Ben Dooks [Thu, 24 Jul 2008 04:31:37 +0000 (21:31 -0700)]
sm501: fixup allocation code to be 64bit resource compliant
As pointed out by Andrew Morton, we have a problem when setting the 64bit
resources option. Alter the allocation routines to remove the need to use
the start and end fields, use the proper HEAD_PANEL/HEAD_CRT and update
the comments.
Note, we also fix the bug where we failed to check the size of the
CRT memory allocation.
[akpm@linux-foundation.org: cleanup] Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ben Dooks [Thu, 24 Jul 2008 04:31:36 +0000 (21:31 -0700)]
sm501: restructure init to allow only 1 fb on an SM501
Add the ability to register only one of the two possible main framebuffer
devices on the SM501 by passing platform data for only the framebuffer
that you are interested in having.
As a side note, we update the init sequence to commonise the code that is
executed twice, and fix a pair of missing frees that we didn't do on
framebuffer exit, such as freeing the fb's cmap.
Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nicolas Ferre [Thu, 24 Jul 2008 04:31:34 +0000 (21:31 -0700)]
atmel_lcdfb: avoid division by zero
Avoid division by zero in atmel_lcdfb_check_var() function.
If pixclock is not specified while passing a var structure in
the check_var() funtion, a division by zero occurs (when
translating pixclock to KHz).
This patch adds a checking of this value and try to choose a
video mode in the modelist.
The mode found in the probe function in added to the modelist.
[akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: Haavard Skinnemoen <hskinnemoen@atmel.com> Cc: Andrew Victor <linux@maxim.org.za> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:34 +0000 (21:31 -0700)]
skeletonfb: update to correct platform driver usage
It updates skeletonfb to new platform driver API. The skeletonfb is
templates for creating new drivers.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Akinobu Mita [Thu, 24 Jul 2008 04:31:33 +0000 (21:31 -0700)]
aty: use memory_read_from_buffer()
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:32 +0000 (21:31 -0700)]
atyfb: fix a cast
The argument to iounmap() is void __iomem *. Fix the cast.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:32 +0000 (21:31 -0700)]
atyfb: report probe errors
Properly propagate errors to the probe function.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:31 +0000 (21:31 -0700)]
atyfb: use a PCI device ID table
Convert atyfb to use a PCI device ID table.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:30 +0000 (21:31 -0700)]
atyfb: correct_chipset() can fail
Atari probe code relies on correct_chipset() failing if the device is not
a mach64 GX/CX. aty_chips[] array would be indexed with -1 in that case.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Olaf Hering [Thu, 24 Jul 2008 04:31:29 +0000 (21:31 -0700)]
atyfb: remove dead code
Remove dead code. This will slightly change the behaviour of the driver
on systems that support backlight control. Previously they would just
turn the backlight off using the backlight control but now the generic LCD
code will also turn off the LCD using the POWER_MANAGEMENT register.
Signed-off-by: Olaf Hering <olaf@aepfle.de> Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:28 +0000 (21:31 -0700)]
fbdev: xoffset, yoffset and yres are unsigned
The xoffset, yoffset and yres members of fb_var_screeninfo are __u32.
Make them unsigned in the code as well.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Ville Syrjala [Thu, 24 Jul 2008 04:31:27 +0000 (21:31 -0700)]
fbdev: width and height are unsigned
The width and height members of fb_var_screeninfo are __u32. The code
initializes them to -1 which seems wrong, and 0 seems like an equally good
default value.
Signed-off-by: Ville Syrjala <syrjala@sci.fi> Cc: "Antonino A. Daplas" <adaplas@pol.net> 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>
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: Anton Vorontsov <avorontsov@ru.mvista.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Frans Pop [Thu, 24 Jul 2008 04:31:26 +0000 (21:31 -0700)]
vfb: only enable if explicitly requested when compiled in
The Kconfig help for the vfb driver says:
Do NOT enable it for normal systems! To protect the innocent, it
has to be enabled explicitly at boot time using the kernel option
`video=vfb:'.
This change lets the code match the description.
Support for vfb:disable is kept for backwards compatibility; vfb:off works
because it is tested at a higher level.
Note: any undefined option (e.g. vfb:enable) will also enable this driver.
The relevant code has been unchanged since before the migration to
git (2.6.12).
This patch fixes bugzilla #9310 and was the root cause behind
http://lkml.org/lkml/2008/5/31/220.
Signed-off-by: Frans Pop <elendil@planet.nl> Cc: Antonino A. Daplas <adaplas@gmail.com> Acked-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Magnus Damm [Thu, 24 Jul 2008 04:31:24 +0000 (21:31 -0700)]
fbdev: SuperH Mobile LCDC Driver
This is the SuperH Mobile LCDC frame buffer driver V2, adding support for
the LCDC block found in SuperH Mobile processors. The hardware supports
up to two LCD panels per LCDC block, and both RGB and SYS interfaces can
be used to hook up LCD panels/modules.
