Withtout this, ring initialization fails reliabily during resume with
[drm:init_ring_common] *ERROR* render ring initialization failed ctl 0001f001 head ffffff8804 tail 00000000 start 000e4000
This is not a complete fix, but it is verified to make the ring
initialization failures during resume much less likely.
We were not able to root-cause this bug (likely HW-specific to Gen4 chips)
yet. This is therefore used as a ducttape before problem is fully
understood and proper fix created, so that people don't suffer from
completely unusable systems in the meantime.
The discussion and debugging is happening at
https://bugs.freedesktop.org/show_bug.cgi?id=76554
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This particular nasty presented itself while trying to register the
intelfb device (intel_fbdev.c). During the process of registering the device
the driver will disable the crtc via i9xx_crtc_disable. These will
also disable the panel using the generic mipi panel functions in
dsi_mod_vbt_generic.c. The stale MIPI generic data sequence pointers would
cause a crash within those functions. However, all of this is happening
while console_lock is held from do_register_framebuffer inside fbcon.c. Which
means that you got kernel log and just the device appearing to reboot/hang for
no apparent reason.
The fault started from the FB_EVENT_FB_REGISTERED event using the
fb_notifier_call_chain call in fbcon.c.
This regression has been introduced in
commit d3b542fcfc
Author: Shobhit Kumar <shobhit.kumar@intel.com>
Date: Mon Apr 14 11:00:34 2014 +0530
drm/i915: Add parsing support for new MIPI blocks in VBT
Cc: Shobhit Kumar <shobhit.kumar@intel.com>
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Reviewed-by: Shobhit Kumar <shobhit.kumar@intel.com>
[danvet: Add regression citation.]
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We might be leaving the PGU Frequency (and thus vnn) high during the suspend.
Flusing the delayed work queue should take care of this.
Signed-off-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
BDW has many other Display Engine interrupts and GT interrupts registers.
Collecting it properly on gpu_error_state.
On debugfs all was properly listed already but besides we were also listing old
DEIER and GTIER that doesn't exist on BDW anymore. This was causing
unclaimed register messages
v2: Fix small issues of first version and don't read DEIER regs when pipe's
power well is disabled
v3: bikeshed accepted: use enum pipe pipe instead of int i for pipe interection
v4: Ben notice previous version was checking for display_power_enabled without
using propper locks. Using _unlocked version isn't reliable and we cannot
get this registers when power well is off. So let's avoid getting all DE_IER
per pipe for now. If someone think this is an useful information it can be
added later.
v5: Ben: put back debugfs stuff that might be coverred by pm_get and use
gen >= 8 trying to predict future.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81701
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: (v3) Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
If the actual head has progressed forward inside a batch (request),
don't accumulate hangcheck score.
As the hangcheck score in increased only by acthd jumping backwards,
the result is that we only declare an active batch as stuck if it is
trapped inside a loop. Or that the looping will dominate the batch
progression so that it overcomes the bonus that forward progress gives.
v2: Improved commit message (Chris Wilson)
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
[danvet: s/active_loop/active (loop)/ as requested by Chris.]
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
On Broadwell, any PIPE_CONTROL with the "State Cache Invalidate" bit set
must be preceded by a PIPE_CONTROL with the "CS Stall" bit set.
Documented on the BSpec 3D workarounds page.
Reviewed-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
[vsyrjala: add chv w/a note too]
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We'll want to reuse this for a workaround.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[danvet: Rmove now unused int.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The DDL registers can hold 7bit numbers. Make the most of those seven
bits by adjusting the threshold where we switch between the 64 vs. 32
precision multipliers.
Also we compute 'entries' to make the decision about precision, and then
we recompute the same value to calculate the actual drain latency. Just
use the already calculate 'entries' there.
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
GTIER and DEIER doesn't have same interface on HSW so this "or" operation
makes the information provided useless.
v2: since we have gtier variable already let's split for everybody
and avoid the strange | op.
Also avoid overriding the value that was set for vlv. In this case I
believe that we should reorganize the whole function, but I'll respect
the comment that ask to not touch the order and let this organization
work to be done later.
v3: moving VLV check to the right place.
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Users often can't do anything about this since their vendors stopped
providing BIOS updates. Also we seem to be able to hack around it
with increased latency values, and thus far the only reports have
been for screens with really high resolutions. So tune it down to a
level where only developers can see it.
