17 Jan, 2010
19 commits
-
Commit f50de2d3 (vmscan: have kswapd sleep for a short interval and double
check it should be asleep) can cause kswapd to enter an infinite loop if
running on a single-CPU system. If all zones are unreclaimble,
sleeping_prematurely return 1 and kswapd will call balance_pgdat() again.
but it's totally meaningless, balance_pgdat() doesn't anything against
unreclaimable zone!Signed-off-by: KOSAKI Motohiro
Cc: Mel Gorman
Reported-by: Will Newton
Reviewed-by: Minchan Kim
Reviewed-by: Rik van Riel
Tested-by: Will Newton
Reviewed-by: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The change in acpi_cpufreq to use smp_call_function_any causes a warning
when it is called since the function erroneously passes the cpu id to
cpumask_of_node rather than the node that the cpu is on. Fix this.cpumask_of_node(3): node > nr_node_ids(1)
Pid: 1, comm: swapper Not tainted 2.6.33-rc3-00097-g2c1f189 #223
Call Trace:
[] cpumask_of_node+0x23/0x58
[] smp_call_function_any+0x65/0xfa
[] ? do_drv_read+0x0/0x2f
[] get_cur_val+0xb0/0x102
[] get_cur_freq_on_cpu+0x74/0xc5
[] acpi_cpufreq_cpu_init+0x417/0x515
[] ? __down_write+0xb/0xd
[] cpufreq_add_dev+0x278/0x922Signed-off-by: David John
Cc: Suresh Siddha
Cc: Rusty Russell
Cc: Thomas Gleixner
Cc: "H. Peter Anvin"
Cc: Ingo Molnar
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add BUILD_BUG_ON_NOT_POWER_OF_2()
When code relies on a constant being a power of 2:
#define FOO 512 /* must be a power of 2 */
it would be nice to be able to do:
BUILD_BUG_ON(!is_power_of_2(FOO));
However applying an inline function does not result in a compile-time
constant that can be used with BUILD_BUG_ON(), so trying that gives
results in:error: bit-field '' width not an integer constant
As suggested by akpm, rather than monkeying around with is_power_of_2()
and risking gcc warts about constant expressions, just create a macro
BUILD_BUG_ON_NOT_POWER_OF_2() to encapsulate this common requirement.Signed-off-by: Roland Dreier
Cc: Bart Van Assche
Cc: David Dillow
Cc: "Robert P. J. Day"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The current check for 'backward merging' within add_active_range() does
not seem correct. start_pfn must be compared against
early_node_map[i].start_pfn (and NOT against .end_pfn) to find out whether
the new region is backward-mergeable with the existing range.Signed-off-by: Kazuhisa Ichikawa
Acked-by: David Rientjes
Cc: KOSAKI Motohiro
Cc: Mel Gorman
Cc: Christoph Lameter
Cc: Johannes Weiner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
On my first try using them I missed that the fifos need to be power of
two, resulting in a runtime bug. Document that requirement everywhere
(and fix one grammar bug)Signed-off-by: Andi Kleen
Acked-by: Stefani Seibold
Cc: Roland Dreier
Cc: Dmitry Torokhov
Cc: Andy Walls
Cc: Vikram Dhillon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Simple inline that checks if kfifo_init() has been executed on a fifo.
