Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933904AbcJUMvV (ORCPT ); Fri, 21 Oct 2016 08:51:21 -0400 Received: from mail-lf0-f67.google.com ([209.85.215.67]:33773 "EHLO mail-lf0-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933531AbcJUMvR (ORCPT ); Fri, 21 Oct 2016 08:51:17 -0400 Date: Fri, 21 Oct 2016 14:51:12 +0200 From: Daniel Vetter To: Gustavo Padovan Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org, Daniel Stone , Daniel Vetter , Rob Clark , Greg Hackmann , John Harrison , laurent.pinchart@ideasonboard.com, seanpaul@google.com, marcheu@google.com, m.chehab@samsung.com, Sumit Semwal , Maarten Lankhorst , Gustavo Padovan Subject: Re: [PATCH v5 1/4] drm/fence: add in-fences support Message-ID: <20161021125112.GC20761@phenom.ffwll.local> Mail-Followup-To: Gustavo Padovan , dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org, Daniel Stone , Rob Clark , Greg Hackmann , John Harrison , laurent.pinchart@ideasonboard.com, seanpaul@google.com, marcheu@google.com, m.chehab@samsung.com, Sumit Semwal , Maarten Lankhorst , Gustavo Padovan References: <1476975005-30441-1-git-send-email-gustavo@padovan.org> <1476975005-30441-2-git-send-email-gustavo@padovan.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1476975005-30441-2-git-send-email-gustavo@padovan.org> X-Operating-System: Linux phenom 4.6.0-1-amd64 User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 8440 Lines: 225 On Thu, Oct 20, 2016 at 12:50:02PM -0200, Gustavo Padovan wrote: > From: Gustavo Padovan > > There is now a new property called FENCE_FD attached to every plane > state that receives the sync_file fd from userspace via the atomic commit > IOCTL. > > The fd is then translated to a fence (that may be a fence_collection > subclass or just a normal fence) and then used by DRM to fence_wait() for > all fences in the sync_file to signal. So it only commits when all > framebuffers are ready to scanout. > > v2: Comments by Daniel Vetter: > - remove set state->fence = NULL in destroy phase > - accept fence -1 as valid and just return 0 > - do not call fence_get() - sync_file_fences_get() already calls it > - fence_put() if state->fence is already set, in case userspace > set the property more than once. > > v3: WARN_ON if fence is set but state has no FB > > v4: Comment from Maarten Lankhorst > - allow set fence with no related fb > > v5: rename FENCE_FD to IN_FENCE_FD > > Signed-off-by: Gustavo Padovan Since you rename plane_state->fence to in_fence (why exactly?) this breaks compilation for imx and msm. Also there's the question of which fence should win, if there's both explicit and implicit fences. I think the best way to solve that is by adding a helper to set the implicit fence, which drivers are supposed to call. That helper only sets the fence if there's not yet an explicit fence attached. New sequence would be: - patch1: roll out that helper (e.g. drm_atomic_plane_state_set_fence or whatever) for all drivers using it. - patch2 (this one here): add explicit fences and change the helper to cooporate correctly. > --- > drivers/gpu/drm/Kconfig | 1 + > drivers/gpu/drm/drm_atomic.c | 14 ++++++++++++++ > drivers/gpu/drm/drm_atomic_helper.c | 13 +++++++------ > drivers/gpu/drm/drm_crtc.c | 6 ++++++ > drivers/gpu/drm/drm_plane.c | 1 + > include/drm/drm_crtc.h | 5 +++++ > include/drm/drm_plane.h | 4 ++-- > 7 files changed, 36 insertions(+), 8 deletions(-) > > diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig > index 483059a..43cb33d 100644 > --- a/drivers/gpu/drm/Kconfig > +++ b/drivers/gpu/drm/Kconfig > @@ -12,6 +12,7 @@ menuconfig DRM > select I2C > select I2C_ALGOBIT > select DMA_SHARED_BUFFER > + select SYNC_FILE > help > Kernel-level support for the Direct Rendering Infrastructure (DRI) > introduced in XFree86 4.0. If you say Y here, you need to select > diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c > index 5dd7054..b3284b2 100644 > --- a/drivers/gpu/drm/drm_atomic.c > +++ b/drivers/gpu/drm/drm_atomic.c > @@ -30,6 +30,7 @@ > #include > #include > #include > +#include > > #include "drm_crtc_internal.h" > > @@ -686,6 +687,17 @@ int drm_atomic_plane_set_property(struct drm_plane *plane, > drm_atomic_set_fb_for_plane(state, fb); > if (fb) > drm_framebuffer_unreference(fb); > + } else if (property == config->prop_in_fence_fd) { > + if (U642I64(val) == -1) > + return 0; > + > + if (state->in_fence) > + fence_put(state->in_fence); > + > + state->in_fence = sync_file_get_fence(val); > + if (!state->in_fence) > + return -EINVAL; > + > } else if (property == config->prop_crtc_id) { > struct drm_crtc *crtc = drm_crtc_find(dev, val); > return