If the device is asleep (no GT wakeref), we know the GPU is already idle.
If we add an early return, we can avoid touching registers and checking
hw state outside of the assumed GT wakelock. This prevents causing such
errors whilst debugging:

[ 2613.401647] RPM wakelock ref not held during HW access
[ 2613.401684] ------------[ cut here ]------------
[ 2613.401720] WARNING: CPU: 5 PID: 7739 at drivers/gpu/drm/i915/intel_drv.h:1787 gen6_read32+0x21f/0x2b0 [i915]
[ 2613.401731] Modules linked in: snd_hda_intel i915 vgem snd_hda_codec_hdmi x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_realtek coretemp snd_hda_codec_generic crct10dif_pclmul crc32_pclmul ghash_clmulni_intel snd_hda_codec snd_hwdep snd_hda_core snd_pcm r8169 mii mei_me lpc_ich mei prime_numbers [last unloaded: i915]
[ 2613.401823] CPU: 5 PID: 7739 Comm: drv_missed_irq Tainted: G     U          4.12.0-rc2-CI-CI_DRM_421+ #1
[ 2613.401825] Hardware name: MSI MS-7924/Z97M-G43(MS-7924), BIOS V1.12 02/15/2016
[ 2613.401840] task: ffff880409e3a740 task.stack: ffffc900084dc000
[ 2613.401861] RIP: 0010:gen6_read32+0x21f/0x2b0 [i915]
[ 2613.401863] RSP: 0018:ffffc900084dfce8 EFLAGS: 00010292
[ 2613.401869] RAX: 000000000000002a RBX: ffff8804016a8000 RCX: 0000000000000006
[ 2613.401871] RDX: 0000000000000006 RSI: ffffffff81cbf2d9 RDI: ffffffff81c9e3a7
[ 2613.401874] RBP: ffffc900084dfd18 R08: ffff880409e3afc8 R09: 0000000000000000
[ 2613.401877] R10: 000000008a1c483f R11: 0000000000000000 R12: 000000000000209c
[ 2613.401879] R13: 0000000000000001 R14: ffff8804016a8000 R15: ffff8804016ac150
[ 2613.401882] FS:  00007f39ef3dd8c0(0000) GS:ffff88041fb40000(0000) knlGS:0000000000000000
[ 2613.401885] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 2613.401887] CR2: 00000000023717c8 CR3: 00000002e7b34000 CR4: 00000000001406e0
[ 2613.401889] Call Trace:
[ 2613.401912]  intel_engine_is_idle+0x76/0x90 [i915]
[ 2613.401931]  i915_gem_wait_for_idle+0xe6/0x1e0 [i915]
[ 2613.401951]  fault_irq_set+0x40/0x90 [i915]
[ 2613.401970]  i915_ring_test_irq_set+0x42/0x50 [i915]
[ 2613.401976]  simple_attr_write+0xc7/0xe0
[ 2613.401981]  full_proxy_write+0x4f/0x70
[ 2613.401987]  __vfs_write+0x23/0x120
[ 2613.401992]  ? rcu_read_lock_sched_held+0x75/0x80
[ 2613.401996]  ? rcu_sync_lockdep_assert+0x2a/0x50
[ 2613.401999]  ? __sb_start_write+0xfa/0x1f0
[ 2613.402004]  vfs_write+0xc5/0x1d0
[ 2613.402008]  ? trace_hardirqs_on_caller+0xe7/0x1c0
[ 2613.402013]  SyS_write+0x44/0xb0
[ 2613.402020]  entry_SYSCALL_64_fastpath+0x1c/0xb1
[ 2613.402022] RIP: 0033:0x7f39eded6670
[ 2613.402025] RSP: 002b:00007fffdcdcb1a8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[ 2613.402030] RAX: ffffffffffffffda RBX: ffffffff81470203 RCX: 00007f39eded6670
[ 2613.402033] RDX: 0000000000000001 RSI: 000000000041bc33 RDI: 0000000000000006
[ 2613.402036] RBP: ffffc900084dff88 R08: 00007f39ef3dd8c0 R09: 0000000000000001
[ 2613.402038] R10: 0000000000000000 R11: 0000000000000246 R12: 000000000041bc33
[ 2613.402041] R13: 0000000000000006 R14: 0000000000000000 R15: 0000000000000000
[ 2613.402046]  ? __this_cpu_preempt_check+0x13/0x20
[ 2613.402052] Code: 01 9b fa e0 0f ff e9 28 fe ff ff 80 3d 6a dd 0e 00 00 0f 85 29 fe ff ff 48 c7 c7 48 19 29 a0 c6 05 56 dd 0e 00 01 e8 da 9a fa e0 <0f> ff e9 0f fe ff ff b9 01 00 00 00 ba 01 00 00 00 44 89 e6 48
[ 2613.402199] ---[ end trace 31f0cfa93ab632bf ]---

Fixes: 25112b64 ("drm/i915: Wait for all engines to be idle as part of i915_gem_wait_for_idle()")
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/20170530121334.17364-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

We depend on intel_iommu_gfx_mapped for various workarounds, but that is
only available under an #ifdef CONFIG_INTEL_IOMMU. Refactor all the
cut-and-paste ifdefs to a common routine.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170525121612.2190-1-chris@chris-wilson.co.ukReviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>

Now that drm_[cm]alloc* helpers are simple one line wrappers around
kvmalloc_array and drm_free_large is just kvfree alias we can drop
them and replace by their native forms.

This shouldn't introduce any functional change.

Changes since v1
- fix typo in drivers/gpu//drm/etnaviv/etnaviv_gem.c - noticed by 0day
  build robot
Suggested-by: NDaniel Vetter <daniel@ffwll.ch>
Signed-off-by: Michal Hocko <mhocko@suse.com>drm: drop drm_[cm]alloc* helpers
[danvet: Fixup vgem which grew another user very recently.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: NChristian König <christian.koenig@amd.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517122312.GK18247@dhcp22.suse.cz

The i915_priolist are allocated within an atomic context on a path where
we wish to minimise latency. If we use a dedicated kmem_cache, we have
the advantage of a local freelist from which to service new requests
that should keep the latency impact of an allocation small. Though
currently we expect the majority of requests to be at default priority
(and so hit the preallocate priolist), once userspace starts using
priorities they are likely to use many fine grained policies improving
the utilisation of a private slab.
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>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-9-chris@chris-wilson.co.uk

All the requests at the same priority are executed in FIFO order. They
do not need to be stored in the rbtree themselves, as they are a simple
list within a level. If we move the requests at one priority into a list,
we can then reduce the rbtree to the set of priorities. This should keep
the height of the rbtree small, as the number of active priorities can not
exceed the number of active requests and should be typically only a few.

Currently, we have ~2k possible different priority levels, that may
increase to allow even more fine grained selection. Allocating those in
advance seems a waste (and may be impossible), so we opt for allocating
upon first use, and freeing after its requests are depleted. To avoid
the possibility of an allocation failure causing us to lose a request,
we preallocate the default priority (0) and bump any request to that
priority if we fail to allocate it the appropriate plist. Having a
request (that is ready to run, so not leading to corruption) execute
out-of-order is better than leaking the request (and its dependency
tree) entirely.

