- 07 Nov, 2005 40 commits
-
-
Ville Syrjälä authored
The current init code sets hiprilvl to 0 and maxhipri to 5. According to the specs those values are illegal on both G200 and G400. It also causes distortions on the TV-out at least when CRTC2 is in YUV mode as is the case with DirectFB. This patch resets both values to 0. Signed-off-by:
Ville Syrjälä <syrjala@sci.fi> Signed-off-by:
Petr Vandrovec <petr@vandrovec.name> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Ville Syrjälä authored
No point in spamming the logs with a message about xres rounding. Signed-off-by:
-
Ville Syrjälä authored
Use the CACHEFLUSH register on all chip types. The register is listed in all other specs except 2064W. However I have verified that the register does work on a 2064W despite being marked reserved in the specs. There were no noticeable side effects after writing to the register. Signed-off-by:
-
Ville Syrjälä authored
Add new entries for Mystique AGP with the PCI ID 0x051e. I don't actually have such boards but according to google they do exist. Curiosly X.Org doesn't recognize that PCI ID. And what's even more interesting is that Matrox's own Windows drivers don't recognize it either. After going through about a dozen different versions I did find one older driver that does list this particular ID. It is also listed in the pci.ids file. I'm not sure if non-220 AGP chips exist. I left the chip revision check intact for AGP chips nonetheless. Signed-off-by:
-
Ville Syrjälä authored
wait_event_event_interruptible() uses a private wait queue entry so there's no need for the caller to initialize one. Signed-off-by:
-
Jake Moilanen authored
Here's the PCI ID for the ATI RN50 chip. Signed-off-by:
Jake Moilanen <moilanen@austin.ibm.com> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: "Antonino A. Daplas" <adaplas@hotpop.com> Signed-off-by:
-
Antonino A. Daplas authored
Convert i810fb, nvidiafb and savagefb to use the fb_find_best_display helper when searching for the initial video mode. Signed-off-by:
Antonino Daplas <adaplas@pol.net> Signed-off-by:
-
Antonino A. Daplas authored
Add new helper, fb_find_best_display(), which will search the modelist for the best mode for the attached display. This requires an EDID block that is converted to struct fb_monspecs and a private modelist. The search will be done in this manner: - if 1st detailed timing is preferred, use that - else if dimensions of the display are known, use that to estimate xres and - else if modelist has detailed timings, use the first detailed timing - else, use the very first entry from the modelist Signed-off-by:
-
Antonino A. Daplas authored
Rearrange mode database entries such that preferred timings are entered first, and less preferred timings are entered last. (Detailed, VESA, established/standard). Signed-off-by:
-
Zach Smith authored
I converted the "rl" console font from the kbd utility to be a built-in font for the framebuffer console, and I was wondering if you would be OK with including it. I've generated a font_rl.c file and related minor modifications. I find it's the most visually appealing of the kbd fonts which is why I use it and selected it for conversion. I believe the font is GPL'd. Signed-off-by:
-
Antonino A. Daplas authored
Fix following compile error (From Kernel Bugzilla Bug 5427): include/linux/console_struct.h:53: error: field `vt_mode' has incomplete type Signed-off-by:
-
Antonino A. Daplas authored
- The functions for reading the 1st and 2nd bus are essentially the same, except for the register. Consolidate them all. - According to Nicolas Boichat, there is an undocumented 3rd i2c bus for attaching daughter cards. Add support for this. Signed-off-by:
-
Antonino A. Daplas authored
vesafb occassionally gets the size wrong when setting the mtrr. When X or DRI attempts to set the mtrr, it will fail due to range overlap significantly affecting their performance. Disable mtrr and let the user explicitly enable it with the mtrr:n option. Signed-off-by:
-
Ben Dooks authored
Initialise the .owner field, so that if the driver is built as a module, the system has a link to the owner Signed-off-by:
Ben Dooks <ben-linux@fluff.org> Signed-off-by:
-
Michal Januszewski authored
