07 Jan, 2006
13 commits
-
There are some more places where the use of cputime_t instead of an integer
type and the associated macros is necessary for the virtual cputime accounting
on s390. Affected are the s390 specific appldata code and BSD process
accounting.Signed-off-by: Martin Schwidefsky
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This makes the swsusp_info structure become the header of the image in the
literal sense (ie. it is saved to the swap and read before any other image
data with the help of the swsusp's swap map structure, so generally it is
treated in the same way as the rest of the image).The main thing it does is to make swsusp_header contain the offset of the swap
map used to track the image data pages rather than the offset of swsusp_info.
Simultaneously, swsusp_info becomes the first image page written to the swap.The other changes are generally consequences of the above with a few
exceptions (there's some consolidation in the image reading part as a few
functions turn into trivial wrappers around something else).Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This changes the handling of swap partitions by swsusp to avoid locking of the
swap devices that are not used for suspend and, consequently, simplifies the
code.Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Make the suspend image size limit tunable via /sys/power/image_size.
It is necessary for systems on which there is a limited amount of swap
available for suspend. It can also be useful for optimizing performance of
swsusp on systems with 1 GB of RAM or more.Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Limit the size of the suspend image to approx. 500 MB, which should
improve the overall performance of swsusp on systems with more than 1 GB of
RAM.It introduces the constant IMAGE_SIZE that can be set to the preferred size
of the image (in MB) and modifies the memory-shrinking part of swsusp to
take this constant into account (500 is the default value of IMAGE_SIZE).Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
These two prototypes are already present in sched.h, remove duplicate
version.Signed-off-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch fixes a problem with the function enough_free_mem() used by
swsusp to verify if there is a sufficient number of memory pages available
to it to create and save the suspend image.Namely, enough_free_mem() uses nr_free_pages() to obtain the number of free
memory pages, which is incorrect, because this function returns the total
number of free pages, including free highmem pages, and the highmem pages
cannot be used by swsusp for storing the image data.The patch makes enough_free_mem() avoid counting the free highmem
pages as available to swsusp.Signed-off-by: Rafael J. Wysocki
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch makes swsusp free only as much memory as needed to complete the
suspend and not as much as possible. In the most of cases this should speed
up the suspend and make the system much more responsive after resume,
especially if a GUI (eg. X Windows) is used.If needed, the old behavior (ie to free as much memory as possible during
suspend) can be restored by unsetting FAST_FREE in power.hSigned-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch introduces the swap map structure that can be used by swsusp for
keeping tracks of data pages written to the swap. The structure itself is
described in a comment within the patch.The overall idea is to reduce the amount of metadata written to the swap and
to write and read the image pages sequentially, in a file-alike way. This
makes the swap-handling part of swsusp fairly independent of its
snapshot-handling part and will hopefully allow us to completely separate
these two parts in the future.This patch is needed to remove the suspend image size limit imposed by the
limited size of the swsusp_info structure, which is essential for x86-64
systems with more than 512 MB of RAM.Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch removes the image encryption that is only used by swsusp instead of
zeroing the image after resume in order to prevent someone from reading some
confidential data from it in the future and it does not protect the image from
being read by an unauthorized person before resume. The functionality it
provides should really belong to the user space and will possibly be
reimplemented after the swap-handling functionality of swsusp is moved to the
user space.Signed-off-by: Rafael J. Wysocki
Acked-by: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Thanks to Christoph for doing most of the work.
This allows automatic SMP IRQ affinity assignment other than default "all
interrupts on all CPUs" which is rather expensive. This might be useful if
the hardware can be programmed to distribute interrupts among different
CPUs, like Alpha does.Signed-off-by: Ivan Kokshaysky
Cc: Christoph Hellwig
Cc: Richard Henderson
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Make the futex code compilable and usable on NOMMU by making the attempt to
handle page faults conditional on CONFIG_MMU. If this is not enabled, then
we can assume that EFAULT returned from futex_atomic_op_inuser() is not
recoverable, and that the address lies outside of valid memory.handle_mm_fault() is made to BUG if called on NOMMU without attempting to
invoke the actual handler (__handle_mm_fault).Signed-off-by: David Howells
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
- This function returns -EINVAL all the time. Fix.
- Decruftify it a bit too.
- Writing to it doesn't seem to do what it's suppoed to do.
Cc: Pavel Machek
Cc: "Rafael J. Wysocki"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
05 Jan, 2006
6 commits
-
Trivial manual merge fixup for usb_find_interface clashes.
