From bb6d19846d1cee7cc977ce3aa736ec60d7163d3a Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 26 Nov 2015 13:31:42 +0000 Subject: drm/i915: Check the timeout passed to i915_wait_request We have relied upon the sole caller (wait_ioctl) validating the timeout argument. However, when waiting for multiple requests I forgot to ensure that the timeout was still positive on the later requests. This is more simply done inside __i915_wait_request. Fixes regression introduced in commit b47161858ba13c9c7e03333132230d66e008dd55 Author: Chris Wilson Date: Mon Apr 27 13:41:17 2015 +0100 drm/i915: Implement inter-engine read-read optimisations The impact of the regression is 1 jiffie for each extra active ring for a wait_ioctl with a timeout -- I don't think anyone has noticed. Signed-off-by: Chris Wilson Cc: Lionel Landwerlin Cc: Tvrtko Ursulin Cc: Daniel Vetter Reviewed-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1448544702-5594-1-git-send-email-chris@chris-wilson.co.uk Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/i915/i915_gem.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 91bb1fc27420..32e6aade6223 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1210,8 +1210,16 @@ int __i915_wait_request(struct drm_i915_gem_request *req, if (i915_gem_request_completed(req, true)) return 0; - timeout_expire = timeout ? - jiffies + nsecs_to_jiffies_timeout((u64)*timeout) : 0; + timeout_expire = 0; + if (timeout) { + if (WARN_ON(*timeout < 0)) + return -EINVAL; + + if (*timeout == 0) + return -ETIME; + + timeout_expire = jiffies + nsecs_to_jiffies_timeout(*timeout); + } if (INTEL_INFO(dev_priv)->gen >= 6) gen6_rps_boost(dev_priv, rps, req->emitted_jiffies); -- cgit v1.2.3 From 62d622c1f8d34bde3e3b9fd06f15c35d4028a8ff Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 20 Nov 2015 14:16:39 +0000 Subject: drm/i915: Set the map-and-fenceable flag for preallocated objects As we mark the preallocated objects as bound, we should also flag them correctly as being map-and-fenceable (if appropriate!) so that later users do not get confused and try and rebind the pinned vma in order to get a map-and-fenceable binding. Signed-off-by: Chris Wilson Cc: "Goel, Akash" Cc: Daniel Vetter Cc: Jesse Barnes Cc: drm-intel-fixes@lists.freedesktop.org Link: http://patchwork.freedesktop.org/patch/msgid/1448029000-10616-1-git-send-email-chris@chris-wilson.co.uk Reviewed-by: Jesse Barnes Signed-off-by: Daniel Vetter (cherry picked from commit d0710abbcd88b1ff17760e97d74a673e67b49ea1) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem.c | 43 +++++++++++++++++++++++------------------ 1 file changed, 24 insertions(+), 19 deletions(-) (limited to 'drivers/gpu/drm/i915/i915_gem.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 32e6aade6223..3163518ba19a 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -4080,6 +4080,29 @@ i915_vma_misplaced(struct i915_vma *vma, uint32_t alignment, uint64_t flags) return false; } +void __i915_vma_set_map_and_fenceable(struct i915_vma *vma) +{ + struct drm_i915_gem_object *obj = vma->obj; + bool mappable, fenceable; + u32 fence_size, fence_alignment; + + fence_size = i915_gem_get_gtt_size(obj->base.dev, + obj->base.size, + obj->tiling_mode); + fence_alignment = i915_gem_get_gtt_alignment(obj->base.dev, + obj->base.size, + obj->tiling_mode, + true); + + fenceable = (vma->node.size == fence_size && + (vma->node.start & (fence_alignment - 1)) == 0); + + mappable = (vma->node.start + fence_size <= + to_i915(obj->base.dev)->gtt.mappable_end); + + obj->map_and_fenceable = mappable && fenceable; +} + static int i915_gem_object_do_pin(struct drm_i915_gem_object *obj, struct i915_address_space *vm, @@ -4147,25 +4170,7 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj, if (ggtt_view && ggtt_view->type == I915_GGTT_VIEW_NORMAL && (bound ^ vma->bound) & GLOBAL_BIND) { - bool mappable, fenceable; - u32 fence_size, fence_alignment; - - fence_size = i915_gem_get_gtt_size(obj->base.dev, - obj->base.size, - obj->tiling_mode); - fence_alignment = i915_gem_get_gtt_alignment(obj->base.dev, - obj->base.size, - obj->tiling_mode, - true); - - fenceable = (vma->node.size == fence_size && - (vma->node.start & (fence_alignment - 1)) == 0); - - mappable = (vma->node.start + fence_size <= - dev_priv->gtt.mappable_end); - - obj->map_and_fenceable = mappable && fenceable; - + __i915_vma_set_map_and_fenceable(vma); WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable); } -- cgit v1.2.3 From e7571f7fd66c77a760338340adbe41d994fe93ac Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 11 Dec 2015 11:32:57 +0000 Subject: drm/i915: Break busywaiting for requests on pending signals The busywait in __i915_spin_request() does not respect pending signals and so may consume the entire timeslice for the task instead of returning to userspace to handle the signal. In the worst case this could cause a delay in signal processing of 20ms, which would be a noticeable jitter in cursor tracking. If a higher resolution signal was being used, for example to provide fairness of a server timeslices between clients, we could expect to detect