OpenCloudOS-Kernel/drivers/gpu/drm/i915/i915_trace.h

987 lines
26 KiB
C
Raw Normal View History

#if !defined(_I915_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ)
#define _I915_TRACE_H_
#include <linux/stringify.h>
#include <linux/types.h>
#include <linux/tracepoint.h>
#include <drm/drmP.h>
#include "i915_drv.h"
#include "intel_drv.h"
#include "intel_ringbuffer.h"
#undef TRACE_SYSTEM
#define TRACE_SYSTEM i915
#define TRACE_INCLUDE_FILE i915_trace
/* watermark/fifo updates */
TRACE_EVENT(intel_cpu_fifo_underrun,
TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
TP_ARGS(dev_priv, pipe),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
),
TP_fast_assign(
__entry->pipe = pipe;
__entry->frame = dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
__entry->scanline = intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
),
TP_printk("pipe %c, frame=%u, scanline=%u",
pipe_name(__entry->pipe),
__entry->frame, __entry->scanline)
);
TRACE_EVENT(intel_pch_fifo_underrun,
TP_PROTO(struct drm_i915_private *dev_priv, enum transcoder pch_transcoder),
TP_ARGS(dev_priv, pch_transcoder),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
),
TP_fast_assign(
enum pipe pipe = (enum pipe)pch_transcoder;
__entry->pipe = pipe;
__entry->frame = dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
__entry->scanline = intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
),
TP_printk("pch transcoder %c, frame=%u, scanline=%u",
pipe_name(__entry->pipe),
__entry->frame, __entry->scanline)
);
TRACE_EVENT(intel_memory_cxsr,
TP_PROTO(struct drm_i915_private *dev_priv, bool old, bool new),
TP_ARGS(dev_priv, old, new),
TP_STRUCT__entry(
__array(u32, frame, 3)
__array(u32, scanline, 3)
__field(bool, old)
__field(bool, new)
),
TP_fast_assign(
enum pipe pipe;
for_each_pipe(dev_priv, pipe) {
__entry->frame[pipe] =
dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
__entry->scanline[pipe] =
intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
}
__entry->old = old;
__entry->new = new;
),
TP_printk("%s->%s, pipe A: frame=%u, scanline=%u, pipe B: frame=%u, scanline=%u, pipe C: frame=%u, scanline=%u",
onoff(__entry->old), onoff(__entry->new),
__entry->frame[PIPE_A], __entry->scanline[PIPE_A],
__entry->frame[PIPE_B], __entry->scanline[PIPE_B],
__entry->frame[PIPE_C], __entry->scanline[PIPE_C])
);
TRACE_EVENT(vlv_wm,
TP_PROTO(struct intel_crtc *crtc, const struct vlv_wm_values *wm),
TP_ARGS(crtc, wm),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
__field(u32, level)
__field(u32, cxsr)
__field(u32, primary)
__field(u32, sprite0)
__field(u32, sprite1)
__field(u32, cursor)
__field(u32, sr_plane)
__field(u32, sr_cursor)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
crtc->pipe);
__entry->scanline = intel_get_crtc_scanline(crtc);
__entry->level = wm->level;
__entry->cxsr = wm->cxsr;
__entry->primary = wm->pipe[crtc->pipe].plane[PLANE_PRIMARY];
__entry->sprite0 = wm->pipe[crtc->pipe].plane[PLANE_SPRITE0];
__entry->sprite1 = wm->pipe[crtc->pipe].plane[PLANE_SPRITE1];
__entry->cursor = wm->pipe[crtc->pipe].plane[PLANE_CURSOR];
__entry->sr_plane = wm->sr.plane;
__entry->sr_cursor = wm->sr.cursor;
),
TP_printk("pipe %c, frame=%u, scanline=%u, level=%d, cxsr=%d, wm %d/%d/%d/%d, sr %d/%d",
pipe_name(__entry->pipe), __entry->frame,
__entry->scanline, __entry->level, __entry->cxsr,
__entry->primary, __entry->sprite0, __entry->sprite1, __entry->cursor,
__entry->sr_plane, __entry->sr_cursor)
);
TRACE_EVENT(vlv_fifo_size,
TP_PROTO(struct intel_crtc *crtc, u32 sprite0_start, u32 sprite1_start, u32 fifo_size),
TP_ARGS(crtc, sprite0_start, sprite1_start, fifo_size),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
__field(u32, sprite0_start)
__field(u32, sprite1_start)
__field(u32, fifo_size)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
crtc->pipe);
__entry->scanline = intel_get_crtc_scanline(crtc);
__entry->sprite0_start = sprite0_start;
__entry->sprite1_start = sprite1_start;
__entry->fifo_size = fifo_size;
),
TP_printk("pipe %c, frame=%u, scanline=%u, %d/%d/%d",
pipe_name(__entry->pipe), __entry->frame,
__entry->scanline, __entry->sprite0_start,
__entry->sprite1_start, __entry->fifo_size)
);
/* plane updates */
TRACE_EVENT(intel_update_plane,
TP_PROTO(struct drm_plane *plane, struct intel_crtc *crtc),
TP_ARGS(plane, crtc),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(const char *, name)
