At Thu, 19 Apr 2012 20:11:53 +0200,
Takashi Iwai wrote:
>
> At Thu, 19 Apr 2012 13:55:04 -0400,
> Adam Jackson wrote:
> >
> > On Thu, 2012-04-19 at 18:10 +0200, Takashi Iwai wrote:
> >
> > > This patch adds a flag to disable the hotplug during PM operation for
> > > avoiding such a race.
> > >
On Tue, Apr 24, 2012 at 02:05:25PM -0700, Ben Widawsky wrote:
> On Tue, 24 Apr 2012 21:48:47 +0100
> Chris Wilson wrote:
>
> > We were attempting to use a per-ring spinlock whilst modifying global
> > IRQ flags. A recipe for rare missed interrupts.
> >
> > Signed-off-by: Chris Wilson
>
> Regar
On Tue, Apr 24, 2012 at 12:10:09PM -0700, Jesse Barnes wrote:
> On Tue, 24 Apr 2012 16:36:50 +0100
> Chris Wilson wrote:
>
> > Enabling the plane before we have assigned valid address means that it
> > will access random PTE (often with conflicting memory types) and cause
> > GPU lockups. However
On Tue, Apr 24, 2012 at 03:52:35PM +0100, Chris Wilson wrote:
> Currently, we only bump the LRU of an object when we bind into the GTT
> for a page-fault. As the object may be used many times before its
> mapping is zapped, we do not mark it as active as frequently as we
> should. Userspace should
On Tue, Apr 24, 2012 at 06:22:52PM +0100, Chris Wilson wrote:
> This was only used by one external caller who would just be as happy
> with evict-everything, so perform the replacement and make the function
> private.
>
> In the process we note that unbinding the inactive list should not fail,
> a
On Tue, Apr 24, 2012 at 02:51:43PM +0100, Chris Wilson wrote:
> The specs recommend that this bit be set on PineView. No reason is
> given, but it sounds like a powersaving bit that we should expect the
> BIOS to be setting...
>
> v2: Rebase on top of _MASKED_ENABLE_BIT
>
> Signed-off-by: Chris W
On Tue, Apr 24, 2012 at 08:29:42AM +0200, Daniel Vetter wrote:
> Calling these when gem assumes full control of the hw won't end
> in anything else than tears. So be a bit more paranoid here.
>
> Just serves as documentation.
>
> v2: Bail out with ENODEV as suggested by Chris Wilson.
>
> Reviewe
Hi all,
In the light of the recent oops in the error state code I've resurrected my gpu
hangman and brushed it up a bit. Together with the hangman check in i-g-t (which
now also works a bit more reliable) we'll have a basic test of our hangcheck,
reset and error state infrastructure.
Comments, fl
gpu reset is a very important piece of our infrastructure.
Unfortunately we only really it test by actually hanging the gpu,
which often has bad side-effects for the entire system. And the gpu
hang handling code is one of the rather complicated pieces of code we
have, consisting of
- hang detection
This way we can simulate a bunch of gpu hangs and run the error_state
capture code every time (without the need to reload the module).
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_debugfs.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/i9
- reduce the irq disabled section, even for a debugfs file this was
way too long.
- always disable irqs when taking the lock.
v2: Thou shalt not mistake locking for reference counting, so:
- reference count the error_state to protect from concurent freeeing.
This will be only really used in th
- need_display is always true, scrap it.
- don't reacquire the mutex to do nothing after having restored the
gem state.
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_drv.c | 19 +--
1 files changed, 5 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/i
Slightly cleans up the code and could be useful for e.g. Ben
Widawsky's hw context patches.
Cc: Ben Widawsky
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_drv.c | 43 --
1 files changed, 27 insertions(+), 16 deletions(-)
diff --git a/drivers/g
- reset the stop_rings infrastructure while resetting the hw to
avoid angering the hangcheck right away (and potentially declaring
the gpu permanently wedged).
- ignore reset failures when hanging due to the hangman - we don't
have reset code for all generations.
Signed-off-by: Daniel Vette
On Wed, 25 Apr 2012 13:57:07 +0200, Daniel Vetter
wrote:
> Hi all,
>
> In the light of the recent oops in the error state code I've resurrected my
> gpu
> hangman and brushed it up a bit. Together with the hangman check in i-g-t
> (which
> now also works a bit more reliable) we'll have a basic
Slightly cleans up the code and could be useful for e.g. Ben
Widawsky's hw context patches.
v2: New colours!
Cc: Ben Widawsky
Signed-Off-by: Daniel Vetter
---
drivers/gpu/drm/i915/i915_drv.c | 43 --
1 files changed, 27 insertions(+), 16 deletions(-)
diff
On Wed, 25 Apr 2012 13:57:11 +0200
Daniel Vetter wrote:
> - need_display is always true, scrap it.
> - don't reacquire the mutex to do nothing after having restored the
> gem state.
>
Actually I think we generally *don't* need to reset display. It's
currently there just because we haven't tr
On Wed, 25 Apr 2012 15:14:05 +0200
Daniel Vetter wrote:
> Slightly cleans up the code and could be useful for e.g. Ben
> Widawsky's hw context patches.
>
> v2: New colours!
>
> Cc: Ben Widawsky
> Signed-Off-by: Daniel Vetter
I have this queued for both HW context, and DPF, but you've beaten
On Fri, 20 Apr 2012 21:03:35 +0200, Daniel Vetter
wrote:
> If we try to do that and the scanlines just wouldn't advance, we
> busy-hang the machine holding the modeset mutex. Not great for
> debugging.
>
> References: https://bugzilla.kernel.org/show_bug.cgi?id=43020
> Signed-off-by: Daniel Vett
i915.enable_backlight=0 can be used to disable i915 backlight control
and the /sys/class/backlight/intel_backlight interface -- useful for
systems where intel_backight conflicts with BIOS backlight control.
BugLink: https://launchpad.net/bugs/954661
Signed-off-by: Kamal Mostafa
---
drivers/gpu/d
From: Robert Hooker
Dell XPS 13z exhibits problems (backlight flashing/pulsating) when
intel_backlight is enabled at all, so disable it.
BugLink: https://launchpad.net/bugs/954661
Signed-off-by: Robert Hooker
Signed-off-by: Kamal Mostafa
---
drivers/gpu/drm/i915/intel_display.c | 17 +++
The GT power programing guide recommends always unsuppressing the down
interrupt when programming any render frequency above the minimum. This
is because the GT unit doesn't emit interrupts whilst it is asleep which
can lead to the down interrupt being lost. Also due to our own
coalescing of interr
On Wed, Apr 25, 2012 at 06:14:37PM +0100, Chris Wilson wrote:
> On Fri, 20 Apr 2012 21:03:35 +0200, Daniel Vetter
> wrote:
> > If we try to do that and the scanlines just wouldn't advance, we
> > busy-hang the machine holding the modeset mutex. Not great for
> > debugging.
> >
> > References: ht
... by writing (anything) to i915_error_state.
This way we can simulate a bunch of gpu hangs and run the error_state
capture code every time (without the need to reload the module).
To make that happen we need to abandon the simple seq_file wrappers
provided by the drm core. While at it, but the
On Wed, Apr 25, 2012 at 09:18:21AM -0700, Jesse Barnes wrote:
> On Wed, 25 Apr 2012 13:57:11 +0200
> Daniel Vetter wrote:
>
> > - need_display is always true, scrap it.
> > - don't reacquire the mutex to do nothing after having restored the
> > gem state.
> >
>
> Actually I think we generally
On Wed, 25 Apr 2012 22:27:12 +0200
Daniel Vetter wrote:
> On Wed, Apr 25, 2012 at 09:18:21AM -0700, Jesse Barnes wrote:
> > On Wed, 25 Apr 2012 13:57:11 +0200
> > Daniel Vetter wrote:
> >
> > > - need_display is always true, scrap it.
> > > - don't reacquire the mutex to do nothing after having
On Tue, 24 Apr 2012 22:59:45 +0100
Chris Wilson wrote:
> On later gen3, you are able to select the meaning of the FlipPending
> status bit in IIR and change it to FlipDone. This was sometimes done by
> the BIOS leading to confusion on just how pageflipping worked on gen3.
> Simplify the implement
On Tue, 24 Apr 2012 22:59:46 +0100
Chris Wilson wrote:
> And a couple of miscellaneous cleanups to the main body of the IRQ loop;
> move per-loop condition variables within the scope of the loop and move
> the old DRI1 breadcrumb to the tail of the function and so only execute
> it once.
>
> Sig
On Tue, 24 Apr 2012 22:59:47 +0100
Chris Wilson wrote:
> A couple of miscellaneous cleanups as well to move per-loop condition
> variables within the scope of the loop and the update of the DRI1
> breadcrumb to the tail of the function.
>
> Signed-off-by: Chris Wilson
> ---
Reviewed-by: Jesse
On Tue, 24 Apr 2012 22:59:48 +0100
Chris Wilson wrote:
> Signed-off-by: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_irq.c |4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> index 878f1c3..a4e074
On Tue, 24 Apr 2012 22:59:49 +0100
Chris Wilson wrote:
> Bring the for-each-pipe loops together so that the code is easier on the
> eyes.
>
> Signed-off-by: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_irq.c | 11 ---
> 1 file changed, 4 insertions(+), 7 deletions(-)
>
> diff --gi
On Tue, 24 Apr 2012 22:59:50 +0100
Chris Wilson wrote:
> We appear to allow too many pending pageflips as evidenced by an
> apparent pin-leak. So borrow the pageflip completion logic from i8xx for
> handling PendingFlip in a robust manner.
>
> v2: Address Jesse's reminders about the nuances of g
On Tue, 24 Apr 2012 22:59:51 +0100
Chris Wilson wrote:
> Since there is only one remaining user of I915_INTERRUPT_ENABLE_FIX,
> expand it at the callsite. Quoting Jesse Barnes:
>
> "I'd really like to get rid of these defines at the top of i915_irq.c.
> Some are unused and the others just make y
On Tue, 24 Apr 2012 22:59:52 +0100
Chris Wilson wrote:
> Signed-off-by: Chris Wilson
> ---
> drivers/gpu/drm/i915/i915_irq.c | 11 ---
> 1 file changed, 11 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> index a1150b7..d3500c7 100644
On Wed, Apr 25, 2012 at 01:34:45PM -0700, Jesse Barnes wrote:
> On Wed, 25 Apr 2012 22:27:12 +0200
> Daniel Vetter wrote:
>
> > On Wed, Apr 25, 2012 at 09:18:21AM -0700, Jesse Barnes wrote:
> > > On Wed, 25 Apr 2012 13:57:11 +0200
> > > Daniel Vetter wrote:
> > >
> > > > - need_display is alway
On Wed, 25 Apr 2012 13:55:12 -0700, Jesse Barnes
wrote:
> On Tue, 24 Apr 2012 22:59:51 +0100
> Chris Wilson wrote:
> > + dev_priv->irq_mask = ~(I915_ASLE_INTERRUPT |
> > + I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
> > + I915_DISPLAY_PIPE_B_EVENT_IN
On Wed, 25 Apr 2012 22:27:12 +0200, Daniel Vetter wrote:
> On Wed, Apr 25, 2012 at 09:18:21AM -0700, Jesse Barnes wrote:
> > On Wed, 25 Apr 2012 13:57:11 +0200
> > Daniel Vetter wrote:
> >
> > > - need_display is always true, scrap it.
> > > - don't reacquire the mutex to do nothing after having
On Wed, Apr 25, 2012 at 08:57, Daniel Vetter wrote:
> gpu reset is a very important piece of our infrastructure.
> Unfortunately we only really it test by actually hanging the gpu,
> which often has bad side-effects for the entire system. And the gpu
> hang handling code is one of the rather comp
We finished a new round of kernel testing. The version of kernel is:
Kernel: (drm-intel-testing)d5eef3cd2b702d875e35673f46f8c6e0e874fb03
Some additional commit info:
Merge: 0e31d4e 99d9acd
Author: Daniel Vetter
Date: Sat Apr 21 17:50:35 2012 +0200
We covered the platform IvyBridge, SandyBridge,
39 matches
Mail list logo
