[prev in list] [next in list] [prev in thread] [next in thread] 

List:       busybox
Subject:    [PATCH 2/2] fbsplash: don't assume 16bpp means RGB565
From:       Linus Walleij <triad () df ! lth ! se>
Date:       2012-07-01 20:34:55
Message-ID: 1341174895-1711-1-git-send-email-triad () df ! lth ! se
[Download RAW message or body]

The current handling of the 16bpp mode hardcodes the bit
arrangement for 16bpp mode to be of the RGB565 type, but my
system (ARM PL110 on the Integrator/CP) has RGBA5551
arrangement and will give wrong colors. This patch makes the
code actually respect the bit sizes and offset arrangement
from the screen configuration (in turn read from the modes
configuration file and set by fbset).

Signed-off-by: Linus Walleij <triad@df.lth.se>
---
 miscutils/fbsplash.c |   15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/miscutils/fbsplash.c b/miscutils/fbsplash.c
index 988439b..a9581a8 100644
--- a/miscutils/fbsplash.c
+++ b/miscutils/fbsplash.c
@@ -166,10 +166,17 @@ static unsigned fb_pixel_value(unsigned r, unsigned g, unsigned b)
 		return r + g + b;
 	}
 	if (G.bytes_per_pixel == 2) {
-		r = (r & 0xf8) << 8; // 5-bit red
-		g = (g & 0xfc) << 3; // 6-bit green
-		b =  b >> 3;         // 5-bit blue
-		return r + g + b;
+		/* First shift out unused bits */
+		r &= 0xff;
+		g &= 0xff;
+		b &= 0xff;
+		r >>= (8 - G.scr_var.red.length);
+		g >>= (8 - G.scr_var.green.length);
+		b >>= (8 - G.scr_var.blue.length);
+		/* Then shift the remaining bits to their offset */
+		return (r << G.scr_var.red.offset) +
+			(g << G.scr_var.green.offset) +
+			(b << G.scr_var.blue.offset);
 	}
 	// RGB 888
 	return b + (g << 8) + (r << 16);
-- 
1.7.10.4

_______________________________________________
busybox mailing list
busybox@busybox.net
http://lists.busybox.net/mailman/listinfo/busybox
[prev in list] [next in list] [prev in thread] [next in thread]