We want to hide the latency of releasing objects and their backing
storage from the submission, so we move the actual free to a worker.
This allows us to switch to struct_mutex freeing of the object in the
next patch.

Furthermore, if we know that the object we are dereferencing remains valid
for the duration of our access, we can forgo the usual synchronisation
barriers and atomic reference counting. To ensure this we defer freeing
an object til after an RCU grace period, such that any lookup of the
object within an RCU read critical section will remain valid until
after we exit that critical section. We also employ this delay for
rate-limiting the serialisation on reallocation - we have to slow down
object creation in order to prevent resource starvation (in particular,
files).

v2: Return early in i915_gem_tiling() ioctl to skip over superfluous
work on error.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-19-chris@chris-wilson.co.uk

Break the allocation of the backing storage away from struct_mutex into
a per-object lock. This allows parallel page allocation, provided we can
do so outside of struct_mutex (i.e. set-domain-ioctl, pwrite, GTT
fault), i.e. before execbuf! The increased cost of the atomic counters
are hidden behind i915_vma_pin() for the typical case of execbuf, i.e.
as the object is typically bound between execbufs, the page_pin_count is
static. The cost will be felt around set-domain and pwrite, but offset
by the improvement from reduced struct_mutex contention.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-14-chris@chris-wilson.co.uk

The plan is to move obj->pages out from under the struct_mutex into its
own per-object lock. We need to prune any assumption of the struct_mutex
from the get_pages/put_pages backends, and to make it easier we pass
around the sg_table to operate on rather than indirectly via the obj.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-13-chris@chris-wilson.co.uk

The plan is to make obtaining the backing storage for the object avoid
struct_mutex (i.e. use its own locking). The first step is to update the
API so that normal users only call pin/unpin whilst working on the
backing storage.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-12-chris@chris-wilson.co.uk

Treat a framebuffer reference with the same priority as an active
reference whilst shrinking. Framebuffers are likely to be reused and
typically cost more to migrate to and from GPU memory (on LLC
architectures we need to clflush), so defer the temptation to purge them
during a kswapd run until we have run out of cheap buffers.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJohn Harrison <john.c.harrison@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161012124824.23521-1-chris@chris-wilson.co.uk

In the next patch we want to handle reset directly by a locked waiter in
order to avoid issues with returning before the reset is handled. To
handle the reset, we must first know whether we hold the struct_mutex.
If we do not hold the struct_mtuex we can not perform the reset, but we do
not block the reset worker either (and so we can just continue to wait for
request completion) - otherwise we must relinquish the mutex.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-10-chris@chris-wilson.co.uk

We need finer control over wakeup behaviour during i915_wait_request(),
so expand the current bool interruptible to a bitmask.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-9-chris@chris-wilson.co.uk

After removing the user of this wart, we can remove the wart entirely.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470388464-28458-10-git-send-email-chris@chris-wilson.co.uk

We can now wait for the GPU (all engines) to become idle without
requiring the struct_mutex. Inside the shrinker, we need to currently
take the struct_mutex in order to purge objects and to purge the objects
we need the GPU to be idle - causing a stall whilst we hold the
struct_mutex. We can hide most of that stall by performing the wait
before taking the struct_mutex and only doing essential waits for
new rendering on objects to be freed.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470388464-28458-8-git-send-email-chris@chris-wilson.co.uk

The principal motivation for this was to try and eliminate the
struct_mutex from i915_gem_suspend - but we still need to hold the mutex
current for the i915_gem_context_lost(). (The issue there is that there
may be an indirect lockdep cycle between cpu_hotplug (i.e. suspend) and
struct_mutex via the stop_machine().) For the moment, enabling last
request tracking for the engine, allows us to do busyness checking and
waiting without requiring the struct_mutex - which is useful in its own
right.

As a side-effect of having a robust means for tracking engine busyness,
we can replace our other busyness heuristic, that of comparing against
the last submitted seqno. For paranoid reasons, we have a semi-ordered
check of that seqno inside the hangchecker, which we can now improve to
an ordered check of the engine's busyness (removing a locked xchg in the
process).

v2: Pass along "bool interruptible" as being unlocked we cannot rely on
i915->mm.interruptible being stable or even under our control.
v3: Replace check Ironlake i915_gpu_busy() with the common precalculated value
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470388464-28458-6-git-send-email-chris@chris-wilson.co.uk

If we enable RCU for the requests (providing a grace period where we can
inspect a "dead" request before it is freed), we can allow callers to
carefully perform lockless lookup of an active request.