drm_atomic_set_crtc_for_plane(state, crtc); > @@ -745,6 +757,8 @@ drm_atomic_plane_get_property(struct drm_plane *plane, > > if (property == config->prop_fb_id) { > *val = (state->fb) ? state->fb->base.id : 0; > + } else if (property == config->prop_in_fence_fd) { > + *val = -1; > } else if (property == config->prop_crtc_id) { > *val = (state->crtc) ? state->crtc->base.id : 0; > } else if (property == config->prop_crtc_x) { > diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c > index 07b432f..73464b1 100644 > --- a/drivers/gpu/drm/drm_atomic_helper.c > +++ b/drivers/gpu/drm/drm_atomic_helper.c > @@ -1031,22 +1031,20 @@ int drm_atomic_helper_wait_for_fences(struct drm_device *dev, > if (!pre_swap) > plane_state = plane->state; > > - if (!plane_state->fence) > + if (!plane_state->in_fence) > continue; > > - WARN_ON(!plane_state->fb); > - > /* > * If waiting for fences pre-swap (ie: nonblock), userspace can > * still interrupt the operation. Instead of blocking until the > * timer expires, make the wait interruptible. > */ > - ret = fence_wait(plane_state->fence, pre_swap); > + ret = fence_wait(plane_state->in_fence, pre_swap); > if (ret) > return ret; > > - fence_put(plane_state->fence); > - plane_state->fence = NULL; > + fence_put(plane_state->in_fence); > + plane_state->in_fence = NULL; > } > > return 0; > @@ -3114,6 +3112,9 @@ void __drm_atomic_helper_plane_destroy_state(struct drm_plane_state *state) > { > if (state->fb) > drm_framebuffer_unreference(state->fb); > + > + if (state->in_fence) > + fence_put(state->in_fence); > } > EXPORT_SYMBOL(__drm_atomic_helper_plane_destroy_state); > > diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c > index 60403bf..fcb6453 100644 > --- a/drivers/gpu/drm/drm_crtc.c > +++ b/drivers/gpu/drm/drm_crtc.c > @@ -397,6 +397,12 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > return -ENOMEM; > dev->mode_config.prop_fb_id = prop; > > + prop = drm_property_create_signed_range(dev, DRM_MODE_PROP_ATOMIC, > + "IN_FENCE_FD", -1, INT_MAX); > + if (!prop) > + return -ENOMEM; > + dev->mode_config.prop_in_fence_fd = prop; > + > prop = drm_property_create_object(dev, DRM_MODE_PROP_ATOMIC, > "CRTC_ID", DRM_MODE_OBJECT_CRTC); > if (!prop) > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > index 249c0ae..3957ef8 100644 > --- a/drivers/gpu/drm/drm_plane.c > +++ b/drivers/gpu/drm/drm_plane.c > @@ -137,6 +137,7 @@ int drm_universal_plane_init(struct drm_device *dev, struct drm_plane *plane, > > if (drm_core_check_feature(dev, DRIVER_ATOMIC)) { > drm_object_attach_property(&plane->base, config->prop_fb_id, 0); > + drm_object_attach_property(&plane->base, config->prop_in_fence_fd, -1); > drm_object_attach_property(&plane->base, config->prop_crtc_id, 0); > drm_object_attach_property(&plane->base, config->prop_crtc_x, 0); > drm_object_attach_property(&plane->base, config->prop_crtc_y, 0); > diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h > index 284c1b3..279132e 100644 > --- a/include/drm/drm_crtc.h > +++ b/include/drm/drm_crtc.h > @@ -1206,6 +1206,11 @@ struct drm_mode_config { > */ > struct drm_property *prop_fb_id; > /** > + * @prop_in_fence_fd: Sync File fd representing the incoming fences > + * for a Plane. > + */ > + struct drm_property *prop_in_fence_fd; > + /** > * @prop_crtc_id: Default atomic plane property to specify the > * &drm_crtc. > */ > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > index 0235390..3cb9bf1 100644 > --- a/include/drm/drm_plane.h > +++ b/include/drm/drm_plane.h > @@ -34,7 +34,7 @@ struct drm_crtc; > * @plane: backpointer to the plane > * @crtc: currently bound CRTC, NULL if disabled > * @fb: currently bound framebuffer > - * @fence: optional fence to wait for before scanning out @fb > + * @in_fence: optional fence to wait for before scanning out @fb > * @crtc_x: left position of visible portion of plane on crtc > * @crtc_y: upper position of visible portion of plane on crtc > * @crtc_w: width of visible portion of plane on crtc > @@ -59,7 +59,7 @@ struct drm_plane_state { > > struct drm_crtc *crtc; /* do not write directly, use drm_atomic_set_crtc_for_plane() */ > struct drm_framebuffer *fb; /* do not write directly, use drm_atomic_set_fb_for_plane() */ > - struct fence *fence; > + struct fence *in_fence; I think it'd be good to move the kerneldoc into an in-line comment here and expand a bit upon the expected semantics. This is a pretty important part of the atomic uabi! Otherwise looks all good. -Daniel > > /* Signed dest location allows it to be partially off screen */ > int32_t crtc_x, crtc_y; > -- > 2.5.5 > -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch