As an optimization, follow-up SRST used to be skipped if
classification wasn't requested even when hardreset requested it via
-EAGAIN. However, some hardresets can't wait for device readiness and
skipping SRST can cause timeout or other failures during revalidation.
Always perform follow-up SRST if hardreset returns -EAGAIN. This
makes reset paths more predictable and thus less error-prone.
While at it, move hardreset error checking such that it's done right
after hardreset is finished. This simplifies followup SRST condition
check a bit and makes the reset path easier to modify.
Because fbcon_set_all_vcs()->FBCON_SWAP() uses display->rotate == 0 instead
of fbcon_ops->rotate, and vc_resize() has no effect because it is called with
new_cols/rows == ->vc_cols/rows.
Tested on 2.6.26.5-45.fc9.x86_64, but
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git seems to
have the same problem.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> 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: Greg Kroah-Hartman <gregkh@suse.de>
A coding error present since b43legacy was incorporated into the
kernel has prevented the driver from using the rate-setting mechanism
of mac80211. The driver has been forced to remain at a 1 Mb/s rate.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When we do a seekdir() or equivalent, we usually end up doing a
FindFirst call and then call FindNext until we get to the offset that we
want. The problem is that when we call FindNext, the code usually
doesn't have the proper info (mostly, the filename of the entry from the
last search) to resume the search.
Add a "last_entry" field to the cifs_search_info that points to the last
entry in the search. We calculate this pointer by using the
LastNameOffset field from the search parms that are returned. We then
use that info to do a cifs_save_resume_key before we call CIFSFindNext.
This patch allows CIFS to reliably pass the "telldir" connectathon test.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
While working on the new version of the code for SCHED_SPORADIC I
noticed something strange in the present throttling mechanism. More
specifically in the throttling timer handler in sched_rt.c
(do_sched_rt_period_timer()) and in rt_rq_enqueue().
The problem is that, when unthrottling a runqueue, rt_rq_enqueue() only
asks for rescheduling if the runqueue has a sched_entity associated to
it (i.e., rt_rq->rt_se != NULL).
Now, if the runqueue is the root rq (which has a rt_se = NULL)
rescheduling does not take place, and it is delayed to some undefined
instant in the future.
This imply some random bandwidth usage by the RT tasks under throttling.
For instance, setting rt_runtime_us/rt_period_us = 950ms/1000ms an RT
task will get less than 95%. In our tests we got something varying
between 70% to 95%.
Using smaller time values, e.g., 95ms/100ms, things are even worse, and
I can see values also going down to 20-25%!!
The tests we performed are simply running 'yes' as a SCHED_FIFO task,
and checking the CPU usage with top, but we can investigate thoroughly
if you think it is needed.
Things go much better, for us, with the attached patch... Don't know if
it is the best approach, but it solved the issue for us.
Signed-off-by: Dario Faggioli <raistlin@linux.it> Signed-off-by: Michael Trimarchi <trimarchimichael@yahoo.it> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The x86 implementation of early_ioremap has an off by one error. If we get
an object which ends on the first byte of a page we undermap by one page and
this causes a crash on boot with the ASUS P5QL whose DMI table happens to fit
this alignment.
Stefan Bader [Sat, 27 Sep 2008 15:07:30 +0000 (11:07 -0400)]
x86: Reserve FIRST_DEVICE_VECTOR in used_vectors bitmap.
Not in upstream above 2.6.27 due to change in the way this code works
(has been fixed differently there.)
Someone from the community found out, that after repeatedly unloading
and loading a device driver that uses MSI IRQs, the system eventually
assigned the vector initially reserved for IRQ0 to the device driver.
The reason for this is, that although IRQ0 is tied to the
FIRST_DEVICE_VECTOR when declaring the irq_vector table, the
corresponding bit in the used_vectors map is not set. So, if vectors are
released and assigned often enough, the vector will get assigned to
another interrupt. This happens more often with MSI interrupts as those
are exclusively using a vector.
Fix this by setting the bit for the FIRST_DEVICE_VECTOR in the bitmap.
When running a 31-bit ptrace, on either an s390 or s390x kernel,
reads and writes into a padding area in struct user_regs_struct32
will result in a kernel panic.