However, by enabling deferred freeing of requests, we can potentially
hog a lot of memory when dealing with tens of thousands of requests per
second - with a quick insertion of a synchronize_rcu() inside our
shrinker callback, that issue disappears.

v2: Currently, it is our responsibility to handle reclaim i.e. to avoid
hogging memory with the delayed slab frees. At the moment, we wait for a
grace period in the shrinker, and block for all RCU callbacks on oom.
Suggested alternatives focus on flushing our RCU callback when we have a
certain number of outstanding request frees, and blocking on that flush
after a second high watermark. (So rather than wait for the system to
run out of memory, we stop issuing requests - both are nondeterministic.)

Paul E. McKenney wrote:

Another approach is synchronize_rcu() after some largish number of
requests.  The advantage of this approach is that it throttles the
production of callbacks at the source.  The corresponding disadvantage
is that it slows things up.

Another approach is to use call_rcu(), but if the previous call_rcu()
is still in flight, block waiting for it.  Yet another approach is
the get_state_synchronize_rcu() / cond_synchronize_rcu() pair.  The
idea is to do something like this:

        cond_synchronize_rcu(cookie);
        cookie = get_state_synchronize_rcu();

You would of course do an initial get_state_synchronize_rcu() to
get things going.  This would not block unless there was less than
one grace period's worth of time between invocations.  But this
assumes a busy system, where there is almost always a grace period
in flight.  But you can make that happen as follows:

        cond_synchronize_rcu(cookie);
        cookie = get_state_synchronize_rcu();
        call_rcu(&my_rcu_head, noop_function);

Note that you need additional code to make sure that the old callback
has completed before doing a new one.  Setting and clearing a flag
with appropriate memory ordering control suffices (e.g,. smp_load_acquire()
and smp_store_release()).

v3: More comments on compiler and processor order of operations within
the RCU lookup and discover we can use rcu_access_pointer() here instead.

v4: Wrap i915_gem_active_get_rcu() to take the rcu_read_lock itself.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: "Goel, Akash" <akash.goel@intel.com>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1470324762-2545-25-git-send-email-chris@chris-wilson.co.uk

We are motivated to avoid using a bitfield for obj->active for a couple
of reasons. Firstly, we wish to document our lockless read of obj->active
using READ_ONCE inside i915_gem_busy_ioctl() and that requires an
integral type (i.e. not a bitfield). Secondly, gcc produces abysmal code
when presented with a bitfield and that shows up high on the profiles of
request tracking (mainly due to excess memory traffic as it converts
the bitfield to a register and back and generates frequent AGI in the
process).

v2: BIT, break up a long line in compute the other engines, new paint
for i915_gem_object_is_active (now i915_gem_object_get_active).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470324762-2545-23-git-send-email-chris@chris-wilson.co.uk

In preparation to perform some magic to speed up i915_vma_pin(), which
is among the hottest of hot paths in execbuf, refactor all the bitfields
accessed by i915_vma_pin() into a single unified set of flags.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470324762-2545-16-git-send-email-chris@chris-wilson.co.uk

When we call i915_vma_unbind(), we will wait upon outstanding rendering.
This will also trigger a retirement phase, which may update the object
lists. If, we extend request tracking to the VMA itself (rather than
keep it at the encompassing object), then there is a potential that the
obj->vma_list be modified for other elements upon i915_vma_unbind(). As
a result, if we walk over the object list and call i915_vma_unbind(), we
need to be prepared for that list to change.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470293567-10811-8-git-send-email-chris@chris-wilson.co.uk