__field(u32, frame)
__field(u32, scanline)
__array(int, src, 4)
__array(int, dst, 4)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->name = plane->name;
__entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
crtc->pipe);
__entry->scanline = intel_get_crtc_scanline(crtc);
memcpy(__entry->src, &plane->state->src, sizeof(__entry->src));
memcpy(__entry->dst, &plane->state->dst, sizeof(__entry->dst));
),
TP_printk("pipe %c, plane %s, frame=%u, scanline=%u, " DRM_RECT_FP_FMT " -> " DRM_RECT_FMT,
pipe_name(__entry->pipe), __entry->name,
__entry->frame, __entry->scanline,
DRM_RECT_FP_ARG((const struct drm_rect *)__entry->src),
DRM_RECT_ARG((const struct drm_rect *)__entry->dst))
);
TRACE_EVENT(intel_disable_plane,
TP_PROTO(struct drm_plane *plane, struct intel_crtc *crtc),
TP_ARGS(plane, crtc),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(const char *, name)
__field(u32, frame)
__field(u32, scanline)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->name = plane->name;
__entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
crtc->pipe);
__entry->scanline = intel_get_crtc_scanline(crtc);
),
TP_printk("pipe %c, plane %s, frame=%u, scanline=%u",
pipe_name(__entry->pipe), __entry->name,
__entry->frame, __entry->scanline)
);
/* pipe updates */
TRACE_EVENT(i915_pipe_update_start,
TP_PROTO(struct intel_crtc *crtc),
TP_ARGS(crtc),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
__field(u32, min)
__field(u32, max)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
crtc->pipe);
__entry->scanline = intel_get_crtc_scanline(crtc);
__entry->min = crtc->debug.min_vbl;
__entry->max = crtc->debug.max_vbl;
),
TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
pipe_name(__entry->pipe), __entry->frame,
__entry->scanline, __entry->min, __entry->max)
);
TRACE_EVENT(i915_pipe_update_vblank_evaded,
TP_PROTO(struct intel_crtc *crtc),
TP_ARGS(crtc),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
__field(u32, min)
__field(u32, max)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->frame = crtc->debug.start_vbl_count;
__entry->scanline = crtc->debug.scanline_start;
__entry->min = crtc->debug.min_vbl;
__entry->max = crtc->debug.max_vbl;
),
TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
pipe_name(__entry->pipe), __entry->frame,
__entry->scanline, __entry->min, __entry->max)
);
TRACE_EVENT(i915_pipe_update_end,
TP_PROTO(struct intel_crtc *crtc, u32 frame, int scanline_end),
TP_ARGS(crtc, frame, scanline_end),
TP_STRUCT__entry(
__field(enum pipe, pipe)
__field(u32, frame)
__field(u32, scanline)
),
TP_fast_assign(
__entry->pipe = crtc->pipe;
__entry->frame = frame;
__entry->scanline = scanline_end;
),
TP_printk("pipe %c, frame=%u, scanline=%u",
pipe_name(__entry->pipe), __entry->frame,
__entry->scanline)
);
/* object tracking */
TRACE_EVENT(i915_gem_object_create,
TP_PROTO(struct drm_i915_gem_object *obj),
TP_ARGS(obj),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
__field(u32, size)
),
TP_fast_assign(
__entry->obj = obj;
__entry->size = obj->base.size;
),
TP_printk("obj=%p, size=%u", __entry->obj, __entry->size)
);
TRACE_EVENT(i915_gem_shrink,
TP_PROTO(struct drm_i915_private *i915, unsigned long target, unsigned flags),
TP_ARGS(i915, target, flags),
TP_STRUCT__entry(
__field(int, dev)
__field(unsigned long, target)
__field(unsigned, flags)
),
TP_fast_assign(
__entry->dev = i915->drm.primary->index;
__entry->target = target;
__entry->flags = flags;
),
TP_printk("dev=%d, target=%lu, flags=%x",
__entry->dev, __entry->target, __entry->flags)
);
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
TRACE_EVENT(i915_vma_bind,
TP_PROTO(struct i915_vma *vma, unsigned flags),
TP_ARGS(vma, flags),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__field(struct i915_address_space *, vm)
__field(u64, offset)
__field(u32, size)
__field(unsigned, flags)
),
TP_fast_assign(
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__entry->obj = vma->obj;
__entry->vm = vma->vm;
__entry->offset = vma->node.start;
__entry->size = vma->node.size;
__entry->flags = flags;
),
TP_printk("obj=%p, offset=%016llx size=%x%s vm=%p",
__entry->obj, __entry->offset, __entry->size,
__entry->flags & PIN_MAPPABLE ? ", mappable" : "",
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__entry->vm)
);
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
TRACE_EVENT(i915_vma_unbind,
TP_PROTO(struct i915_vma *vma),
TP_ARGS(vma),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__field(struct i915_address_space *, vm)
__field(u64, offset)
__field(u32, size)
),