This is also known as CVE-2008-1514.
Test case available here:
http://sources.redhat.com/cgi-bin/cvsweb.cgi/~checkout~/tests/ptrace-tests/tests/user-area-padding.c?cvsroot=systemtap
Steps to reproduce:
1) wget the above
2) gcc -o user-area-padding-31bit user-area-padding.c -Wall -ggdb2 -D_GNU_SOURCE -m31
3) ./user-area-padding-31bit
<panic>
Test status
-----------
Without patch, both s390 and s390x kernels panic. With patch, the test case,
as well as the gdb testsuite, pass without incident, padding area reads
returning zero, writes ignored.
Nb: original version returned -EINVAL on write attempts, which broke the
gdb test and made the test case slightly unhappy, Jan Kratochvil suggested
the change to return 0 on write attempts.
Signed-off-by: Jarod Wilson <jarod@redhat.com> Tested-by: Jan Kratochvil <jan.kratochvil@redhat.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Moritz Muehlenhoff <jmm@debian.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Balbir Singh [Sun, 5 Oct 2008 16:43:37 +0000 (17:43 +0100)]
mm owner: fix race between swapoff and exit
[Here's a backport of 2.6.27-rc8's 31a78f23bac0069004e69f98808b6988baccb6b6
to 2.6.26 or 2.6.26.5: I wouldn't trouble -stable for the (root only)
swapoff case which uncovered the bug, but the /proc/<pid>/<mmstats> case
is open to all, so I think worth plugging in the next 2.6.26-stable.
- Hugh]
There's a race between mm->owner assignment and swapoff, more easily
seen when task slab poisoning is turned on. The condition occurs when
try_to_unuse() runs in parallel with an exiting task. A similar race
can occur with callers of get_task_mm(), such as /proc/<pid>/<mmstats>
or ptrace or page migration.
CPU0 CPU1
try_to_unuse
looks at mm = task0->mm
increments mm->mm_users
task 0 exits
mm->owner needs to be updated, but no
new owner is found (mm_users > 1, but
no other task has task->mm = task0->mm)
mm_update_next_owner() leaves
mmput(mm) decrements mm->mm_users
task0 freed
dereferencing mm->owner fails
The fix is to notify the subsystem via mm_owner_changed callback(),
if no new owner is found, by specifying the new task as NULL.
Jiri Slaby:
mm->owner was set to NULL prior to calling cgroup_mm_owner_callbacks(), but
must be set after that, so as not to pass NULL as old owner causing oops.
Daisuke Nishimura:
mm_update_next_owner() may set mm->owner to NULL, but mem_cgroup_from_task()
and its callers need to take account of this situation to avoid oops.
Hugh Dickins:
Lockdep warning and hang below exec_mmap() when testing these patches.
exit_mm() up_reads mmap_sem before calling mm_update_next_owner(),
so exec_mmap() now needs to do the same. And with that repositioning,
there's now no point in mm_need_new_owner() allowing for NULL mm.
When userspace uses SIGIO notification and forgets to disable it before
closing file descriptor, rtc->async_queue contains stale pointer to struct
file. When user space enables again SIGIO notification in different
process, kernel dereferences this (poisoned) pointer and crashes.
So disable SIGIO notification on close.
Kernel panic:
(second run of qemu (requires echo 1024 > /sys/class/rtc/rtc0/max_user_freq))
Commit 22af89aa0c0b4012a7431114a340efd3665a7617 ("fbcon: replace mono_col
macro with static inline") changed the order of operations for computing
monochrome color values. This generates 0xffff000f instead of 0x0000000f
for a 4 bit monochrome color, leading to image corruption if it is passed
to cfb_imageblit or other similar functions. Fix it up.
Cc: Harvey Harrison <harvey.harrison@gmail.com> 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: Greg Kroah-Hartman <gregkh@suse.de>
Just like in the arch/sparc64/kernel/of_device.c code fix commit 071d7f4c3b411beae08d27656e958070c43b78b4 ("sparc64: Fix disappearing
PCI devices on e3500.") we have to check the OF device node name for
"pci" instead of relying upon the 'device_type' property being there
on all PCI bridges.