Since we may have VMA allocated for an object, but we interrupted their
binding, there is a disparity between have elements on the obj->vma_list
and being bound. i915_gem_obj_bound_any() does this check, but this is
not rigorously observed - add an explicit count to make it easier.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470293567-10811-7-git-send-email-chris@chris-wilson.co.uk

list_first_entry_or_null() can generate better code than using
if (!list_empty()) {ptr = list_first_entry()) ..., so put it to use.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1469432687-22756-3-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1469530913-17180-2-git-send-email-chris@chris-wilson.co.uk

Ultimately wraps kref_put(), so adopt its nomenclature for consistency
with other subsystems.

s/drm_gem_object_unreference/i915_gem_object_put/
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1469005202-9659-6-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1469017917-15134-5-git-send-email-chris@chris-wilson.co.uk

Ultimately wraps kref_get(), so adopt its nomenclature for consistency
with other subsystems.

s/drm_gem_object_reference/i915_gem_object_get/
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1469005202-9659-5-git-send-email-chris@chris-wilson.co.ukReviewed-by: NDave Gordon <david.s.gordon@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1469017917-15134-4-git-send-email-chris@chris-wilson.co.uk

In commit 7608a43d ("locking/mutexes: Use MUTEX_SPIN_ON_OWNER when
appropriate") the owner field in the mutex was updated from being
dependent upon CONFIG_SMP to using optimistic spin. Update our peek
function to suite.

Fixes:7608a43d ("locking/mutexes: Use MUTEX_SPIN_ON_OWNER...")
Reported-by: NHong Liu <hong.liu@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1468244777-4888-1-git-send-email-chris@chris-wilson.co.ukReviewed-by: NMatthew Auld <matthew.auld@intel.com>
(cherry picked from commit 4f074a53)
Cc: stable@vger.kernel.org
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

In commit 7608a43d ("locking/mutexes: Use MUTEX_SPIN_ON_OWNER when
appropriate") the owner field in the mutex was updated from being
dependent upon CONFIG_SMP to using optimistic spin. Update our peek
function to suite.

Fixes:7608a43d ("locking/mutexes: Use MUTEX_SPIN_ON_OWNER...")
Reported-by: NHong Liu <hong.liu@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1468244777-4888-1-git-send-email-chris@chris-wilson.co.ukReviewed-by: NMatthew Auld <matthew.auld@intel.com>

Since drm_i915_private is now a subclass of drm_device we do not need to
chase the drm_i915_private->dev backpointer and can instead simply
access drm_i915_private->drm directly.

   text	   data	    bss	    dec	    hex	filename
1068757	   4565	    416	1073738	 10624a	drivers/gpu/drm/i915/i915.ko
1066949	   4565	    416	1071930	 105b3a	drivers/gpu/drm/i915/i915.ko

Created by the coccinelle script:
@@
struct drm_i915_private *d;
identifier i;
@@
(
- d->dev->i
+ d->drm.i
|
- d->dev
+ &d->drm
)

and for good measure the dev_priv->dev backpointer was removed entirely.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NMatthew Auld <matthew.auld@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1467711623-2905-4-git-send-email-chris@chris-wilson.co.uk

As we inspect obj->active to decide how many objects we can shrink (we
only shrink idle objects), it helps to flush the active lists first
in order to have a more accurate count of available objects.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1467390209-3576-1-git-send-email-chris@chris-wilson.co.uk

We only need to force a switch to the kernel context placeholder during
eviction. All other uses of i915_gpu_idle() just want to wait until
existing work on the GPU is idle. Rename i915_gpu_idle() to
i915_gem_wait_for_idle() to avoid any implications about "parking" the
context first.

v2: Tweak an error message if the wait fails for the ilk vtd w/a
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1466776558-21516-6-git-send-email-chris@chris-wilson.co.uk

   text	   data	    bss	    dec	    hex	filename
6309351	3578714	 696320	10584385	 a18141	vmlinux
6308391	3578714	 696320	10583425	 a17d81	vmlinux

Almost 1KiB of code reduction.

v2: More s/INTEL_INFO()->gen/INTEL_GEN()/ and IS_GENx() conversions

   text	   data	    bss	    dec	    hex	filename
6304579	3578778	 696320	10579677	 a16edd	vmlinux
6303427	3578778	 696320	10578525	 a16a5d	vmlinux

