forge/linux-pinenote
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
linux-pinenote/drivers/video/fbdev/via/viamode.c

384 lines
17 KiB
C
Raw Normal View History

viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
/*
* Copyright 1998-2008 VIA Technologies, Inc. All Rights Reserved.
* Copyright 2001-2008 S3 Graphics, Inc. All Rights Reserved.
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public
* License as published by the Free Software Foundation;
* either version 2, or (at your option) any later version.
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTIES OR REPRESENTATIONS; without even
* the implied warranty of MERCHANTABILITY or FITNESS FOR
* A PARTICULAR PURPOSE.See the GNU General Public License
* for more details.
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc.,
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*/
viafb: move some include files to include/linux These are the files which should be available to subdevices compiled outside of drivers/video/via. Cc: ScottFang@viatech.com.cn Cc: JosephChan@via.com.tw Cc: Harald Welte <laforge@gnumonks.org> Acked-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Signed-off-by: Jonathan Corbet <corbet@lwn.net>
2010-05-05 14:44:55 -06:00
#include <linux/via-core.h>
viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
#include "global.h"
struct io_reg CN400_ModeXregs[] = { {VIASR, SR10, 0xFF, 0x01},
{VIASR, SR15, 0x02, 0x02},
{VIASR, SR16, 0xBF, 0x08},
{VIASR, SR17, 0xFF, 0x1F},
{VIASR, SR18, 0xFF, 0x4E},
{VIASR, SR1A, 0xFB, 0x08},
{VIASR, SR1E, 0x0F, 0x01},
{VIASR, SR2A, 0xFF, 0x00},
{VIACR, CR32, 0xFF, 0x00},
{VIACR, CR33, 0xFF, 0x00},
{VIACR, CR35, 0xFF, 0x00},
{VIACR, CR36, 0x08, 0x00},
{VIACR, CR69, 0xFF, 0x00},
{VIACR, CR6A, 0xFF, 0x40},
{VIACR, CR6B, 0xFF, 0x00},
{VIACR, CR88, 0xFF, 0x40}, /* LCD Panel Type */
{VIACR, CR89, 0xFF, 0x00}, /* LCD Timing Control 0 */
{VIACR, CR8A, 0xFF, 0x88}, /* LCD Timing Control 1 */
{VIACR, CR8B, 0xFF, 0x69}, /* LCD Power Sequence Control 0 */
{VIACR, CR8C, 0xFF, 0x57}, /* LCD Power Sequence Control 1 */
{VIACR, CR8D, 0xFF, 0x00}, /* LCD Power Sequence Control 2 */
{VIACR, CR8E, 0xFF, 0x7B}, /* LCD Power Sequence Control 3 */
{VIACR, CR8F, 0xFF, 0x03}, /* LCD Power Sequence Control 4 */
{VIACR, CR90, 0xFF, 0x30}, /* LCD Power Sequence Control 5 */
{VIACR, CR91, 0xFF, 0xA0}, /* 24/12 bit LVDS Data off */
{VIACR, CR96, 0xFF, 0x00},
{VIACR, CR97, 0xFF, 0x00},
{VIACR, CR99, 0xFF, 0x00},
{VIACR, CR9B, 0xFF, 0x00}
};
/* Video Mode Table for VT3314 chipset*/
/* Common Setting for Video Mode */
struct io_reg CN700_ModeXregs[] = { {VIASR, SR10, 0xFF, 0x01},
{VIASR, SR15, 0x02, 0x02},
{VIASR, SR16, 0xBF, 0x08},
{VIASR, SR17, 0xFF, 0x1F},
{VIASR, SR18, 0xFF, 0x4E},
{VIASR, SR1A, 0xFB, 0x82},
{VIASR, SR1B, 0xFF, 0xF0},
{VIASR, SR1F, 0xFF, 0x00},
{VIASR, SR1E, 0xFF, 0x01},
{VIASR, SR22, 0xFF, 0x1F},
{VIASR, SR2A, 0x0F, 0x00},
{VIASR, SR2E, 0xFF, 0xFF},
{VIASR, SR3F, 0xFF, 0xFF},
{VIASR, SR40, 0xF7, 0x00},
{VIASR, CR30, 0xFF, 0x04},
{VIACR, CR32, 0xFF, 0x00},
{VIACR, CR33, 0x7F, 0x00},
{VIACR, CR35, 0xFF, 0x00},
{VIACR, CR36, 0xFF, 0x31},
{VIACR, CR41, 0xFF, 0x80},
{VIACR, CR42, 0xFF, 0x00},
{VIACR, CR55, 0x80, 0x00},
{VIACR, CR5D, 0x80, 0x00}, /*Horizontal Retrace Start bit[11] should be 0*/
{VIACR, CR68, 0xFF, 0x67}, /* Default FIFO For IGA2 */
{VIACR, CR69, 0xFF, 0x00},
{VIACR, CR6A, 0xFD, 0x40},
{VIACR, CR6B, 0xFF, 0x00},
{VIACR, CR77, 0xFF, 0x00}, /* LCD scaling Factor */
{VIACR, CR78, 0xFF, 0x00}, /* LCD scaling Factor */
{VIACR, CR79, 0xFF, 0x00}, /* LCD scaling Factor */
{VIACR, CR9F, 0x03, 0x00}, /* LCD scaling Factor */
{VIACR, CR88, 0xFF, 0x40}, /* LCD Panel Type */
{VIACR, CR89, 0xFF, 0x00}, /* LCD Timing Control 0 */
{VIACR, CR8A, 0xFF, 0x88}, /* LCD Timing Control 1 */