The device driver is a regular platform driver, so LCD configuration and
board specific hooks are passed to the driver using platform data. LCD
modules using SYS interface often require special configuration using the
SYS bus, and to solve this cleanly the driver provides SYS interface
operations to the board code.
Tested on sh7723 and sh7722 processors with a SYS16A QVGA panel and WVGA
panels using RGB16 and RGB18 interfaces.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Acked-by: Paul Mundt <lethal@linux-sh.org> Reviewed-by: Krzysztof Helt <krzysztof.h1@poczta.fm> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:24 +0000 (21:31 -0700)]
tdfxfb: remove ypan checks done by a higher layer
These checks and assignments are done by a higher layer so remove them
from the driver.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:23 +0000 (21:31 -0700)]
vga16fb: source code improvement
Use constants and functions from the vga.h file. Also add module
description.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:22 +0000 (21:31 -0700)]
tdfxfb: add mode_option module parameter
Small step toward unification of mode setting parameter. This is required
to fix the Bugzilla's bug 9847
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:22 +0000 (21:31 -0700)]
tridentfb: documentation update
Make the tridentfb documentation closer to current state of the tridentfb
driver. Fix also some formatting.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:21 +0000 (21:31 -0700)]
uvesafb: change mode parameter to mode_option
Make more drivers use the "mode_option" parameter. This one is quite new
so drop the old "mode" parameter before someone starts using it seriously.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:21 +0000 (21:31 -0700)]
fbcon: make logo_height a local variable
Make logo_height variable local in the only function it is used.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nicolas Ferre [Thu, 24 Jul 2008 04:31:20 +0000 (21:31 -0700)]
atmel_lcdfb: FIFO underflow management
Manage atmel_lcdfb FIFO underflow
Resetting the LCD and DMA allows to fix screen shifting after a FIFO
underflow. It follows reset sequence from errata "LCD Screen Shifting
After a Reset".
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: Haavard Skinnemoen <hskinnemoen@atmel.com> Cc: Andrew Victor <linux@maxim.org.za> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Tony Breeds [Thu, 24 Jul 2008 04:31:16 +0000 (21:31 -0700)]
drivers/video/aty/radeon_base.c: notify user if sysfs_create_bin_file() failed
Current kernel builds warn about:
drivers/video/aty/radeon_base.c: In function 'radeonfb_pci_register':
drivers/video/aty/radeon_base.c:2334: warning: ignoring return value of 'sysfs_create_bin_file', declared with attribute warn_unused_result
drivers/video/aty/radeon_base.c:2336: warning: ignoring return value of 'sysfs_create_bin_file', declared with attribute warn_unused_result
Do minimal checking of these functions and issue a warning if either
fails. They don't seem to be critical..
[akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Tony Breeds <tony@bakeyournoodle.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The semaphore ctrlr_sem is used as a mutex. Convert it to the mutex API.
Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net> Cc: Daniel Mack <daniel@caiaq.de> Cc: Eric Miao <eric.miao@marvell.com> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Roland Kletzing [Thu, 24 Jul 2008 04:31:10 +0000 (21:31 -0700)]
mdacon messing up default vc's - set default to vc13-16 again
mdacon incorrectly detects MDA hardware on systems without such graphics card.
One may load this module by chance, for example when doing some systematical
module-testing, and if there is no Monochrome Display Adapter attached ,
module init renders vc1-16 completely unusable.
I and others have run into this more than once. see [Bug 224522 - modprobe
mdacon freezes machine -> https://bugzilla.novell.com/show_bug.cgi?id=224522 ]
for example
Apparently proper MDA detection seems to be broken for a long time - seems to
be related to those #ifdef TEST_MDA_B statements added by Edward Betts.
this commit back in 2002 made things even worse :
http://git.kernel.org/?p=linux/kernel/git/tglx/history.git;a=commit;h=c72757b49c88914433244757fb4967fc63546685
It changed default vc allocation from 13-16 to 1-16 for no apparent reason
(!?) , and with that (and without X), mdacon grabs the vc you`re currently
sitting on and locks you out.
this is from Kconfig :
>config MDA_CONSOLE
> depends on !M68K && !PARISC && ISA
> tristate "MDA text console (dual-headed) (EXPERIMENTAL)"
> ---help---
> Say Y here if you have an old MDA or monochrome Hercules graphics
> adapter in your system acting as a second head ( = video card). You
> will then be able to use two monitors with your Linux system. Do not
> say Y here if your MDA card is the primary card in your system; the
> normal VGA driver will handle it.
As we can see mdacon is just meant as an additional driver for dual-head
setup, and since kernel 2.4.36 still defaults to vc13-16 , setting the default
back to that value again shouldn`t do any harm.
Hereby i'm reverting that change, setting default back to to vc13-16 again.
Besides the fact that mdacon may be rarely or never be used these days and
could perhaps put to trash anyway (pre-dinosaur hardware!), indeed this is not
a real solution, but at least it removes the unfortunate side-effect of
messing up the vc you`re working on.
