- 16 Oct, 2007 40 commits
-
-
Krzysztof Helt authored
This patch replaces busy waiting with cpu_relax() call. This makes scrolling faster. Signed-off-by:
Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by:
Antonino Daplas <adaplas@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Jesper Juhl authored
An update to Documentation/fb/00-INDEX is long overdue. This patch adds entries for new files in the directory and removes entries for files that no longer exist. The files are now also sorted alphabetically. Signed-off-by:
Jesper Juhl <jesper.juhl@gmail.com> Signed-off-by:
-
Geert Uytterhoeven authored
ps3fb: Add a `mode' parameter to ps3fb_get_res_table(), as in some cases it should check the full-screen flag of the _new_ video mode instead of the current video mode. This bug caused spurious mode change failures when switching between fullscreen and non-fullscreen modes using fbset, while ps3-video-mode worked fine. Signed-off-by:
Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com> CC: Geoff Levand <geoffrey.levand@am.sony.com> Signed-off-by:
-
Masashi Kimoto authored
Fix the black and white stripes issue with WUXGA monitor via HDMI. It seems that when we use a VESA resolution from boot up, the AV settings driver could not be initialized correctly. In that case we need a dummy packet before VESA setting. [Geert] Refactored to reduce code duplication. Signed-off-by:
-
Pavel Pisa authored
Allow generic frame-buffer code to correctly write texts and blit images for 1, 2 and 4 bit per pixel frame-buffer organizations when pixels in bytes are organized to in opposite order than bytes in long type. Overhead should be reasonable. If option is not selected, than compiler should eliminate completely all overhead. The feature is disabled at compile time if CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set. [adaplas] Convert helper functions to macros if feature is not enabled. Signed-off-by:
Pavel Pisa <pisa@cmp.felk.cvut.cz> Signed-off-by:
-
Antonino A. Daplas authored
There are a few nvidia hardware where the i2c port assignments are reversed, i.e., the 1st i2c port is assigned to the secondary display and the 2nd i2c port to the primary display. In most cases, if only 1 display is attached, or if only 1 flatpanel and 1 analog display is attached, the port reversal is of no consequence. However if 2 flatpanels are attached, it can cause display problems. There is no sane way of determining if the hardware reversed the i2c port assignment, so the simplest fix is to add a boot/module option, "reverse_i2c to explicitly reverse the i2c port. This also restores i2c ordering back to the pre-2.6.22 state. Signed-off-by:
Antonino Daplas <adaplas@gail.com> Signed-off-by:
-
Michal Januszewski authored
Currently, if the refresh rate is not specified, fb_find_mode() returns the first known video mode with the requested resolution, which provides no guarantees wrt the refresh rate. Change this so that the mode with the highest refresh rate is returned when the driver provides a custom video mode database and the monitor limits, and a mode with the safe 60 Hz refresh rate otherwise. Signed-off-by:
Michal Januszewski <spock@gentoo.org> Signed-off-by:
-
Krzysztof Halasa authored
Intel framebuffer now supports interlaced video modes. Signed-off-by:
Krzysztof Halasa <khc@pm.waw.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Ilya Yanok authored
Changed things: 1. vmalloc()/vfree() replaced with rvmalloc()/rvfree() (taken from drivers/media/video/se401.c) 2. mmap method implemented (mostly taken from drivers/media/video/se401.c) 3. smem_start and smem_len fields of struct fb_fix_screeninfo initialized. (smem_start initialized with virtual address, don't know if it is really bad...) [adaplas: sparse warning fix] Signed-off-by: -
Krzysztof Helt authored
This patch: - disallows setting of virtual height above 2048 as it does not work for accelerated copyarea and imageblit - fixes blinking of pm2v cursor by pushing cursor outside the display (x=2047) - fixes hardware cursor position on Permedia 2 chips with panning enabled - enables hardware cursor by default Signed-off-by:
-
Krzysztof Helt authored
This patch adds hardware cursor support for the Permedia 2 chip. Signed-off-by:
-
Krzysztof Helt authored
This patch fixes white spaces, redudant definitions and formating in the pm3fb header file. Signed-off-by:
-
Krzysztof Helt authored
This patch removes: - redundant fields from the cirrusfb_regs structure - one redundant header - fixes two includes ("" to <>) Signed-off-by: -
Krzysztof Helt authored
This patch does some "short-range" code improvements like merging identical switch clauses, replacing conditional branches with calculation of values, merging only once-used functions into place they are called from. Signed-off-by:
-
Krzysztof Helt authored
This patch removes unused or redundant fields from cirrusfb_info structure. Signed-off-by:
-
Krzysztof Helt authored
This patch replaces "typedef enum" with "enum type" constructs. Using the typedef is error for the checkpatch.pl script. Signed-off-by:
-
Krzysztof Helt authored
This patch fixes over 850 errors and warnings pointed out by the checkpatch.pl script. Signed-off-by:
-
Krzysztof Helt authored