{VIACR, CR8B, 0xFF, 0x5D}, /* LCD Power Sequence Control 0 */
{VIACR, CR8C, 0xFF, 0x2B}, /* LCD Power Sequence Control 1 */
{VIACR, CR8D, 0xFF, 0x6F}, /* LCD Power Sequence Control 2 */
{VIACR, CR8E, 0xFF, 0x2B}, /* LCD Power Sequence Control 3 */
{VIACR, CR8F, 0xFF, 0x01}, /* LCD Power Sequence Control 4 */
{VIACR, CR90, 0xFF, 0x01}, /* LCD Power Sequence Control 5 */
{VIACR, CR91, 0xFF, 0xA0}, /* 24/12 bit LVDS Data off */
{VIACR, CR96, 0xFF, 0x00},
{VIACR, CR97, 0xFF, 0x00},
{VIACR, CR99, 0xFF, 0x00},
{VIACR, CR9B, 0xFF, 0x00},
{VIACR, CR9D, 0xFF, 0x80},
{VIACR, CR9E, 0xFF, 0x80}
};
struct io_reg KM400_ModeXregs[] = {
{VIASR, SR10, 0xFF, 0x01}, /* Unlock Register */
{VIASR, SR16, 0xFF, 0x08}, /* Display FIFO threshold Control */
{VIASR, SR17, 0xFF, 0x1F}, /* Display FIFO Control */
{VIASR, SR18, 0xFF, 0x4E}, /* GFX PREQ threshold */
{VIASR, SR1A, 0xFF, 0x0a}, /* GFX PREQ threshold */
{VIASR, SR1F, 0xFF, 0x00}, /* Memory Control 0 */
{VIASR, SR1B, 0xFF, 0xF0}, /* Power Management Control 0 */
{VIASR, SR1E, 0xFF, 0x01}, /* Power Management Control */
{VIASR, SR20, 0xFF, 0x00}, /* Sequencer Arbiter Control 0 */
{VIASR, SR21, 0xFF, 0x00}, /* Sequencer Arbiter Control 1 */
{VIASR, SR22, 0xFF, 0x1F}, /* Display Arbiter Control 1 */
{VIASR, SR2A, 0xFF, 0x00}, /* Power Management Control 5 */
{VIASR, SR2D, 0xFF, 0xFF}, /* Power Management Control 1 */
{VIASR, SR2E, 0xFF, 0xFF}, /* Power Management Control 2 */
{VIACR, CR33, 0xFF, 0x00},
{VIACR, CR55, 0x80, 0x00},
{VIACR, CR5D, 0x80, 0x00},
{VIACR, CR36, 0xFF, 0x01}, /* Power Mangement 3 */
{VIACR, CR68, 0xFF, 0x67}, /* Default FIFO For IGA2 */
{VIACR, CR6A, 0x20, 0x20}, /* Extended FIFO On */
{VIACR, CR88, 0xFF, 0x40}, /* LCD Panel Type */
{VIACR, CR89, 0xFF, 0x00}, /* LCD Timing Control 0 */
{VIACR, CR8A, 0xFF, 0x88}, /* LCD Timing Control 1 */
{VIACR, CR8B, 0xFF, 0x2D}, /* LCD Power Sequence Control 0 */
{VIACR, CR8C, 0xFF, 0x2D}, /* LCD Power Sequence Control 1 */
{VIACR, CR8D, 0xFF, 0xC8}, /* LCD Power Sequence Control 2 */
{VIACR, CR8E, 0xFF, 0x36}, /* LCD Power Sequence Control 3 */
{VIACR, CR8F, 0xFF, 0x00}, /* LCD Power Sequence Control 4 */
{VIACR, CR90, 0xFF, 0x10}, /* LCD Power Sequence Control 5 */
{VIACR, CR91, 0xFF, 0xA0}, /* 24/12 bit LVDS Data off */
{VIACR, CR96, 0xFF, 0x03}, /* DVP0 ; DVP0 Clock Skew */
{VIACR, CR97, 0xFF, 0x03}, /* DFP high ; DFPH Clock Skew */
{VIACR, CR99, 0xFF, 0x03}, /* DFP low ; DFPL Clock Skew*/
{VIACR, CR9B, 0xFF, 0x07} /* DVI on DVP1 ; DVP1 Clock Skew*/
};
/* For VT3324: Common Setting for Video Mode */
struct io_reg CX700_ModeXregs[] = { {VIASR, SR10, 0xFF, 0x01},
{VIASR, SR15, 0x02, 0x02},
{VIASR, SR16, 0xBF, 0x08},
{VIASR, SR17, 0xFF, 0x1F},
{VIASR, SR18, 0xFF, 0x4E},
{VIASR, SR1A, 0xFB, 0x08},
{VIASR, SR1B, 0xFF, 0xF0},
{VIASR, SR1E, 0xFF, 0x01},
{VIASR, SR2A, 0xFF, 0x00},
viafb: delete clock and PLL initialization We do this also in the real program code so there is no reason to do it here too (and here it's hardly readable). Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-03-26 03:17:42 +00:00
{VIASR, SR2D, 0xC0, 0xC0}, /* delayed E3_ECK */
viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
{VIACR, CR32, 0xFF, 0x00},
{VIACR, CR33, 0xFF, 0x00},
{VIACR, CR35, 0xFF, 0x00},
{VIACR, CR36, 0x08, 0x00},
{VIACR, CR47, 0xC8, 0x00}, /* Clear VCK Plus. */
{VIACR, CR69, 0xFF, 0x00},
{VIACR, CR6A, 0xFF, 0x40},
{VIACR, CR6B, 0xFF, 0x00},
{VIACR, CR88, 0xFF, 0x40}, /* LCD Panel Type */
{VIACR, CR89, 0xFF, 0x00}, /* LCD Timing Control 0 */
{VIACR, CR8A, 0xFF, 0x88}, /* LCD Timing Control 1 */
{VIACR, CRD4, 0xFF, 0x81}, /* Second power sequence control */
{VIACR, CR8B, 0xFF, 0x5D}, /* LCD Power Sequence Control 0 */
{VIACR, CR8C, 0xFF, 0x2B}, /* LCD Power Sequence Control 1 */
{VIACR, CR8D, 0xFF, 0x6F}, /* LCD Power Sequence Control 2 */