some unfairness between clients (i.e. some windows not updating as fast as others). This issue was noticed when inspecting a report of poor interactivity resulting from excessively high __i915_spin_request usage. Fixes regression from commit 2def4ad99befa25775dd2f714fdd4d92faec6e34 [v4.2] Author: Chris Wilson Date: Tue Apr 7 16:20:41 2015 +0100 drm/i915: Optimistically spin for the request completion v2: Try to assess the impact of the bug Signed-off-by: Chris Wilson Reviewed-by: Tvrtko Ursulin Cc: Jens Axboe Cc; "Rogozhkin, Dmitry V" Cc: Daniel Vetter Cc: Tvrtko Ursulin Cc: Eero Tamminen Cc: "Rantala, Valtteri" Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-2-git-send-email-chris@chris-wilson.co.uk (cherry picked from commit 91b0c352ace9afec1fb51590c7b8bd60e0eb9fbd) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/i915/i915_gem.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 3163518ba19a..06631e130efc 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1146,7 +1146,7 @@ static bool missed_irq(struct drm_i915_private *dev_priv, return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings); } -static int __i915_spin_request(struct drm_i915_gem_request *req) +static int __i915_spin_request(struct drm_i915_gem_request *req, int state) { unsigned long timeout; @@ -1158,6 +1158,9 @@ static int __i915_spin_request(struct drm_i915_gem_request *req) if (i915_gem_request_completed(req, true)) return 0; + if (signal_pending_state(state, current)) + break; + if (time_after_eq(jiffies, timeout)) break; @@ -1197,6 +1200,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req, struct drm_i915_private *dev_priv = dev->dev_private; const bool irq_test_in_progress = ACCESS_ONCE(dev_priv->gpu_error.test_irq_rings) & intel_ring_flag(ring); + int state = interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE; DEFINE_WAIT(wait); unsigned long timeout_expire; s64 before, now; @@ -1229,7 +1233,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req, before = ktime_get_raw_ns(); /* Optimistic spin for the next jiffie before touching IRQs */ - ret = __i915_spin_request(req); + ret = __i915_spin_request(req, state); if (ret == 0) goto out; @@ -1241,8 +1245,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req, for (;;) { struct timer_list timer; - prepare_to_wait(&ring->irq_queue, &wait, - interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE); + prepare_to_wait(&ring->irq_queue, &wait, state); /* We need to check whether any gpu reset happened in between * the caller grabbing the seqno and now ... */ @@ -1260,7 +1263,7 @@ int __i915_wait_request(struct drm_i915_gem_request *req, break; } - if (interruptible && signal_pending(current)) { + if (signal_pending_state(state, current)) { ret = -ERESTARTSYS; break; } -- cgit v1.2.3 From f87a780f07b22b6dc4642dbaf44af65112076cb8 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 11 Dec 2015 11:32:58 +0000 Subject: drm/i915: Limit the busy wait on requests to 5us not 10ms! When waiting for high frequency requests, the finite amount of time required to set up the irq and wait upon it limits the response rate. By busywaiting on the request completion for a short while we can service the high frequency waits as quick as possible. However, if it is a slow request, we want to sleep as quickly as possible. The tradeoff between waiting and sleeping is roughly the time it takes to sleep on a request, on the order of a microsecond. Based on measurements of synchronous workloads from across big core and little atom, I have set the limit for busywaiting as 10 microseconds. In most of the synchronous cases, we can reduce the limit down to as little as 2 miscroseconds, but that leaves quite a few test cases regressing by factors of 3 and more. The code currently uses the jiffie clock, but that is far too coarse (on the order of 10 milliseconds) and results in poor interactivity as the CPU ends up being hogged by slow requests. To get microsecond resolution we need to use a high resolution timer. The cheapest of which is polling local_clock(), but that is only valid on the same CPU. If we switch CPUs because the task was preempted, we can also use that as an indicator that the system is too busy to waste cycles on spinning and we should sleep instead. __i915_spin_request was introduced in commit 2def4ad99befa25775dd2f714fdd4d92faec6e34 [v4.2] Author: Chris Wilson Date: Tue Apr 7 16:20:41 2015 +0100 drm/i915: Optimistically spin for the request completion v2: Drop full u64 for unsigned long - the timer is 32bit wraparound safe, so we can use native register sizes on smaller architectures. Mention the approximate microseconds units for elapsed time and add some extra comments describing the reason for busywaiting. v3: Raise the limit to 10us v4: Now 5us. Reported-by: Jens Axboe Link: https://lkml.org/lkml/2015/11/12/621 Reviewed-by: Tvrtko Ursulin Cc: "Rogozhkin, Dmitry V" Cc: Daniel Vetter Cc: Tvrtko Ursulin Cc: Eero Tamminen Cc: "Rantala, Valtteri" Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-3-git-send-email-chris@chris-wilson.co.uk (cherry picked from commit ca5b721e238226af1d767103ac852aeb8e4c0764) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem.c | 47 +++++++++++++++++++++++++++++++++++++++-- 1 file changed, 45 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/i915/i915_gem.