The bulk of the change is to convert the growing list of rings into an
array so that the relationship between the rings and the semaphore sync
registers can be easily computed.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
With this change, every batchbuffer can use all available fences (save
pinned and scanout, of course) without ever stalling the gpu!
In theory. Currently the actual pipelined update of the register is
disabled due to some stability issues. However, just the deferred update
is a significant win.
Based on a series of patches by Daniel Vetter.
The premise is that before every access to a buffer through the GTT we
have to declare whether we need a register or not. If the access is by
the GPU, a pipelined update to the register is made via the ringbuffer,
and we track the last seqno of the batches that access it. If by the
CPU we wait for the last GPU access and update the register (either
to clear or to set it for the current buffer).
One advantage of being able to pipeline changes is that we can defer the
actual updating of the fence register until we first need to access the
object through the GTT, i.e. we can eliminate the stall on set_tiling.
This is important as the userspace bo cache does not track the tiling
status of active buffers which generate frequent stalls on gen3 when
enabling tiling for an already bound buffer.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This makes the various rings more consistent by removing the anomalous
handing of the rendering ring execbuffer dispatch.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Besides the minimal improvement in reducing the execbuffer overhead, the
real benefit is clarifying a few routines.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
A number of dragons have been seen lurking within the execbuffer code.
The first step is then to isolate them from the rest and begin to
scrutinise them in depth. Suggested by Daniel Vetter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Simply remove our accounting of objects inside the aperture, keeping
only track of what is in the aperture and its current usage. This
removes the over-complication of BUGs that were attempting to keep the
accounting correct and also removes the overhead of the accounting on
the hot-paths.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The pipe control object is allocated by the device for the sole use of the
render ringbuffer. Move this detail from the general code to the render
ring buffer initialisation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Having seen the effects of erroneous fencing on the batchbuffer, a
useful sanity check is to record the fence registers at the time of an
error.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This still uses the agp functions to actually reinstate the mappings
(with a gross hack to make agp cooperate), but it wires everything
up correctly for the switchover.
The call to agp_rebind_memory can be dropped because all non-kms drivers
do all their rebinding on EnterVT.
v2: Be more paranoid and flush the chipset cache after restoring gtt
mappings.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This is required to restore gtt mappings on resume when agp is gone.
The right way to do this would be to make sturct drm_mm_node embeddable
and use the allocation list maintained by the drm memory manager. But
that's a bigger project. Getting rid of the per bo agp_mem will save
more memory than this wastes, anyway.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The GATT is a write-only set of registers, reading from them in the
manner of i915_gtt_to_phys() is supposed to be undefined. However a
simple solution exists as we allocate linear memory from the stolen
area, we can simply add the block offset to the base register. As a
side-effect we recover all the unused stolen GTT entries and so enlarge
our aperture.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We were reading our 64-bit value in I915_READ64 and returning 32 bits
of it. The restoration of fence regs at resume then had a zero end
value, and the fence had no effect.
Version 2: Split register access functions into per-size versions
Sharing code between different sizes seemed reasonable when we only
needed a single copy, but as 64-bit access requires its own version,
it makes sense to just split them out for each size.
Reported-by: Peter Clifton <pcjc2@cam.ac.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
[ickle: use a macro to create the various read/write routines]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
When trying to diagnose mysterious errors on resume, capture the
display register contents as well.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The pinned buffers are useful for diagnosing errors in setting up state
for the chipset, which may not necessarily be 'active' at the time of
the error, e.g. the cursor buffer object.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
An old and oft reported bug, is that of the GPU hanging on a
MI_WAIT_FOR_EVENT following a mode switch. The cause is that the GPU is
waiting on a scanline counter on an inactive pipe, and so waits for a
very long time until eventually the user reboots his machine.
We can prevent this either by moving the WAIT into the kernel and
thereby incurring considerable cost on every swapbuffers, or by waiting
for the GPU to retire the last batch that accesses the framebuffer
before installing a new one. As mode switches are much rarer than swap
buffers, this looks like an easy choice.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28964
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29252
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Before reading ring register, set FORCE_WAKE bit to prevent GT core
power down to low power state, otherwise we may read stale values.
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
[ickle: added a udelay which seemed to do the trick on my SNB]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we use POSTING_READ to flush the write to the register before
proceeding, we do not care what the return value is and similar we do
not care for the read to be recorded whilst tracing register
read/writes.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This will be used later to hide the frequently written registers
from debug traces in order to increase the signal-to-noise.