-
lib/lib.a(kobject_uevent.o)(.text+0x25f): In function `kobject_uevent':
: undefined reference to `__alloc_skb'
lib/lib.a(kobject_uevent.o)(.text+0x2a1): In function `kobject_uevent':
: undefined reference to `skb_over_panic'
lib/lib.a(kobject_uevent.o)(.text+0x31d): In function `kobject_uevent':
: undefined reference to `skb_over_panic'
lib/lib.a(kobject_uevent.o)(.text+0x356): In function `kobject_uevent':
: undefined reference to `netlink_broadcast'
lib/lib.a(kobject_uevent.o)(.init.text+0x9): In function `kobject_uevent_init':
: undefined reference to `netlink_kernel_create'
make: *** [.tmp_vmlinux1] Error 1Netlink is unconditionally enabled if CONFIG_NET, so that's OK.
kobject_uevent.o is compiled even if !CONFIG_HOTPLUG, which is lazy.
Let's compound the sin.
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman -
Leave the overloaded "hotplug" word to susbsystems which are handling
real devices. The driver core does not "plug" anything, it just exports
the state to userspace and generates events.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
This deprecates the /proc/sys/kernel/hotplug file, as all
this stuff should be in /sys some day, right? :)
In /sys/kernel/ we have now uevent_seqnum and uevent_helper.
The seqnum is no longer used by udev, as the version for this
kernel depends on netlink which events will never get
out-of-order.Recent udev versions disable the /sbin/hotplug helper with
an init script, cause it leads to OOM on big boxes by running
hundreds of shells in parallel. It should be done now by:
echo "" > /sys/kernel/uevent_helper(Note that "-n" does not work, cause neighter proc nor sysfs
support truncate().)Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
It makes zero sense to have hotplug, but not the netlink
events enabled today. Remove this option and merge the
kobject_uevent.h header into the kobject.h header file.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman
03 Jan, 2006
2 commits
-
This patch removes all references to the bouncing address
rddunlap@osdl.org and one dead web page from the kernel.Signed-off-by: Adrian Bunk
Acked-by: Randy Dunlap -
Signed-off-by: Sam Ravnborg
01 Jan, 2006
1 commit
-
This is a slightly more complete fix for the previous minimal sysctl
string fix. It always terminates the returned string with a NUL, even
if the full result wouldn't fit in the user-supplied buffer.The returned length is the full untruncated length, so that you can
tell when truncation has occurred.Signed-off-by: Linus Torvalds
31 Dec, 2005
2 commits
-
For the sysctl syscall, if the user wants to get the old value of a
sysctl entry and set a new value for it in the same syscall, the old
value is always overwritten by the new value if the sysctl entry is of
string type and if the user sets its strategy to sysctl_string. This
issue lies in the strategy being run twice if the strategy is set to
sysctl_string, the general strategy sysctl_string always returns 0 if
success.Such strategy routines as sysctl_jiffies and sysctl_jiffies_ms return 1
because they do read and write for the sysctl entry.The strategy routine sysctl_string return 0 although it actually read
and write the sysctl entry.According to my analysis, if a strategy routine do read and write, it
should return 1, if it just does some necessary check but not read and
write, it should return 0, for example sysctl_intvec.Signed-off-by: Yi Yang
Signed-off-by: Linus Torvalds -
If the string was too long to fit in the user-supplied buffer,
the sysctl layer would zero-terminate it by writing past the
end of the buffer. Don't do that.Noticed by Yi Yang
Signed-off-by: Linus Torvalds
25 Dec, 2005
1 commit
-
Fix a memory ordering problem that occurs on IA64. The "store" to q->lock_ptr
in wake_futex() can become visible before wake_up_all() clears the lock in the
futex_q.Signed-off-by: Jack Steiner
Acked-by: Ingo Molnar
Signed-off-by: Linus Torvalds
21 Dec, 2005
1 commit
-
All the work was done to setup the file and maintain the file handles but
the access functions were zeroed out due to the #ifdef. Removing the
#ifdef allows full access to all the parameters when CONFIG_MODULES=n.akpm: put it back again, but use CONFIG_SYSFS instead.
Signed-off-by: Jason Wessel
Signed-off-by: Andrew Morton
Signed-off-by: Adrian Bunk
Signed-off-by: Linus Torvalds
13 Dec, 2005
9 commits
-
When multiple probes are registered at the same address and if due to some
recursion (probe getting triggered within a probe handler), we skip calling
pre_handlers and just increment nmissed field.The below patch make sure it walks the list for multiple probes case.