TP_fast_assign(
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__entry->obj = vma->obj;
__entry->vm = vma->vm;
__entry->offset = vma->node.start;
__entry->size = vma->node.size;
),
TP_printk("obj=%p, offset=%016llx size=%x vm=%p",
drm/i915: plumb VM into bind/unbind code As alluded to in several patches, and it will be reiterated later... A VMA is an abstraction for a GEM BO bound into an address space. Therefore it stands to reason, that the existing bind, and unbind are the ones which will be the most impacted. This patch implements this, and updates all callers which weren't already updated in the series (because it was too messy). This patch represents the bulk of an earlier, larger patch. I've pulled out a bunch of things by the request of Daniel. The history is preserved for posterity with the email convention of ">" One big change from the original patch aside from a bunch of cropping is I've created an i915_vma_unbind() function. That is because we always have the VMA anyway, and doing an extra lookup is useful. There is a caveat, we retain an i915_gem_object_ggtt_unbind, for the global cases which might not talk in VMAs. > drm/i915: plumb VM into object operations > > This patch was formerly known as: > "drm/i915: Create VMAs (part 3) - plumbing" > > This patch adds a VM argument, bind/unbind, and the object > offset/size/color getters/setters. It preserves the old ggtt helper > functions because things still need, and will continue to need them. > > Some code will still need to be ported over after this. > > v2: Fix purge to pick an object and unbind all vmas > This was doable because of the global bound list change. > > v3: With the commit to actually pin/unpin pages in place, there is no > longer a need to check if unbind succeeded before calling put_pages(). > Make put_pages only BUG() after checking pin count. > > v4: Rebased on top of the new hangcheck work by Mika > plumbed eb_destroy also > Many checkpatch related fixes > > v5: Very large rebase > > v6: > Change BUG_ON to WARN_ON (Daniel) > Rename vm to ggtt in preallocate stolen, since it is always ggtt when > dealing with stolen memory. (Daniel) > list_for_each will short-circuit already (Daniel) > remove superflous space (Daniel) > Use per object list of vmas (Daniel) > Make obj_bound_any() use obj_bound for each vm (Ben) > s/bind_to_gtt/bind_to_vm/ (Ben) > > Fixed up the inactive shrinker. As Daniel noticed the code could > potentially count the same object multiple times. While it's not > possible in the current case, since 1 object can only ever be bound into > 1 address space thus far - we may as well try to get something more > future proof in place now. With a prep patch before this to switch over > to using the bound list + inactive check, we're now able to carry that > forward for every address space an object is bound into. Signed-off-by: Ben Widawsky <ben@bwidawsk.net> [danvet: Rebase on top of the loss of "drm/i915: Cleanup more of VMA in destroy".] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-01 08:00:10 +08:00
__entry->obj, __entry->offset, __entry->size, __entry->vm)
);
TRACE_EVENT(i915_gem_object_pwrite,
TP_PROTO(struct drm_i915_gem_object *obj, u32 offset, u32 len),
TP_ARGS(obj, offset, len),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
__field(u32, offset)
__field(u32, len)
),
TP_fast_assign(
__entry->obj = obj;
__entry->offset = offset;
__entry->len = len;
),
TP_printk("obj=%p, offset=%u, len=%u",
__entry->obj, __entry->offset, __entry->len)
);
TRACE_EVENT(i915_gem_object_pread,
TP_PROTO(struct drm_i915_gem_object *obj, u32 offset, u32 len),
TP_ARGS(obj, offset, len),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
__field(u32, offset)
__field(u32, len)
),
TP_fast_assign(
__entry->obj = obj;
__entry->offset = offset;
__entry->len = len;
),
TP_printk("obj=%p, offset=%u, len=%u",
__entry->obj, __entry->offset, __entry->len)
);
TRACE_EVENT(i915_gem_object_fault,
TP_PROTO(struct drm_i915_gem_object *obj, u32 index, bool gtt, bool write),
TP_ARGS(obj, index, gtt, write),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
__field(u32, index)
__field(bool, gtt)
__field(bool, write)
),
TP_fast_assign(
__entry->obj = obj;
__entry->index = index;
__entry->gtt = gtt;
__entry->write = write;
),
TP_printk("obj=%p, %s index=%u %s",
__entry->obj,
__entry->gtt ? "GTT" : "CPU",
__entry->index,
__entry->write ? ", writable" : "")
);
DECLARE_EVENT_CLASS(i915_gem_object,