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 06631e130efc..8719fa2ae7e7 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1146,14 +1146,57 @@ static bool missed_irq(struct drm_i915_private *dev_priv, return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings); } +static unsigned long local_clock_us(unsigned *cpu) +{ + unsigned long t; + + /* Cheaply and approximately convert from nanoseconds to microseconds. + * The result and subsequent calculations are also defined in the same + * approximate microseconds units. The principal source of timing + * error here is from the simple truncation. + * + * Note that local_clock() is only defined wrt to the current CPU; + * the comparisons are no longer valid if we switch CPUs. Instead of + * blocking preemption for the entire busywait, we can detect the CPU + * switch and use that as indicator of system load and a reason to + * stop busywaiting, see busywait_stop(). + */ + *cpu = get_cpu(); + t = local_clock() >> 10; + put_cpu(); + + return t; +} + +static bool busywait_stop(unsigned long timeout, unsigned cpu) +{ + unsigned this_cpu; + + if (time_after(local_clock_us(&this_cpu), timeout)) + return true; + + return this_cpu != cpu; +} + static int __i915_spin_request(struct drm_i915_gem_request *req, int state) { unsigned long timeout; + unsigned cpu; + + /* When waiting for high frequency requests, e.g. during synchronous + * rendering split between the CPU and GPU, the finite amount of time + * required to set up the irq and wait upon it limits the response + * rate. By busywaiting on the request completion for a short while we + * can service the high frequency waits as quick as possible. However, + * if it is a slow request, we want to sleep as quickly as possible. + * The tradeoff between waiting and sleeping is roughly the time it + * takes to sleep on a request, on the order of a microsecond. + */ if (i915_gem_request_get_ring(req)->irq_refcount) return -EBUSY; - timeout = jiffies + 1; + timeout = local_clock_us(&cpu) + 5; while (!need_resched()) { if (i915_gem_request_completed(req, true)) return 0; @@ -1161,7 +1204,7 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state) if (signal_pending_state(state, current)) break; - if (time_after_eq(jiffies, timeout)) + if (busywait_stop(timeout, cpu)) break; cpu_relax_lowlatency(); -- cgit v1.2.3 From 0f0cd472062eca6f9fac8be0cd5585f9a2df1ab2 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 11 Dec 2015 11:32:59 +0000 Subject: drm/i915: Only spin whilst waiting on the current request Limit busywaiting only to the request currently being processed by the GPU. If the request is not currently being processed by the GPU, there is a very low likelihood of it being completed within the 2 microsecond spin timeout and so we will just be wasting CPU cycles. v2: Check for logical inversion when rebasing - we were incorrectly checking for this request being active, and instead busywaiting for when the GPU was not yet processing the request of interest. v3: Try another colour for the seqno names. v4: Another colour for the function names. v5: Remove the forced coherency when checking for the active request. On reflection and plenty of recent experimentation, the issue is not a cache coherency problem - but an irq/seqno ordering problem (timing issue). Here, we do not need the w/a to force ordering of the read with an interrupt. Signed-off-by: Chris Wilson Reviewed-by: Tvrtko Ursulin Cc: "Rogozhkin, Dmitry V" Cc: Daniel Vetter Cc: Tvrtko Ursulin Cc: Eero Tamminen Cc: "Rantala, Valtteri" Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-4-git-send-email-chris@chris-wilson.co.uk (cherry picked from commit 821485dc2ad665f136c57ee589bf7a8210160fe2) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/i915/i915_gem.c') diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 8719fa2ae7e7..f56af0aaafde 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1193,9 +1193,13 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state) * takes to sleep on a request, on the order of a microsecond. */ - if (i915_gem_request_get_ring(req)->irq_refcount) + if (req->ring->irq_refcount) return -EBUSY; + /* Only spin if we know the GPU is processing this request */ + if (!i915_gem_request_started(req, true)) + return -EAGAIN; + timeout = local_clock_us(&cpu) + 5; while (!need_resched()) { if (i915_gem_request_completed(req, true)) @@ -1209,6 +1213,7 @@ static int __i915_spin_request(struct drm_i915_gem_request *req, int state) cpu_relax_lowlatency(); } + if (i915_gem_request_completed(req, false)) return 0; @@ -2600,6 +2605,7 @@ void __i915_add_request(struct drm_i915_gem_request *request, request->batch_obj = obj; request->emitted_jiffies = jiffies; + request->previous_seqno = ring->last_submitted_seqno; ring->last_submitted_seqno = request->seqno; list_add_tail(&request->list, &ring->request_list); -- cgit v1.2.3