Tested by Meelis Roos, and confirmed to make the PCI QFE devices
reappear on the E3500 system.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The OF device layer builds properties by matching bus types and
applying 'range' properties as appropriate, up to the root.
The match for "PCI" busses is looking at the 'device_type' property,
and this does work %99 of the time.
But on an E3500 system with a PCI QFE card, the DEC 21153 bridge
sitting above the QFE network interface devices has a 'name' of "pci",
but it completely lacks a 'device_type' property. So we don't match
it as a PCI bus, and subsequently we end up with no resource values at
all for the devices sitting under that DEC bridge.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The previous patch in response to the recursive locking on IPsec
reception is broken as it tries to drop the BH socket lock while in
user context.
This patch fixes it by shrinking the section protected by the
socket lock to sock_queue_rcv_skb only. The only reason we added
the lock is for the accounting which happens in that function.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
If INIT-ACK is received with SupportedExtensions parameter which
indicates that the peer does not support AUTH, the packet will be
silently ignore, and sctp_process_init() do cleanup all of the
transports in the association.
When T1-Init timer is expires, OOPS happen while we try to choose
a different init transport.
The solution is to only clean up the non-active transports, i.e
the ones that the peer added. However, that introduces a problem
with sctp_connectx(), because we don't mark the proper state for
the transports provided by the user. So, we'll simply mark
user-provided transports as ACTIVE. That will allow INIT
retransmissions to work properly in the sctp_connectx() context
and prevent the crash.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Do not enable peer features like addip and auth, if they
are administratively disabled localy. If the peer resports
that he supports something that we don't, neither end can
use it so enabling it is pointless. This solves a problem
when talking to a peer that has auth and addip enabled while
we do not. Found by Andrei Pelinescu-Onciul <andrei@iptel.org>.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
We're never supposed to shrink the headroom or tailroom. In fact,
shrinking the headroom is a fatal action.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
/**
* nla_ok - check if the netlink attribute fits into the remaining bytes
* @nla: netlink attribute
* @remaining: number of bytes remaining in attribute stream
*/
static inline int nla_ok(const struct nlattr *nla, int remaining)
{
return remaining >= sizeof(*nla) &&
nla->nla_len >= sizeof(*nla) &&
nla->nla_len <= remaining;
}
It turns out that remaining can become negative due to alignment in
nla_next(). But GCC promotes "remaining" to unsigned in the test
against sizeof(*nla) above. Therefore the test succeeds, and the
nla_for_each_attr() may access memory outside the received buffer.
A short example illustrating this point is here:
#include <stdio.h>
main(void)
{
printf("%d\n", -1 >= sizeof(int));
}
...which prints "1".
This patch adds a cast in front of the sizeof so that GCC will make
a signed comparison and fix the illegal memory dereference. With the
patch applied, there is no kmemcheck report.
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com> Acked-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The reset_task function in the niu driver does not reset the tx and rx
buffers properly. This leads to panic on reset. This patch is a
modified implementation of the previously posted fix.
Signed-off-by: Santwona Behera <santwona.behera@sun.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This fixes kernel bugzilla 11469: "TUN with 1024 neighbours:
ip6_dst_lookup_tail NULL crash"
dst->neighbour is not necessarily hooked up at this point
in the processing path, so blindly dereferencing it is
the wrong thing to do. This NULL check exists in other
similar paths and this case was just an oversight.
Also fix the completely wrong and confusing indentation
here while we're at it.
Based upon a patch by Evgeniy Polyakov.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
ip6_dst_blackhole_ops.kmem_cachep is not expected to be NULL (i.e. to
be initialized) when dst_alloc() is called from ip6_dst_blackhole().
Otherwise, it results in the following (xfrm_larval_drop is now set to
1 by default):
Signed-off-by: Arnaud Ebalard <arno@natisbad.org> Acked-by: Benjamin Thery <benjamin.thery@bull.net> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
PCMCIA abuses dev->private_data in the probe methods. Unfortunately it
continues to abuse it after calling drv->probe() which leads to crashes and
other nasties (such as bogus probes of multifunction devices) giving errors like
pcmcia: registering new device pcmcia0.1
kernel: 0.1: GetNextTuple: No more items
Extract the passed data before calling the driver probe function that way
we don't blow up when the driver reuses dev->private_data as its right.