Also drop some of the end-user fluff.
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Power users spot this and then get adventurous and try to adjust
module driver options. Nothing good ever came out of that, so
hide it better.
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since I've reworked psr support to no longer require x-tiling we don't
check any state protected by the Giant GEM Lock. So drop that check.
Also boo for lockdep_assert_held for not yelling when lockdep is
disabled.
Cc: Paulo Zanoni <przanoni@gmail.com>
Reported-by: Paulo Zanoni <przanoni@gmail.com>
Acked-by: Paulo Zanoni <przanoni@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Fix signal_offset when recording semaphore state on BDW.
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Just like during booting the BIOS can leave the VDD bit enabled after
system resume. So apply the same state sanitization there too. This
fixes a problem where after resume the port power domain refcount gets
unbalanced.
v2:
- unchanged
v3:
- call edp sanitizing from the encoder reset handler (Daniel)
Reported-and-tested-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Check in vlv_crtc_clock_get if DPLL is enabled before calling dpio read.
It will not be enabled for DSI and avoid dpio read WARN dumps.
Absence of ->get_config was causing other WARN dumps as well. Update
dpll_hw_state as well correctly
v2: Address review comments by Daniel
- Check if DPLL is enabled rather than checking pipe output type
- set adjusted_mode->flags to 0 in compute_config rather than using
pipe_config->quirks
- Add helper function in intel_dsi_pll.c and use that in intel_dsi.c
- updated dpll_hw_state correctly
- Updated commit message and title
v3: Address review comments by Imre
- Proper masking of P1, M1 fields while computing divisors
- assert in case of bpp mismatch
- guard for divide by 0 while computing pclk
- Use ARRAY_SIZE instead of direct calculation
Signed-off-by: Shobhit Kumar <shobhit.kumar@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This will be needed by an upcoming patch too that needs to sanitize the
VDD state during resume. The additional async disabling is only needed
for the resume path, here it doesn't make a difference since we enable
VDD right after the sanitize call.
v2:
- don't set intel_dp ptr for non-eDP encoders (Ville)
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Ensure that the DSI packets for a particular sequence are completely
sent before going ahead in the enabling or disabling of the panel
Signed-off-by: Shobhit Kumar <shobhit.kumar@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Gcc warns that addr might be used uninitialized. It may not, but I see
why gcc gets confused.
Additionally, hiding code with side-effects inside WARN_ON() argument
seems uncool, so I moved it outside.
Signed-off-by: Pavel Machek <pavel@ucw.cz>
[danvet: Add obligatory /* shuts up gcc */ comment.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Final feature pull for 3.17.
drm-intel-next-2014-07-25:
- Ditch UMS support (well just the config option for now)
- Prep work for future platforms (Sonika Jindal, Damien)
- runtime pm/soix fixes (Paulo, Jesse)
- psr tracking improvements, locking fixes, now enabled by default!
- rps fixes for chv (Deepak, Ville)
- drm core patches for rotation support (Ville, Sagar Kamble) - the i915 parts
unfortunately didn't make it yet
- userptr fixes (Chris)
- minimum backlight brightness (Jani), acked long ago by Matthew Garret on irc -
I've forgotten about this patch :(
QA is a bit unhappy about the DP MST stuff since it broke hpd testing a
bit, but otherwise looks sane. I've backmerged drm-next to resolve
conflicts with the mst stuff, which means the new tag itself doesn't
contain the overview as usual.
* tag 'drm-intel-next-2014-07-25-merged' of git://anongit.freedesktop.org/drm-intel: (75 commits)
drm/i915/userptr: Keep spin_lock/unlock in the same block
drm/i915: Allow overlapping userptr objects
drm/i915: Ditch UMS config option
drm/i915: respect the VBT minimum backlight brightness
drm/i915: extract backlight minimum brightness from VBT
drm/i915: Replace HAS_PCH_SPLIT which incorrectly lets some platforms in
drm/i915: Returning from increase/decrease of pllclock when invalid
drm/i915: Setting legacy palette correctly for different platforms
drm/i915: Avoid incorrect returning for some platforms
drm/i915: Writing proper check for reading of pipe status reg
drm/i915: Returning the right VGA control reg for platforms
drm/i915: Allowing changing of wm latencies for valid platforms
drm/i915: Adding HAS_GMCH_DISPLAY macro
drm/i915: Fix possible overflow when recording semaphore states.
drm/i915: Do not unmap object unless no other VMAs reference it
drm/i915: remove plane/cursor/pipe assertions from intel_crtc_disable
drm/i915: Reorder ctx unref on ppgtt cleanup
drm/i915/error: Check the potential ctx obj's vm
drm/i915: Fix printing proper min/min/rpe values in debugfs
drm/i915: BDW can also detect unclaimed registers
...