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Add two tracepoints at I915_WRITE/READ for tracing down all the
register write and read.
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
a00b10c360 "Only enforce fence limits inside the GTT" also
added a fenceable/mappable disdinction when binding/pinning buffers.
This only complicates the code with no pratical gain:
- In execbuffer this matters on for g33/pineview, as this is the only
chip that needs fences and has an unmappable gtt area. But fences
are only possible in the mappable part of the gtt, so need_fence
implies need_mappable. And need_mappable is only set independantly
with relocations which implies (for sane userspace) that the buffer
is untiled.
- The overlay code is only really used on i8xx, which doesn't have
unmappable gtt. And it doesn't support tiled buffers, currently.
- For all other buffers it's a bug to pass in a tiled bo.
In short, this disdinction doesn't have any practical gain.
I've also reverted mapping the overlay and context pages as possibly
unmappable. It's not worth being overtly clever here, all the big
gains from unmappable are for execbuf bos.
Also add a comment for a clever optimization that confused me
while reading the original patch by Chris Wilson.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Take two passes to evict everything whilst searching for sufficient free
space to bind the batchbuffer. After searching for sufficient free space
using LRU eviction, evict everything that is purgeable and try again.
Only then if there is insufficient free space (or the GTT is too badly
fragmented) evict everything from the aperture and try one last time.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
So long as we adhere to the fence registers rules for alignment and no
overlaps (including with unfenced accesses to linear memory) and account
for the tiled access in our size allocation, we do not have to allocate
the full fenced region for the object. This allows us to fight the bloat
tiling imposed on pre-i965 chipsets and frees up RAM for real use. [Inside
the GTT we still suffer the additional alignment constraints, so it doesn't
magic allow us to render larger scenes without stalls -- we need the
expanded GTT and fence pipelining to overcome those...]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
By using read_cache_page() for individual pages during pwrite/pread we
can eliminate an unnecessary large allocation (and immediate free) of
obj->pages. Also this eliminates any potential nesting of get/put pages,
simplifying the code and preparing the path for greater things.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Since we rarely use the mmap_offset and it is easily computable from the
obj->map_list.hash, remove it.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Eliminate the racy device unload by embedding a shrinker into each
device. Smaller, simpler code.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This holds error state from the main graphics arbiter mainly involving
the DMA engine and address translation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
More precisely: For those that _need_ to be mappable. Also add two
BUG_ONs in fault and pin to check the consistency of the mappable
flag.
Changes in v2:
- Add tracking of gtt mappable space (to notice mappable/unmappable
balancing issues).
- Improve the mappable working set tracking by tracking fault and pin
separately.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
At least the part that's currently enabled by the BIOS.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Like before add a parameter mappable (also to gem_object_pin) and
set it depending upon the context. Only bos that are brought into
the gtt due to an execbuffer call can be put into the unmappable
part of the gtt, everything else (especially pinned objects) need
to be put into the mappable part of the gtt.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Add a mappable parameter to i915_gem_evict_something to distinguish
the two cases (non-restricted vs. mappable gtt allocations). No
functional changes because the mappable limit is set to the end of
the gtt currently.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Preparing the ringbuffer for adding new commands can fail (a timeout
whilst waiting for the GPU to catch up and free some space). So check
for any potential error before overwriting HEAD with new commands, and
propagate that error back to the user where possible.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The ringbuffer keeps a pointer to the parent device, so we can use that
instead of passing around the pointer on the stack.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
... to prevent flush processing of an idle (or even absent) ring.
This fixes a regression during suspend from 87acb0a5.
Reported-and-tested-by: Alexey Fisher <bug-track@fisher-privat.net>
Tested-by: Peter Clifton <pcjc2@cam.ac.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Based on an original patch by Zhenyu Wang, this initializes the BLT ring for
SandyBridge and enables support for user execbuffers.
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
To handle retirements, we need per-ring tracking of active objects.
To handle evictions, we need global tracking of active objects.
As we enable more rings, rebuilding the global list from the individual
per-ring lists quickly grows tiresome and overly complicated. Tracking the
active objects in two lists is the lesser of two evils.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cache the first 4 bytes of DPCD data in the eDP case. It's unlikely to
change and can save us some trouble at link training time.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We need to use some of these values in eDP configurations, so be sure to
fetch them and store them in the i915 private structure.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The _DSM method on the integrated graphics device can tell us which
connectors are muxable, so add support for making the call and parsing
out the connector info.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[ickle: fix compiler warnings for using uninitialized 'result' and
downgrade error message for non-switchable devices]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The issue is that we may become stuck executing a long running shader
and continually attempt to reset the GPU. (Or maybe we tickle some bug
and need to break the vicious cycle.) So if we are detect a second hang
within 5 seconds, give up trying to programme the GPU and report it
wedged.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
When the GPU is reset, the fence registers are invalidated, so release
the objects and clear them out.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Only drm/i915 does the bookkeeping that makes the information useful,
and the information maintained is driver specific, so move it out of the
core and into its single user.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dave Airlie <airlied@redhat.com>
This check only appears to succeed when using GMBUS, so we need to skip
it if we have fallen back to using GPIO bit banging.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Besides a couple of bugs when writing more than a single byte along the
GMBUS, SDVO was completely failing whilst trying to use GMBUS, so use
bit banging instead.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter pointed out that in this case is would be clearer and
cleaner to use a spinlock instead of a mutex to protect the per-file
request list manipulation. Make it so.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Owain Ainsworth reported an issue between the interaction of the
hangcheck and userspace immediately (and permanently) falling back to
s/w rasterisation. In order to break the mutex and begin resetting the
GPU, we must abort the current operation (usually within the wait) and
climb sufficiently far back up the call chain to drop the mutex. In his
implementation, Owain has a loop within the ioctl handler to detect the
hang and then sleep until the error handler has run. I've chosen to
return to userspace and report an EAGAIN which should trigger the
userspace ioctl handler to repeat the call (simply because it felt less
invasive...). Before hitting a wedged GPU, we then wait upon completion
of the error handler.
Reported-by: Owain G. Ainsworth <zerooa@googlemail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Avoid cause latencies in other clients by not taking the global struct
mutex and moving the per-client request manipulation a local per-client
mutex. For example, this allows a compositor to schedule a page-flip
(through X) whilst an OpenGL application is monopolising the GPU.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
First step, lets have a look at the values for troublesome panels and
see if they may be used to improve our link training.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We need to drain the pending flips prior to disabling the pipe during
modeset, and these need to be done in an uninterruptible fashion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This is already performed with the pipelined flush, so by the time we
schedule the flush in the page-flip, the ring is NULL and we OOPs
instead.
Reported-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Keep a list of pinned objects and display it via debugfs. Now all
objects that exist in the GTT are always tracked on one of the
active, flushing, inactive or pinned lists.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
If we have queued a page flip on the current fb and then request a mode
change, wait until the page flip completes before performing the new
request.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Track if the gpu requires the fence for the execution of a batch buffer
and so only wait upon the retirement of the object's last rendering
seqno if the fence is in use by the GPU.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Previously we only tidied up the active bo lists for chipsets were we
would attempt to reset the GPU. However, this action is necessary for
the system to continue and reclaim the dead bo for all chipsets.
Pointed out, in passing, by Owain Ainsworth.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Clear the GPU read domain for the inactive objects on a reset so that
they are correctly invalidated on reuse.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Owain Ainsworth noticed that the reset code failed to clear the flushing
list leaving the driver in an inconsistent state following a hung GPU.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
When flushing the GPU domains,we emit a flush on *both* rings, even
though they share a unified cache. Only emit the flush on the currently
active ring.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Use the GMBUS interface rather than direct bit banging to grab the EDID
over DDC (and for other forms of auxiliary communication with external
display controllers). The hope is that this method will be much faster
and more reliable than bit banging for fetching EDIDs from buggy monitors
or through switches, though we still preserve the bit banging as a
fallback in case GMBUS fails.
Based on an original patch by Jesse Barnes.
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we currently may need to acquire a fence register during a modeset,
we need to be able to do so in an uninterruptible manner. So expose that
parameter to the callers of the fence management code.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This ensures that we do wait upon the flushes to complete if necessary
and avoid the visual tears, whilst enabling pipelined page-flips.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
By reducing the hangcheck frequency we check less often, conserving
resources, and still detect a lock up quickly. On a fast machine with a
slow GPU (like a Core2 paired with a 945G) it is easy for the hangcheck to
misfire as we check too fast.
Also once hung and if we fail to completely reset the chip, we have a
nasty habit of proclaming a hang many times a second and generating a
strobe-like display.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We need to track different state on each generation in order to detect
when we need to refresh the FBC registers.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Use the detection from intel-gtt.ko instead. Hooray!
