Chris Wilson [Fri, 1 Jul 2016 16:23:18 +0000 (17:23 +0100)]
drm/i915: Stop mapping the scratch page into CPU space
After the elimination of using the scratch page for Ironlake's
breadcrumb, we no longer need to kmap the object. We therefore can move
it into the high unmappable space and do not need to force the object to
be coherent (i.e. snooped on !llc platforms).
Chris Wilson [Fri, 1 Jul 2016 16:23:17 +0000 (17:23 +0100)]
drm/i915: Use HWS for seqno tracking everywhere
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)
Chris Wilson [Fri, 1 Jul 2016 16:23:16 +0000 (17:23 +0100)]
drm/i915: Spin after waking up for an interrupt
When waiting for an interrupt (waiting for the engine to complete some
work), we know we are the only waiter to be woken on this engine. We also
know when the GPU has nearly completed our request (or at least started
processing it), so after being woken and we detect that the GPU is
active and working on our request, allow us the bottom-half (the first
waiter who wakes up to handle checking the seqno after the interrupt) to
spin for a very short while to reduce client latencies.
The impact is minimal, there was an improvement to the realtime-vs-many
clients case, but exporting the function proves useful later. However,
it is tempting to adjust irq_seqno_barrier to include the spin. The
problem is first ensuring that the "start-of-request" seqno is coherent
as we use that as our basis for judging when it is ok to spin. If we
could, spinning there could dramatically shorten some sleeps, and allow
us to make the barriers more conservative to handle missed seqno writes
on more platforms (all gen7+ are known to have the occasional issue, at
least).
Chris Wilson [Fri, 1 Jul 2016 16:23:15 +0000 (17:23 +0100)]
drm/i915: Slaughter the thundering i915_wait_request herd
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: Chris 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
Chris Wilson [Fri, 1 Jul 2016 16:23:14 +0000 (17:23 +0100)]
drm/i915: Separate GPU hang waitqueue from advance
Currently __i915_wait_request uses a per-engine wait_queue_t for the dual
purpose of waking after the GPU advances or for waking after an error.
In the future, we may add even more wake sources and require greater
separation, but for now we can conceptually simplify wakeups by separating
the two sources. In particular, this allows us to use different wait-queues
(e.g. one on the engine advancement, a global one for errors and one on
each requests) without any hassle.
Chris Wilson [Fri, 1 Jul 2016 16:23:12 +0000 (17:23 +0100)]
drm/i915: Remove the dedicated hangcheck workqueue
The queue only ever contains at most one item and has no special flags.
It is just a very simple wrapper around the system-wq - a complication
with no benefits.
v2: Use the system_long_wq as we may wish to capture the error state
after detecting the hang - which may take a bit of time.
Chris Wilson [Fri, 1 Jul 2016 16:23:11 +0000 (17:23 +0100)]
drm/i915: Delay queuing hangcheck to wait-request
We can forgo queuing the hangcheck from the start of every request to
until we wait upon a request. This reduces the overhead of every
request, but may increase the latency of detecting a hang. However, if
nothing every waits upon a hang, did it ever hang? It also improves the
robustness of the wait-request by ensuring that the hangchecker is
indeed running before we sleep indefinitely (and thereby ensuring that
we never actually sleep forever waiting for a dead GPU).
As pointed out by Tvrtko, it is possible for a GPU hang to go unnoticed
for as long as nobody is waiting for the GPU. Though this rare, during
that time we may be consuming more power than if we had promptly
recovered, and in the most extreme case we may exhaust all memory before
forcing the hangcheck. Something to be wary off in future.
Chris Wilson [Fri, 1 Jul 2016 16:23:10 +0000 (17:23 +0100)]
drm/i915/shrinker: Flush active on objects before counting
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.
Pooled EU is a bxt only feature and kernel changes are already merged. This
feature is not yet exposed to userspace as the support was not yet
available. Beignet team expressed interest and added patches to use this.
Since we now have a user and patches to use them, expose them from the
kernel side as well.
Cc: Winiarski, Michal <michal.winiarski@intel.com> Cc: Zou, Nanhai <nanhai.zou@intel.com> Cc: Yang, Rong R <rong.r.yang@intel.com> Cc: Tim Gore <tim.gore@intel.com> Cc: Jeff McGee <jeff.mcgee@intel.com> Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com> Acked-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1467369782-25992-1-git-send-email-arun.siluvery@linux.intel.com Acked-by: Jani Nikula <jani.nikula@intel.com>
Imre Deak [Wed, 29 Jun 2016 16:13:55 +0000 (19:13 +0300)]
drm/i915/bxt: Fix sanity check for BIOS RC6 setup
BXT BIOS has two options related to GPU power management: "RC6(Render
Standby)" and "GT PM Support". The assumption so far was that disabling
either of these options would leave RC6 uninitialized. According to my
tests this isn't so: for a proper RC6 setup we only need the "GT PM
Support" option to be enabled while the "RC6" option only controls
whether RC6 is left enabled or not by BIOS. OTOH we were missing a few
checks to ensure a proper RC6 setup. Add these now and don't fail the
sanity check if RC6 is disabled. This fixes a problem where RC6 remains
disabled after reloading the driver, since we explicitly disable RC6
during unloading.
v2:
- Print a debug message about the BIOS enabled RC state. (Sagar)
Chris Wilson [Fri, 1 Jul 2016 08:18:12 +0000 (09:18 +0100)]
drm/i915/ringbuffer: Move all generic engine->dispatch_batchbuffer together
Consolidate the block of default vfuncs for dispatching the batchbuffer.