There should be a benefit to reducing execlists_dequeue() to principally
using a simple list (and reducing the frequency of both rbtree iteration
and balancing on erase) but for typical workloads, request coalescing
should be small enough that we don't notice any change. The main gain is
from improving PI calls to schedule, and the explicit list within a
level should make request unwinding simpler (we just need to insert at
the head of the list rather than the tail and not have to make the
rbtree search more complicated).

v2: Avoid use-after-free when deleting a depleted priolist

v3: Michał found the solution to handling the allocation failure
gracefully. If we disable all priority scheduling following the
allocation failure, those requests will be executed in fifo and we will
ensure that this request and its dependencies are in strict fifo (even
when it doesn't realise it is only a single list). Normal scheduling is
restored once we know the device is idle, until the next failure!
Suggested-by: NMichał Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NMichał Winiarski <michal.winiarski@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-8-chris@chris-wilson.co.uk

add/remove: 1/1 grow/shrink: 5/4 up/down: 391/-578 (-187)
function                                     old     new   delta
execlists_submit_ports                       262     471    +209
port_assign.isra                               -     136    +136
capture                                     6344    6359     +15
reset_common_ring                            438     452     +14
execlists_submit_request                     228     238     +10
gen8_init_common_ring                        334     341      +7
intel_engine_is_idle                         106     105      -1
i915_engine_info                            2314    2290     -24
__i915_gem_set_wedged_BKL                    485     411     -74
intel_lrc_irq_handler                       1789    1604    -185
execlists_update_context                     294       -    -294

The most important change there is the improve to the
intel_lrc_irq_handler and excclist_submit_ports (net improvement since
execlists_update_context is now inlined).

v2: Use the port_api() for guc as well (even though currently we do not
pack any counters in there, yet) and hide all port->request_count inside
the helpers.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-5-chris@chris-wilson.co.uk

Rebrand the current (pointer | bits) pack/unpack utility macros as
explicit bit twiddling for PAGE_SIZE so that we can use the more
flexible underlying macros for different bits.
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/20170517121007.27224-4-chris@chris-wilson.co.uk

ptr_unpack_bits() is a function-like macro, as such it is meant to be
replaceable by a function. In this case, we should be passing in the
out-param as a pointer.

Bizarrely this does affect code generation:

function                                     old     new   delta
i915_gem_object_pin_map                      409     389     -20

An improvement(?) in this case, but one can't help wonder what
strict-aliasing optimisations we are preventing.

The generated code looks identical in using ptr_unpack_bits (no extra
motions to stack, the pointer and bits appear to be kept in registers),
the difference appears to be code ordering and with a reorder it is able
to use smaller forward jumps.
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/20170517121007.27224-3-chris@chris-wilson.co.uk

Track the latest fence waited upon on each context, and only add a new
asynchronous wait if the new fence is more recent than the recorded
fence for that context. This requires us to filter out unordered
timelines, which are noted by DMA_FENCE_NO_CONTEXT. However, in the
absence of a universal identifier, we have to use our own
i915->mm.unordered_timeline token.

v2: Throw around the debug crutches
v3: Inline the likely case of the pre-allocation cache being full.
v4: Drop the pre-allocation support, we can lose the most recent fence
in case of allocation failure -- it just means we may emit more awaits
than strictly necessary but will not break.
v5: Trim allocation size for leaf nodes, they only need an array of u32
not pointers.
v6: Create mock_timeline to tidy selftest writing
v7: s/intel_timeline_sync_get/intel_timeline_sync_is_later/ (Tvrtko)
v8: Prune the stale sync points when we idle.
v9: Include a small benchmark in the kselftests
v10: Separate the idr implementation into its own compartment. (Tvrkto)
v11: Refactor igt_sync kselftests to avoid deep nesting (Tvrkto)
v12: __sync_leaf_idx() to assert that p->height is 0 when checking leaves
v13: kselftests to investigate struct i915_syncmap itself (Tvrtko)
v14: Foray into ascii art graphs
v15: Take into account that the random lookup/insert does 2 prng calls,
not 1, when benchmarking, and use for_each_set_bit() (Tvrtko)
v16: Improved ascii art
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170503093924.5320-4-chris@chris-wilson.co.uk

2 clflushes on two different objects are not ordered, and so do not
belong to the same timeline (context). Either we use a unique context
for each, or we reserve a special global context to mean unordered.
Ideally, we would reserve 0 to mean unordered (DMA_FENCE_NO_CONTEXT) to
have the same semantics everywhere.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170503093924.5320-1-chris@chris-wilson.co.uk

ILK should survive a reset without display corruption.
Suggested-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

HAS_HW_CONTEXTS is misleading condition for GPU reset and CCID,
replace it with Gen specific (to be updated in next patches).

