The request tells us where to read the ringbuf from, so use that
information to simplify the error capture. If no request was active at
the time of the hang, the ring is idle and there is no information
inside the ring pertaining to the hang.

Note carefully that this will reduce the amount of information stored in
the error state - any ring without an active request will not be
recorded.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NDave Gordon <david.s.gordon@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1467616119-4093-8-git-send-email-chris@chris-wilson.co.uk

smatch complains:
	drivers/gpu/drm/i915/i915_gpu_error.c:503 i915_error_state_to_str()
	warn: inconsistent indenting
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1467470166-31717-4-git-send-email-chris@chris-wilson.co.ukReviewed-by: NMatthew Auld <matthew.auld@intel.com>

By using the same address for storing the HWS on every platform, we can
remove the platform specific vfuncs and reduce the get-seqno routine to
a single read of a cached memory location.

v2: Fix semaphore_passed() to look at the signaling engine (not the
waiter's)
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/1467390209-3576-8-git-send-email-chris@chris-wilson.co.uk

One particularly stressful scenario consists of many independent tasks
all competing for GPU time and waiting upon the results (e.g. realtime
transcoding of many, many streams). One bottleneck in particular is that
each client waits on its own results, but every client is woken up after
every batchbuffer - hence the thunder of hooves as then every client must
do its heavyweight dance to read a coherent seqno to see if it is the
lucky one.

Ideally, we only want one client to wake up after the interrupt and
check its request for completion. Since the requests must retire in
order, we can select the first client on the oldest request to be woken.
Once that client has completed his wait, we can then wake up the
next client and so on. However, all clients then incur latency as every
process in the chain may be delayed for scheduling - this may also then
cause some priority inversion. To reduce the latency, when a client
is added or removed from the list, we scan the tree for completed
seqno and wake up all the completed waiters in parallel.

Using igt/benchmarks/gem_latency, we can demonstrate this effect. The
benchmark measures the number of GPU cycles between completion of a
batch and the client waking up from a call to wait-ioctl. With many
concurrent waiters, with each on a different request, we observe that
the wakeup latency before the patch scales nearly linearly with the
number of waiters (before external factors kick in making the scaling much
worse). After applying the patch, we can see that only the single waiter
for the request is being woken up, providing a constant wakeup latency
for every operation. However, the situation is not quite as rosy for
many waiters on the same request, though to the best of my knowledge this
is much less likely in practice. Here, we can observe that the
concurrent waiters incur extra latency from being woken up by the
solitary bottom-half, rather than directly by the interrupt. This
appears to be scheduler induced (having discounted adverse effects from
having a rbtree walk/erase in the wakeup path), each additional
wake_up_process() costs approximately 1us on big core. Another effect of
performing the secondary wakeups from the first bottom-half is the
incurred delay this imposes on high priority threads - rather than
immediately returning to userspace and leaving the interrupt handler to
wake the others.

To offset the delay incurred with additional waiters on a request, we
could use a hybrid scheme that did a quick read in the interrupt handler
and dequeued all the completed waiters (incurring the overhead in the
interrupt handler, not the best plan either as we then incur GPU
submission latency) but we would still have to wake up the bottom-half
every time to do the heavyweight slow read. Or we could only kick the
waiters on the seqno with the same priority as the current task (i.e. in
the realtime waiter scenario, only it is woken up immediately by the
interrupt and simply queues the next waiter before returning to userspace,
minimising its delay at the expense of the chain, and also reducing
contention on its scheduler runqueue). This is effective at avoid long
pauses in the interrupt handler and at avoiding the extra latency in
realtime/high-priority waiters.