{VIACR, CR8E, 0xFF, 0x2B}, /* LCD Power Sequence Control 3 */
{VIACR, CR8F, 0xFF, 0x01}, /* LCD Power Sequence Control 4 */
{VIACR, CR90, 0xFF, 0x01}, /* LCD Power Sequence Control 5 */
{VIACR, CR91, 0xFF, 0x80}, /* 24/12 bit LVDS Data off */
{VIACR, CR96, 0xFF, 0x00},
{VIACR, CR97, 0xFF, 0x00},
{VIACR, CR99, 0xFF, 0x00},
viafb: remove LVDS initialization At least for VX800 this initialization is not very good as some parts of the register are written with reserved values. This makes the display go white in some configurations and not usable until the framebuffer is removed. It's better to not initialize it as it allows to use a previously (by BIOS) correctly configured display. This patch makes some displays work but might cause problems on others. This is bad but can not be easily avoided. If this causes some regressions it's probably the best to fix it in the 'active' display setup code. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Harald Welte <laforge@gnumonks.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:22 -07:00
{VIACR, CR9B, 0xFF, 0x00}
viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
};
viafb: add support for the VX855 chipset Add support for a new VIA integrated graphics chipset, the VX855. Signed-off-by: HaraldWelte <HaraldWelte@viatech.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:35 -07:00
struct io_reg VX855_ModeXregs[] = {
{VIASR, SR10, 0xFF, 0x01},
{VIASR, SR15, 0x02, 0x02},
{VIASR, SR16, 0xBF, 0x08},
{VIASR, SR17, 0xFF, 0x1F},
{VIASR, SR18, 0xFF, 0x4E},
{VIASR, SR1A, 0xFB, 0x08},
{VIASR, SR1B, 0xFF, 0xF0},
{VIASR, SR1E, 0x07, 0x01},
{VIASR, SR2A, 0xF0, 0x00},
{VIASR, SR58, 0xFF, 0x00},
{VIASR, SR59, 0xFF, 0x00},
viafb: delete clock and PLL initialization We do this also in the real program code so there is no reason to do it here too (and here it's hardly readable). Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-03-26 03:17:42 +00:00
{VIASR, SR2D, 0xC0, 0xC0}, /* delayed E3_ECK */
viafb: add support for the VX855 chipset Add support for a new VIA integrated graphics chipset, the VX855. Signed-off-by: HaraldWelte <HaraldWelte@viatech.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:35 -07:00
{VIACR, CR32, 0xFF, 0x00},
{VIACR, CR33, 0x7F, 0x00},
{VIACR, CR35, 0xFF, 0x00},
{VIACR, CR36, 0x08, 0x00},
{VIACR, CR69, 0xFF, 0x00},
{VIACR, CR6A, 0xFD, 0x60},
{VIACR, CR6B, 0xFF, 0x00},
{VIACR, CR88, 0xFF, 0x40}, /* LCD Panel Type */
{VIACR, CR89, 0xFF, 0x00}, /* LCD Timing Control 0 */
{VIACR, CR8A, 0xFF, 0x88}, /* LCD Timing Control 1 */
{VIACR, CRD4, 0xFF, 0x81}, /* Second power sequence control */
{VIACR, CR91, 0xFF, 0x80}, /* 24/12 bit LVDS Data off */
{VIACR, CR96, 0xFF, 0x00},
{VIACR, CR97, 0xFF, 0x00},
{VIACR, CR99, 0xFF, 0x00},
{VIACR, CR9B, 0xFF, 0x00},
{VIACR, CRD2, 0xFF, 0xFF} /* TMDS/LVDS control register. */
};
viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
/* Video Mode Table */
/* Common Setting for Video Mode */
struct io_reg CLE266_ModeXregs[] = { {VIASR, SR1E, 0xF0, 0x00},
{VIASR, SR2A, 0x0F, 0x00},
{VIASR, SR15, 0x02, 0x02},
{VIASR, SR16, 0xBF, 0x08},
{VIASR, SR17, 0xFF, 0x1F},
{VIASR, SR18, 0xFF, 0x4E},
{VIASR, SR1A, 0xFB, 0x08},
{VIACR, CR32, 0xFF, 0x00},
{VIACR, CR35, 0xFF, 0x00},
{VIACR, CR36, 0x08, 0x00},
{VIACR, CR6A, 0xFF, 0x80},
{VIACR, CR6A, 0xFF, 0xC0},
{VIACR, CR55, 0x80, 0x00},
{VIACR, CR5D, 0x80, 0x00},
{VIAGR, GR20, 0xFF, 0x00},
{VIAGR, GR21, 0xFF, 0x00},
{VIAGR, GR22, 0xFF, 0x00},
};
/* Mode:1024X768 */
struct io_reg PM1024x768[] = { {VIASR, 0x16, 0xBF, 0x0C},
{VIASR, 0x18, 0xFF, 0x4C}
};
struct patch_table res_patch_table[] = {