TP_PROTO(struct drm_i915_gem_object *obj),
TP_ARGS(obj),
TP_STRUCT__entry(
__field(struct drm_i915_gem_object *, obj)
),
TP_fast_assign(
__entry->obj = obj;
),
TP_printk("obj=%p", __entry->obj)
);
DEFINE_EVENT(i915_gem_object, i915_gem_object_clflush,
TP_PROTO(struct drm_i915_gem_object *obj),
TP_ARGS(obj)
);
DEFINE_EVENT(i915_gem_object, i915_gem_object_destroy,
TP_PROTO(struct drm_i915_gem_object *obj),
TP_ARGS(obj)
);
TRACE_EVENT(i915_gem_evict,
TP_PROTO(struct i915_address_space *vm, u32 size, u32 align, unsigned int flags),
TP_ARGS(vm, size, align, flags),
TP_STRUCT__entry(
__field(u32, dev)
__field(struct i915_address_space *, vm)
__field(u32, size)
__field(u32, align)
__field(unsigned int, flags)
),
TP_fast_assign(
__entry->dev = vm->i915->drm.primary->index;
__entry->vm = vm;
__entry->size = size;
__entry->align = align;
__entry->flags = flags;
),
TP_printk("dev=%d, vm=%p, size=%d, align=%d %s",
__entry->dev, __entry->vm, __entry->size, __entry->align,
__entry->flags & PIN_MAPPABLE ? ", mappable" : "")
);
TRACE_EVENT(i915_gem_evict_everything,
drm/i915: Track unbound pages When dealing with a working set larger than the GATT, or even the mappable aperture when touching through the GTT, we end up with evicting objects only to rebind them at a new offset again later. Moving an object into and out of the GTT requires clflushing the pages, thus causing a double-clflush penalty for rebinding. To avoid having to clflush on rebinding, we can track the pages as they are evicted from the GTT and only relinquish those pages on memory pressure. As usual, if it were not for the handling of out-of-memory condition and having to manually shrink our own bo caches, it would be a net reduction of code. Alas. Note: The patch also contains a few changes to the last-hope evict_everything logic in i916_gem_execbuffer.c - we no longer try to only evict the purgeable stuff in a first try (since that's superflous and only helps in OOM corner-cases, not fragmented-gtt trashing situations). Also, the extraction of the get_pages retry loop from bind_to_gtt (and other callsites) to get_pages should imo have been a separate patch. v2: Ditch the newly added put_pages (for unbound objects only) in i915_gem_reset. A quick irc discussion hasn't revealed any important reason for this, so if we need this, I'd like to have a git blame'able explanation for it. v3: Undo the s/drm_malloc_ab/kmalloc/ in get_pages that Chris noticed. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> [danvet: Split out code movements and rant a bit in the commit message with a few Notes. Done v2] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-08-20 17:40:46 +08:00
TP_PROTO(struct drm_device *dev),
TP_ARGS(dev),
TP_STRUCT__entry(
__field(u32, dev)
),
TP_fast_assign(
__entry->dev = dev->primary->index;
),
drm/i915: Track unbound pages When dealing with a working set larger than the GATT, or even the mappable aperture when touching through the GTT, we end up with evicting objects only to rebind them at a new offset again later. Moving an object into and out of the GTT requires clflushing the pages, thus causing a double-clflush penalty for rebinding. To avoid having to clflush on rebinding, we can track the pages as they are evicted from the GTT and only relinquish those pages on memory pressure. As usual, if it were not for the handling of out-of-memory condition and having to manually shrink our own bo caches, it would be a net reduction of code. Alas. Note: The patch also contains a few changes to the last-hope evict_everything logic in i916_gem_execbuffer.c - we no longer try to only evict the purgeable stuff in a first try (since that's superflous and only helps in OOM corner-cases, not fragmented-gtt trashing situations). Also, the extraction of the get_pages retry loop from bind_to_gtt (and other callsites) to get_pages should imo have been a separate patch. v2: Ditch the newly added put_pages (for unbound objects only) in i915_gem_reset. A quick irc discussion hasn't revealed any important reason for this, so if we need this, I'd like to have a git blame'able explanation for it. v3: Undo the s/drm_malloc_ab/kmalloc/ in get_pages that Chris noticed. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> [danvet: Split out code movements and rant a bit in the commit message with a few Notes. Done v2] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-08-20 17:40:46 +08:00
