Jan 2012 17:00:52 +0100, Daniel Vetter daniel@ffwll.ch wrote:
If MSLEEP is indeed unused, I think it should just die - we have way too much yelling cruft from dri1 yonder lying around that hides important details like this. Otherwise
It was used, they just happened to all be redundant or mispellings of wait-for-vblank. All the remaining sleeps now look to be away from the atomic paths, and so MSLEEP can die for the time being. -Chris
So we have a few places where the drm drivers would like to sleep to be nice to the system, mainly in the modesetting paths, but we also have two cases were atomic modesetting must take place, panic writing and kernel debugger. So provide a central inline to determine if a sleep or delay should be used and use this in the intel and radeon drivers.
Based on patch from Michel Dänzer michel.daenzer@amd.com
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=43941
Signed-off-by: Dave Airlie airlied@redhat.com
If MSLEEP is indeed unused, I think it should just die - we have way too much yelling cruft from dri1 yonder lying around that hides important details like this. Otherwise
Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
I've pushed the version that just drops MSLEEP since nobody used it.
Thanks, Dave.
dri-devel@lists.freedesktop.org