viafb: split global index up This is the first step to remove an artificial global index that was used in two ways: 1. As a pseudo index in the mode table. Pseudo as you had to search through the table to find the referenced entry. This was replaced by using a pointer to the entry. 2. As a shortcut to compare a combination of horizontal and vertical resolution at the same time. This was replaced by a "(hres<<16) | vres" which is good enough for now and the near future. If vres or hres become greater than 2^16 this might indeed cause problems but this solution allows to split this indexing mess up without the requirement to do even more code changes. This is a big change that will allow more clean ups. It should be a bit faster but that is probably not relevant for normal operation. No regressions expected but as this is a relatively big step heavy testing is appreciated. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-10 15:21:28 -08:00
{ARRAY_SIZE(PM1024x768), PM1024x768}
viafb: via_i2c.c, via_i2c.h, viamode.c, viamode.h via_i2c.c, via_i2c.h: Implement i2c specification. viamode.c, viamode.c: all support modes information. Signed-off-by: Joseph Chan <josephchan@via.com.tw> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-15 22:03:29 -07:00
};
/* struct VPITTable {
unsigned char Misc;
unsigned char SR[StdSR];
unsigned char CR[StdCR];
unsigned char GR[StdGR];
unsigned char AR[StdAR];
};*/
struct VPITTable VPIT = {
/* Msic */
0xC7,
/* Sequencer */
{0x01, 0x0F, 0x00, 0x0E},
/* Graphic Controller */
{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05, 0x0F, 0xFF},
/* Attribute Controller */
{0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
0x01, 0x00, 0x0F, 0x00}
};
/********************/
/* Mode Table */
/********************/
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
static const struct fb_videomode viafb_modes[] = {
{NULL, 60, 480, 640, 40285, 72, 24, 19, 1, 48, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 640, 480, 39682, 48, 16, 33, 10, 96, 2, 0, 0, 0},
{NULL, 75, 640, 480, 31746, 120, 16, 16, 1, 64, 3, 0, 0, 0},
{NULL, 85, 640, 480, 27780, 80, 56, 25, 1, 56, 3, 0, 0, 0},
{NULL, 100, 640, 480, 23167, 104, 40, 25, 1, 64, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 120, 640, 480, 19081, 104, 40, 31, 1, 64, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 720, 480, 37426, 88, 16, 13, 1, 72, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 720, 576, 30611, 96, 24, 17, 1, 72, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 800, 600, 25131, 88, 40, 23, 1, 128, 4, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 800, 600, 20202, 160, 16, 21, 1, 80, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 85, 800, 600, 17790, 152, 32, 27, 1, 64, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 100, 800, 600, 14667, 136, 48, 32, 1, 88, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 120, 800, 600, 11911, 144, 56, 39, 1, 88, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 800, 480, 33602, 96, 24, 10, 3, 72, 7, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 848, 480, 31565, 104, 24, 12, 3, 80, 5, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 856, 480, 31517, 104, 16, 13, 1, 88, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1024, 512, 24218, 136, 32, 15, 1, 104, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1024, 600, 20423, 144, 40, 18, 1, 104, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1024, 768, 15385, 160, 24, 29, 3, 136, 6, 0, 0, 0},