Signed-off-by: Roland Kletzing <devzero@web.de> Cc: James Simmons <jsimmons@infradead.org> Cc: "Antonino A. Daplas" <adaplas@pol.net> Cc: Tim Schmielau <tim@physik3.uni-rostock.de> Cc: Jan Engelhardt <jengelh@gmx.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:09 +0000 (21:31 -0700)]
tridentfb: remove warning message that cyblafb driver should be used
The tridentfb driver should handle now all chipsets handled by the cyblafb
driver. Remove the message which claims that support will be removed.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:08 +0000 (21:31 -0700)]
tridentfb: add imageblit acceleration for Blade3D family
Add imageblit acceleration for the Blade3D family of cores. The code is
based on code from the cyblafb driver.
It is a step toward assimilating back the cyblafb driver into the
tridentfb driver. The cyblafb driver handles a subfamily of the Trident
Blade3d cores.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:08 +0000 (21:31 -0700)]
tridentfb: Blade3D clock fixes
This patch fixes following problems:
- does not allow the m parameter to reach 0 as
it locks the graphics core (power cycle needed)
- for the newer chips (with new clock registers)
does not allow of n / m ratio below 4 as it gives
unstable image on the Blade3D core
- extend shift parameter (k) range to 2 for the newer
chips to cope with the n /m >= 4 limit at low resolution
(bandwidth) modes
- prefer modes with higher n / m ratio (higher k values)
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:07 +0000 (21:31 -0700)]
tridentfb: y-panning fixes
The Trident cards uses only 20-bit address of screen start in double
words. This allows addressing for only 4MB of video memory so check this.
Also remove some redundant checks and assignments.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:06 +0000 (21:31 -0700)]
tridentfb: fix 224 color logo at 8 bpp
Fix depth setting for 8 bpp mode. The nice 224 color logo is not
displayed in 8 bpp depth without this fix.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:06 +0000 (21:31 -0700)]
tridentfb: fix console freeze when switching from X11
This patch fixes two problems when acceleration is enabled:
- console switch from the Xorg locks up the computer
because the Xorg code locks some registers and disables
the mmio mode, so reenable these in the
tridentfb_set_par() and enable_mmio()
- blacklist the Image975 chipset from setting PCI burst
mode. This helps with random lock ups of the
framebuffer on this chip. The same fix is probably
needed for the Xorg as well.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:05 +0000 (21:31 -0700)]
tridentfb: source code improvements
This patch contains general source code improvments:
- more simple functions are inline
- removes some meaningless output and the VERSION
string as it is no use
- eng_par is moved into the tridentfb_par
- removed small section of code for CyberBladeXPAi1
which is maybe right for only one resolution
and refresh rate and is probably redundant now
- other minor improvements
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:04 +0000 (21:31 -0700)]
tridentfb: acceleration constants change
This patch replaces deprecated constant FB_ACCELF_TEXT with
FBINFO_HWACCEL_DISABLED and adds constants for Trident families of
accelerators.
The FBINFO_HWACCEL_DISABLED is correctly used so noaccel parameter works
now.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:04 +0000 (21:31 -0700)]
tridentfb: various pixclock and timing improvements
This patch fixes few issues related to timings and pixclock generation:
- disallow the pixclocks with numerator lower than
double denominator. This fixes display instability
for some modes.
- choose the pixelclock with the highest
numerator and denominator values. This improve
image quality and fixes display instability
for some modes.
- make interlaced modes work.
- set synchronization pulses polarization
correctly.
- horizontal synchronization timing are now
the same as generated by X.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:03 +0000 (21:31 -0700)]
tridentfb: acceleration bug fixes
This patch fixes two problems when acceleration is enabled:
- bit for bitblt direction is corrected
so scrolling down works as expected on 3DImage chips
- initialization of acceleration is done later
this helps with initial console malfuntion (on Blade3D
chips) well documented here:
http://marc.info/?l=linux-fbdev-users&m=111386953124478&w=2
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:02 +0000 (21:31 -0700)]
tridentfb: acceleration code improvements
This patch brings various acceleration improvements:
- set copyarea/fillrect for non-accelerated framebuffer (fix)
- remove 15 bpp depth handling to simplify code as it hardly
works (15 bpp handling was obviously missing in some switches)
- add fb_sync call and move waiting before accelerated function
to make acceleration more asynchronous to cpu (few % of speed
improvement)
- add cpu_relax() call in waiting loops
- make longer register names and name more registers
- move registers' definition to header
- general code improvements (shortening, simplifying)
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:01 +0000 (21:31 -0700)]
tridentfb: add acceleration for TGUI families
This patch adds acceleration for TGUI 9440 and 96xx chips. These chips
requires line length to be power of 2, so this is also changed.
It also moves the troubling enable_mmio() function to its final
destination.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:01 +0000 (21:31 -0700)]
tridentfb: fix hi-color modes for TGUI 9440
The TGUI 9440 requires doubling clock for 16bpp (hi-color) modes.