HAS_HW_CONTEXTS in i915_l3_write is bogus because each HAS_L3_DPF
match also has .has_hw_contexts = 1 set.

This leads to us being able to get rid of the property completely.

v2:
- Keep the checks at Gen6 for no functional change (Ville)
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>

Many sightings report the greater prevalence of allocation failures.
This is all due to the incorrect use of mapping_gfp_constraint(), so
remove it in favour of just querying the mapping_gfp_mask() which are
the exact gfp_t we wanted in the first place.

We still do expect a higher chance of reporting ENOMEM, as that is the
intention of using __GFP_NORETRY -- to fail rather than oom after having
reclaimed from our bo caches, and having done a direct|kswapd reclaim
pass.
Reported-by: NJason Ekstrand <jason.ekstrand@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100594
Fixes: 24f8e00a ("drm/i915: Prefer to report ENOMEM rather than incur the oom for gfx allocations")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170405221514.23251-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
(cherry picked from commit b268d9fe)
Signed-off-by: NJani Nikula <jani.nikula@intel.com>

A group of Linux kernel hackers reported chasing a bug that resulted
from their assumption that SLAB_DESTROY_BY_RCU provided an existence
guarantee, that is, that no block from such a slab would be reallocated
during an RCU read-side critical section.  Of course, that is not the
case.  Instead, SLAB_DESTROY_BY_RCU only prevents freeing of an entire
slab of blocks.

However, there is a phrase for this, namely "type safety".  This commit
therefore renames SLAB_DESTROY_BY_RCU to SLAB_TYPESAFE_BY_RCU in order
to avoid future instances of this sort of confusion.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: <linux-mm@kvack.org>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NVlastimil Babka <vbabka@suse.cz>
[ paulmck: Add comments mentioning the old name, as requested by Eric
  Dumazet, in order to help people familiar with the old name find
  the new one. ]
Acked-by: NDavid Rientjes <rientjes@google.com>

When discussing a new WC mmap, we based the interface upon the
assumption that GTT was fully coherent. How naive! Commits 3b5724d7
("drm/i915: Wait for writes through the GTT to land before reading
back") and ed4596ea ("drm/i915/guc: WA to address the Ringbuffer
coherency issue") demonstrate that writes through the GTT are indeed
delayed and may be overtaken by direct WC access. To be safe, if
userspace is mixing WC mmaps with other potential GTT access (pwrite,
GTT mmaps) it should use set_domain(WC).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96563
Testcase: igt/gem_pwrite/small-gtt*
Testcase: igt/drv_selftest/coherency
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/20170412110111.26626-2-chris@chris-wilson.co.uk

In the next patch, we will introduce a new cache domain for
differentiating between GTT access and direct WC access. This will
require us to include WC in our write_domain flushes. Rather than
duplicate a third function, combine the existing two into one and
flushing WC writes will then be automatically handled as well.

v2: Be smarter and clearer by passing in the write domains to flush (Joonas)
v3: One missed ~ in v2 conversion
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170412110111.26626-1-chris@chris-wilson.co.uk

We want to refer to the index of the engine consistently throughout the
userspace ABI. We already have such an index through the execbuffer
engine specifier, that needs to be able to refer to each engine
specifically, so rename it the index to uabi_id to reflect its
generality beyond execbuf.
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/20170411124306.15448-1-chris@chris-wilson.co.uk

i915 is currently doing a full GPU reset at the end of
i915_gem_suspend() followed by GuC suspend in i915_drm_suspend(). This
GPU reset clobbers the GuC, causing the suspend request to then fail,
leaving the GuC in an undefined state. We need to tell the GuC to
suspend before we do the direct intel_gpu_reset().