The issue of the endless reprogramming loop due to a too small
min_delta_ns was fixed with the previous updates of the clock events
code, but we had no information about the spread of this problem. I
added a WARN_ON to get automated information via kerneloops.org and to
get some direct reports, which allowed me to analyse the affected
machines.
The WARN_ON has served its purpose and would be annoying for a release
kernel. Remove it and just keep the information about the increase of
the min_delta_ns value.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
We have a bug in the calculation of the next jiffie to trigger the RTC
synchronisation. The aim here is to run sync_cmos_clock() as close as
possible to the middle of a second. Which means we want this function to
be called less than or equal to half a jiffie away from when now.tv_nsec
equals 5e8 (500000000).
If this is not the case for a given call to the function, for this purpose
instead of updating the RTC we calculate the offset in nanoseconds to the
next point in time where now.tv_nsec will be equal 5e8. The calculated
offset is then converted to jiffies as these are the unit used by the
timer.
Hovewer timespec_to_jiffies() used here uses a ceil()-type rounding mode,
where the resulting value is rounded up. As a result the range of
now.tv_nsec when the timer will trigger is from 5e8 to 5e8 + TICK_NSEC
rather than the desired 5e8 - TICK_NSEC / 2 to 5e8 + TICK_NSEC / 2.
As a result if for example sync_cmos_clock() happens to be called at the
time when now.tv_nsec is between 5e8 + TICK_NSEC / 2 and 5e8 to 5e8 +
TICK_NSEC, it will simply be rescheduled HZ jiffies later, falling in the
same range of now.tv_nsec again. Similarly for cases offsetted by an
integer multiple of TICK_NSEC.
This change addresses the problem by subtracting TICK_NSEC / 2 from the
nanosecond offset to the next point in time where now.tv_nsec will be
equal 5e8, effectively shifting the following rounding in
timespec_to_jiffies() so that it produces a rounded-to-nearest result.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
After fixing the u32 thinko I sill had occasional hickups on ATI chipsets
with small deltas. There seems to be a delay between writing the compare
register and the transffer to the internal register which triggers the
interrupt. Reading back the value makes sure, that it hit the internal
match register befor we compare against the counter value.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Until the C1E patches arrived there where no users of periodic broadcast
before switching to oneshot mode. Now we need to trigger a possible
waiter for a periodic broadcast when switching to oneshot mode.
Otherwise we can starve them for ever.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The minimum reprogramming delta was hardcoded in HPET ticks,
which is stupid as it does not work with faster running HPETs.
The C1E idle patches made this prominent on AMD/RS690 chipsets,
where the HPET runs with 25MHz. Set it to 5us which seems to be
a reasonable value and fixes the problems on the bug reporters
machines. We have a further sanity check now in the clock events,
which increases the delta when it is not sufficient.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br> Tested-by: Dmitry Nezhevenko <dion@inhex.net> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The C1E/HPET bug reports on AMDX2/RS690 systems where tracked down to a
too small value of the HPET minumum delta for programming an event.
The clockevents code needs to enforce an interrupt event on the clock event
device in some cases. The enforcement code was stupid and naive, as it just
added the minimum delta to the current time and tried to reprogram the device.
When the minimum delta is too small, then this loops forever.
Add a sanity check. Allow reprogramming to fail 3 times, then print a warning
and double the minimum delta value to make sure, that this does not happen again.
Use the same function for both tick-oneshot and tick-broadcast code.
While chasing the C1E/HPET bugreports I went through the clock events
code inch by inch and found that the broadcast device can be initialized
and shutdown multiple times. Multiple shutdowns are not critical, but
useless waste of time. Multiple initializations are simply broken. Another
CPU might have the device in use already after the first initialization and
the second init could just render it unusable again.