Now over 1KiB!
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1462545621-30125-3-git-send-email-chris@chris-wilson.co.uk

i915_gem_shrink() will scan the bound list only if device is not
suspended but in OOM failure scenario it becomes absolutely necessary
to release as much memory as possible. Also in allocation failure from
vmap address space, it is incumbent on the Driver to reap all its
vmaps. So, adding rpm get/put in i915_gem_shrinker_oom() and
i915_gem_shrinker_vmap() to ensure shrinking of bound objects as well.
Signed-off-by: NPraveen Paneri <praveen.paneri@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1462178429-13449-2-git-send-email-praveen.paneri@intel.comSigned-off-by: NChris Wilson <chris@chris-wilson.co.uk>

When the system is running low on memory, gem shrinker is invoked.
In this process objects will be unbounded from GTT and unbinding process
will require access to GTT(GTTADR) and also to fence register potentially.
That requires a resume of gfx device, if suspended, in the shrinker path.
Considering the power leakage due to intermediate resume, perform unbinding
operation only if device is already runtime active.

v2: Use newly implemented intel_runtime_pm_get_if_in_use (Chris)
Signed-off-by: NAkash Goel <akash.goel@intel.com>
Signed-off-by: NPraveen Paneri <praveen.paneri@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1462178429-13449-1-git-send-email-praveen.paneri@intel.comSigned-off-by: NChris Wilson <chris@chris-wilson.co.uk>

By tracking the iomapping on the VMA itself, we can share that area
between multiple users. Also by only revoking the iomapping upon
unbinding from the mappable portion of the GGTT, we can keep that iomap
across multiple invocations (e.g. execlists context pinning).

Note that by moving the iounnmap tracking to the VMA, we actually end up
fixing a leak of the iomapping in intel_fbdev.

v1.5: Rebase prompted by Tvrtko
v2: Drop dev_priv parameter, we can recover the i915_ggtt from the vma.
v3: Move handling of ioremap space exhaustion to vmap_purge and also
allow vmallocs to recover old iomaps. Add Tvrtko's kerneldoc.
v4: Fix a use-after-free in shrinker and rearrange i915_vma_iomap
v5: Back to i915_vm_to_ggtt
v6: Use i915_vma_pin_iomap and i915_vma_unpin_iomap to mark critical
sections and ensure the VMA cannot be reaped whilst mapped.
v7: Move i915_vma_iounmap so that consumers of the API are not tempted,
and add iomem annotations
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461833819-3991-5-git-send-email-chris@chris-wilson.co.uk

Since we can only swap out shmemfs objects, those are the only ones that
can influence the ability of the shrinker to free pages. Currently, all
non-shmemfs objects have a raised pages_pin_count to protect them from
the shrinker, so this just makes the logic for can_release_pages()
clearer (and safer in future so that we don't over estimate our ability
to free up pages from future non-swappable objects).
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461150592-27818-3-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>

Inside the shrinker we call can_release_pages() to indicate whether or
not we can make forward progress in freeing up memory by unbinding that
object. When adding our report to oom, we should be using the same
logic.

Whilst here, change the reporting from bytes to pages so that it looks
smaller to the user!, is consistent with the neighbouring oom report
itself which displays counts in pages, and makes the unsigned long
overflow less likely.

v2: Split oversized format string into two lines
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461150592-27818-2-git-send-email-chris@chris-wilson.co.ukReviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

When iterating over the bound list, we expect all objects there to have
their pages pinned (by the bound VMA). So only report those objects with
additional pin count on their pages as "pinned". These should be those
objects used for display and hardware access.
Reported-by: NAkash Goel <akash.goel@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Akash Goel <akash.goel@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461150592-27818-1-git-send-email-chris@chris-wilson.co.uk

When called because we have run out of vmap address space, we only need
to recover objects that have vmappings and not all.

v2: Start using is_vmalloc_addr()
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1460113874-17366-5-git-send-email-chris@chris-wilson.co.ukReviewed-by: NTvrtko Ursulin <tvrtko.ursulin@linux.intel.com>

Both the oom and vmap notifier callbacks have a loop to acquire the
struct_mutex and set the device as uninterruptible, within a certain
time. Refactor the common code into a pair of functions.
Suggested-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1459848145-24042-1-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