On HSW+, the digital encoders are shared between HDMI and DP outputs,
with one encoder masquerading as both. The VBT should tell us if we need
to have DP or HDMI support on a particular port, but if we don't have DP
support and we enable the DP hpd pulse handler then we cause an oops.
Don't hook up the DP hpd handling if we don't have a DP port.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81856
Reported-by: Intel QA Team.
Signed-off-by: Dave Airlie <airlied@redhat.com> # v1
[ickle: Fix the error handling after a malloc failure]
Reviewed-by: Dave Airlie <airlied@redhat.com>
Cc: Paulo Zanoni <przanoni@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Pull in drm-next with Dave's DP MST support so that I can merge some
conflicting patches which also touch the driver load sequencing around
interrupt handling.
Conflicts:
drivers/gpu/drm/i915/intel_display.c
drivers/gpu/drm/i915/intel_dp.c
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move the code around in order to acquire and release the spinlock in the
same function and in the same block. This keeps static analysers happy
and the reader sane.
Suggested-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Whilst I strongly advise against doing so for the implicit coherency
issues between the multiple buffer objects accessing the same backing
store, it nevertheless is a valid use case, akin to mmaping the same
file multiple times.
The reason why we forbade it earlier was that our use of the interval
tree for fast invalidation upon vma changes excluded overlapping
objects. So in the case where the user wishes to create such pairs of
overlapping objects, we degrade the range invalidation to walkin the
linear list of objects associated with the mm.
A situation where overlapping objects could arise is the lax implementation
of MIT-SHM Pixmaps in the xserver. A second situation is where the user
wishes to have different access modes to a region of memory (e.g. access
through a read-only userptr buffer and through a normal userptr buffer).
v2: Compile for mmu-notifiers after tweaking
v3: Rename is_linear/has_linear
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Li, Victor Y" <victor.y.li@intel.com>
Cc: "Kelley, Sean V" <sean.v.kelley@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: "Gong, Zhipeng" <zhipeng.gong@intel.com>
Cc: Akash Goel <akash.goel@intel.com>
Cc: "Volkin, Bradley D" <bradley.d.volkin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Let's march ahead with the deprecation plan laid out in
commit b30324adaf
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed Nov 13 22:11:25 2013 +0100
drm/i915: Deprecated UMS support
Thus far no regression report yet, so the transparent fallback plan
seems to pan out.
Cc: Dave Airlie <airlied@gmail.com>
Cc: David Herrmann <dh.herrmann@gmail.com>
Suggested-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Dave Airlie <airlied@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This chunk was no longer required from what I can see, or
at least it is doing the wrong thing, as I confused
intel_connector->encoder and connector->encoder. Drop it
for now, to remove the warnings at bootup.
Signed-off-by: Dave Airlie <airlied@redhat.com>
When I moved the irq disable down to after display disable,
I didn't realise the gt suspend also required irqs off, so move it
down as well.
Fixes WARNs seen at suspend/resume time.
v2: moved the rps flush down as well.
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
An object can only have an active gtt mapping if it is currently bound
into the global gtt. Therefore we can simply walk the list of all bound
objects and check the flag upon those for an active gtt mapping.
From commit 48018a57a8
Author: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Fri Dec 13 15:22:31 2013 -0200
drm/i915: release the GTT mmaps when going into D3
Also note that the WARN is inappropriate for this function as GPU
activity is orthogonal to GTT mmap status. Rather it is the caller that
relies upon this condition and so it should assert that the GPU is idle
itself.
References: https://bugs.freedesktop.org/show_bug.cgi?id=80081
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
[danvet: cherry-pick from -next to -fixes.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Historically we've exposed the full backlight PWM duty cycle range to
the userspace, in the name of "mechanism, not policy". However, it turns
out there are both panels and board designs where there is a minimum
duty cycle that is required for proper operation. The minimum duty cycle
is available in the VBT.
