On Mon, Oct 16, 2023 at 11:08:03AM +0300, Jouni Högander wrote:
> We are preparing for Xe driver. Xe driver doesn't have i915_sw_fence
> implementation. Lets drop i915_sw_fence usage from display code and
> use dma_fence interfaces directly.
>
> For this purpose stack dma fences from related objects into old and new
> plane states using drm_gem_plane_helper_prepare_fb. Then wait for these
> stacked fences during atomic commit.
>
> There is no be need for separate GPU reset handling in
> intel_atomic_commit_fence_wait as the fences are signaled when GPU hang is
> detected and GPU is being reset.
>
> Cc: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
> Cc: Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>
> Cc: José Roberto de Souza <jose.souza@xxxxxxxxx>
>
> Signed-off-by: Jouni Högander <jouni.hogander@xxxxxxxxx>
> ---
> drivers/gpu/drm/i915/display/intel_atomic.c | 3 -
> .../gpu/drm/i915/display/intel_atomic_plane.c | 49 +++---------
> drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++-------------
> .../drm/i915/display/intel_display_types.h | 2 -
> 4 files changed, 37 insertions(+), 95 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
> index 5d18145da279..ec0d5168b503 100644
> --- a/drivers/gpu/drm/i915/display/intel_atomic.c
> +++ b/drivers/gpu/drm/i915/display/intel_atomic.c
> @@ -331,9 +331,6 @@ void intel_atomic_state_free(struct drm_atomic_state *_state)
>
> drm_atomic_state_default_release(&state->base);
> kfree(state->global_objs);
> -
> - i915_sw_fence_fini(&state->commit_ready);
> -
> kfree(state);
> }
>
> diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
> index b1074350616c..d4f9168ec42c 100644
> --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c
> +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
> @@ -32,6 +32,7 @@
> */
>
> #include <drm/drm_atomic_helper.h>
> +#include <drm/drm_gem_atomic_helper.h>
> #include <drm/drm_blend.h>
> #include <drm/drm_fourcc.h>
>
> @@ -1035,7 +1036,7 @@ intel_prepare_plane_fb(struct drm_plane *_plane,
> struct intel_atomic_state *state =
> to_intel_atomic_state(new_plane_state->uapi.state);
> struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
> - const struct intel_plane_state *old_plane_state =
> + struct intel_plane_state *old_plane_state =
> intel_atomic_get_old_plane_state(state, plane);
> struct drm_i915_gem_object *obj = intel_fb_obj(new_plane_state->hw.fb);
> struct drm_i915_gem_object *old_obj = intel_fb_obj(old_plane_state->hw.fb);
> @@ -1057,56 +1058,30 @@ intel_prepare_plane_fb(struct drm_plane *_plane,
> * This should only fail upon a hung GPU, in which case we
> * can safely continue.
> */
> - if (new_crtc_state && intel_crtc_needs_modeset(new_crtc_state)) {
> - ret = i915_sw_fence_await_reservation(&state->commit_ready,
> - old_obj->base.resv,
> - false, 0,
> - GFP_KERNEL);
> + if (new_crtc_state && intel_crtc_needs_modeset(new_crtc_state) &&
> + !dma_resv_test_signaled(old_obj->base.resv,
> + dma_resv_usage_rw(false))) {
> + ret = drm_gem_plane_helper_prepare_fb(_plane, &old_plane_state->uapi);
> if (ret < 0)
> return ret;
> }
> }
>
> - if (new_plane_state->uapi.fence) { /* explicit fencing */
> - i915_gem_fence_wait_priority(new_plane_state->uapi.fence,
> - &attr);
> - ret = i915_sw_fence_await_dma_fence(&state->commit_ready,
> - new_plane_state->uapi.fence,
> - i915_fence_timeout(dev_priv),
> - GFP_KERNEL);
> - if (ret < 0)
> - return ret;
> - }
> -
> if (!obj)
> return 0;
>
> -
> ret = intel_plane_pin_fb(new_plane_state);
> if (ret)
> return ret;
>
> - i915_gem_object_wait_priority(obj, 0, &attr);
> -
> - if (!new_plane_state->uapi.fence) { /* implicit fencing */
> - struct dma_resv_iter cursor;
> - struct dma_fence *fence;
> + ret = drm_gem_plane_helper_prepare_fb(_plane, &new_plane_state->uapi);
I don't think we can use that as is due to bigjoiner stuff.
I think we'd need a slightly lower level variant that takes
the fb+fence in explicitly instead of the full plane state.
And I suppose we already have a slight bug here where only the
master pipe's plane will consult the explicit fence and the rest
will take the implicit sync path.