{NULL, 75, 1024, 768, 12703, 176, 16, 28, 1, 96, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 85, 1024, 768, 10581, 208, 48, 36, 1, 96, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 100, 1024, 768, 8825, 184, 72, 42, 1, 112, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1152, 864, 9259, 256, 64, 32, 1, 128, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 768, 12478, 200, 64, 23, 1, 136, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 50, 1280, 768, 15342, 184, 56, 19, 1, 128, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 960, 600, 21964, 128, 32, 15, 3, 96, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1000, 600, 20803, 144, 40, 18, 1, 104, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1024, 576, 21278, 144, 40, 17, 1, 104, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1088, 612, 18825, 152, 48, 16, 3, 104, 5, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1152, 720, 14974, 168, 56, 19, 3, 112, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1200, 720, 14248, 184, 56, 22, 1, 128, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 49, 1200, 900, 17703, 21, 11, 1, 1, 32, 10, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 600, 16259, 184, 56, 18, 1, 128, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 800, 11938, 200, 72, 22, 3, 128, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 960, 9259, 312, 96, 36, 1, 112, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 1024, 9262, 248, 48, 38, 1, 112, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1280, 1024, 7409, 248, 16, 38, 1, 144, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 85, 1280, 1024, 6351, 224, 64, 44, 1, 160, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1360, 768, 11759, 208, 72, 22, 3, 136, 5, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1368, 768, 11646, 216, 72, 23, 1, 144, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 50, 1368, 768, 14301, 200, 56, 19, 1, 144, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1368, 768, 11646, 216, 72, 23, 1, 144, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1440, 900, 9372, 232, 80, 25, 3, 152, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1440, 900, 7311, 248, 96, 33, 3, 152, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1440, 1040, 7993, 248, 96, 33, 1, 152, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1600, 900, 8449, 256, 88, 26, 3, 168, 5, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1600, 1024, 7333, 272, 104, 32, 1, 168, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1600, 1200, 6172, 304, 64, 46, 1, 192, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1600, 1200, 4938, 304, 64, 46, 1, 192, 3, FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1680, 1050, 6832, 280, 104, 30, 3, 176, 6, 0, 0, 0},
{NULL, 75, 1680, 1050, 5339, 296, 120, 40, 3, 176, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1792, 1344, 4883, 328, 128, 46, 1, 200, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1856, 1392, 4581, 352, 96, 43, 1, 224, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1920, 1440, 4273, 344, 128, 56, 1, 208, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1920, 1440, 3367, 352, 144, 56, 1, 224, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 2048, 1536, 3738, 376, 152, 49, 3, 224, 4, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1280, 720, 13484, 216, 112, 20, 5, 40, 5, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 50, 1280, 720, 16538, 176, 48, 17, 1, 128, 3, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1920, 1080, 5776, 328, 128, 32, 3, 200, 5, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1920, 1200, 5164, 336, 136, 36, 3, 200, 6, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 60, 1400, 1050, 8210, 232, 88, 32, 3, 144, 4, FB_SYNC_VERT_HIGH_ACT, 0, 0},
{NULL, 75, 1400, 1050, 6398, 248, 104, 42, 3, 144, 4, FB_SYNC_VERT_HIGH_ACT, 0, 0} };
static const struct fb_videomode viafb_rb_modes[] = {
{NULL, 60, 1360, 768, 13879, 80, 48, 14, 3, 32, 5, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1440, 900, 11249, 80, 48, 17, 3, 32, 6, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1400, 1050, 9892, 80, 48, 23, 3, 32, 4, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1600, 900, 10226, 80, 48, 18, 3, 32, 5, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1680, 1050, 8387, 80, 48, 21, 3, 32, 6, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1920, 1080, 7212, 80, 48, 23, 3, 32, 5, FB_SYNC_HOR_HIGH_ACT, 0, 0},