v2: Commit message update. (Chris, Daniele)

Fixes: 1c777c5d ("drm/i915/hsw: Fix GPU hang during resume from S3-devices state")
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NSagar Arun Kamble <sagar.a.kamble@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491387710-20553-1-git-send-email-sagar.a.kamble@intel.comReviewed-by: NDaniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
(cherry picked from commit fd089233)
Signed-off-by: NJani Nikula <jani.nikula@intel.com>

As we may have very many objects to free, check to see if the task needs
to be rescheduled whilst freeing them.
Suggested-by: NAndrea Arcangeli <aarcange@redhat.com>
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/20170407102552.5781-4-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Before freeing the next batch of objects from the worker, check if the
worker's timeslice has expired and if so, defer the next batch to the
next invocation of the worker.
Suggested-by: NAndrea Arcangeli <aarcange@redhat.com>
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/20170407102552.5781-3-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

By using the same structure for both interruptible and
uninterruptible locking in shrinker code, combined with the
information that mm.interruptible is only being written to, the
code can be greatly simplified.

Also removing the i915_gem_ prefix from the locking functions so
that nobody in their wildest dreams considers exporting them.
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1491562175-27680-1-git-send-email-joonas.lahtinen@linux.intel.com

As we call into the shrinker during freeze, we may have freed more
objects since we idled during i915_gem_suspend. Make sure we flush the
i915_gem_free_objects worker prior to saving the unwanted pages into the
hibernation image.
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/20170407102552.5781-2-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

The shrinker is prepared to be called unlocked (and at other times with
struct_mutex held for DIRECT_RECLAIM) so we can skip acquiring the
struct_mutex prior to calling the shrinker during freeze. This improves
our ability to shrink as we can be more aggressive when we know the
caller isn't holding struct_mutex.
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/20170407102552.5781-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Many sightings report the greater prevalence of allocation failures.
This is all due to the incorrect use of mapping_gfp_constraint(), so
remove it in favour of just querying the mapping_gfp_mask() which are
the exact gfp_t we wanted in the first place.

We still do expect a higher chance of reporting ENOMEM, as that is the
intention of using __GFP_NORETRY -- to fail rather than oom after having
reclaimed from our bo caches, and having done a direct|kswapd reclaim
pass.
Reported-by: NJason Ekstrand <jason.ekstrand@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100594
Fixes: 24f8e00a ("drm/i915: Prefer to report ENOMEM rather than incur the oom for gfx allocations")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170405221514.23251-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

i915 is currently doing a full GPU reset at the end of
i915_gem_suspend() followed by GuC suspend in i915_drm_suspend(). This
GPU reset clobbers the GuC, causing the suspend request to then fail,
leaving the GuC in an undefined state. We need to tell the GuC to
suspend before we do the direct intel_gpu_reset().

v2: Commit message update. (Chris, Daniele)

Fixes: 1c777c5d ("drm/i915/hsw: Fix GPU hang during resume from S3-devices state")
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NSagar Arun Kamble <sagar.a.kamble@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491387710-20553-1-git-send-email-sagar.a.kamble@intel.comReviewed-by: NDaniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Since this can be triggered by simply attempting a huge object,
a WARN_ON is not appropriate.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330163130.24141-1-tvrtko.ursulin@linux.intel.com

Make i915_gem_wait_for_idle() be a little heavier in order to try and
guarantee that the GPU is indeed idle (by checking each engine
individually is idle, i.e. all writes are complete and the rings
stopped) after waiting for in-flight requests to be completed.

v2: And return the final error.
v3: Break the wait_for() out from under the WARN -- the macro expansion
is hideous and unreadable in the warning message
v4: If wait_for_engine() fails the result is catastrophic, mark the
device as wedged and wait for the repair team.