TP_printk("dev=%d", __entry->dev)
);
TRACE_EVENT(i915_gem_evict_vm,
TP_PROTO(struct i915_address_space *vm),
TP_ARGS(vm),
TP_STRUCT__entry(
__field(u32, dev)
__field(struct i915_address_space *, vm)
),
TP_fast_assign(
__entry->dev = vm->i915->drm.primary->index;
__entry->vm = vm;
),
TP_printk("dev=%d, vm=%p", __entry->dev, __entry->vm)
);
TRACE_EVENT(i915_gem_evict_node,
TP_PROTO(struct i915_address_space *vm, struct drm_mm_node *node, unsigned int flags),
TP_ARGS(vm, node, flags),
drm/i915: Fix i915_gem_evict_for_vma (soft-pinning) Soft-pinning depends upon being able to check for availabilty of an interval and evict overlapping object from a drm_mm range manager very quickly. Currently it uses a linear list, and so performance is dire and not suitable as a general replacement. Worse, the current code will oops if it tries to evict an active buffer. It also helps if the routine reports the correct error codes as expected by its callers and emits a tracepoint upon use. For posterity since the wrong patch was pushed (i.e. that missed these key points and had known bugs), this is the changelog that should have been on commit 506a8e87d8d2 ("drm/i915: Add soft-pinning API for execbuffer"): 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. This extends the DRM_IOCTL_I915_GEM_EXECBUFFER2 to do the following: * if the user supplies a virtual address via the execobject->offset *and* sets the EXEC_OBJECT_PINNED flag in execobject->flags, then that object is placed at that offset in the address space selected by the context specifier in execbuffer. * the location must be aligned to the GTT page size, 4096 bytes * as the object is placed exactly as specified, it may be used by this execbuffer call without relocations pointing to it It may fail to do so if: * EINVAL is returned if the object does not have a 4096 byte aligned address * the object conflicts with another pinned object (either pinned by hardware in that address space, e.g. scanouts in the aliasing ppgtt) or within the same batch. EBUSY is returned if the location is pinned by hardware EINVAL is returned if the location is already in use by the batch * EINVAL is returned if the object conflicts with its own alignment (as meets the hardware requirements) or if the placement of the object does not fit within the address space All other execbuffer errors apply. Presence of this execbuf extension may be queried by passing I915_PARAM_HAS_EXEC_SOFTPIN to DRM_IOCTL_I915_GETPARAM and checking for a reported value of 1 (or greater). v2: Combine the hole/adjusted-hole ENOSPC checks v3: More color, more splitting, more blurb. Fixes: 506a8e87d8d2 ("drm/i915: Add soft-pinning API for execbuffer") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161205142941.21965-2-chris@chris-wilson.co.uk
2016-12-05 22:29:37 +08:00
TP_STRUCT__entry(
__field(u32, dev)
__field(struct i915_address_space *, vm)
__field(u64, start)
__field(u64, size)
__field(unsigned long, color)
__field(unsigned int, flags)
),
TP_fast_assign(
__entry->dev = vm->i915->drm.primary->index;
__entry->vm = vm;
__entry->start = node->start;
__entry->size = node->size;
__entry->color = node->color;
drm/i915: Fix i915_gem_evict_for_vma (soft-pinning) Soft-pinning depends upon being able to check for availabilty of an interval and evict overlapping object from a drm_mm range manager very quickly. Currently it uses a linear list, and so performance is dire and not suitable as a general replacement. Worse, the current code will oops if it tries to evict an active buffer. It also helps if the routine reports the correct error codes as expected by its callers and emits a tracepoint upon use. For posterity since the wrong patch was pushed (i.e. that missed these key points and had known bugs), this is the changelog that should have been on commit 506a8e87d8d2 ("drm/i915: Add soft-pinning API for execbuffer"): 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. This extends the DRM_IOCTL_I915_GEM_EXECBUFFER2 to do the following: * if the user supplies a virtual address via the execobject->offset *and* sets the EXEC_OBJECT_PINNED flag in execobject->flags, then that object is placed at that offset in the address space selected by the context specifier in execbuffer. * the location must be aligned to the GTT page size, 4096 bytes * as the object is placed exactly as specified, it may be used by this execbuffer call without relocations pointing to it It may fail to do so if: * EINVAL is