This patch changes mode selection to always prefer default mode if possible and always honor type of display set by default mode settings. This patch is required in case the same display modes were defined for different panels (e.g. STN and TFT). Signed-off-by:
-
Krzysztof Helt authored
This patch removes lcdcon1 register field from the s3c2410fb_display as all bits are calculated from other fields. Signed-off-by:
-
Krzysztof Helt authored
This patch adds pixelclock field to the s3c2410fb_display structure and make use of it in the driver. The Bast machine defined 9 modes but pixclock and margin values are defined only for the 640x480 modes so I removed other modes. This patch also fixes wrong display type constant for the SMDK2440 board. Signed-off-by:
-
Krzysztof Helt authored
This patch: - moves more display mode preparations to s3c2410fb_check_var() - reduces number of fields in s3c2410fb_info - removes redundant values setting in s3c2410fb_probe() - removes static mach_info pointer - releases fb_info structure in s3c2410fb_remove() - changes s3c2410fb_init to __init from __devinit - fixes few typos in comments and removes unused includes Signed-off-by:
-
Krzysztof Helt authored
This patch replaces <asm/uaccess.h> with <linux/uaccess.h> after the checkpatch.pl hint. The include of <asm/uaccess.h> is removed if the driver does not use it. Signed-off-by:
-
Ralf Baechle authored
<linux/selection.h> assumes that struct tty_struct has previously been included. If not, this pile of warnings will result: CC [M] drivers/video/console/newport_con.o In file included from drivers/video/console/newport_con.c:18: include/linux/selection.h:16: warning: 'struct tty_struct' declared inside param eter list include/linux/selection.h:16: warning: its scope is only this definition or decl aration, which is probably not what you want include/linux/selection.h:17: warning: 'struct tty_struct' declared inside param eter list include/linux/selection.h:20: warning: 'struct tty_struct' declared inside param eter list Fixed by adding a forward declaration of struct tty_struct. Signed-off-by:
Ralf Baechle <ralf@linux-mips.org> Signed-off-by:
-
Ralf Baechle authored
Akpm's patch "newport_con warning fix" got me to look at the console drivers again and one thing that I noticed was that none of the fonts was using ISO initializers for it's fonts. Signed-off-by:
-
Geert Uytterhoeven authored
atyfb: Move the pseudo palette into the card-specific atyfb_par, so it's no longer shared among multiple cards Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
-
Krzysztof Helt authored
This patch sets correct bits related to the byte ordering of the framebuffer. This was tested on little endian kernel only. The big endian kernel may require different settings. The patch also adds 32 bpp mode which is called 24 bpp by Samsung. One pixel takes 32 bits but only 24 bits are used in this mode. Signed-off-by:
-
Krzysztof Helt authored
This patch adds missing virtual register offsets where appropriate. This fixes crashes in the driver. Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl Signed-off-by:
-
Krzysztof Helt authored
This patch removes unused lcdcon2 and lcdcon3 register value from the s3c2410fb_display structure. Signed-off-by:
-
Krzysztof Helt authored
This patch adds synchronization pulse lenght fields to the s3c2410fb_display structure and makes use of them in the driver. Signed-off-by:
-
Krzysztof Helt authored
This patch makes use of vertical margins fields in the s3c2410fb_display structure. Signed-off-by:
-
Krzysztof Helt authored
This patch adds vertical margins values to all s3c24xx platform. Signed-off-by:
-
Krzysztof Helt authored
This patch removes unused lcdcon3 register from the s3c2410fb_display structure. Signed-off-by:
-
Krzysztof Helt authored
This patch makes use of margins fields in the s3c2410fb_display structure. Signed-off-by:
-
Krzysztof Helt authored
This patch adds margins fields to the s3c2410fb_display structure. It also sets display type and horizontal margins in all platform files that use the s3c2410fb driver. Signed-off-by:
-
Krzysztof Helt authored
This patch adds a new structure to describe and handle more than one panel (display mode) for the s3c2410 framebuffer. This structure is added after the pxafb driver. Signed-off-by:
-
Krzysztof Helt authored
This patch removes redundant fb field from the s3c2410fb_info structure. This breaks circular reference fb_info -> s3c2410fb_info -> fb_info again. Signed-off-by:
-
Krzysztof Helt authored
This patch cleans up the driver a bit. It contains coding style fixes (pointed by Lindent and checkpatch), white space clean ups and few minor code improvements. Signed-off-by:
-
Krzysztof Helt authored
This patch adds hardware cursor support to the pm3fb driver. Signed-off-by:
-
Krzysztof Helt authored
This patch adds hardware cursor support for Permedia 2V chips. The hardware cursor is disabled by default. It does not blink - the same issue is mentioned in the x11 driver. Signed-off-by:
-
Geert Uytterhoeven authored
Looking at the code flow, `int first' in fb_create_modedb() should be static. [adaplas] Better for 'int first' to be moved outside the loop. [akpm@linux-foundation.org: kill stray semicolon] Signed-off-by:
-
