All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Zimmermann <tzimmermann@suse.de>
To: airlied@redhat.com, daniel@ffwll.ch
Cc: Thomas Zimmermann <tzimmermann@suse.de>, dri-devel@lists.freedesktop.org
Subject: [PATCH v2 09/10] drm/ast: Store each HW cursor offset after pinning the rsp BO
Date: Tue,  9 Feb 2021 14:46:31 +0100	[thread overview]
Message-ID: <20210209134632.12157-10-tzimmermann@suse.de> (raw)
In-Reply-To: <20210209134632.12157-1-tzimmermann@suse.de>

As HW cursor BOs never move, we can store the offset in VRAM in the
cursor-plane's HWC state. This removes the last possible source of
runtime errors from atomic_update.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
---
 drivers/gpu/drm/ast/ast_drv.h  |  1 +
 drivers/gpu/drm/ast/ast_mode.c | 21 +++++++++++----------
 2 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 22193cfde255..e82ab8628770 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -108,6 +108,7 @@ struct ast_cursor_plane {
 	struct {
 		struct drm_gem_vram_object *gbo;
 		struct dma_buf_map map;
+		u64 off;
 	} hwc[AST_DEFAULT_HWC_NUM];
 
 	unsigned int next_hwc_index;
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index b9b9badcee00..e8de86d23805 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -823,26 +823,19 @@ ast_cursor_plane_helper_atomic_update(struct drm_plane *plane,
 	struct drm_plane_state *state = plane->state;
 	struct drm_framebuffer *fb = state->fb;
 	struct ast_private *ast = to_ast_private(plane->dev);
-	struct drm_device *dev = &ast->base;
-	struct drm_gem_vram_object *gbo =
-		ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].gbo;
+	u64 dst_off =
+		ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].off;
 	unsigned int offset_x, offset_y;
-	s64 off;
 	struct dma_buf_map map;
 	u16 x, y;
 	u8 x_offset, y_offset;
 	u8 __iomem *dst;
 	u8 __iomem *sig;
 
-	gbo = ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].gbo;
 	map = ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].map;
 
 	if (state->fb != old_state->fb) {
-		/* A new cursor image was installed. */
-		off = drm_gem_vram_offset(gbo);
-		if (drm_WARN_ON_ONCE(dev, off < 0))
-			return; /* Bug: we didn't pin the cursor HW BO to VRAM. */
-		ast_set_cursor_base(ast, off);
+		ast_set_cursor_base(ast, dst_off);
 
 		++ast_cursor_plane->next_hwc_index;
 		ast_cursor_plane->next_hwc_index %= ARRAY_SIZE(ast_cursor_plane->hwc);
@@ -931,6 +924,7 @@ static int ast_cursor_plane_init(struct ast_private *ast)
 	struct drm_gem_vram_object *gbo;
 	struct dma_buf_map map;
 	int ret;
+	s64 off;
 
 	/*
 	 * Allocate backing storage for cursors. The BOs are permanently
@@ -952,8 +946,14 @@ static int ast_cursor_plane_init(struct ast_private *ast)
 		ret = drm_gem_vram_vmap(gbo, &map);
 		if (ret)
 			goto err_drm_gem_vram_unpin;
+		off = drm_gem_vram_offset(gbo);
+		if (off < 0) {
+			ret = off;
+			goto err_drm_gem_vram_vunmap;
+		}
 		ast_cursor_plane->hwc[i].gbo = gbo;
 		ast_cursor_plane->hwc[i].map = map;
+		ast_cursor_plane->hwc[i].off = off;
 	}
 
 	/*
@@ -979,6 +979,7 @@ static int ast_cursor_plane_init(struct ast_private *ast)
 		--i;
 		gbo = ast_cursor_plane->hwc[i].gbo;
 		map = ast_cursor_plane->hwc[i].map;
+err_drm_gem_vram_vunmap:
 		drm_gem_vram_vunmap(gbo, &map);
 err_drm_gem_vram_unpin:
 		drm_gem_vram_unpin(gbo);
-- 
2.30.0

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  parent reply	other threads:[~2021-02-09 13:46 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-09 13:46 [PATCH v2 00/10] drm/ast: Clean-up cursor-plane updates Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 01/10] drm/ast: Add constants for VGACRCB register bits Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 02/10] drm/ast: Fix invalid usage of AST_MAX_HWC_WIDTH in cursor atomic_check Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 03/10] drm/ast: Initialize planes in helper functions Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 04/10] drm/ast: Allocate HW cursor BOs during cursor-plane initialization Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 05/10] drm/ast: Inline ast cursor-update functions into modesetting code Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 06/10] drm/ast: Add cursor-plane data structure Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 07/10] drm/ast: Store cursor BOs in cursor plane Thomas Zimmermann
2021-02-09 13:46 ` [PATCH v2 08/10] drm/ast: Map HW cursor BOs permanently Thomas Zimmermann
2021-02-09 13:46 ` Thomas Zimmermann [this message]
2021-02-09 13:46 ` [PATCH v2 10/10] drm/ast: Move all of the cursor-update functionality to atomic_update Thomas Zimmermann
2021-02-17  9:53 ` [PATCH v2 00/10] drm/ast: Clean-up cursor-plane updates Gerd Hoffmann

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210209134632.12157-10-tzimmermann@suse.de \
    --to=tzimmermann@suse.de \
    --cc=airlied@redhat.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.