Also move the stolen mem allocator to the other gtt stuff in dev_prv->mem.
v2: Chris Wilson noted that my error handling was crap. Fix it. He also
said that this fixes a problem on his i845. Indeed, i915_probe_agp
misses a special case for i830/i845 stolen mem detection.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=25476
Cc: stable@kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Add a few definitions to it that are already shared and that will
be shared in the future (like the number of stolen entries).
No functional changes in here.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
There were two instances of code to control the panel backlight and
neither handled the complete set of device variations.
Fixes:
Bug 29716 - [GM965] Regression: Backlight resets to minimum when changing resolution
https://bugs.freedesktop.org/show_bug.cgi?id=29716
And a bug on one of my PineView boxes which overflowed the backlight
value.
Incorporates part of a similar patch by Matthew Garrett that exposes a
native Intel backlight controller.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This spinlock only served debugging purposes in a time when we could not
be sure of the mutex ever being released upon a GPU hang. As we now
should be able rely on hangcheck to do the job for us (and that error
reporting should not itself require the struct mutex) we can kill the
incomplete attempt at protection.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Alexander reported that the compilation of intel_overlay.c was failing
due to an inclusion that was only valid with CONFIG_DEBUG_FS. As the
whole error reporting is only useful with debugfs enabled, remove all
the redundant error state collection code when compiling without
CONFIG_DEBUG_FS.
Reported-by: Alexander Lam <lambchop468@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
By allocating the request prior to writing to the ringbuffer, we can
abort the operation without leaving the GPU in an inconsistent state.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
... instead of threading flush_domains through the execbuf code to
i915_add_request.
With this change 2 small cleanups are possible (likewise the majority
of the patch):
- The flush_domains parameter of i915_add_request is always 0. Drop it
and the corresponding logic.
- Ditto for the seqno param of i915_gem_process_flushing_list.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Previously I thought that one interrupt per batchbuffer should be
enough. Now tedious benchmarking showed this to be wrong.
Therefore track whether any commands have been isssued with a future
seqno (like pipelined fencing changes or flushes). If this is the case
emit a request before issueing the batchbuffer.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Useful for capturing register read/write traces to send to the hw guys.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
It is recommended that we use the Video BIOS tables that were copied
into the OpRegion during POST when initialising the driver. This saves
us from having to furtle around inside the ROM ourselves and possibly
allows the vBIOS to adjust the tables prior to initialisation.
On some systems, such as the Samsung N210, there is no accessible VBIOS
and the only means of finding the VBT is through the OpRegion.
v2: Rearrange the code so that ASLE is enabled along with ACPI
v3: Enable OpRegion parsing even without ACPI
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Garrett <mjg@redhat.com>
It's part of the generic Intel driver infrastructure so rename it in
prepreparation for using it for VBT.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
ums-gem code correctly cancels the retire work (at lastclose time),
kms does not do so. Fix this by canceling the work right after ideling
the gpu.
While staring at the code I noticed that the work function is not
static. Fix this, too.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
RC6 power state requires a logical render context in place for saving
render context.
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
i830 requires 32bpp cursors to be aligned to 16KB, so we have to expose
the alignment parameter to i915_gem_attach_phys_object().
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
v2: Hook in DP paths to keep FULLSCREEN panel fitting on eDP.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Based in a large part upon Daniel Vetter's implementation and adapted
for handling multiple rings in a single pass.
This should lead to better gtt usage and fixes the page-fault-of-doom
triggered. The fairness is provided by scanning through the GTT space
amalgamating space in rendering order. As soon as we have a contiguous
space in the GTT large enough for the new object (and its alignment),
evict any object which lies within that space. This should keep more
objects resident in the GTT.
Doing throughput testing on a PineView machine with cairo-perf-trace
indicates that there is very little difference with the new LRU scan,
perhaps a small improvement... Except oddly for the poppler trace.