The patch also moves back enable_mmio() call to the right position.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:00 +0000 (21:31 -0700)]
tridentfb: preserve memory type settings
Do not overwrite bits which contain memory type settings. It removes
noise pixels ("snow") on Blade3D and 3DImage chips.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:31:00 +0000 (21:31 -0700)]
tridentfb: improve check_var function
Do some additional checks (like pixelclock versus ramdac speed) to
eliminate modes which do not work.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:59 +0000 (21:30 -0700)]
tridentfb: fix unitialized pseudo_palette
Initialize the pseudo_palette pointer properly. This fixes crash when
16bpp or 32bpp mode is selected.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:58 +0000 (21:30 -0700)]
tridentfb: add TGUI 9440 support
Add support for TGUI 9440 chip.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:58 +0000 (21:30 -0700)]
tridentfb: improved register values on TGUI 9680
Improved values for some registers after Xorg Trident driver. The main
problem was that values set by BIOS have been ignored.
This patch completely remove random pixels ("snow") on the TGUI 9680 and
9440 (not supported yet by the driver). It does not help with the "snow"
on 3DImage and Blade3D cards.
There is also small improvement in timing calculations (hblank start and
vblank start)
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:57 +0000 (21:30 -0700)]
tridentfb: improve probe function
Add missing release of allocated fb_info structure and move enable_mmio() to
fix error path.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:56 +0000 (21:30 -0700)]
tridentfb: fix clock settings for older Trident 96XX chips
The Xorg code shows that Trident models 9660, 9680 and 9682 require a
different clock setting method. Add the second clock setting method for older
models.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:56 +0000 (21:30 -0700)]
tridentfb: use mmio access for clock setting
Use the mmio outb function instead of direct one. The mmio registers are
already mapped (in the probe function).
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:55 +0000 (21:30 -0700)]
tridentfb: fix timing calculations
Fix broken timings calculations. This patch helps with following
problems:
- no left part of screen visible (up to half of the screen)
- monitor's frequencies are not the ones intended for selected modes
- if mode with resoultion y > 1024 is selected at least once then
all modes with y < 1024 are "out of sync" (no display)
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:54 +0000 (21:30 -0700)]
tridentfb: make use of functions and constants from the vga.h
Make use of functions and constants from the vga.h header to compact the code
and make it more readable.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:54 +0000 (21:30 -0700)]
tridentfb: move global acceleration hooks into structure
This patch moves acceleration hooks into the tridentfb_par structure and
removes global hooks.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:53 +0000 (21:30 -0700)]
tridentfb: convert is_blade and is_xp macros into functions
This patch converts the is_blade() and is_xp() macros into local functions.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:53 +0000 (21:30 -0700)]
tridentfb: move global flat panel variable into structure
This patch moves flat panel indicator into tridentfb_par structure and removes
related global variables and macros.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:52 +0000 (21:30 -0700)]
tridentfb: move global chip_id into structure
This patch moves the chip_id into tridentfb_par structure and removes global
chip_id related constants.
It also bumps version of the driver to 0.7.9
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:51 +0000 (21:30 -0700)]
tridentfb: move global pseudo palette into structure
This patch moves pseudo palette int tridentfb_par structure and removes global
default_var.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:51 +0000 (21:30 -0700)]
tridentfb: convert fb_info into allocated one
This patch converts fb_info structure from global variable to allocatable one.
The global default_par is moved into function variable.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Krzysztof Helt [Thu, 24 Jul 2008 04:30:50 +0000 (21:30 -0700)]
tridentfb: replace macros with functions
This patch replaces macros with static functions and puts tridentfb_par
pointer as the first argument of these functions. These is a step toward
multihead support.
Additionally, bogus TRIDENT_MMIO define is removed as the driver supports
graphics cards only through the mmio mode.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Basic FB driver for the carmine chip. The driver registers two FB devices for
the two possible screens. The DRAM settings can be be switched via Kconfig
(between eval board and custom).
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:47 +0000 (21:30 -0700)]
rtc-cmos: avoid spurious irqs
This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus
RTC update IRQs reported) for rtc-cmos, in two ways:
- When HPET is stealing the IRQs, use the first IRQ to grab
the seconds counter which will be monitored (instead of
using whatever was previously in that memory);
- In sane IRQ handling modes, scrub out old IRQ status before
enabling IRQs.
That latter is done by tightening up IRQ handling for rtc-cmos everywhere,
also ensuring that when HPET is used it's the only thing triggering IRQ
reports to userspace; net object shrink.
Also fix a bogus HPET message related to its RTC emulation.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Report-by: W Unruh <unruh@physics.ubc.ca> Cc: Andrew Victor <avictor.za@gmail.com> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:46 +0000 (21:30 -0700)]
rtc-at91rm9200: avoid spurious irqs
This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus
RTC update IRQs reported) for rtc-at91rm9200 by scrubbing old IRQ status
before enabling IRQs.
It also removes nonfunctional periodic IRQ support from this driver;
only update IRQs are reported, or provided by the hardware.