Just a minor tweak on top of Tvrtko's great job of tidying up the vfunc
initialisation.
Introduce a function which initializes vfuncs mostly common
across engines and move write_tail initialization in it since
only one engine overrides the default.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 30 Jun 2016 14:33:36 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:44 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:45 +0000 (15:33 +0100)]
drm/i915: Perform Sandybridge BSD tail write under the forcewake
Since we have a sequence of register reads and writes, we can reduce the
latency of starting the BSD ring by performing all the mmio operations
under the same forcewake wakeref.
Chris Wilson [Thu, 30 Jun 2016 14:33:43 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:42 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:41 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:40 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:39 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:38 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:37 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:35 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:34 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:33 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:32 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:31 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:30 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:29 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:28 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:27 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:26 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:25 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:24 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:23 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:22 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:21 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:20 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:19 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:18 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:17 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:16 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:15 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:14 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:13 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:12 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:11 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:10 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:09 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:08 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:07 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:06 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:05 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:04 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:03 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:02 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:01 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:33:00 +0000 (15:33 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:59 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:58 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:57 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:56 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:55 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:54 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:53 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:52 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:51 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:50 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:49 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:48 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:47 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:46 +0000 (15:32 +0100)]
drm/i915: Convert wait_for(I915_READ(reg)) to intel_wait_for_register()
By using the out-of-line intel_wait_for_register() not only do we can
efficiency from using the hybrid wait_for() contained within, but we
avoid code bloat from the numerous inlined loops, in total (all patches):
text data bss dec hex filename 1078551 4557 416 1083524 108884 drivers/gpu/drm/i915/i915.ko 1070775 4557 416 1075748 106a24 drivers/gpu/drm/i915/i915.ko
Chris Wilson [Thu, 30 Jun 2016 14:32:45 +0000 (15:32 +0100)]
drm/i915: Convert sandybridge_pcode_*() to use intel_wait_for_register()
We want to replace the inline wait_for() with an out-of-line hybrid
busy/sleep wait_for() in the hopes of speeding up the communication wit
the PCode unit.
Indeed, on my i5-2500s, __gen6_update_ring_freq improves from
6,080,661ns to 8172ns.
v2: Missed using _fw variants for sandybridge_pcode_read()
Chris Wilson [Thu, 30 Jun 2016 14:32:44 +0000 (15:32 +0100)]
drm/i915: Use a hybrid scheme for fast register waits
Ville Syrjälä reported that in the majority of wait_for(I915_READ()) he
inspect, most completed within the first couple of reads and that the
delay between those wait_for() reads was the ratelimiting step for many
code paths. For example, __gen6_update_ring_freq() was blamed for
slowing down boot by many milliseconds, but under Ville's scrutiny the
issue was just excessive delay waiting for sandybridge_pcode_write().
We can eliminate the wait by initially using a busyspin upon the register
read and only fallback to the sleeping loop in cases where the hardware
is indeed too slow. A threshold of 2 microseconds is used as the initial
ballpark.
To avoid excessive code bloating from converting every wait_for() into a
hybrid busy/sleep loop, we extend wait_for_register_fw() and export it
for use by other callers.
Something appears to be off in the timing, but as far as I can tell it
is not along the event delivery path. The net effect appears to be
rendering flicker (the current render buffer appears on the scanout,
with what appears to be active rendering for a fraction of a frame) and
is causing me a headache.
The cursor is also being stalled by page flips, causing a "heavy mouse"
and jitter.
Daniel Stone did find what appears to the cause of the tearing, in
https://lists.freedesktop.org/archives/intel-gfx/2016-June/099466.html
That is the parameter passed to intel_atomic_commit_tail is the
old_state but we need the new_state to wait upon.
That leaves the question of how the CRC based tests didn't spot the
error (how can we improve our tests?), the issue of legacy cursor
stalling flips, and the issue of flips stalling the cursor. For the
moment, step back until the condundrum of new/old state is reviewed
along with more tests!
Reported-and-tested-by: Steven Newbury <steve@snewbury.org.uk> Reported-by: Rafael Ristovski <rafael.ristovski@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96593
Testcase: igt/kms_cursor_legacy/basic-cursor-vs-flip Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1466772243-21879-1-git-send-email-chris@chris-wilson.co.uk Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Chris Wilson [Wed, 29 Jun 2016 14:51:14 +0000 (15:51 +0100)]
drm/i915: Remove request->reset_counter
Since commit 2ed53a94d8cb ("drm/i915: On GPU reset, set the HWS
breadcrumb to the last seqno") once a hang is completed, the seqno is
advanced past all current requests. With this we know that if we wake up
from waiting for a request, if a hang has occurred and reset completed,
our request will be considered complete (i.e.
i915_gem_request_completed() returns true). Therefore we only need to
worry about the situation where a hang has occurred, but not yet reset,
where we may need to release our struct_mutex. Since we don't need to
detect the completed reset using the global gpu_error->reset_counter
anymore, we do not need to track the reset_counter epoch inside the
request.
Tvrtko Ursulin [Tue, 28 Jun 2016 11:51:49 +0000 (12:51 +0100)]
drm/i915/debug: Select PREEMPT_COUNT when enabling debugging
Required to enable correct wait_for_atomic checks.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>