In tick_oneshot_setup we program the device to the given next_event,
but we do not check the return value. We need to make sure that the
device is programmed enforced so the interrupt handler engine starts
working. Split out the reprogramming function from tick_program_event()
and call it with the device, which was handed in to tick_setup_oneshot().
Set the force argument, so the devices is firing an interrupt.
The reprogramming of the periodic broadcast handler was broken,
when the first programming returned -ETIME. The clockevents code
stores the new expiry value in the clock events device next_event field
only when the programming time has not been elapsed yet. The loop in
question calculates the new expiry value from the next_event value
and therefor never increases.
There is a ordering related problem with clockevents code, due to which
clockevents_register_device() called after tickless/highres switch
will not work. The new clockevent ends up with clockevents_handle_noop as
event handler, resulting in no timer activity.
The problematic path seems to be
* old device already has hrtimer_interrupt as the event_handler
* new clockevent device registers with a higher rating
* tick_check_new_device() is called
* clockevents_exchange_device() gets called
* old->event_handler is set to clockevents_handle_noop
* tick_setup_device() is called for the new device
* which sets new->event_handler using the old->event_handler which is noop.
Change the ordering so that new device inherits the proper handler.
This does not have any issue in normal case as most likely all the clockevent
devices are setup before the highres switch. But, can potentially be affecting
some corner case where HPET force detect happens after the highres switch.
This was a problem with HPET in MSI mode code that we have been experimenting
with.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Signed-off-by: Shaohua Li <shaohua.li@intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This is happening because of an improper usage of strcmp() in the
e820 parsing code. The strcmp() always returns !0 and never resets the
value for e820.nr_map and returns an incorrect user-defined map.
When EC is in Polling mode, OS will check the EC status continually by using
the following source code:
clear_bit(EC_FLAGS_WAIT_GPE, &ec->flags);
while (time_before(jiffies, delay)) {
if (acpi_ec_check_status(ec, event))
return 0;
msleep(1);
}
But msleep is realized by the function of schedule_timeout. At the same time
although one process is already waken up by some events, it won't be scheduled
immediately. So maybe there exists the following phenomena:
a. The current jiffies is already after the predefined jiffies.
But before timeout happens, OS has no chance to check the EC
status again.
b. If preemptible schedule is enabled, maybe preempt schedule will happen
before checking loop. When the process is resumed again, maybe
timeout already happens, which means that OS has no chance to check
the EC status.
In such case maybe EC status is already what OS expects when timeout happens.
But OS has no chance to check the EC status and regards it as AE_TIME.
So it will be more appropriate that OS will try to check the EC status again
when timeout happens. If the EC status is what we expect, it won't be regarded
as timeout. Only when the EC status is not what we expect, it will be regarded
as timeout, which means that EC controller can't give a response in time.
There is a race with dirty page accounting where a page may not properly
be accounted for.
clear_page_dirty_for_io() calls page_mkclean; then TestClearPageDirty.
page_mkclean walks the rmaps for that page, and for each one it cleans and
write protects the pte if it was dirty. It uses page_check_address to
find the pte. That function has a shortcut to avoid the ptl if the pte is
not present. Unfortunately, the pte can be switched to not-present then
back to present by other code while holding the page table lock -- this
should not be a signal for page_mkclean to ignore that pte, because it may
be dirty.
For example, powerpc64's set_pte_at will clear a previously present pte
before setting it to the desired value. There may also be other code in
core mm or in arch which do similar things.
The consequence of the bug is loss of data integrity due to msync, and
loss of dirty page accounting accuracy. XIP's __xip_unmap could easily
also be unreliable (depending on the exact XIP locking scheme), which can
lead to data corruption.
Fix this by having an option to always take ptl to check the pte in
page_check_address.
It's possible to retain this optimization for page_referenced and
try_to_unmap.
Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Jared Hulbert <jaredeh@gmail.com> Cc: Carsten Otte <cotte@freenet.de> Cc: Hugh Dickins <hugh@veritas.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Cc: Chuck Ebbert <cebbert@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
> Now some warnings:
>
> ------------[ cut here ]------------
> WARNING: at /uio/arkimedes/s29/vegardno/git-working/linux-2.6/kernel/smp.c:328 s
> mp_call_function_mask+0x194/0x1a0()
The usual problem: the suspend function when interrupts are
already disabled calls smp_call_function which is not allowed with
interrupt off. But at this point all the other CPUs should be already
down anyways, so it should be enough to just drop that.