I suspect some other RTCs probably have versions of #11112; it's easy to
overlook, since most non-RTC drivers don't care about spurious IRQs:
they're not reported to userspace.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Report-by: W Unruh <unruh@physics.ubc.ca> Cc: Andrew Victor <avictor.za@gmail.com> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ben Dooks [Thu, 24 Jul 2008 04:30:45 +0000 (21:30 -0700)]
rtc: rtc-s3c: update IRQ handling
The rtc-s3c.c driver has been using its own ioctl() handling to deal with
alarm and periodic interrupts to handle what should now be done with the
rtc core code.
Change to using the .irq_set_freq and .irq_set_state driver entries and
remove the .ioctl handling.
Signed-off-by: Ben Dooks <ben-linux@fluff.org> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:43 +0000 (21:30 -0700)]
rtc-cmos: improve HPET IRQ glue
Resolve http://bugzilla.kernel.org/show_bug.cgi?id=11051 and other bugs
related to the way the HPET glue code in rtc-cmos was incomplete and
inconsistent:
* Switch the approach so that the basic driver code flow isn't
changed by having HPET ... instead, just have HPET shadow the
RTC_CONTROL irq enables and RTC_FREQ_SELECT data. It's only
coping with IRQ thievery, after all.
* Do that consistently (!!) to avoid problems when the HPET code
is out of sync with the real RTC intent. Examples include:
- cmos_procfs(), which now reports correct data
- cmos_irq_set_state() ... also removing the previous PIE_{ON,OFF}
ioctl support so only one code path manages "periodic" IRQs
- cmos_do_shutdown() ... currently a "just in case" change.
- cmos_suspend() and cmos_resume() ... also handling a bug that
was specific to HPET's IRQ thievery, where the alarm wasn't
disabled after waking the system
* Always call that HPET code under the RTC spinlock (it doesn't do
its own locking)
Also clean up the HPET glue:
* Add some comments explaining what's going on.
* Switch to having just one #ifdef for the HPET glue, and inline
functions (not #defines) to avoid some compiler warnings.
* Have the probe message also report when HPET IRQs are involved
This still leaves various holes in the HPET glue, like the emulated update
IRQs being out of sync with the RTC, alarms never using day or month
matches, and many extra IRQs (at 64 Hz).
[akpm@linux-foundation.org: fix build] Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Tomas Janousek <tomi@nomi.cz> Cc: Bernhard Walle <bwalle@suse.de> Cc: Carlos R. Mafra <crmafra@ift.unesp.br> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:37 +0000 (21:30 -0700)]
rtc: BCD codeshrink
This updates <linux/bcd.h> to define the key routines as constant
functions, which the macros will then call. Newer code can now call
bcd2bin() instead of SCREAMING BCD2BIN() TO THE FOUR WINDS.
This lets each driver shrink their codespace by using N function calls to
a single (global) copy of those routines, instead of N inlined copies of
these functions per driver.
These routines aren't used in speed-critical code. Almost all callers are
in the RTC framework. Typical per-driver savings is near 300 bytes.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Adrian Bunk <bunk@kernel.org> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:36 +0000 (21:30 -0700)]
rtc: ds1305/ds1306 driver
Support the Dallas/Maxim DS1305 and DS1306 RTC chips. These use SPI, and
support alarms, NVRAM, and a trickle charger for use when their backup
power supply is a supercap or rechargeable cell.
This basic driver doesn't yet support suspend/resume or wakealarms.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Kim B. Heino [Thu, 24 Jul 2008 04:30:34 +0000 (21:30 -0700)]
rtc: add support for ST M41T94 SPI RTC
This patch adds kernel driver for M41T94 RTC chip connected via SPI.
I've tested it on two different AT91-based hardwares.
This is third revision of the patch: some comments made by
Alessandro Zummo fixed.
Revision two added support for century bit and fixes.
Signed-off-by: Kim B. Heino <Kim.Heino@bluegiga.com> Signed-off-by: Alessandro Zummo <a.zummo@towertech.it> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Brownell [Thu, 24 Jul 2008 04:30:33 +0000 (21:30 -0700)]
rtc: remove BKL for ioctl()
Remove implicit use of BKL in ioctl() from the RTC framework.
Instead, the rtc->ops_lock is used. That's the same lock that already
protects the RTC operations when they're issued through the exported
rtc_*() calls in drivers/rtc/interface.c ... making this a bugfix, not
just a cleanup, since both ioctl calls and set_alarm() need to update IRQ
enable flags and that implies a common lock (which RTC drivers as a rule
do not provide on their own).
A new comment at the declaration of "struct rtc_class_ops" summarizes
current locking rules. It's not clear to me that the exceptions listed
there should exist ... if not, those are pre-existing problems which can
be fixed in a patch that doesn't relate to BKL removal.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Jonathan Corbet <corbet@lwn.net> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Replace printk(KERN_INFO ...) calls with appropriate pr_info(...)
equivalents.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Cc: Alessandro Zummo <a.zummo@towertech.it> Cc: Alexander Bigga <ab@mycable.de> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
rtc: m41t80: sort header inclusions for readability
Sort the header inclusions for readability. No functional changes.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Cc: Alessandro Zummo <a.zummo@towertech.it> Cc: Alexander Bigga <ab@mycable.de> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:29 +0000 (21:30 -0700)]
autofs4: remove unused ioctls
The ioctls AUTOFS_IOC_TOGGLEREGHOST and AUTOFS_IOC_ASKREGHOST were added
several years ago but what they were intended for has never been
implemented (as far as I'm aware noone uses them) so remove them.