If the core runs out of vmap address space, it will call a notifier in
case any driver can reap some of its vmaps. As i915.ko is possibily
holding onto vmap address space that could be recovered, hook into the
notifier chain and try and reap objects holding onto vmaps.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Roman Pen <r.peniaev@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: linux-mm@kvack.org
Cc: linux-kernel@vger.kernel.org
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kahola <mika.kahola@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1459777603-23618-4-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Since we only attempt to purge an object if can_release_pages() report
true, we should also only add it to the count of potential recoverable
pages when can_release_pages() is true.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Akash Goel <akash.goel@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1459777603-23618-2-git-send-email-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Elsewhere we have adopted the convention of using '_link' to denote
elements in the list (and '_list' for the actual list_head itself), and
that the name should indicate which list the link belongs to (and
preferrably not just where the link is being stored).

s/vma_link/obj_link/ (we iterate over obj->vma_list)
s/mm_list/vm_link/ (we iterate over vm->[in]active_list)
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>

Factor out the common GEM shrinker clean-up code and call the shrinker
init function from the same function from where the corresponding
shrinker clean-up function is called. Also add sanity checking to the
shrinker and OOM registration calls.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NDavid Weinehall <david.weinehall@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1453209992-25995-4-git-send-email-imre.deak@intel.com

If the system has no available swap pages, we cannot make forward
progress in the shrinker by releasing active pages, only by releasing
purgeable pages which are immediately reaped. Take total_swap_pages into
account when counting up available objects to be shrunk and subsequently
shrinking them. By doing so, we avoid unbinding objects that cannot be
shrunk and so wasting CPU cycles flushing those objects from the GPU to
the system and then immediately back again (as they will more than
likely be reused shortly after).

Based on a patch by Akash Goel.

v2: frontswap registers extra swap pages available for the system, so it
is already include in the count of available swap pages.

v3: Use get_nr_swap_pages() to query the currently available amount of
swap space. This should also stop us from shrinking the GPU buffers if
we ever run out of swap space. Though at that point, we would expect the
oom-notifier to be running and failing miserably...
Reported-by: NAkash Goel <akash.goel@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: linux-mm@kvack.org
Cc: Akash Goel <akash.goel@intel.com>
Cc: sourab.gupta@intel.com
Link: http://patchwork.freedesktop.org/patch/msgid/1449244734-25733-2-git-send-email-chris@chris-wilson.co.ukAcked-by: NJohannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

I've botched this in

commit eb0b44ad
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed Mar 18 14:47:59 2015 +0100

    drm/i915: kerneldoc for i915_gem_shrinker.c

so let's fix it.
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Signed-off-by: NJani Nikula <jani.nikula@intel.com>

Exclude active GPU pages from the purview of the background shrinker
(kswapd), as these cause uncontrollable GPU stalls. Given that the
shrinker is rerun until the freelists are satisfied, we should have
opportunity in subsequent passes to recover the pages once idle. If the
machine does run out of memory entirely, we have the forced idling in the
oom-notifier as a means of releasing all the pages we can before an oom
is prematurely executed.

Note that this relies upon an up-front retire_requests to keep the
inactive list in shape, which was added in a previous patch, mostly as
execlist ctx pinning band-aids.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NDamien Lespiau <damien.lespiau@intel.com>
[danvet: Add note about retire_requests.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We can forgo an evict-everything here as the shrinker operation itself
will unbind any vma as required. If we explicitly idle the GPU through a
switch to the default context, we not only create a request in an
illegal context (e.g. whilst shrinking during execbuf with a request
already allocated), but switching to the default context will not free
up the memory backing the active contexts - unless in the unlikely
situation that context had already been closed (and just kept arrive by
being the current context). The saving is near zero and the danger real.

To compensate for the loss of the forced retire, add a couple of
retire-requests to i915_gem_shirnk() - this should help free up any
transitive cache from the requests.

Note that the second retire_requests is for the benefit of the
hand-rolled execlist ctx active tracking: We need to manually kick
requests to get those unpinned again. Once that's fixed we can try to
remove this again.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
[danvet: Add summary of why we need a pile of retire_requests.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>