linux-kernel - [PATCH v6 2/3] drm/panel: simple: Use display

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20190711203455.125667-3-dianders@chromium.org>
Date:   Thu, 11 Jul 2019 13:34:54 -0700
From:   Douglas Anderson <dianders@...omium.org>
To:     Thierry Reding <thierry.reding@...il.com>,
        Heiko Stuebner <heiko@...ech.de>,
        Sean Paul <seanpaul@...omium.org>
Cc:     Enric Balletbo i Serra <enric.balletbo@...labora.com>,
        Ezequiel Garcia <ezequiel@...labora.com>,
        linux-rockchip@...ts.infradead.org,
        dri-devel@...ts.freedesktop.org,
        Boris Brezillon <boris.brezillon@...labora.com>,
        mka@...omium.org, Rob Herring <robh+dt@...nel.org>,
        Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        Douglas Anderson <dianders@...omium.org>,
        linux-kernel@...r.kernel.org, Sam Ravnborg <sam@...nborg.org>,
        David Airlie <airlied@...ux.ie>,
        Daniel Vetter <daniel@...ll.ch>
Subject: [PATCH v6 2/3] drm/panel: simple: Use display_timing for Innolux n116bge

Convert the Innolux n116bge from using a fixed mode to specifying a
display timing with min/typ/max values.

Note that the n116bge's datasheet doesn't fit too well into DRM's way
of specifying things.  Specifically the panel's datasheet just
specifies the vertical blanking period and horizontal blanking period
and doesn't break things out.  For now we'll leave everything as a
fixed value but just allow adjusting the pixel clock.  I've added a
comment on what the datasheet claims so someone could later expand
things to fit their needs if they wanted to test other blanking
periods.

The goal here is to be able to specify the panel timings in the device
tree for several rk3288 Chromebooks (like rk3288-veryon-jerry).  These
Chromebooks have all been running in the downstream kernel with the
standard porches and sync lengths but just with a slightly slower
pixel clock because the 76.42 MHz clock is not achievable from the
fixed PLL that was available.  These Chromebooks only achieve a
refresh rate of ~58 Hz.  While it's probable that we could adjust the
timings to achieve 60 Hz it's probably wisest to match what's been
running on these devices all these years.

I'll note that though the upstream kernel has always tried to achieve
76.42 MHz, it has actually been running at 74.25 MHz also since the
video processor is parented off the same fixed PLL.

Changes in v4:
 - display_timing for Innolux n116bge new for v4.
Changes in v5:
 - Added Heiko's Tested-by
Changes in v6:
 - Rebased to drm-misc next
 - Added tags

Signed-off-by: Douglas Anderson <dianders@...omium.org>
Tested-by: Heiko Stuebner <heiko@...ech.de>
Acked-by: Thierry Reding <thierry.reding@...il.com>
---

 drivers/gpu/drm/panel/panel-simple.c | 37 +++++++++++++++++-----------
 1 file changed, 23 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 1bee197821ef..602809f6da6a 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -1702,23 +1702,32 @@ static const struct panel_desc innolux_g121x1_l03 = {
 	},
 };

-static const struct drm_display_mode innolux_n116bge_mode = {
-	.clock = 76420,
-	.hdisplay = 1366,
-	.hsync_start = 1366 + 136,
-	.hsync_end = 1366 + 136 + 30,
-	.htotal = 1366 + 136 + 30 + 60,
-	.vdisplay = 768,
-	.vsync_start = 768 + 8,
-	.vsync_end = 768 + 8 + 12,
-	.vtotal = 768 + 8 + 12 + 12,
-	.vrefresh = 60,
-	.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+/*
+ * Datasheet specifies that at 60 Hz refresh rate:
+ * - total horizontal time: { 1506, 1592, 1716 }
+ * - total vertical time: { 788, 800, 868 }
+ *
+ * ...but doesn't go into exactly how that should be split into a front
+ * porch, back porch, or sync length.  For now we'll leave a single setting
+ * here which allows a bit of tweaking of the pixel clock at the expense of
+ * refresh rate.
+ */
+static const struct display_timing innolux_n116bge_timing = {
+	.pixelclock = { 72600000, 76420000, 80240000 },
+	.hactive = { 1366, 1366, 1366 },
+	.hfront_porch = { 136, 136, 136 },
+	.hback_porch = { 60, 60, 60 },
+	.hsync_len = { 30, 30, 30 },
+	.vactive = { 768, 768, 768 },
+	.vfront_porch = { 8, 8, 8 },
+	.vback_porch = { 12, 12, 12 },
+	.vsync_len = { 12, 12, 12 },
+	.flags = DISPLAY_FLAGS_VSYNC_LOW | DISPLAY_FLAGS_HSYNC_LOW,
 };

 static const struct panel_desc innolux_n116bge = {
-	.modes = &innolux_n116bge_mode,
-	.num_modes = 1,
+	.timings = &innolux_n116bge_timing,
+	.num_timings = 1,
 	.bpc = 6,
 	.size = {
 		.width = 256,
-- 
2.22.0.410.gd8fdbe21b5-goog