Signed-off-by: Ian Kent <raven@themaw.net> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:28 +0000 (21:30 -0700)]
autofs4: reorganize expire pending wait function calls
This patch re-orgnirzes the checking for and waiting on active expires and
elininates redundant checks.
Signed-off-by: Ian Kent <raven@themaw.net> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:28 +0000 (21:30 -0700)]
autofs4: fix direct mount pending expire race - correction
Appologies, somehow I seem to have sent an out dated version of this
patch. Here is an additional patch that brings the patch up to date.
Signed-off-by: Ian Kent <raven@themaw.net> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:27 +0000 (21:30 -0700)]
autofs4: fix direct mount pending expire race
For direct and offset type mounts that are covered by another mount we
cannot check the AUTOFS_INF_EXPIRING flag during a path walk which leads
to lookups walking into an expiring mount while it is being expired.
For example, for the direct multi-mount map entry with a couple of
offsets:
an autofs trigger mount is mounted on /race/mm1 and when accessed it is
over mounted and trigger mounts made for /race/mm1/om1 and /race/mm1/om2.
So it isn't possible for path walks to see the expiring flag at all and
they happily walk into the file system while it is expiring.
When expiring these mounts follow_down() must stop at the autofs mount and
all processes must block in the ->follow_link() method (except the daemon)
until the expire is complete. This is done by decrementing the d_mounted
field of the autofs trigger mount root dentry until the expire is
completed. In ->follow_link() all processes wait on the expire and the
mount following is completed for the daemon until the expire is complete.
Signed-off-by: Ian Kent <raven@themaw.net> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:26 +0000 (21:30 -0700)]
autofs4: fix indirect mount pending expire race
The selection of a dentry for expiration and the setting of the
AUTOFS_INF_EXPIRING flag isn't done atomically which can lead to lookups
walking into an expiring mount.
What happens is that an expire is initiated by the daemon and a dentry is
selected for expire but, since there is no lock held between the selection
and setting of the expiring flag, a process may find the flag clear and
continue walking into the mount tree at the same time the daemon attempts
the expire it.
Signed-off-by: Ian Kent <raven@themaw.net> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:25 +0000 (21:30 -0700)]
autofs4: fix pending checks
There are two cases for which a dentry that has a pending mount request
does not wait for completion. One is via autofs4_revalidate() and the
other via autofs4_follow_link().
In revalidate, after the mount point directory is created, but before the
mount is done, the check in try_to_fill_dentry() can can fail to send the
dentry to the wait queue since the dentry is positive and the lookup flags
may contain only LOOKUP_FOLLOW. Although we don't trigger a mount for the
LOOKUP_FOLLOW flag, if ther's one pending we might as well wait and use
the mounted dentry for the lookup.
In autofs4_follow_link() the dentry is not checked to see if it is pending
so it may fail to call try_to_fill_dentry() and not wait for mount
completion.
A dentry that is pending must always be sent to the wait queue.
Signed-off-by: Ian Kent <raven@themaw.net> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:24 +0000 (21:30 -0700)]
autofs4: cleanup redundant readir code
The mount triggering functionality of readdir and related functions is no
longer used (and is quite broken as well). The unused portions have been
removed.
Signed-off-by: Ian Kent <raven@themaw.net> Reviewed-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:23 +0000 (21:30 -0700)]
autofs4: indirect dentry must almost always be positive
We have been seeing mount requests comming to the automount daemon for
keys of the form "<map key>/<non key directory>" which are lookups for
invalid map keys. But we can check for this in the kernel module and
return a fail immediately, without having to send a request to the daemon.
It is possible to recognise these requests are invalid based on whether
the request dentry is negative and its relation to the autofs file system
root.
For example, given the indirect multi-mount map entry:
For a request to mount idm1, IS_ROOT((idm1)->d_parent) will be always be
true and the dentry may be negative. But directories idm1/mm1 and
idm1/mm2 will always be created as part of the mount request for idm1. So
any mount request within idm1 itself must have a positive dentry otherwise
the map key is invalid.
In version 4 these multi-mount entries are all mounted and umounted as a
single request and in version 5 the directories idm1/mm1 and idm1/mm2 are
created and an autofs fs mounted on them to act as a mount trigger so the
above is also true.
This also holds true for the autofs version 4 pseudo direct mount feature.
When this feature is used without the "--ghost" option automount(8) will
create internal submounts as we go down the map key paths which are
essentially normal indirect mounts for which the above holds. If the
"--ghost" option is given the directories for map keys are created at
daemon startup so valid map entries correspond to postive dentries in the
autofs fs.
autofs version 5 direct mount maps are similar except that the IS_ROOT
check is not needed. This has been addressed in a previous patch tittled
"autofs4 - detect invalid direct mount requests".