The backlight class sysfs interface does not make any promises to the
userspace about the physical meaning of the range
0..max_brightness. Specifically there is no guarantee that 0 means off;
indeed for acpi_backlight 0 usually is not off, but the minimum
acceptable value.
Respect the minimum backlight, and expose the range acceptable to the
hardware as 0..max_brightness to the userspace via the backlight class
device; 0 means the minimum acceptable enabled value. To switch off the
backlight, the user must disable the encoder.
As a side effect, make the backlight class device max brightness and
physical PWM modulation frequency (i.e. max duty cycle)
independent. This allows a follow-up patch to virtualize the max value
exposed to the userspace.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[danvet: s/BUG_ON/WARN_ON/]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
v2: Adding IS_G4X instead of gen < 5 as suggested by Daniel
Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
semaphore _sync_seqno, _seqno and _mbox are smaller than number of rings.
This optimization is to remove the ring itself from the list and the logic to do that
is at intel_ring_sync_index as below:
/*
* rcs -> 0 = vcs, 1 = bcs, 2 = vecs, 3 = vcs2;
* vcs -> 0 = bcs, 1 = vecs, 2 = vcs2, 3 = rcs;
* bcs -> 0 = vecs, 1 = vcs2. 2 = rcs, 3 = vcs;
* vecs -> 0 = vcs2, 1 = rcs, 2 = vcs, 3 = bcs;
* vcs2 -> 0 = rcs, 1 = vcs, 2 = bcs, 3 = vecs;
*/
v2: Skip when from == to (Damien).
v3: avoid computing idx when from == to (Damien).
use ring == to instead of ring->id == to->id (Damien).
use continue instead of return (Rodrigo).
v4: avoid all unecessary computation (Damien).
reduce idx to loop scope (Damien).
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When using an IOMMU, GEM objects are mapped by their DMA address as the
physical address is unknown. This depends on the underlying IOMMU
driver to map and unmap the physical pages properly as defined in
intel_iommu.c.
The current code will tell the IOMMU to unmap the GEM BO's pages on the
destruction of the first VMA that "maps" that BO. This is clearly wrong
as there may be other VMAs "mapping" that BO (using flink). The scanout
is one such example.
The patch fixes this issue by only unmapping the DMA maps when there are
no more VMAs mapping that object. This is equivalent to when an object
is considered unbound as can be seen by the code. On the first VMA that
again because bound, we will remap.
An alternate solution would be to move the dma mapping to object
creation and destrubtion. I am not sure if this is considered an
unfriendly thing to do.
Some notes to backporters trying to backport full PPGTT:
The bug can never be hit without enabling the IOMMU. The existing code
will also do the right thing when the object is shared via dmabuf. The
failure should be demonstrable with flink. In cases when not using
intel_iommu_strict it is likely (likely, as defined by: off the top of
my head) on current workloads to *not* hit this bug since we often
teardown all VMAs for an object shared across multiple VMs. We also
finish access to that object before the first dma_unmapping.
intel_iommu_strict with flinked buffers is likely to hit this issue.
Signed-off-by: Armin Reese <armin.c.reese@intel.com>
[danvet: Add the excellent commit message provided by Ben.]
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since we merged runtime PM support for DPMS, it is possible that these
assertions will be called when the power wells are disabled but a mode
is "set", resulting in "failed assertion" and "device suspended while
reading register" WARNs.
To reproduce the bug: disable all screens using mode unset, do a
modeset on one screen, disable it using DPMS, then try to do a mode
unset on it again to see the WARNs.
v2: The first version of this patch changed the assertions to also
check the power domains. Daniel suggested that it would be better to
just remove the assertions: "The modeset state checker
will already notice when we've failed to turn off the pipe. And we
check cursors and plane state in the enable sequence, too. Since we
use these asserts a lot to lock down the precise modeset sequence I
actually prefer if they're a bit dumb and don't check the power
wells."
Testcase: igt/rpm_rpm/dpms-mode-unset-lpsp
Testcase: igt/rpm_rpm/dpms-mode-unset-non-lpsp
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The comment [which was mine] is wrong. The context object can never be
bound in a PPGTT because it is only capable of living in the Global GTT.
So, remove the comment, and reorder the unref. What's nice about the
latter is it keeps the context object alive past the PPGTT. This makes
the destroy ordering symmetric with the creation ordering.