v2: Convert from a kworker per engine into a dedicated kthread for the
bottom-half.
v3: Rename request members and tweak comments.
v4: Use a per-engine spinlock in the breadcrumbs bottom-half.
v5: Fix race in locklessly checking waiter status and kicking the task on
adding a new waiter.
v6: Fix deciding when to force the timer to hide missing interrupts.
v7: Move the bottom-half from the kthread to the first client process.
v8: Reword a few comments
v9: Break the busy loop when the interrupt is unmasked or has fired.
v10: Comments, unnecessary churn, better debugging from Tvrtko
v11: Wake all completed waiters on removing the current bottom-half to
reduce the latency of waking up a herd of clients all waiting on the
same request.
v12: Rearrange missed-interrupt fault injection so that it works with
igt/drv_missed_irq_hang
v13: Rename intel_breadcrumb and friends to intel_wait in preparation
for signal handling.
v14: RCU commentary, assert_spin_locked
v15: Hide BUG_ON behind the compiler; report on gem_latency findings.
v16: Sort seqno-groups by priority so that first-waiter has the highest
task priority (and so avoid priority inversion).
v17: Add waiters to post-mortem GPU hang state.
v18: Return early for a completed wait after acquiring the spinlock.
Avoids adding ourselves to the tree if the is already complete, and
skips the awkward question of why we don't do completion wakeups for
waits earlier than or equal to ourselves.
v19: Prepare for init_breadcrumbs to fail. Later patches may want to
allocate during init, so be prepared to propagate back the error code.

Testcase: igt/gem_concurrent_blit
Testcase: igt/benchmarks/gem_latency
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
Cc: "Gong, Zhipeng" <zhipeng.gong@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: "Goel, Akash" <akash.goel@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> #v18
Link: http://patchwork.freedesktop.org/patch/msgid/1467390209-3576-6-git-send-email-chris@chris-wilson.co.uk

This way optimization from a previous patch works even better.

v2: Rebase.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>

   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

In order to simplify future patches, extract the
lazy_coherency optimisation our of the engine->get_seqno() vfunc into
its own callback.

v2: Rename the barrier to engine->irq_seqno_barrier to try and better
reflect that the barrier is only required after the user interrupt before
reading the seqno (to ensure that the seqno update lands in time as we
do not have strict seqno-irq ordering on all platforms).

Reviewed-by: Dave Gordon <david.s.gordon@intel.com> [#v2]

v3: Comments for hangcheck paranoia. Mika wanted to keep the extra
barrier inside the hangcheck, just in case. I can argue that it doesn't
provide a barrier against anything, but the side-effects of applying the
barrier may prevent a false declaration of a hung GPU.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1460195877-20520-2-git-send-email-chris@chris-wilson.co.uk

It's useful to look at the last seqno submitted on a particular engine
and compare it against the HWS value to check for irregularities.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1460010558-10705-1-git-send-email-chris@chris-wilson.co.uk

dev_priv is what the macro works hard to extract, pass it directly.

> sed 's/\([A-Z].*(dev_priv\)->dev)/\1)/g'

v2:
- Include all wrapper macros too (Chris)

v3:
- Include sed cmdline (Chris)

v4:
- Break long line
- Rebase

Cc: Chris Wilson <chris@chris-wilson.co.uk>
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/1460016485-8089-1-git-send-email-joonas.lahtinen@linux.intel.com

Refer to the GGTT VM consistently as "ggtt->base" instead of just "ggtt",
"vm" or indirectly through other variables like "dev_priv->ggtt.base"
to avoid confusion with the i915_ggtt object itself and PPGTT VMs.

Refer to the GGTT as "ggtt" instead of indirectly through chaining.

As a bonus gets rid of the long-standing i915_obj_to_ggtt vs.
i915_gem_obj_to_ggtt conflict, due to removal of i915_obj_to_ggtt!

v2:
- Added some more after grepping sources with Chris

v3:
- Refer to GGTT VM through ggtt->base consistently instead of ggtt_vm
  (Chris)

v4:
- Convert all dev_priv->ggtt->foo accesses to ggtt->foo.

v5:
- Make patch checker happy

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

Equivalent to the existing for_each_engine() macro, this will replace
the latter wherever the third argument *is* actually wanted (in most
places, it is not used). The third argument is renamed to emphasise
that it is an engine id (type enum intel_engine_id). All the callers of
the macro that actually need the third argument are updated to use this
version, and the argument (generally 'i') is also updated to be 'id'.
Other callers (where the third argument is unused) are untouched for
now; they will be updated in the next patch.
Signed-off-by: NDave Gordon <david.s.gordon@intel.com>
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

In preparation for engine reset, the wedged argument of i915_handle_error()
is extended to reflect as a mask of engines that are hung. This is further
passed down to error state capture functions which are also updated.