Reference:
Bug 15911 - Intermittent X crash (freeze)
https://bugzilla.kernel.org/show_bug.cgi?id=15911
Bug 20152 - cannot view JPG in firefox when running UXA
https://bugs.freedesktop.org/show_bug.cgi?id=20152
Bug 24369 - Hang when scrolling firefox page with window in front
https://bugs.freedesktop.org/show_bug.cgi?id=24369
Bug 28478 - Intermittent graphics lockups due to overflow/loop
https://bugs.freedesktop.org/show_bug.cgi?id=28478
v2: Attempt to clarify the logic and order of eviction through the use
of comments and macros.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Eric Anholt <eric@anholt.net>
The eviction code is the gnarly underbelly of memory management, and is
clearer if kept separated from the normal domain management in GEM.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
This will be used by the eviction logic to maintain fairness between the
rings.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
v2: Add the interrupt status and address.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Eric Anholt <eric@anholt.net>
Using dev_priv__ avoids sparse complaining about shadowed variables in
the *LP_RING() macros.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
This avoids the excess flush and requests on idle rings (and spamming
the debug log ;-)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
If during the freeing of an object the unbind is interrupted by a system
call, which is quite possible if we have outstanding GPU writes that
must be flushed, the unbind is silently aborted. This still leaves the
AGP region and backing pages allocated, and perhaps more importantly,
the object remains upon the various lists exposing us to memory
corruption.
I think this is the cause behind the use-after-free, such as
Bug 15664 - Graphics hang and kernel backtrace when starting Azureus
with Compiz enabled
https://bugzilla.kernel.org/show_bug.cgi?id=15664
v2: Daniel Vetter reminded me that kernel space programming is never easy.
We cannot simply spin to clear the pending signal and so must deferred
the freeing of the object until later.
v3: Run from the top level retire requests.
v4: Tested with P(return -ERESTARTSYS)=.5 from i915_gem_do_wait_request()
v5: Rebase against Eric's for-linus tree.
v6: Refactor, split and add a comment about avoiding unbounded recursion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Eric Anholt <eric@anholt.net>
Combine the iteration over active render rings into a common function.
This is in preparation for reusing the idle function to also retire
deferred free requests.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
The original i965, including the revised G35 and Q35, requires an
alignment of 128K for the display surface with linear memory, so
increase the requirement from 64k for these chipsets. For the later
chipsets in the i965 family, only a 4k alignment is required. (So
long as we do not start performing asynchronous flips.)
Note the impact of this should be slight as on i965 we should be using a
tiled frontbuffer for anything up to a 4096x4096 display.
v2: compilation fixes and note that the docs do not exclude the G35 from
the extra alignment.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
References:
Bug 26691 - Spurious hangcheck whilst executing a long shader over a
large vertex buffer
https://bugs.freedesktop.org/show_bug.cgi?id=26691
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
drivers/gpu/drm/i915/i915_drv.h|676 col 19| warning: dubious bitfield without explicit `signed' or `unsigned'
drivers/gpu/drm/i915/i915_drv.h|712 col 19| warning: dubious bitfield without explicit `signed' or `unsigned'
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
About 0.2W power can be saved on one HP laptop.
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
For self-refresh mode WM calculation's "line time" should use
mode's htotal instead of hdisplay. "surface width" is the hdisplay
for display plane and 64 for cursor plane.
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Ported over from the old UMS list. Unfortunately they're still
necessary especially on older laptop platforms.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=22126.
Tested-by: Xavier <shiningxc@gmail.com>
Tested-by: Diego Escalante Urrelo <diegoe@gnome.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
We're really supposed to do this to avoid trouble with underflows when
multiple planes are active.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=26987.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Tested-by: fangxun <xunx.fang@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel:
drm/i915: fix page flip finish vs. prepare on plane B
drm/i915: change default panel fitting mode to preserve aspect ratio
drm/i915: fix uninitialized variable warning in i915_setup_compression()
drm/i915: take struct_mutex in i915_dma_cleanup()
drm/i915: Fix CRT hotplug regression in 2.6.35-rc1
i915: fix ironlake edp panel setup (v4)
drm/i915: don't access FW_BLC_SELF on 965G
drm/i915: Account for space on the ring buffer consumed whilst wrapping.
drm/i915: gen3 page flipping fixes
drm/i915: don't queue flips during a flip pending event
drm/i915: Fix incorrect intel_ring_begin size in BSD ringbuffer.
drm/i915: Turn on 945 self-refresh only if single CRTC is active
drm/i915/gen4: Fix interrupt setup ordering
drm/i915: Use RSEN instead of HTPLG for tfp410 monitor detection.
drm/i915: Move non-phys cursors into the GTT
Revert "drm/i915: Don't enable pipe/plane/VCO early (wait for DPMS on)."