> + if (ret < 0)
> + goto unpin_fb;
>
> - ret = i915_sw_fence_await_reservation(&state->commit_ready,
> - obj->base.resv, false,
> - i915_fence_timeout(dev_priv),
> - GFP_KERNEL);
> - if (ret < 0)
> - goto unpin_fb;
> + if (new_plane_state->uapi.fence) {
> + i915_gem_fence_wait_priority(new_plane_state->uapi.fence,
> + &attr);
>
> - dma_resv_iter_begin(&cursor, obj->base.resv,
> - DMA_RESV_USAGE_WRITE);
> - dma_resv_for_each_fence_unlocked(&cursor, fence) {
> - intel_display_rps_boost_after_vblank(new_plane_state->hw.crtc,
> - fence);
> - }
> - dma_resv_iter_end(&cursor);
> - } else {
> intel_display_rps_boost_after_vblank(new_plane_state->hw.crtc,
> new_plane_state->uapi.fence);
> }
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index 28d85e1e858e..48df0559df00 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -48,6 +48,7 @@
> #include "g4x_dp.h"
> #include "g4x_hdmi.h"
> #include "hsw_ips.h"
> +#include "i915_config.h"
> #include "i915_drv.h"
> #include "i915_reg.h"
> #include "i915_utils.h"
> @@ -7056,29 +7057,34 @@ void intel_atomic_helper_free_state_worker(struct work_struct *work)
>
> static void intel_atomic_commit_fence_wait(struct intel_atomic_state *intel_state)
> {
> - struct wait_queue_entry wait_fence, wait_reset;
> - struct drm_i915_private *dev_priv = to_i915(intel_state->base.dev);
> -
> - init_wait_entry(&wait_fence, 0);
> - init_wait_entry(&wait_reset, 0);
> - for (;;) {
> - prepare_to_wait(&intel_state->commit_ready.wait,
> - &wait_fence, TASK_UNINTERRUPTIBLE);
> - prepare_to_wait(bit_waitqueue(&to_gt(dev_priv)->reset.flags,
> - I915_RESET_MODESET),
> - &wait_reset, TASK_UNINTERRUPTIBLE);
> + struct drm_i915_private *i915 = to_i915(intel_state->base.dev);
> + struct drm_plane *plane;
> + struct drm_plane_state *new_plane_state, *old_plane_state;
> + int ret, i;
>
> + for_each_oldnew_plane_in_state(&intel_state->base, plane,
> + old_plane_state, new_plane_state, i) {
> + if (new_plane_state->fence) {
> + ret = dma_fence_wait_timeout(new_plane_state->fence,
> + false,
> + i915_fence_timeout(i915));
> + if (ret <= 0)
> + break;
>
> - if (i915_sw_fence_done(&intel_state->commit_ready) ||
> - test_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags))
> - break;
> + dma_fence_put(new_plane_state->fence);
> + new_plane_state->fence = NULL;
> + }
> + if (old_plane_state->fence) {
> + ret = dma_fence_wait_timeout(old_plane_state->fence,
> + false,
> + i915_fence_timeout(i915));
> + if (ret <= 0)
> + break;
>
> - schedule();
> + dma_fence_put(old_plane_state->fence);
> + old_plane_state->fence = NULL;
> + }
> }
> - finish_wait(&intel_state->commit_ready.wait, &wait_fence);
> - finish_wait(bit_waitqueue(&to_gt(dev_priv)->reset.flags,
> - I915_RESET_MODESET),
> - &wait_reset);
> }
>
> static void intel_atomic_cleanup_work(struct work_struct *work)
> @@ -7370,32 +7376,6 @@ static void intel_atomic_commit_work(struct work_struct *work)
> intel_atomic_commit_tail(state);
> }
>
> -static int
> -intel_atomic_commit_ready(struct i915_sw_fence *fence,
> - enum i915_sw_fence_notify notify)
> -{
> - struct intel_atomic_state *state =
> - container_of(fence, struct intel_atomic_state, commit_ready);
> -
> - switch (notify) {
> - case FENCE_COMPLETE:
> - /* we do blocking waits in the worker, nothing to do here */
> - break;
> - case FENCE_FREE:
> - {
> - struct drm_i915_private *i915 = to_i915(state->base.dev);
> - struct intel_atomic_helper *helper =
> - &i915->display.atomic_helper;
> -
> - if (llist_add(&state->freed, &helper->free_list))
> - queue_work(i915->unordered_wq, &helper->free_work);
> - break;
> - }
> - }
> -
> - return NOTIFY_DONE;
> -}
> -
> static void intel_atomic_track_fbs(struct intel_atomic_state *state)
> {
> struct intel_plane_state *old_plane_state, *new_plane_state;
> @@ -7418,10 +7398,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
>
> state->wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
>
> - drm_atomic_state_get(&state->base);
> - i915_sw_fence_init(&state->commit_ready,
> - intel_atomic_commit_ready);
> -
> /*
> * The intel_legacy_cursor_update() fast path takes care
> * of avoiding the vblank waits for simple cursor
> @@ -7454,7 +7430,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
> if (ret) {
> drm_dbg_atomic(&dev_priv->drm,
> "Preparing state failed with %i\n", ret);
> - i915_sw_fence_commit(&state->commit_ready);
> intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref);
> return ret;
> }
> @@ -7470,8 +7445,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
> struct intel_crtc *crtc;
> int i;
>
> - i915_sw_fence_commit(&state->commit_ready);
> -
> for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i)
> intel_color_cleanup_commit(new_crtc_state);
>
> @@ -7485,7 +7458,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
> drm_atomic_state_get(&state->base);
> INIT_WORK(&state->base.commit_work, intel_atomic_commit_work);
>
> - i915_sw_fence_commit(&state->commit_ready);
> if (nonblock && state->modeset) {
> queue_work(dev_priv->display.wq.modeset, &state->base.commit_work);
> } else if (nonblock) {
> diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
> index 8d8b2f8d37a9..a685c527c7d1 100644
> --- a/drivers/gpu/drm/i915/display/intel_display_types.h
> +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
> @@ -670,8 +670,6 @@ struct intel_atomic_state {
>
> bool rps_interactive;
>
> - struct i915_sw_fence commit_ready;
> -
> struct llist_node freed;
> };
>
> --
> 2.34.1
--
Ville Syrjälä
Intel