{NULL, 60, 1920, 1200, 6488, 80, 48, 26, 3, 32, 6, FB_SYNC_HOR_HIGH_ACT, 0, 0} };
viafb: split global index up This is the first step to remove an artificial global index that was used in two ways: 1. As a pseudo index in the mode table. Pseudo as you had to search through the table to find the referenced entry. This was replaced by using a pointer to the entry. 2. As a shortcut to compare a combination of horizontal and vertical resolution at the same time. This was replaced by a "(hres<<16) | vres" which is good enough for now and the near future. If vres or hres become greater than 2^16 this might indeed cause problems but this solution allows to split this indexing mess up without the requirement to do even more code changes. This is a big change that will allow more clean ups. It should be a bit faster but that is probably not relevant for normal operation. No regressions expected but as this is a relatively big step heavy testing is appreciated. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-10 15:21:28 -08:00
viafb: clean up viamode.h Remove unneeeded declarations from the header and makes it more maintainable by evaluating the array size in the file the array exist in and exporting it via variables. This removes the need to keep the array size in the header in sync with the real array size. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Harald Welte <laforge@gnumonks.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:16 -07:00
int NUM_TOTAL_CN400_ModeXregs = ARRAY_SIZE(CN400_ModeXregs);
int NUM_TOTAL_CN700_ModeXregs = ARRAY_SIZE(CN700_ModeXregs);
int NUM_TOTAL_KM400_ModeXregs = ARRAY_SIZE(KM400_ModeXregs);
int NUM_TOTAL_CX700_ModeXregs = ARRAY_SIZE(CX700_ModeXregs);
viafb: add support for the VX855 chipset Add support for a new VIA integrated graphics chipset, the VX855. Signed-off-by: HaraldWelte <HaraldWelte@viatech.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:35 -07:00
int NUM_TOTAL_VX855_ModeXregs = ARRAY_SIZE(VX855_ModeXregs);
viafb: clean up viamode.h Remove unneeeded declarations from the header and makes it more maintainable by evaluating the array size in the file the array exist in and exporting it via variables. This removes the need to keep the array size in the header in sync with the real array size. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Harald Welte <laforge@gnumonks.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-09-22 16:47:16 -07:00
int NUM_TOTAL_CLE266_ModeXregs = ARRAY_SIZE(CLE266_ModeXregs);
int NUM_TOTAL_PATCH_MODE = ARRAY_SIZE(res_patch_table);
viafb: split global index up This is the first step to remove an artificial global index that was used in two ways: 1. As a pseudo index in the mode table. Pseudo as you had to search through the table to find the referenced entry. This was replaced by using a pointer to the entry. 2. As a shortcut to compare a combination of horizontal and vertical resolution at the same time. This was replaced by a "(hres<<16) | vres" which is good enough for now and the near future. If vres or hres become greater than 2^16 this might indeed cause problems but this solution allows to split this indexing mess up without the requirement to do even more code changes. This is a big change that will allow more clean ups. It should be a bit faster but that is probably not relevant for normal operation. No regressions expected but as this is a relatively big step heavy testing is appreciated. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-10 15:21:28 -08:00
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
static const struct fb_videomode *get_best_mode(