Engine reset recovery mechanism uses this mask and schedules recovery work
for those particular engines.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NTomas Elf <tomas.elf@intel.com>
Signed-off-by: NArun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1458331676-567-3-git-send-email-arun.siluvery@linux.intel.com

Refer to Global GTT consistently as GGTT, thus rename dev_priv->gtt
to dev_priv->ggtt and struct i915_gtt to struct i915_ggtt.

Fix a couple of whitespace problems while at it.

v2:
- Fix a typo in commit message.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NJoonas Lahtinen <joonas.lahtinen@linux.intel.com>

Some trivial ones, first pass done with Coccinelle:

@@
@@
(
- I915_NUM_RINGS
+ I915_NUM_ENGINES
|
- intel_ring_flag
+ intel_engine_flag
|
- for_each_ring
+ for_each_engine
|
- i915_gem_request_get_ring
+ i915_gem_request_get_engine
|
- intel_ring_idle
+ intel_engine_idle
|
- i915_gem_reset_ring_status
+ i915_gem_reset_engine_status
|
- i915_gem_reset_ring_cleanup
+ i915_gem_reset_engine_cleanup
|
- init_ring_lists
+ init_engine_lists
)

But that didn't fully work so I cleaned it up with:

for f in *.[hc]; do sed -i -e s/I915_NUM_RINGS/I915_NUM_ENGINES/ $f; done
for f in *.[hc]; do sed -i -e s/i915_gem_request_get_ring/i915_gem_request_get_engine/ $f; done
for f in *.[hc]; do sed -i -e s/intel_ring_flag/intel_engine_flag/ $f; done
for f in *.[hc]; do sed -i -e s/intel_ring_idle/intel_engine_idle/ $f; done
for f in *.[hc]; do sed -i -e s/init_ring_lists/init_engine_lists/ $f; done
for f in *.[hc]; do sed -i -e s/i915_gem_reset_ring_cleanup/i915_gem_reset_engine_cleanup/ $f; done
for f in *.[hc]; do sed -i -e s/i915_gem_reset_ring_status/i915_gem_reset_engine_status/ $f; done

v2: Rebase.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

below and a couple manual fixups.

@@
identifier I, J;
@@
struct I {
...
- struct intel_engine_cs *J;
+ struct intel_engine_cs *engine;
...
}
@@
identifier I, J;
@@
struct I {
...
- struct intel_engine_cs J;
+ struct intel_engine_cs engine;
...
}
@@
struct drm_i915_private *d;
@@
(
- d->ring
+ d->engine
)
@@
struct i915_execbuffer_params *p;
@@
(
- p->ring
+ p->engine
)
@@
struct intel_ringbuffer *r;
@@
(
- r->ring
+ r->engine
)
@@
struct drm_i915_gem_request *req;
@@
(
- req->ring
+ req->engine
)

v2: Script missed the tracepoint code - fixed up by hand.
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

@@
identifier func;
@@
func(..., struct intel_engine_cs *
- ring
+ engine
, ...)
{
<...
- ring
+ engine
...>
}
@@
identifier func;
type T;
@@
T func(..., struct intel_engine_cs *
- ring
+ engine
, ...);
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

Done by the Coccinelle script below plus a manual
intervention to GEN8_RING_SEMAPHORE_INIT.

@@
expression E;
@@
- struct intel_engine_cs *ring = E;
+ struct intel_engine_cs *engine = E;
<+...
- ring
+ engine
...+>
@@
@@
- struct intel_engine_cs *ring;
+ struct intel_engine_cs *engine;
<+...
- ring
+ engine
...+>
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

With full-ppgtt, it takes the GPU an eon to traverse the entire 256PiB
address space, causing a loop to be detected. Under the current scheme,
if ACTHD walks off the end of a batch buffer and into an empty
address space, we "never" detect the hang. If we always increment the
score as the ACTHD is progressing then we will eventually timeout (after
~46.5s (31 * 1.5s) without advancing onto a new batch). To counter act
this, increase the amount we reduce the score for good batches, so that
only a series of almost-bad batches trigger a full reset. DoS detection
suffers slightly but series of long running shader tests will benefit.

Based on a patch from Chris Wilson.