This is useful for walking all per CPU fifos, when some of them might not
have been brought up yet.Signed-off-by: Andi Kleen
Acked-by: Stefani Seibold
Cc: Roland Dreier
Cc: Dmitry Torokhov
Cc: Andy Walls
Cc: Vikram Dhillon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
In some upcoming code it's useful to peek into a FIFO without permanentely
removing data. This patch implements a new kfifo_out_peek() to do this.Signed-off-by: Andi Kleen
Acked-by: Stefani Seibold
Cc: Roland Dreier
Cc: Dmitry Torokhov
Cc: Andy Walls
Cc: Vikram Dhillon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Right now for kfifo_*_user it's not easily possible to distingush between
a user copy failing and the FIFO not containing enough data. The problem
is that both conditions are multiplexed into the same return code.Avoid this by moving the "copy length" into a separate output parameter
and only return 0/-EFAULT in the main return value.I didn't fully adapt the weird "record" variants, those seem
to be unused anyways and were rather messy (should they be just removed?)I would appreciate some double checking if I did all the conversions
correctly.Signed-off-by: Andi Kleen
Cc: Stefani Seibold
Cc: Roland Dreier
Cc: Dmitry Torokhov
Cc: Andy Walls
Cc: Vikram Dhillon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The pointers to user buffers are currently unsigned char *, which requires
a lot of casting in the caller for any non-char typed buffers. Use void *
instead.Signed-off-by: Andi Kleen
Acked-by: Stefani Seibold
Cc: Roland Dreier
Cc: Dmitry Torokhov
Cc: Andy Walls
Cc: Vikram Dhillon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
I get a few dozen of these warnings when using
gcc (GCC) 4.4.1 20090725 (Red Hat 4.4.1-2):In file included from mmotm-2010-0113-1217/init/do_mounts.c:5:
mmotm-2010-0113-1217/include/linux/tty.h: In function 'tty_port_get':
mmotm-2010-0113-1217/include/linux/tty.h:469: warning: '______f' is static but declared in inline function 'tty_port_get' which is not staticso make the function static inline.
[akpm@linux-foundation.org: may as well convert tty_port_users() also]
Signed-off-by: Randy Dunlap
Cc: Alan Cox
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
EDAC MC0: INTERNAL ERROR: channel-b out of range (4 >= 4)
Kernel panic - not syncing: EDAC MC0: Uncorrected Error (XEN) Domain 0 crashed: 'noreboot' set - not rebooting.This happens because FERR_NF_FBD bit 28 is not updated on i5000. Due to
that, both bits 28 and 29 may be equal to one, returning channel = 3. As
this value is invalid, EDAC core generates the panic.Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14568
Signed-off-by: Tamas Vincze
Signed-off-by: Mauro Carvalho Chehab
Signed-off-by: Doug Thompson
Cc:
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The m68knommu coldfire pit clocksource looks like it was incorrectly
marked as a continuous clocksource. Running with it marked as a
continuous clocksource could cause hangs when the system switches to
highres mode or enables nohz.This patch removes the CLOCK_SOURCE_IS_CONTINUOUS flag on the coldfire pit
clocksource. This will disallow systems using this clocksource from
entering oneshot mode (disabling highres timers and nohz).Signed-off-by: John Stultz
Acked-by: Greg Ungerer
Cc: Steven King
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This is a new two finger touch Fujitsu Wacom Tablet PC.
Signed-off-by: Ping Cheng
Cc: Alan Cox
Cc: Dmitry Torokhov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
When I try to use markup_oops.pl in x86, I always get:
cat 1 | perl markup_oops.pl ./vmlinux
objdump: --start-address: bad number: NaN
No matching code foundThis is because in line:
if ($line =~ /EIP is at ([a-zA-Z0-9\_]+)\+0x([0-9a-f]+)\/[a-f0-9]/) {
$function = $1;
$func_offset = $2;
}$func_offset will get a number like "0x2"
But in follow code:
my $decodestart = Math::BigInt->from_hex("0x$target") -
Math::BigInt->from_hex("0x$func_offset");It add other ox to ox2. Then this value will be set to NaN.
So I made a small patch to fix it.
Signed-off-by: Hui Zhu
Acked-by: Arjan van de Ven
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix a regression in hardware acceleration which made the accelerated
framebuffer unusable on some chips. These need extra initialization and
an extra flag which is no longer needed/available on current chips.Signed-off-by: Erik-Jan Post
Signed-off-by: Florian Tobias Schandinat
Cc: Scott Fang
Cc: Joseph Chan
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Reorder viafb_set_par to allow using the updated variables in
viafb_setmode. This fixes a regression that prevented proper runtime mode
changes.Signed-off-by: Erik-Jan Post
Signed-off-by: Florian Tobias Schandinat
Cc: Scott Fang
Cc: Joseph Chan
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Although I'd consider this a hardware bug, as there is hardware out that
for whatever reason does not support hardware cursors on LCD output we
have to care about it in the driver. This fixes a regression (invisible
cursor) introduced by:viafb: cleanup viafb_cursor
Signed-off-by: Florian Tobias Schandinat
Reported-by: Julian Wollrath
Tested-by: Julian Wollrath
Cc: Scott Fang
Cc: Joseph Chan
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix a wrong optimization in include/linux/kfifo.h which could cause a race
in kfifo_out_locked.Signed-off-by: Stefani Seibold
Reported-by: Johan Hovold
Cc: Pete Zaitcev
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel:
drm/i915: enable 36bit physical address for hardware status page
drm/i915: fix eDP pipe mask
drm/i915: fix pixel color depth setting on eDP
drm/i915: parse eDP panel color depth from VBT block
drm/i915: disable LVDS downclock by default
drm/i915: Fix the incorrect cursor A bit definition in DSPFW2 register
drm/i915: Remove chatty execbuf failure message.