returned if the object does not have a 4096 byte aligned address * the object conflicts with another pinned object (either pinned by hardware in that address space, e.g. scanouts in the aliasing ppgtt) or within the same batch. EBUSY is returned if the location is pinned by hardware EINVAL is returned if the location is already in use by the batch * EINVAL is returned if the object conflicts with its own alignment (as meets the hardware requirements) or if the placement of the object does not fit within the address space All other execbuffer errors apply. Presence of this execbuf extension may be queried by passing I915_PARAM_HAS_EXEC_SOFTPIN to DRM_IOCTL_I915_GETPARAM and checking for a reported value of 1 (or greater). v2: Combine the hole/adjusted-hole ENOSPC checks v3: More color, more splitting, more blurb. Fixes: 506a8e87d8d2 ("drm/i915: Add soft-pinning API for execbuffer") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161205142941.21965-2-chris@chris-wilson.co.uk
2016-12-05 22:29:37 +08:00
__entry->flags = flags;
),
TP_printk("dev=%d, vm=%p, start=%llx size=%llx, color=%lx, flags=%x",
__entry->dev, __entry->vm,
__entry->start, __entry->size,
__entry->color, __entry->flags)
);
TRACE_EVENT(i915_gem_ring_sync_to,
TP_PROTO(struct drm_i915_gem_request *to,
struct drm_i915_gem_request *from),
TP_ARGS(to, from),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, sync_from)
__field(u32, sync_to)
__field(u32, seqno)
),
TP_fast_assign(
__entry->dev = from->i915->drm.primary->index;
__entry->sync_from = from->engine->id;
__entry->sync_to = to->engine->id;
__entry->seqno = from->global_seqno;
),
TP_printk("dev=%u, sync-from=%u, sync-to=%u, seqno=%u",
__entry->dev,
__entry->sync_from, __entry->sync_to,
__entry->seqno)
);
TRACE_EVENT(i915_gem_request_queue,
TP_PROTO(struct drm_i915_gem_request *req, u32 flags),
TP_ARGS(req, flags),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ring)
__field(u32, ctx)
__field(u32, seqno)
__field(u32, flags)
),
TP_fast_assign(
__entry->dev = req->i915->drm.primary->index;
__entry->ring = req->engine->id;
__entry->ctx = req->fence.context;
__entry->seqno = req->fence.seqno;
__entry->flags = flags;
),
TP_printk("dev=%u, ring=%u, ctx=%u, seqno=%u, flags=0x%x",
__entry->dev, __entry->ring, __entry->ctx, __entry->seqno,
__entry->flags)
);
TRACE_EVENT(i915_gem_ring_flush,
TP_PROTO(struct drm_i915_gem_request *req, u32 invalidate, u32 flush),
TP_ARGS(req, invalidate, flush),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ring)
__field(u32, invalidate)
__field(u32, flush)
),
TP_fast_assign(
__entry->dev = req->i915->drm.primary->index;
__entry->ring = req->engine->id;
__entry->invalidate = invalidate;
__entry->flush = flush;
),
TP_printk("dev=%u, ring=%x, invalidate=%04x, flush=%04x",
__entry->dev, __entry->ring,
__entry->invalidate, __entry->flush)
);
DECLARE_EVENT_CLASS(i915_gem_request,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ctx)
__field(u32, ring)
__field(u32, seqno)
__field(u32, global)
),
TP_fast_assign(
__entry->dev = req->i915->drm.primary->index;
__entry->ring = req->engine->id;
__entry->ctx = req->fence.context;
__entry->seqno = req->fence.seqno;
__entry->global = req->global_seqno;
),
TP_printk("dev=%u, ring=%u, ctx=%u, seqno=%u, global=%u",
__entry->dev, __entry->ring, __entry->ctx, __entry->seqno,
__entry->global)
);
DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
#if defined(CONFIG_DRM_I915_LOW_LEVEL_TRACEPOINTS)
DEFINE_EVENT(i915_gem_request, i915_gem_request_submit,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
DEFINE_EVENT(i915_gem_request, i915_gem_request_execute,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
DECLARE_EVENT_CLASS(i915_gem_request_hw,
TP_PROTO(struct drm_i915_gem_request *req,
unsigned int port),
TP_ARGS(req, port),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ring)
__field(u32, seqno)
__field(u32, global_seqno)
__field(u32, ctx)
__field(u32, port)
),
TP_fast_assign(
__entry->dev = req->i915->drm.primary->index;
__entry->ring = req->engine->id;
__entry->ctx = req->fence.context;
__entry->seqno = req->fence.seqno;
__entry->global_seqno = req->global_seqno;
__entry->port = port;
),
TP_printk("dev=%u, ring=%u, ctx=%u, seqno=%u, global=%u, port=%u",
__entry->dev, __entry->ring, __entry->ctx,
__entry->seqno, __entry->global_seqno,
__entry->port)
);
DEFINE_EVENT(i915_gem_request_hw, i915_gem_request_in,
TP_PROTO(struct drm_i915_gem_request *req, unsigned int port),