References: https://bugs.freedesktop.org/show_bug.cgi?id=98836Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330145041.9005-4-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

As we now distinguish everywhere that can call
i915_gem_retire_requests() following a successful wait_for_idle, we can
remove the duplication by moving that call into i915_gem_wait_for_idle()
itself.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330145041.9005-3-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

If the driver is wedged, HW state may be very inconsistent and
report that it is still busy, even though we have stopped using it. This
can lead to a double *ERROR* rather than a graceful cleanup after
wedging.
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/20170330145041.9005-2-chris@chris-wilson.co.uk

As we declare an engine as wedged, we mark all of its active requests as
in error. However, we don't want to mark successfully completed requests
as in error, which requires us to retire those requests first.
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/20170330145041.9005-1-chris@chris-wilson.co.uk

The should happen as soon as possible, but always within the logic that
depends on it (and not interrupting the top-level driver control flow).

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: NOscar Mateo <oscar.mateo@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1490720027-23234-1-git-send-email-oscar.mateo@intel.com

We can assume that if the device is asleep then all pending GTT writes
will have been posted, and so we can defer the flush from
i915_gem_object_flush_gtt_write_domain()

[ 1957.462568] WARNING: CPU: 0 PID: 6132 at drivers/gpu/drm/i915/intel_drv.h:1742 fwtable_read32+0x123/0x150 [i915]
[ 1957.462582] RPM wakelock ref not held during HW access
[ 1957.462583] Modules linked in: i915 intel_gtt drm_kms_helper prime_numbers
[ 1957.462607] CPU: 0 PID: 6132 Comm: gem_concurrent_ Tainted: G     U          4.11.0-rc1+ #464
[ 1957.462619] Hardware name:                  /        , BIOS PYBSWCEL.86A.0027.2015.0507.1758 05/07/2015
[ 1957.462630] Call Trace:
[ 1957.462646]  dump_stack+0x4d/0x6f
[ 1957.462657]  __warn+0xc1/0xe0
[ 1957.462667]  warn_slowpath_fmt+0x4a/0x50
[ 1957.462709]  fwtable_read32+0x123/0x150 [i915]
[ 1957.462750]  i915_gem_object_flush_gtt_write_domain+0x43/0x70 [i915]
[ 1957.462791]  i915_gem_object_set_to_cpu_domain+0x46/0xa0 [i915]
[ 1957.462831]  i915_gem_set_domain_ioctl+0x15d/0x220 [i915]
[ 1957.462843]  drm_ioctl+0x1d7/0x440
[ 1957.462885]  ? i915_gem_obj_prepare_shmem_write+0x1d0/0x1d0 [i915]
[ 1957.462896]  ? pick_next_task_fair+0x436/0x440
[ 1957.462906]  ? mntput+0x1f/0x30
[ 1957.462915]  do_vfs_ioctl+0x8f/0x5c0
[ 1957.462925]  ? __schedule+0x16f/0x5f0
[ 1957.462935]  ? ____fput+0x9/0x10
[ 1957.462943]  SyS_ioctl+0x3c/0x70
[ 1957.462952]  entry_SYSCALL_64_fastpath+0x17/0x98
[ 1957.462961] RIP: 0033:0x7fc542179ca7
[ 1957.462968] RSP: 002b:00007ffeef12ff98 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[ 1957.462982] RAX: ffffffffffffffda RBX: 00007ffeef1301d0 RCX: 00007fc542179ca7
[ 1957.462990] RDX: 00007ffeef12ffd0 RSI: 00000000400c645f RDI: 0000000000000003
[ 1957.462999] RBP: 0000000000000003 R08: 000055f433bc7c40 R09: 000000000000002c
[ 1957.463006] R10: 0000000000000073 R11: 0000000000000246 R12: 0000000000000018
[ 1957.463015] R13: 000055f432c89d20 R14: 000055f432c87690 R15: 0000000000000000

Fixes: 3b5724d7 ("drm/i915: Wait for writes through the GTT to land before reading back")
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/20170323150053.28582-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Starting with intel_guc_loader, down to intel_guc_submission
and finally to intel_guc_log.