drm/i915: remove loop in Ironlake interrupt handler
drm/i915: Don't wait interruptible for possible plane buffer flush
drm/i915: try another possible DDC bus for the SDVO device with multiple outputs
drm/i915: Read the response after issuing DDC bus switch command
drm/i915: Don't use the child device parsed from VBT to setup HDMI/DP
drm/i915: Fix resume regression on MSI Wind U100 w/o KMS
drm/i915: Fix Ironlake M/N/P ranges to match the spec
drm/i915: Use find_pll function to calculate DPLL setting for LVDS downclock
drm/i915: Add HP nx9020/SamsungSX20S to ACPI LID quirk list
drm/i915: disable TV hotplug status checkTrivial conflicts in drivers/gpu/drm/i915/i915_drv.c due to i915
non-modeset suspend fix with different comment.
16 Jan, 2010
15 commits
-
* 'for-linus/samsung' of git://git.fluff.org/bjdooks/linux:
ARM: MINI2440: Fixup __initdata usage
ARM: MINI2440: Fix crash on boot due to improper __initdata qualifier
ARM: SMDK6410: Specify no GPIO for B_PWR_5V regulator
ARM: S3C: NAND: Check the existence of nr_map before copying -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp:
amd64_edac: Ensure index stays within bounds in amd64_get_scrub_rate -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: sentelic - fix left/right horizontal scroll mapping
Input: pmouse - move Sentelic probe down the list
Input: add compat support for sysfs and /proc capabilities output
Input: i8042 - add Dritek quirk for Acer Aspire 5610.
Input: xbox - do not use GFP_KERNEL under spinlock
Input: psmouse - fix Synaptics detection when protocol is disabled
Input: bcm5974 - report ABS_MT events
Input: davinci_keyscan - add device_enable method to platform data
Input: evdev - be less aggressive about sending SIGIO notifies
Input: atkbd - fix canceling event_work in disconnect
Input: serio - fix potential deadlock when unbinding drivers
Input: gf2k - fix &&/|| confusion in gf2k_connect() -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha-2.6:
alpha: cpumask_of_node() should handle -1 as a node
alpha: add myself as a maintainer, and drop mention of 2.4 -
* 'sh/for-2.6.33' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
serial: sh-sci: using correct fifo size for SCIF and SCIFA ports.
sh: mach-ecovec24: Add motion sensor driver support. -
This enables possible 36bit address mask on 965G that use physical
address for hw status page.Signed-off-by: Zhenyu Wang
Signed-off-by: Eric Anholt -
inotify will WARN() if it finds that the idr and the fsnotify internals
somehow got out of sync. It was only supposed to do this once but due
to this stupid bug it would warn every single time a problem was
detected.Signed-off-by: Eric Paris
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds -
Since commit 7e790dd5fc937bc8d2400c30a05e32a9e9eef276 ("inotify: fix
error paths in inotify_update_watch") inotify changed the manor in which
it gave watch descriptors back to userspace. Previous to this commit
inotify acted like the following:inotify_add_watch(X, Y, Z) = 1
inotify_rm_watch(X, 1);
inotify_add_watch(X, Y, Z) = 2but after this patch inotify would return watch descriptors like so:
inotify_add_watch(X, Y, Z) = 1
inotify_rm_watch(X, 1);
inotify_add_watch(X, Y, Z) = 1which I saw as equivalent to opening an fd where
open(file) = 1;
close(1);
open(file) = 1;seemed perfectly reasonable. The issue is that quite a bit of userspace
apparently relies on the behavior in which watch descriptors will not be
quickly reused. KDE relies on it, I know some selinux packages rely on
it, and I have heard complaints from other random sources such as debian
bug 558981.Although the man page implies what we do is ok, we broke userspace so
this patch almost reverts us to the old behavior. It is still slightly
racey and I have patches that would fix that, but they are rather large
and this will fix it for all real world cases. The race is as follows:- task1 creates a watch and blocks in idr_new_watch() before it updates
the hint.