Testcase: igt/drv_hangman/hangcheck-unterminated
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NMika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1456930109-21532-1-git-send-email-mika.kuoppala@intel.com

execute during context save/restore, good to have them in error state.

v2: use wa_ctx->size and print only size values (Mika)
v3: simplify conditions when recording and freeing object (Chris)
v4: resolve checkpatch errors (Tvrtko)

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: NMika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NArun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1456831476-10782-1-git-send-email-arun.siluvery@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>

Revision id along with device id is useful in better identification of the HW
and its limitations so include this detail in error state.

v2: make it clear that it is PCI revision and We might as well dump PCI
subsystem details while we update this (Ville, Chris).

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NArun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1454001521-7701-1-git-send-email-arun.siluvery@linux.intel.com

Now that we've eliminated a lot of uses of ring->default_context,
we can eliminate the pointer itself.

All the engines share the same default intel_context, so we can just
keep a single reference to it in the dev_priv structure rather than one
in each of the engine[] elements. This make refcounting more sensible
too, as we now have a refcount of one for the one pointer, rather than
a refcount of one but multiple pointers.

From an idea by Chris Wilson.

v2:	transform an extra instance of ring->default_context introduced by
    42f1cae8 drm/i915: Restore inhibiting the load of the default context
    That patch's commentary includes:
	v2: Mark the global default context as uninitialized on GPU reset so
	    that the context-local workarounds are reloaded upon re-enabling
    The code implementing that now also benefits from the replacement of
    the multiple (per-ring) pointers to the default context with a single
    pointer to the unique kernel context.

v4:	Rebased, remove underused local (Nick Hoath)
Signed-off-by: NDave Gordon <david.s.gordon@intel.com>
Reviewed-by: NNick Hoath <nicholas.hoath@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1453230175-19330-3-git-send-email-david.s.gordon@intel.comSigned-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Make I915_READ and I915_WRITE more type safe by wrapping the register
offset in a struct. This should eliminate most of the fumbles we've had
with misplaced parens.

This only takes care of normal mmio registers. We could extend the idea
to other register types and define each with its own struct. That way
you wouldn't be able to accidentally pass the wrong thing to a specific
register access function.

The gpio_reg setup is probably the ugliest thing left. But I figure I'd
just leave it for now, and wait for some divine inspiration to strike
before making it nice.

As for the generated code, it's actually a bit better sometimes. Eg.
looking at i915_irq_handler(), we can see the following change:
  lea    0x70024(%rdx,%rax,1),%r9d
  mov    $0x1,%edx
- movslq %r9d,%r9
- mov    %r9,%rsi
- mov    %r9,-0x58(%rbp)
- callq  *0xd8(%rbx)
+ mov    %r9d,%esi
+ mov    %r9d,-0x48(%rbp)
 callq  *0xd8(%rbx)

So previously gcc thought the register offset might be signed and
decided to sign extend it, just in case. The rest appears to be
mostly just minor shuffling of instructions.

v2: i915_mmio_reg_{offset,equal,valid}() helpers added
    s/_REG/_MMIO/ in the register defines
    mo more switch statements left to worry about
    ring_emit stuff got sorted in a prep patch
    cmd parser, lrc context and w/a batch buildup also in prep patch
    vgpu stuff cleaned up and moved to a prep patch
    all other unrelated changes split out
v3: Rebased due to BXT DSI/BLC, MOCS, etc.
v4: Rebased due to churn, s/i915_mmio_reg_t/i915_reg_t/
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1447853606-2751-1-git-send-email-ville.syrjala@linux.intel.com

Use the RING_PSMI_CTL define insted of hand rolling the register offset.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446672017-24497-17-git-send-email-ville.syrjala@linux.intel.comReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

We have had one case where buggy csr/dmc firmware version influenced
gt side and caused a hang. Add dmc firmware loading state and
version to error state.

v2: - Rebased on top of Damien's patches
    - included fw load state
v3: include dmc info only if platform supports it (Chris)
v4: move *csr to branch scope (Chris)
v5: remove dependency to csr_state

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v4)
Signed-off-by: NMika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446124879-22240-1-git-send-email-mika.kuoppala@intel.com
Tested-by: Daniel Stone <daniels@collabora.com> # SKL
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>

Change FORCEWAKE & co. reads for the error state to use I915_READ_FW().
Reading a FORCEWAKE register using a function that can frob forcewake
just seems wrong.