Create:
1. Create context
2. Create PPGTT
Destroy:
1. Destroy PPGTT
2. Destroy context
As far as I know, this does not fix a bug. The code previously kept the
context data structure, only the object was gone. As the code was,
nothing tried to use the object after this point.
NOTE: If in the future we have cases where the PPGTT can/should outlive
the context (which doesn't occur today, but the code permits it), this
ordering does not matter. Even if this occurs, as it stands now, we do
not expect that to be the normal case, and having this order makes
debugging a bit easier if we're tracking object lifetimes for the
context vs ppgtt
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
[danvet: Resolve conflict with Oscar's execlist prep patches.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The bound list is global (all objects which back the VMAs are stored
here). Recently the BUG() in the offset lookup was demoted to a WARN,
but the fault actually lies in the caller, here.
This bug has existed since the initial introduction of PPGTT (however,
it was fixed in unmerged patches to fix up the error state).
Note: The reason for the BUG_ON to WARN_ON demotion was _not_ to
duct-tape over this bug here but another but triggerable without
ppgtt. See the commit for details:
commit f25748ea73
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue Jun 17 22:34:38 2014 +0200
drm/i915: Don't BUG_ON in i915_gem_obj_offset
A WARN_ON is perfectly fine.
The BUG in here seems to be the cause behind hard-hangs when I cat the
i915_gem_pageflip debugfs file (which calls this from an irq
spinlock). But only while running a full igt run after a while. I
still need to root cause the underlying issue.
I'll also start reject patches which add new BUG_ON but don't come
with a really good justification for it. The general rule really
should be to just WARN and hope the driver survives for long enough.
v2: Make the WARN a bit more useful per Chris' suggestion.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
[danvet: Clarfy that the WARN_ON (former BUG_ON) in ggtt_offset caught
more than just this bug fixed in this patch here.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This was fumbled while trying to use the cached min/min/rpe values in
the vlv debugfs code.
This is a regression from
commit 03af20458a
Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date: Sat Jun 28 02:03:53 2014 +0300
drm/i915: Use the cached min/min/rpe values in the vlv debugfs code
Signed-off-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
By the time I wrote this patch, it allowed me to catch some problems.
But due to patch reordering - in order to prevent fake "regression"
reports - this patch may be merged after the fixes of the problems
identified by this patch.
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The current code only runs when we do an I915_WRITE operation. It
checks if the unclaimed register flag is set before we do the
operation, and then it checks it again after we do the operation. This
double check allows us to find out if the I915_WRITE operation in
question is the bad one, or if some previous code is the bad one. When
it finds a problem, our code uses DRM_ERROR to signal it.
The good thing about the current code is that it detects the problem,
so at least we can know we did something wrong. The problem is that
even though we find the problem, we don't really have much information
to actually debug it. So whenever I see one of these DRM_ERROR
messages on my systems, the first thing I do is apply a patch to
change the DRM_ERROR to a WARN and also check for unclaimed registers
on I915_READ operations. This local patch makes things even slower,
but it usually helps a lot in finding the bad code.
The first point here is that since the current code is only useful to
detect whether we have a problem or not, but it is not really good to
find the cause of the problem, I don't think we should be checking
both before and after every I915_WRITE operation: just doing the check
once should be enough for us to quickly detect problems. With this
change, the code that runs by default for every single user will only
do 1 read operation for every single I915_WRITE, instead of 2. This
patch does this change.
The second point is that the local patch I have should be upstream,
but since it makes things slower it should be disabled by default. So
I added the i915.mmio_debug option to enable it.
So after this patch, this is what will happen:
- By default, we will try to detect unclaimed registers once after
every I915_WRITE operation. Previously we tried twice for every
I915_WRITE.
- When we find an unclaimed register we will still print a DRM_ERROR
message, but we will now tell the user to try again with
i915.mmio_debug=1.
- When we use i915.mmio_debug=1 we will try to find unclaimed
registers both before and after every I915_READ and I915_WRITE
operation, and we will print stack traces in case we find them.
This should really help locating the exact point of the bad code
(or at least finding out that i915.ko is not the problem).
This commit also opens space for really-slow register debugging
operations on other platforms. In theory we can now add lots and lots
of debug code behind i915.mmio_debug, enable this option on our tests,
and catch more problems.
v2: - Remove not-so-useful comments (Daniel)
- Fix the param definition macros (Rodrigo)
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
After this point, we'll modify it with the runtime routines.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