For example, given the direct multi-mount map entry:
An autofs fs is mounted on /test/dm1 as a trigger mount and when a mount
is triggered for /test/dm1, the multi-mount offset directories
/test/dm1/mm1 and /test/dm1/mm2 are created and an autofs fs is mounted on
them to act as mount triggers. So valid direct mount requests must always
have a positive dentry if they correspond to a valid map entry.
Signed-off-by: Ian Kent <raven@themaw.net> Acked-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:22 +0000 (21:30 -0700)]
autofs4: detect invalid direct mount requests
autofs v5 direct and offset mounts within an autofs filesystem are
triggered by existing autofs triger mounts so the mount point dentry must
be positive. If the mount point dentry is negative then the trigger
doesn't exist so we can return fail immediately.
Signed-off-by: Ian Kent <raven@themaw.net> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:21 +0000 (21:30 -0700)]
autofs4: fix waitq memory leak
If an autofs mount becomes catatonic before autofs4_wait_release() is
called the wait queue counter will not be decremented down to zero and the
entry will never be freed. There are also races decrementing the wait
counter in the wait release function. To deal with this the counter needs
to be updated while holding the wait queue mutex and waiters need to be
woken up unconditionally when the wait is removed from the queue to ensure
we eventually free the wait.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:20 +0000 (21:30 -0700)]
autofs4: check kernel communication pipe is valid for write
It is possible for an autofs mount to become catatonic (and for the daemon
communication pipe to become NULL) after a wait has been initiallized but
before the request has been sent to the daemon. We need to check for this
before sending the request packet.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:19 +0000 (21:30 -0700)]
autofs4: add missing kfree
It see that the patch tittled "autofs4 - fix pending mount race" is
missing a change that I had recently made.
It's missing a kfree for the case mutex_lock_interruptible() fails
to aquire the wait queue mutex.
Signed-off-by: Ian Kent <raven@themaw.net> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:19 +0000 (21:30 -0700)]
autofs4: fix pending mount race
Close a race between a pending mount that is about to finish and a new
lookup for the same directory.
Process P1 triggers a mount of directory foo. It sets
DCACHE_AUTOFS_PENDING in the ->lookup routine, creates a waitq entry for
'foo', and calls out to the daemon to perform the mount. The autofs
daemon will then create the directory 'foo', using a new dentry that will
be hashed in the dcache.
Before the mount completes, another process, P2, tries to walk into the
'foo' directory. The vfs path walking code finds an entry for 'foo' and
calls the revalidate method. Revalidate finds that the entry is not
PENDING (because PENDING was never set on the dentry created by the
mkdir), but it does find the directory is empty. Revalidate calls
try_to_fill_dentry, which sets the PENDING flag and then calls into the
autofs4 wait code to trigger or wait for a mount of 'foo'. The wait code
finds the entry for 'foo' and goes to sleep waiting for the completion of
the mount.
Yet another process, P3, tries to walk into the 'foo' directory. This
process again finds a dentry in the dcache for 'foo', and calls into the
autofs revalidate code.
The revalidate code finds that the PENDING flag is set, and so calls
try_to_fill_dentry.
a) try_to_fill_dentry sets the PENDING flag redundantly for this
dentry, then calls into the autofs4 wait code.
b) the autofs4 wait code takes the waitq mutex and searches for an
entry for 'foo'
Between a and b, P1 is woken up because the mount completed. P1 takes the
wait queue mutex, clears the PENDING flag from the dentry, and removes the
waitqueue entry for 'foo' from the list.
When it releases the waitq mutex, P3 (eventually) acquires it. At this
time, it looks for an existing waitq for 'foo', finds none, and so creates
a new one and calls out to the daemon to mount the 'foo' directory.
Now, the reason that three processes are required to trigger this race is
that, because the PENDING flag is not set on the dentry created by mkdir,
the window for the race would be way to slim for it to ever occur.
Basically, between the testing of d_mountpoint(dentry) and the taking of
the waitq mutex, the mount would have to complete and the daemon would
have to be woken up, and that in turn would have to wake up P1. This is
simply impossible. Add the third process, though, and it becomes slightly
more likely.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:17 +0000 (21:30 -0700)]
autofs4: fix waitq locking
The autofs4_catatonic_mode() function accesses the wait queue without any
locking but can be called at any time. This could lead to a possible
double free of the name field of the wait and a double fput of the daemon
communication pipe or an fput of a NULL file pointer.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jeff Moyer [Thu, 24 Jul 2008 04:30:16 +0000 (21:30 -0700)]
autofs4: use struct qstr in waitq.c
The autofs_wait_queue already contains all of the fields of the
struct qstr, so change it into a qstr.