There is a check to skip grabbing the forcewake for accessing FORCEWAKE
in intel_uncore.c, but there's no such check for FORCEWAKE_MT. So no
idea what is currently happening with FORCEWAKE_MT reads. FORCEWAKE_VLV
is fortunately outside the forcewake range anyway, so no actual issue
with that one.

So let's just make the rule that you can't access FORCEWAKE registers with
the normal I915_READ() stuff, and we can drop the extra FORCEWAKE check
from NEEDS_FORCEWAKE(). While at it use NEEDS_FORCEWAKE() on BDW, where
it was skipped for whatever bikeshed reason that I've already forgotten.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1445517300-28173-3-git-send-email-ville.syrjala@linux.intel.comReviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

Since we're not synchronizing the ring request list during error state capture
the request list state might change between the time the corresponding error
request list was allocated and dimensioned to the time when the ring request
list is actually captured into the error state. If this happens then do an
early exit and be aware that the captured error state might not be fully
reliable.

* v2:
- Chris Wilson: Removed WARN_ON from size check since having the error state
  request list and the live driver request list diverge like this is a
  legitimate behaviour.

- Tomas Elf: Removed update of num_request field since this made no sense. Just
  exit and move on.

* v3:
- Chris Wilson: Removed error message at the point of early exit. The user is
  not interested in any state changes happening during the error state capture,
  only in the state that we're trying to capture at the point of the error.
Signed-off-by: NTomas Elf <tomas.elf@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This reverts commit 51056723.

I somehow managed to combine a patch from Tomas Elf with a totally
unrelated commit message from Chris Wilson. Let's revert this and
reapply properly.
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>

Userspace can pass in an offset that it presumes the object is located
at. The kernel will then do its utmost to fit the object into that
location. The assumption is that userspace is handling its own object
locations (for example along with full-ppgtt) and that the kernel will
rarely have to make space for the user's requests.

v2: Fix i915_gem_evict_range() (now evict_for_vma) to handle ordinary
and fixed objects within the same batch
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: "Daniel, Thomas" <thomas.daniel@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This register was added on GEN4, by the name INSTDONE_1 whereas the GEN6
specification calls it INSTDONE_2. Keep the original name with a
platform prefix to make it clearer which INSTDONE register instance this
is. Also add a comment about the SNB alternative name.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We have a bunch of INSTDONE registers for different platforms and
purposes and it's not immediately clear which instance they are just by
looking at the register name. This one was added on GEN2, where it was
the only INSTDONE register, so mark it as such.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We use 3 different names to refer to the same render ring INSTDONE
register. This can be confusing when comparing two parts of the code
accessing the register via different names. Although the GEN4 version's
layout is different, we treat it the same way as the GEN7+ version, in
that we simply read it out during error capture. So remove the
duplicates and leave a comment about the GEN4 difference.

Note that there is also a GEN2 version of this register, but that's on a
different address so not handled in this patch.
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

v2: Hide the 945 vs. rest of gen2/3 difference in the macro
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Looks like this was introduced in:
commit d1675198
Author: Alex Dai <yu.dai@intel.com>
Date:   Wed Aug 12 15:43:43 2015 +0100

    drm/i915: Integrate GuC-based command submission

This patch assumed LRC contexts and HWS layout, which is incorrect on
platforms without execlists.  This can lead to a crash in GPU error
state readout on those platforms.

I don't see a bug filed for this, but there may be one that I haven't
found.

v2: fixup offset handling for error capture fix (Dave)

Cc: Alex Dai <yu.dai@intel.com>
Reviewed-by: NDave Gordon <david.s.gordon@intel.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Add a common function to return "yes" or "no" string based on the
argument, and drop the local versions of it.
Signed-off-by: NJani Nikula <jani.nikula@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

GuC-based submission is mostly the same as execlist mode, up to
intel_logical_ring_advance_and_submit(), where the context being
dispatched would be added to the execlist queue; at this point
we submit the context to the GuC backend instead.