Currently the fb_find_nearest_mode() function finds a mode with screen resolution closest to that described by the 'var' argument and with some arbitrary refresh rate (eg. in the following sequence of refresh rates: 70 60 53 85 75, 53 is selected). This patch fixes the function so that it looks for the closest mode as far as both resolution and refresh rate are concerned. The function's first argument is changed to fb_videomode so that the refresh rate can be specified by the caller, as fb_var_screeninfo doesn't have any fields that could directly hold this data. Signed-off-by:
Michal Januszewski <spock@gentoo.org> Signed-off-by:
-
Antonino A. Daplas authored
Use do {} while(0) for empty reverse_order() macro. Signed-off-by: -
Antonino A. Daplas authored
Some EDID blocks set the flag "prefer first detailed timing" without providing any detailed timing at all. Clear this flag if the block does not provide detailed timings. Signed-off-by:
-
Peter 'p2' De Schrijver authored
Attached is a small patch which configures the correct memory clock and timings on the Elsa Winner 2000 Office pm2 based card. This is necessary when the card is used on a platform which does not support PC style BIOS initialization. Signed-off-by:
-
Antonino A. Daplas authored
Use MMIO registers instead of banging the VGA IO registers. Signed-off-by:
-
Antonino A. Daplas authored
Reported by: Jean-Philippe Guérard (Bugzilla Bug 1782) "I've tried with video=atyfb:debug and video=atyfb:debug,mode:1280x600, \ nomtrr. In both case, the screen stays black, but seems divided into 4 vertical bands. Some white lines pop up randomly on each vertical band." The problem is a combination of an incorrect xclk plus lack of timing information. The adapter is attached to an LCD device that can do 1280x600 (which is not a standard resolution). The global mode database does not have an entry for it. Fortunately, the Video BIOS contains the complete timing info for this display, however, atyfb is not making use of it. Add support to get the timing information from the BIOS, if available. Signed-off-by:
-
Antonino A. Daplas authored
Fix out-of-bounds bug. The pseudopalette has room only for 16 entries, thus, write only the first 16 entries to the pseudopalette. Signed-off-by:
-
Antonino A. Daplas authored
Remove software clipping from imageblit, fillrect and copyarea. Clipping is not needed because the console layer assures that reads/writes doest not happen beyond the extents of the framebuffer. And software clipping tends to hide bugs, if they do exist. Signed-off-by:
-
Antonino A. Daplas authored
If machine has more than 1 driver installed, and they all drive the same hardware, it's possible that the driver's fb_release() method will attempt to restore the hardware state to the initial state. This will leave the new driver in an undefined state. To prevent this problem, initialize the new driver by calling fb_set_par() when the old driver is released by fbcon. Signed-off-by: Antonino Daplas<adaplas@pol.net> Signed-off-by:
-
Antonino A. Daplas authored
Instead of manually filling up the fields in struct fb_var_screeninfo, use the display_to_var() helper. Signed-off-by:
-
Antonino A. Daplas authored
Lot's of redundant code scattered throughout fbcon.c. Consolidate them all into one function, fbcon_update_softback(). Signed-off-by:
-
Antonino A. Daplas authored
According to Jon Smirl, filling in the field fb_cursor with soft_cursor for drivers that do not support hardware cursors is redundant. The soft_cursor function is usable by all drivers because it is just a wrapper around fb_imageblit. And because soft_cursor is an fbcon-specific hook, the file is moved to the console directory. Thus, drivers that do not support hardware cursors can leave the fb_cursor field blank. For drivers that do, they can fill up this field with their own version. The end result is a smaller code size. And if the framebuffer console is not loaded, module/kernel size is also reduced because the soft_cursor module will also not be loaded. Signed-off-by:
-
Jean Delvare authored
Prevent spurious recompilations of the radeonfb driver when I2C/DDC support is not included and i2c header files are modified. Signed-off-by:
Jean Delvare <khali@linux-fr.org> Acked-by:
-
Scott MacKenzie authored
Add partial support for GMA900 within the i915GM chipset. Signed-off-by:
Scott MacKenzie <irrational@poboxes.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by:
-
Benjamin Herrenschmidt authored