This patch, from Jeff Moyer, has been modified a liitle by myself.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:15 +0000 (21:30 -0700)]
autofs4: use lookup intent flags to trigger mounts
When an open(2) call is made on an autofs mount point directory that
already exists and the O_DIRECTORY flag is not used the needed mount
callback to the daemon is not done. This leads to the path walk
continuing resulting in a callback to the daemon with an incorrect
key. open(2) is called without O_DIRECTORY by the "find" utility but
this should be handled properly anyway.
This happens because autofs needs to use the lookup flags to decide
when to callback to the daemon to perform a mount to prevent mount
storms. For example, an autofs indirect mount map that has the "browse"
option will have the mount point directories are pre-created and the
stat(2) call made by a color ls against each directory will cause all
these directories to be mounted. It is unfortunate we need to resort
to this but mount maps can be quite large. Additionally, if a user
manually umounts an autofs indirect mount the directory isn't removed
which also leads to this situation.
To resolve this autofs needs to use the lookup intent flags to enable
it to make this decision. This patch adds this check and triggers a
call back if any of the lookup intent flags are set as all these calls
warrant a mount attempt be requested.
I know that external VFS code which uses the lookup flags is something
that the VFS would like to eliminate but I have no choice as I can't
see any other way to do this. A VFS dentry or inode operation callback
which returns the lookup "type" (requires a definition) would be
sufficient. But this change is needed now and I'm not aware of the form
that coming VFS changes will take so I'm not willing to propose anything
along these lines.
If anyone can provide an alternate method I would be happy to use it.
[akpm@linux-foundation.org: fix build for concurrent VFS changes] Signed-off-by: Ian Kent <raven@themaw.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:14 +0000 (21:30 -0700)]
autofs4: don't release directory mutex if called in oz_mode
Since we now delay hashing of dentrys until the ->mkdir() call, droping
and re-taking the directory mutex within the ->lookup() function when we
are being called by user space is not needed. This can lead to a race
when other processes are attempting to access the same directory during
mount point directory creation.
In this case we need to hang onto the mutex to ensure we don't get user
processes trying to create a mount request for a newly created dentry
after the mount point entry has already been created. This ensures that
when we need to check a dentry passed to autofs4_wait(), if it is hashed,
it is always the mount point dentry and not a new dentry created by
another lookup during directory creation.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:13 +0000 (21:30 -0700)]
autofs4: fix symlink name allocation
The length of the symlink name has been moved but it needs to be set
before allocating space for it in the dentry info struct. This corrects a
mistake in a recent patch.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:12 +0000 (21:30 -0700)]
autofs4: use look aside list for lookups
A while ago a patch to resolve a deadlock during directory creation was
merged. This delayed the hashing of lookup dentrys until the ->mkdir()
(or ->symlink()) operation completed to ensure we always went through
->lookup() instead of also having processes go through ->revalidate() so
our VFS locking remained consistent.
Now we are seeing a couple of side affects of that change in situations
with heavy mount activity.
Two cases have been identified:
1) When a mount request is triggered, due to the delayed hashing, the
directory created by user space for the mount point doesn't have the
DCACHE_AUTOFS_PENDING flag set. In the case of an autofs multi-mount
where a tree of mount point directories are created this can lead to
the path walk continuing rather than the dentry being sent to the wait
queue to wait for request completion. This is because, if the pending
flag isn't set, the criteria for deciding this is a mount in progress
fails to hold, namely that the dentry is not a mount point and has no
subdirectories.
2) A mount request dentry is initially created negative and unhashed.
It remains this way until the ->mkdir() callback completes. Since it
is unhashed a fresh dentry is used when the user space mount request
creates the mount point directory. This leaves the original dentry
negative and unhashed. But revalidate has no way to tell the VFS that
the dentry has changed, other than to force another ->lookup() by
returning false, which is at best wastefull and at worst not possible.
This results in an -ENOENT return from the original path walk when in
fact the mount succeeded.
To resolve this we need to ensure that the same dentry is used in all
calls to ->lookup() during the course of a mount request. This patch
achieves that by adding the initial dentry to a look aside list and
removes it at ->mkdir() or ->symlink() completion (or when the dentry is
released), since these are the only create operations autofs4 supports.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:11 +0000 (21:30 -0700)]
autofs4: revert - redo lookup in ttfd
This patch series enables the use of a single dentry for lookups prior to
the dentry being hashed and so we no longer need to redo the lookup. This
patch reverts the patch of commit 033790449ba9c4dcf8478a87693d33df625c23b5.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ian Kent [Thu, 24 Jul 2008 04:30:09 +0000 (21:30 -0700)]
autofs4: don't make expiring dentry negative
Correct the error of making a positive dentry negative after it has been
instantiated.
The code that makes this error attempts to re-use the dentry from a
concurrent expire and mount to resolve a race and the dentry used for the
lookup must be negative for mounts to trigger in the required cases. The
fact is that the dentry doesn't need to be re-used because all that is
needed is to preserve the flag that indicates an expire is still
incomplete at the time of the mount request.
This change uses the the dentry to check the flag and wait for the expire
to complete then discards it instead of attempting to re-use it.
Signed-off-by: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / linux-2.6 / log