There are, however, a few other changes also required, notably:
1.  Contexts must be pinned at GGTT addresses accessible by the GuC
    i.e. NOT in the range [0..WOPCM_SIZE), so we have to add the
    PIN_OFFSET_BIAS flag to the relevant GGTT-pinning calls.

2.  The GuC's TLB must be invalidated after a context is pinned at
    a new GGTT address.

3.  GuC firmware uses the one page before Ring Context as shared data.
    Therefore, whenever driver wants to get base address of LRC, we
    will offset one page for it. LRC_PPHWSP_PN is defined as the page
    number of LRCA.

4.  In the work queue used to pass requests to the GuC, the GuC
    firmware requires the ring-tail-offset to be represented as an
    11-bit value, expressed in QWords. Therefore, the ringbuffer
    size must be reduced to the representable range (4 pages).

v2:
    Defer adding #defines until needed [Chris Wilson]
    Rationalise type declarations [Chris Wilson]

v4:
    Squashed kerneldoc patch into here [Daniel Vetter]

v5:
    Update request->tail in code common to both GuC and execlist modes.
    Add a private version of lr_context_update(), as sharing the
        execlist version leads to race conditions when the CPU and
        the GuC both update TAIL in the context image.
    Conversion of error-captured HWS page to string must account
        for offset from start of object to actual HWS (LRC_PPHWSP_PN).

Issue: VIZ-4884
Signed-off-by: NAlex Dai <yu.dai@intel.com>
Signed-off-by: NDave Gordon <david.s.gordon@intel.com>
Reviewed-by: NTom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

v2: For semaphore errors, object is mapped to GGTT and offset will not
be > 4GB, print only lower 32-bits (Akash)
v3: Print gtt_offset in groups of 32-bit (Chris)

Cc: Akash Goel <akash.goel@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NMichel Thierry <michel.thierry@intel.com>
Reviewed-by: NAkash Goel <akash.goel@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The IOMMU for Intel graphics has historically had many issues resulting
in random GPU hangs. Lets include its status when capturing the GPU hang
error state for post-mortem analysis.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Currently, we only track the last request globally across all engines.
This prevents us from issuing concurrent read requests on e.g. the RCS
and BCS engines (or more likely the render and media engines). Without
semaphores, we incur costly stalls as we synchronise between rings -
greatly impacting the current performance of Broadwell versus Haswell in
certain workloads (like video decode). With the introduction of
reference counted requests, it is much easier to track the last request
per ring, as well as the last global write request so that we can
optimise inter-engine read read requests (as well as better optimise
certain CPU waits).

v2: Fix inverted readonly condition for nonblocking waits.
v3: Handle non-continguous engine array after waits
v4: Rebase, tidy, rewrite ring list debugging
v5: Use obj->active as a bitfield, it looks cool
v6: Micro-optimise, mostly involving moving code around
v7: Fix retire-requests-upto for execlists (and multiple rq->ringbuf)
v8: Rebase
v9: Refactor i915_gem_object_sync() to allow the compiler to better
optimise it.

Benchmark: igt/gem_read_read_speed
hsw:gt3e (with semaphores):
Before: Time to read-read 1024k:		275.794µs
After:  Time to read-read 1024k:		123.260µs

hsw:gt3e (w/o semaphores):
Before: Time to read-read 1024k:		230.433µs
After:  Time to read-read 1024k:		124.593µs

bdw-u (w/o semaphores):             Before          After
Time to read-read 1x1:            26.274µs       10.350µs
Time to read-read 128x128:        40.097µs       21.366µs
Time to read-read 256x256:        77.087µs       42.608µs
Time to read-read 512x512:       281.999µs      181.155µs
Time to read-read 1024x1024:    1196.141µs     1118.223µs
Time to read-read 2048x2048:    5639.072µs     5225.837µs
Time to read-read 4096x4096:   22401.662µs    21137.067µs
Time to read-read 8192x8192:   89617.735µs    85637.681µs

Testcase: igt/gem_concurrent_blit (read-read and friends)
Cc: Lionel Landwerlin <lionel.g.landwerlin@linux.intel.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> [v8]
[danvet: s/\<rq\>/req/g]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

wa_batchbuffer is part of some error states. Make sure it
is freed.
Signed-off-by: NMika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>