This patch should fix that problem at least by fixing cpu hotplug&
suspend support.
The fdiv detection code writes s32 integer into
the boot_cpu_data.fdiv_bug.
However, the boot_cpu_data.fdiv_bug is only char (s8)
field so the detection overwrites already set fields for
other bugs, e.g. the f00f bug field.
Use local s32 variable to receive result.
This is a partial fix to Bugzilla #9928 - fixes wrong
information about the f00f bug (tested) and probably
for coma bug (I have no cpu to test this).
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Cc: Chuck Ebbert <cebbert@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Remove the rt2x00 singlethreaded workqueue and move
the link tuner and packet filter scheduled work to
the ieee80211_hw->workqueue again.
The only exception is the interface scheduled work
handler which uses the mac80211 interface iterator
under the RTNL lock. This work needs to be handled
on the kernel workqueue to prevent lockdep issues.
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
vsmp_patch has been marked with __init ever since pvops, however,
apply_paravirt can be called during module load causing calls to
freed memory location.
Since apply_paravirt can only be called during bootup and module load,
mark vsmp patch with "__init_or_module"
The callers of sg_copy_buffer must disable interrupts before calling
it (since it uses kmap_atomic). Some callers use it on
interrupt-disabled code but some need to take the trouble to disable
interrupts just for this. No wonder they forget about it and we hit a
bug like:
http://bugzilla.kernel.org/show_bug.cgi?id=11529
James said that it might be better to disable interrupts inside the
function rather than risk the callers getting it wrong.
The ocfs2_stack_driver_request() function failed to increment the
refcount of an already-active stack. It only did the increment on the
first reference. Whoops.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Tested-by: Marcos Matsunaga <marcos.matsunaga@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Spencer reported a problem where utime and stime were going negative despite
the fixes in commit b27f03d4bdc145a09fb7b0c0e004b29f1ee555fa. The suspected
reason for the problem is that signal_struct maintains it's own utime and
stime (of exited tasks), these are not updated using the new task_utime()
routine, hence sig->utime can go backwards and cause the same problem
to occur (sig->utime, adds tsk->utime and not task_utime()). This patch
fixes the problem
TODO: using max(task->prev_utime, derived utime) works for now, but a more
generic solution is to implement cputime_max() and use the cputime_gt()
function for comparison.
Reported by Milan Broz <mbroz@redhat.com>, commit 18ce3751 inadvertently
made submit_bh() discard the barrier bit for a WRITE_SYNC request. Fix
that up.
After investigating a JRE failure, I found this bug was introduced a
long time ago, and had already managed to survive another bugfix which
occurred on the same line. The result is a total failure of the JRE due
to LDT selectors not working properly.
This one took a long time to rear up because LDT usage is not very
common, but the bug is quite serious. It got introduced along with
another bug, already fixed, by 75b8bb3e56ca09a467fbbe5229bc68627f7445be
On Thu, Jul 24, 2008 at 03:43:44PM -0700, Linus Torvalds wrote:
> So how about this patch as a starting point? This is the RightThing(tm) to
> do regardless, and if it then makes it easier to do some other cleanups,
> we should do it first. What do you think?
restore_fpu_checking() calls init_fpu() in error conditions.
While this is wrong(as our main intention is to clear the fpu state of
the thread), this was benign before commit 92d140e21f1 ("x86: fix taking
DNA during 64bit sigreturn").
Post commit 92d140e21f1, live FPU registers may not belong to this
process at this error scenario.
In the error condition for restore_fpu_checking() (especially during the
64bit signal return), we are doing init_fpu(), which saves the live FPU
register state (possibly belonging to some other process context) into
the thread struct (through unlazy_fpu() in init_fpu()). This is wrong
and can leak the FPU data.
For the signal handler restore error condition in restore_i387(), clear
the fpu state present in the thread struct(before ultimately sending a
SIGSEGV for badframe).