nvidiafb didn't fully hook-up the code it borrowed from X for doing flat panel dithering (this is useful for 6 bits panels). This adds a driver option to force it, and by default "reads" the current value from the chip to get the firmware setting. It significantly improves the quality of images on the iMac G5 I have here (though the X driver doesn't yet "read" the current value and defaults to 0, so you have to add Option "FBDither" "true" to your X config file to get that, I'll try to fix X.org to "read" the default unless specified asap). Signed-off-by:
-
Andrew Morton authored
drivers/video/nvidia/nv_setup.c: In function `NVCommonSetup': drivers/video/nvidia/nv_setup.c:408: warning: statement with no effect drivers/video/nvidia/nv_setup.c:496: warning: statement with no effect drivers/video/nvidia/nv_setup.c:504: warning: statement with no effect Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: "Antonino A. Daplas" <adaplas@hotpop.com> Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch fixes nvifiafb mode setting code to be closer to what the X driver does, which actually makes it work on the 5200FX I have access to. It also fix the routine that gets the EDID from Open Firmware on PPC, it was broken in various ways and would crash at boot. Compared to the patch I posted to linux-fbdev last week, this one just changes a printk to be closer to the other ones in the driver. Signed-off-by:
-
Randy Dunlap authored
Arrange frame buffer menu: - puts all Epson drivers together - removes split of FB_PXA and FB_PXA_PARAMETERS by FB_W100 - results in PXA, W100, Epson, S3C2410, & Virtual FB drivers being presented at the same menu level as all other FB drivers Signed-off-by:
Randy Dunlap <rdunlap@xenotime.net> Acked-by:
-
Randy.Dunlap authored
Frame buffer driver help text changes: - Move S3 Trio next to S3 Savage; - add or clarify help text for several FB drivers; - add help text for FB console; - add help text for bootup logos; Acked-by:
-
NeilBrown authored
If an RPC socket is serving multiple programs, then the pg_authenticate of the first program in the list is called, instead of pg_authenticate for the program to be run. This does not cause a problem with any programs in the current kernel, but could confuse future code. Also set pg_authenticate for nfsd_acl_program incase it ever gets used. Signed-off-by:
Neil Brown <neilb@suse.de> Signed-off-by:
-
NeilBrown authored
There are a couple of tests which could possibly be confused by extremely large numbers appearing in 'xdr' packets. I think the closest to an exploit you could get would be writing random data from a free page into a file - i.e. leak data out of kernel space. I'm fairly sure they cannot be used for remote compromise. Signed-off-by:
-
NeilBrown authored
Provide a file in the NFSD filesystem that allows setting and querying of which version of NFS are being exported. Changes are only allowed while no server is running. Signed-off-by:
Steve Dickson <steved@redhat.com> Signed-off-by:
-
NeilBrown authored
Most files in the nfsd filesystems are transaction files. You write a request, and read a response. For some (e.g. 'threads') it makes sense to just be able to read and get the current value. This functionality did exist but was broken recently when someone modified nfsctl.c without going through the maintainer. This patch fixes the regression. Signed-off-by:
-
NeilBrown authored
This is a somewhat cosmetic fix to keep the SpecFS validation test from complaining. SpecFS want's to try chmod on symlinks, and ext3 and reiser (at least) return ENOTSUPP. Probably both sides are being silly, but it is easiest to simply make it a non-issue and filter out chmod requests on symlinks at the nfsd level. Signed-off-by:
Olaf Kirch <okir@suse.de> Signed-off-by:
J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by:
-
Ronald S. Bultje authored
Fix the warning "Debug: sleeping function called from invalid context at include/asm/semaphore.h:102" that the zr36067 driver emits every time an application using JPEG capture starts up (e.g. mjpegtools' lavrec). The warning is harmless, but clogs up the dmesg output. This was logged as bugzilla #5403. (Thanks to Christian Casteyde for helping me in fixing this long-standing annoyance.) Signed-off-by:
Ronald S. Bultje <rbultje@ronald.bitfreak.net> Signed-off-by:
-
Martin Bachem authored
- cleanup source - remove nonfunctional code parts Signed-off-by:
Karsten Keil <kkeil@suse.de> Signed-off-by:
-