- task2 creates a watch and updates the hint.
- task1 updates the hint with it's older wd
- task removes the watch created by task2
- task adds a new watch and will reuse the wd originally given to task2it requires moving some locking around the hint (last_wd) but this should
solve it for the real world and be -stable safe.As a side effect this patch papers over a bug in the lib/idr code which
is causing a large number WARN's to pop on people's system and many
reports in kerneloops.org. I'm working on the root cause of that idr
bug seperately but this should make inotify immune to that issue.Signed-off-by: Eric Paris
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds -
eDP could be on pipe A or B.
Signed-off-by: Zhenyu Wang
Signed-off-by: Eric Anholt -
Original DP mode_valid check didn't take pixel color depth into account,
which made one 1600x900 eDP panel's mode check invalid because of overclock,
but actually this 6bpc panel does can work with x1 lane at 2.7G. This one
trys to take bpp value properly both in mode validation and mode setting.Signed-off-by: Zhenyu Wang
Signed-off-by: Eric Anholt -
Signed-off-by: Zhenyu Wang
Signed-off-by: Eric Anholt -
Many platform support this feature, and it can provide significant
power savings when the reduced refresh rate is low. However, on some
platforms a secondary (reduced) timing is provided but not actually
supported by the hardware. This results in undesirable flicker at
runtime.So disable the feature by default, but allow users to opt-in to the
reduced clock behavior with a new module parameter, lvds_downclock,
that can be set to 1 to enable the feature.Signed-off-by: Jesse Barnes
Signed-off-by: Eric Anholt -
Signed-off-by: Zhao Yakui
Reviewed-by: Eric Anholt
Signed-off-by: Eric Anholt -
Suggested-by: Chris Wilson >
Acked-by: Jesse Barnes (in principle)
Signed-off-by: Eric Anholt -
On Ironlake, there is an interrupt master control bit. With the bit
disabled before clearing IIR, we do not need to handle extra interrupt
in a loop. This patch removes the loop in Ironlake interrupt handler.
It fixed irq lost issue on some Ironlake platforms.Cc: Stable Team
Signed-off-by: Zou Nan hai
Signed-off-by: Zhenyu Wang
Signed-off-by: Eric Anholt
15 Jan, 2010
6 commits
-
Add a missing iterator variable thus fixing the conditional of the
for-loop in amd64_get_scrub_rate().Signed-off-by: Roel Kluin
Signed-off-by: Borislav Petkov -
Remove some of the __initdata tags which are currently inappropriate for
platform_device and some of the platform data. These can be returned once
support for copying platform devices and data is added.Signed-off-by: Ben Dooks
-
This patch fix mini2440 crash on boot due to improper __initdata
qualifier on mini2440_led1_pdata.Signed-off-by: Uri Yosef
Signed-off-by: Ben Dooks -
Since the fixed voltage regulator grew support for GPIO based
enables and GPIO 0 is valid on some systems we need to specify
that there is no valid GPIO enable control.Signed-off-by: Mark Brown
Signed-off-by: Ben Dooks -
The sh-sci driver used the wrong fifosize for PORT_SCIFA and PORT_SCIF
ports. If an incorrect size is used, the serial core will enforce an
early shutdown on the port, especially with baudrates < 9600.Signed-off-by: Markus Pietrek
Signed-off-by: Paul Mundt -
This patch adds support for the lis3lv02d motion sensor connected via
i2c on the Ecovec board. Tested with evtest.Signed-off-by: NISHIMOTO Hiroki
Signed-off-by: Paul Mundt