TP_ARGS(req, port)
);
DEFINE_EVENT(i915_gem_request, i915_gem_request_out,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
#else
#if !defined(TRACE_HEADER_MULTI_READ)
static inline void
trace_i915_gem_request_submit(struct drm_i915_gem_request *req)
{
}
static inline void
trace_i915_gem_request_execute(struct drm_i915_gem_request *req)
{
}
static inline void
trace_i915_gem_request_in(struct drm_i915_gem_request *req, unsigned int port)
{
}
static inline void
trace_i915_gem_request_out(struct drm_i915_gem_request *req)
{
}
#endif
#endif
TRACE_EVENT(intel_engine_notify,
TP_PROTO(struct intel_engine_cs *engine, bool waiters),
TP_ARGS(engine, waiters),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ring)
__field(u32, seqno)
__field(bool, waiters)
),
TP_fast_assign(
__entry->dev = engine->i915->drm.primary->index;
__entry->ring = engine->id;
__entry->seqno = intel_engine_get_seqno(engine);
__entry->waiters = waiters;
),
TP_printk("dev=%u, ring=%u, seqno=%u, waiters=%u",
__entry->dev, __entry->ring, __entry->seqno,
__entry->waiters)
);
DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
TRACE_EVENT(i915_gem_request_wait_begin,
TP_PROTO(struct drm_i915_gem_request *req, unsigned int flags),
TP_ARGS(req, flags),
TP_STRUCT__entry(
__field(u32, dev)
__field(u32, ring)
__field(u32, ctx)
__field(u32, seqno)
__field(u32, global)
__field(unsigned int, flags)
),
/* NB: the blocking information is racy since mutex_is_locked
* doesn't check that the current thread holds the lock. The only
* other option would be to pass the boolean information of whether
* or not the class was blocking down through the stack which is
* less desirable.
*/
TP_fast_assign(
__entry->dev = req->i915->drm.primary->index;
__entry->ring = req->engine->id;
__entry->ctx = req->fence.context;
__entry->seqno = req->fence.seqno;
__entry->global = req->global_seqno;
__entry->flags = flags;
),
TP_printk("dev=%u, ring=%u, ctx=%u, seqno=%u, global=%u, blocking=%u, flags=0x%x",
__entry->dev, __entry->ring, __entry->ctx, __entry->seqno,
__entry->global, !!(__entry->flags & I915_WAIT_LOCKED),
__entry->flags)
);
DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end,
TP_PROTO(struct drm_i915_gem_request *req),
TP_ARGS(req)
);
TRACE_EVENT(i915_flip_request,
TP_PROTO(int plane, struct drm_i915_gem_object *obj),
TP_ARGS(plane, obj),
TP_STRUCT__entry(
__field(int, plane)
__field(struct drm_i915_gem_object *, obj)
),
TP_fast_assign(
__entry->plane = plane;
__entry->obj = obj;
),
TP_printk("plane=%d, obj=%p", __entry->plane, __entry->obj)
);
TRACE_EVENT(i915_flip_complete,
TP_PROTO(int plane, struct drm_i915_gem_object *obj),
TP_ARGS(plane, obj),
TP_STRUCT__entry(
__field(int, plane)
__field(struct drm_i915_gem_object *, obj)
),
TP_fast_assign(
__entry->plane = plane;
__entry->obj = obj;
),
TP_printk("plane=%d, obj=%p", __entry->plane, __entry->obj)
);
TRACE_EVENT_CONDITION(i915_reg_rw,
drm/i915: Type safe register read/write 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: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/1447853606-2751-1-git-send-email-ville.syrjala@linux.intel.com
2015-11-18 21:33:26 +08:00
TP_PROTO(bool write, i915_reg_t reg, u64 val, int len, bool trace),
TP_ARGS(write, reg, val, len, trace),
TP_CONDITION(trace),
TP_STRUCT__entry(
__field(u64, val)
__field(u32, reg)
__field(u16, write)
__field(u16, len)
),
TP_fast_assign(
__entry->val = (u64)val;
drm/i915: Type safe register read/write 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: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/1447853606-2751-1-git-send-email-ville.syrjala@linux.intel.com
2015-11-18 21:33:26 +08:00
__entry->reg = i915_mmio_reg_offset(reg);
__entry->write = write;
__entry->len = len;
),
TP_printk("%s reg=0x%x, len=%d, val=(0x%x, 0x%x)",
__entry->write ? "write" : "read",
__entry->reg, __entry->len,
(u32)(__entry->val & 0xffffffff),
(u32)(__entry->val >> 32))
);
TRACE_EVENT(intel_gpu_freq_change,
TP_PROTO(u32 freq),
TP_ARGS(freq),
TP_STRUCT__entry(
__field(u32, freq)
),
TP_fast_assign(
__entry->freq = freq;
),
TP_printk("new_freq=%u", __entry->freq)
);
/**
* DOC: i915_ppgtt_create and i915_ppgtt_release tracepoints
*
* With full ppgtt enabled each process using drm will allocate at least one
* translation table. With these traces it is possible to keep track of the
* allocation and of the lifetime of the tables; this can be used during
* testing/debug to verify that we are not leaking ppgtts.