(Included the "fix page flip finish vs. prepare on plane B" patch from
Jesse on top of the pull request from Eric. -- Linus)
Gen3 chips have slightly different flip commands, and also contain a bit
that indicates whether a "flip pending" interrupt means the flip has
been queued or has been completed.
So implement support for the gen3 flip command, and make sure we use the
flip pending interrupt correctly depending on the value of ECOSKPD bit
0.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
The ring_begin API was taking a number of bytes, while all of our
other begin/end macros take number of dwords. Change the API over to
dwords to prevent future bugs.
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
(regression fix since fbdev/kms rework).
My fb rework didn't remember about the 84/65s.
Reported-by: Ondrej Zary <linux@rainbow-software.org>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Pineview with DDR3 memory has different latencies to enable CxSR.
This patch updates CxSR latency table to add Pineview DDR3 latency
configuration. It also adds one flag "is_ddr3" for checking DDR3
setting in MCHBAR.
Cc: Shaohua Li <shaohua.li@intel.com>
Cc: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Li Peng <peng.li@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
This saves a whooping 7 dwords. Zero functional changes. Because
some of the refcounts are rather tightly calculated, I've put
BUG_ONs in the code to check for overflows.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Eric Anholt <eric@anholt.net>
Add power monitoring support to the i915 driver for use by the IPS
driver. Export the available power info to the IPS driver through a few
new inter-driver hooks. When used together, the IPS driver and this
patch can significantly increase graphics performance on Ironlake class
chips.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[anholt: Fixed 32-bit compile. stupid obfuscating div_u64()]
Signed-off-by: Eric Anholt <eric@anholt.net>
The BSD (bit stream decoder) ring is used for accessing the BSD engine
which decodes video bitstream for H.264 and VC1 on G45+. It is
asynchronous with the render ring and has access to separate parts of
the GPU from it, though the render cache is coherent between the two.
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Xiang Hai hao <haihao.xiang@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
The active list and request list move into the ringbuffer structure,
so each can track its active objects in the order they are in that
ring. The flushing list does not, as it doesn't matter which ring
caused data to end up in the render cache. Objects gain a pointer to
the ring they are active on (if any).
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Xiang Hai hao <haihao.xiang@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Introduces a more complete intel_ring_buffer structure with callbacks
for setup and management of a particular ringbuffer, and converts the
render ring buffer consumers to use it.
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Xiang Hai hao <haihao.xiang@intel.com>
[anholt: Fixed up whitespace fail and rebased against prep patches]
Signed-off-by: Eric Anholt <eric@anholt.net>
This is preparation for supporting multiple ringbuffers on Ironlake.
The non-copy-and-paste changes are:
- de-staticing functions
- I915_GEM_GPU_DOMAINS moving to i915_drv.h to be used by both files.
- i915_gem_add_request had only half its implementation
copy-and-pasted out of the middle of it.
This lru tracks fences, not objects, so move it to where it belongs.
As a side effect, this nicely shrinks drm_i915_gem_object by two
pointers.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Eric Anholt <eric@anholt.net>
Multifunction SDVO cards stopped working after 14571b4, and would report
something that looked remarkably like an ADD2 SPD ROM instead of EDID.
This appears to be because DDC bus selection was utterly horked by that
commit; controlled_output was no longer always a single bit, so
intel_sdvo_select_ddc_bus would pick bus 0, which is (unsurprisingly)
the SPD ROM bus, not a DDC bus.
So, instead of that, let's just use the DDC bus the child device table
tells us to use. I'm guessing at the bitmask and shifting from VBIOS
dumps, but it can't possibly be worse.
cf. https://bugzilla.redhat.com/584229
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Conflicts:
drivers/gpu/drm/i915/i915_dma.c
drivers/gpu/drm/i915/i915_drv.h
drivers/gpu/drm/radeon/r300.c
The BSD ringbuffer support that is landing in this branch
significantly conflicts with the Ironlake PIPE_CONTROL fix on master,
and requires it to be tested successfully anyway.
Since 965, the hardware has supported the PIPE_CONTROL command, which
provides fine grained GPU cache flushing control. On recent chipsets,
this instruction is required for reliable interrupt and sequence number
reporting in the driver.
So add support for this instruction, including workarounds, on Ironlake
and Sandy Bridge hardware.
https://bugs.freedesktop.org/show_bug.cgi?id=27108
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
This keeps the memory manager from complaining when we take it down.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Luckily the change is quite a little bit less invasive than I've
feared.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Chris Wilson