v2:
  - Null execbuf client outside guc_client_free (Daniele)
  - Assert if things try to get allocated twice (Daniele/Joonas)
  - Null guc->log.buf_addr when destroyed (Daniele)
  - Newline between returning success and error labels (Joonas)
  - Remove some unnecessary comments (Joonas)
  - Keep guc_log_create_extras naming convention (Joonas)
  - Helper function guc_log_has_extras (Joonas)
  - No need for separate relay_channel create/destroy. It's just another extra.
  - No need to nullify guc->log.flush_wq when destroyed (Joonas)
  - Hoist the check for has_extras out of guc_log_create_extras (Joonas)
  - Try to do i915_guc_log_register/unregister calls (kind of) symmetric (Daniele)
  - Make sure initel_guc_fini is not called before init is ever called (Daniele)

v3:
  - Remove unnecessary parenthesis (Joonas)
  - Check for logs enabled on debugfs registration
  - Rebase on top of Tvrtko's "Fix request re-submission after reset"

v4:
  - Rebased
  - Comment around enabling/disabling interrupts inside GuC logging (Joonas)
Signed-off-by: NOscar Mateo <oscar.mateo@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Words cannot describe the embarrassment at creating a new gfp_t relaim to
only prevent the oomkiller but allow direct|kswapd reclaim, and then not
use it in the shmem_read_mapping_page_gfp().

Fixes: 24f8e00a ("drm/i915: Prefer to report ENOMEM rather than incur the oom for gfx allocations")
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170322223447.7493-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Since gfx allocations tend to be large, unmovable and disposable, report
the allocation failure back to userspace as an ENOMEM rather than incur
the oomkiller. We have already tried to make room by purging our own
cached gfx objects, and the oomkiller doesn't attribute ownership of gfx
objects so will likely pick the wrong candidate. Instead, let userspace
see the ENOMEM.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170322110521.29930-1-chris@chris-wilson.co.ukReviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The trick of using an uncached mmio read to ensure that the GGTT writes
are flushed does not require us to do the forcewake dance, so avoid it
in the hope of reducing the frequency that we do keep the device forced
awake.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170318104257.694-1-chris@chris-wilson.co.ukReviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>

Use pagecache_write to avoid shmemfs clearing the pages prior to us
immediately overwriting them with our data.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170317194648.12468-2-chris@chris-wilson.co.ukReviewed-by: NMatthew Auld <matthew.auld@intel.com>

Both object creation and backing storage page allocation do not require
struct_mutex, so do not require the caller to take it.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170317194648.12468-1-chris@chris-wilson.co.ukReviewed-by: NMatthew Auld <matthew.auld@intel.com>

When we wedge the device, we override engine->submit_request with a nop
to ensure that all in-flight requests are marked in error. However, igt
would like to unwedge the device to test -EIO handling. This requires us
to flush those in-flight requests and restore the original
engine->submit_request.

v2: Use a vfunc to unify enabling request submission to engines
v3: Split new vfunc to a separate patch.
v4: Make the wait interruptible -- the third party fences we wait upon
may be indefinitely broken, so allow the reset to be aborted.

Fixes: 821ed7df ("drm/i915: Update reset path to fix incomplete requests")
Testcase: igt/gem_eio
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> #v3
Link: http://patchwork.freedesktop.org/patch/msgid/20170316171305.12972-3-chris@chris-wilson.co.uk

I915_RESET_IN_PROGRESS is being used for both signaling the requirement
to i915_mutex_lock_interruptible() to avoid taking the struct_mutex and
to instruct a waiter (already holding the struct_mutex) to perform the
reset. To allow for a little more coordination, split these two meaning
into a couple of distinct flags. I915_RESET_BACKOFF tells
i915_mutex_lock_interruptible() not to acquire the mutex and
I915_RESET_HANDOFF tells the waiter to call i915_reset().
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Acked-by: NMichel Thierry <michel.thierry@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170316171305.12972-1-chris@chris-wilson.co.uk