* These traces identify the ppgtt through the vm pointer, which is also printed
* by the i915_vma_bind and i915_vma_unbind tracepoints.
*/
DECLARE_EVENT_CLASS(i915_ppgtt,
TP_PROTO(struct i915_address_space *vm),
TP_ARGS(vm),
TP_STRUCT__entry(
__field(struct i915_address_space *, vm)
__field(u32, dev)
),
TP_fast_assign(
__entry->vm = vm;
__entry->dev = vm->i915->drm.primary->index;
),
TP_printk("dev=%u, vm=%p", __entry->dev, __entry->vm)
)
DEFINE_EVENT(i915_ppgtt, i915_ppgtt_create,
TP_PROTO(struct i915_address_space *vm),
TP_ARGS(vm)
);
DEFINE_EVENT(i915_ppgtt, i915_ppgtt_release,
TP_PROTO(struct i915_address_space *vm),
TP_ARGS(vm)
);
/**
* DOC: i915_context_create and i915_context_free tracepoints
*
* These tracepoints are used to track creation and deletion of contexts.
* If full ppgtt is enabled, they also print the address of the vm assigned to
* the context.
*/
DECLARE_EVENT_CLASS(i915_context,
TP_PROTO(struct i915_gem_context *ctx),
TP_ARGS(ctx),
TP_STRUCT__entry(
__field(u32, dev)
__field(struct i915_gem_context *, ctx)
__field(u32, hw_id)
__field(struct i915_address_space *, vm)
),
TP_fast_assign(
__entry->dev = ctx->i915->drm.primary->index;
__entry->ctx = ctx;
__entry->hw_id = ctx->hw_id;
__entry->vm = ctx->ppgtt ? &ctx->ppgtt->base : NULL;
),
TP_printk("dev=%u, ctx=%p, ctx_vm=%p, hw_id=%u",
__entry->dev, __entry->ctx, __entry->vm, __entry->hw_id)
)
DEFINE_EVENT(i915_context, i915_context_create,
TP_PROTO(struct i915_gem_context *ctx),
TP_ARGS(ctx)
);
DEFINE_EVENT(i915_context, i915_context_free,
TP_PROTO(struct i915_gem_context *ctx),
TP_ARGS(ctx)
);
/**
* DOC: switch_mm tracepoint
*
* This tracepoint allows tracking of the mm switch, which is an important point
* in the lifetime of the vm in the legacy submission path. This tracepoint is
* called only if full ppgtt is enabled.
*/
TRACE_EVENT(switch_mm,
TP_PROTO(struct intel_engine_cs *engine, struct i915_gem_context *to),
TP_ARGS(engine, to),
TP_STRUCT__entry(
__field(u32, ring)
__field(struct i915_gem_context *, to)
__field(struct i915_address_space *, vm)
__field(u32, dev)
),
TP_fast_assign(
__entry->ring = engine->id;
__entry->to = to;
__entry->vm = to->ppgtt? &to->ppgtt->base : NULL;
__entry->dev = engine->i915->drm.primary->index;
),
TP_printk("dev=%u, ring=%u, ctx=%p, ctx_vm=%p",
__entry->dev, __entry->ring, __entry->to, __entry->vm)
);
#endif /* _I915_TRACE_H_ */
/* This part must be outside protection */
#undef TRACE_INCLUDE_PATH
#define TRACE_INCLUDE_PATH .
#include <trace/define_trace.h>