const struct fb_videomode *modes, int n,
int hres, int vres, int refresh)
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
{
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
const struct fb_videomode *best = NULL;
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
int i;
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
for (i = 0; i < n; i++) {
if (modes[i].xres != hres || modes[i].yres != vres)
continue;
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
if (!best || abs(modes[i].refresh - refresh) <
abs(best->refresh - refresh))
best = &modes[i];
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
}
return best;
}
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
const struct fb_videomode *viafb_get_best_mode(int hres, int vres, int refresh)
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
{
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
return get_best_mode(viafb_modes, ARRAY_SIZE(viafb_modes),
hres, vres, refresh);
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
}
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
const struct fb_videomode *viafb_get_best_rb_mode(int hres, int vres,
int refresh)
viafb: add new funcions to select a single mode This patch introduces 2 new functions for selecting a single mode based on hres, vres and refresh rate and changes some uses to use those. The advantage is that it is less error prone than doing the selection based on refresh rate everywhere and allows replacing the modetable structure. This includes a little change that users may notice: If a refresh rate was given as module parameters but does not exist in the modetable prior to this patch a refresh rate of 60 was assumed and after this patch the closest supported refresh rate to the one provided by the user is used. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2011-08-06 23:04:43 +00:00
{
viafb: modetable conversion This patch converts the modetables used in viafb to - remove the strange thing that sync_end and blanking_end contained the length and not the absolute value - remove hundreds of useless defines - use fb_videomode and not our own definition so modes defined in the subsystem and received via EDID are compatible with ours As the modes are now stored in a flat structure and no longer in a tree like thing the lookup time was increased but as it is a rare event anyway it shouldn't matter. Otherwise the behaviour should be the same as before. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
2012-02-17 07:55:06 +00:00
return get_best_mode(viafb_rb_modes, ARRAY_SIZE(viafb_rb_modes),
hres, vres, refresh);
viafb: split global index up This is the first step to remove an artificial global index that was used in two ways: 1. As a pseudo index in the mode table. Pseudo as you had to search through the table to find the referenced entry. This was replaced by using a pointer to the entry. 2. As a shortcut to compare a combination of horizontal and vertical resolution at the same time. This was replaced by a "(hres<<16) | vres" which is good enough for now and the near future. If vres or hres become greater than 2^16 this might indeed cause problems but this solution allows to split this indexing mess up without the requirement to do even more code changes. This is a big change that will allow more clean ups. It should be a bit faster but that is probably not relevant for normal operation. No regressions expected but as this is a relatively big step heavy testing is appreciated. Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Joseph Chan <JosephChan@via.com.tw> Cc: Scott Fang <ScottFang@viatech.com.cn> Cc: Krzysztof Helt <krzysztof.h1@poczta.fm> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-10 15:21:28 -08:00
}
Reference in a new issue Copy permalink