Without the below patch we get incorrect results of nmissed count for
multiple probe case.Signed-off-by: Anil S Keshavamurthy
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
For Kprobes critical path is the path from debug break exception handler
till the control reaches kprobes exception code. No probes can be
supported in this path as we will end up in recursion.This patch prevents this by moving the below function to safe __kprobes
section onto which no probes can be inserted.Signed-off-by: Anil S Keshavamurthy
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
kauditd was causing suspends to fail because it refused to freeze. Adding
a try_to_freeze() to its sleep loop solves the issue.Signed-off-by: Pierre Ossman
Acked-by: Pavel Machek
Cc: David Woodhouse
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
When registering multiple kprobes at the same address, we leave a small
window where the kprobe hlist will not contain a reference to the
registered kprobe, leading to potentially, a system crash if the breakpoint
is hit on another processor.Patch below now automically relpace the old kprobe with the new
kprobe from the hash list.Signed-off-by: Anil S Keshavamurthy
Acked-by: Ananth N Mavinakayanahalli
Cc: "Paul E. McKenney"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
There are several functions that might seem appropriate for a timestamp:
get_cycles()
current_kernel_time()
do_gettimeofday()Each has problems with combinations of SMP-safety, low resolution, and
monotonicity. This patch adds a new function that returns a monotonic SMP-safe
timestamp with nanosecond resolution where available.Changes:
Split timestamp into separate patch
Moved to kernel/time.c
Renamed to getnstimestamp
Fixed unintended-pointer-arithmetic bugSigned-off-by: Matt Helsley
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Accessing nohz_cpu_mask before incrementing rcp->cur is racy. It can cause
tickless idle CPUs to be included in rsp->cpumask, which will extend
graceperiods unnecessarily.Fix this race. It has been tested using extensions to RCU torture module
that forces various CPUs to become idle.Signed-off-by: Srivatsa Vaddagiri
Cc: Dipankar Sarma
Cc: "Paul E. McKenney"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
While doing some test of RCU torture module, I hit a OOPS in rcu_do_batch,
which was trying to processes callback of a module that was just removed.
This is because we weren't waiting long enough for all callbacks to fire.Signed-off-by: Srivatsa Vaddagiri
Cc: Dipankar Sarma
Acked-by: "Paul E. McKenney"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This introduces a new interface - rcu_barrier() which waits until all
the RCUs queued until this call have been completed.Reiser4 needs this, because we do more than just freeing memory object
in our RCU callback: we also remove it from the list hanging off
super-block. This means, that before freeing reiser4-specific portion
of super-block (during umount) we have to wait until all pending RCU
callbacks are executed.The only change of reiser4 made to the original patch, is exporting of
rcu_barrier().Cc: Hans Reiser
Cc: Vladimir V. Saveliev
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
When a Kprobes are inserted/removed on a modules, the modules must be ref
counted so as not to allow to unload while probes are registered on that
module.Without this patch, the probed module is free to unload, and when the
probing module unregister the probe, the kpobes code while trying to
replace the original instruction might crash.Signed-off-by: Anil S Keshavamurthy
Signed-off-by: Mao Bibo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
30 Nov, 2005
2 commits
-
Fix swsusp on machines not supporting S4. With recent changes, it is not
possible to trigger it using /sys filesystem. Swsusp does not really need
any support from low-level code, it is possible to reboot or halt at the
end of suspend.Signed-off-by: Pavel Machek
Cc: "Brown, Len"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
set_page_dirty() will not cope with being handed a page * which is part of
a compound page, but not the master page in that compound page. This case
can occur via access_process_vm() if you attemp to write to another
process's hugepage memory area using ptrace() (causing an oops or hang).This patch fixes the bug by only calling set_page_dirty() from
access_process_vm() if the page is not a compound page. We already use a
similar fix in bio_set_pages_dirty() for the case of direct io to
hugepages.Signed-off-by: David Gibson
Acked-by: William Irwin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
29 Nov, 2005
3 commits
-
Move the cpuset_fork() call below the write_unlock_irq call in
kernel/fork.c copy_process().Since the cpuset-dual-semaphore-locking-overhaul.patch, the cpuset_fork()
routine acquires task_lock(), so cannot be called while holding the
tasklist_lock for write.Signed-off-by: Paul Jackson
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Tracked this down on an Ultra Enterprise 3000. It's a 6-way machine. Odd
thing about this machine (and it's good for finding bugs like this) is that
the CPU id's are not 0 based. For instance, on my machine the CPU's are
6/7/10/11/14/15.This caused some NULL pointer dereference in kernel/workqueue.c because for
single_threaded workqueue's, it hardcoded the cpu to 0.I changed the 0's to any_online_cpu(cpu_online_mask), which cpumask.h
claims is "First cpu in mask". So this fits the same usage.Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
fix 32bit overflow in timespec_to_sample()
Signed-off-by: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