For the paranoid error condition check in math_state_restore(), send a
SIGSEGV, if we fail to restore the state.
Suresh Siddha wants to fix a possible FPU leakage in error conditions,
but the fact that save/restore_i387() are inlines in a header file makes
that harder to do than necessary. So start off with an obvious cleanup.
This just moves the x86-64 version of save/restore_i387() out of the
header file, and moves it to the only file that it is actually used in:
arch/x86/kernel/signal_64.c. So exposing it in a header file was wrong
to begin with.
[ Side note: I'd like to fix up some of the games we play with the
32-bit version of these functions too, but that's a separate
matter. The 32-bit versions are shared - under different names
at that! - by both the native x86-32 code and the x86-64 32-bit
compatibility code ]
Accesses to CR4 are intercepted even with Nested Paging enabled. But the code
does not check if the guest wants to do a global TLB flush. So this flush gets
lost. This patch adds the check and the flush to svm_set_cr4.
This patch introduces a guest TLB flush on every NPF exit in KVM. This fixes
random segfaults and #UD exceptions in the guest seen under some workloads
(e.g. long running compile workloads or tbench). A kernbench run with and
without that fix showed that it has a slowdown lower than 0.5%
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@qumranet.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The PCM and rawmidi open callbacks have a lock against card->controls_list
but it takes a wrong one, card->controls_rwsem, instead of a right one
card->ctl_files_rwlock. This patch fixes them.
This change also fixes automatically the potential deadlocks due to
mm->mmap_sem in munmap and copy_from/to_user, reported by Sitsofe
Wheeler:
When changing the sample rate, the CMI8788's master clock output becomes
unstable for a short time. The AK4396 needs the master clock to do SPI
writes, so writing to an AK4396 control register directly after a sample
rate change will garble the value. In our case, this leads to the DACs
being misconfigured to I2S sample format, which results in a wrong
output level and horrible distortions on samples louder than -6 dB.
To fix this, we need to wait until the new master clock signal has
become stable before doing SPI writes.
Dell Inspiron 1525 seems to have a buggy BIOS setup and screws up
the recent codec parser, as reported by Oleksandr Natalenko:
http://lkml.org/lkml/2008/9/12/203
This patch adds the working model, dell-3stack, statically.
Josip Rodin noted
(http://article.gmane.org/gmane.linux.ports.sparc/10152) the
driver oopsing during registration of an rport to the
FC-transport layer with a backtrace indicating a dereferencing of
an shost->shost_data equal to NULL. David Miller identified a
small window in driver logic where this could happen:
> Look at how the driver registers the IRQ handler before the host has
> been registered with the SCSI layer.
>
> That leads to a window of time where the shost hasn't been setup
> fully, yet ISRs can come in and trigger DPC thread events, such as
> loop resyncs, which expect the transport area to be setup.
>
> But it won't be setup, because scsi_add_host() hasn't finished yet.
>
> Note that in Josip's crash log, we don't even see the
>
> qla_printk(KERN_INFO, ha, "\n"
> " QLogic Fibre Channel HBA Driver: %s\n"
> " QLogic %s - %s\n"
> " ISP%04X: %s @ %s hdma%c, host#=%ld, fw=%s\n",
> ...
>
> message yet.
>
> Which means that the crash occurs between qla2x00_request_irqs()
> and printing that message.
Close this window by enabling RISC interrupts after the host has
been registered with the SCSI midlayer.
Reported-by: Josip Rodin <joy@entuzijast.net> Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Commit de85422b94ddb23c021126815ea49414047c13dc, 'USB: fix interrupt
disabling for HCDs with shared interrupt handlers' changed usb_add_hcd()
to strip IRQF_DISABLED from irqflags prior to calling request_irq()
with the justification that such a removal was necessary for shared
interrupts to work properly. Unfortunately, the change in that commit
unconditionally removes the IRQF_DISABLED flag, causing problems on
platforms that don't use a shared interrupt but require IRQF_DISABLED.
This change adds a check for IRQF_SHARED prior to removing the
IRQF_DISABLED flag.
Fixes the PS3 system startup hang reported with recent Fedora and
OpenSUSE kernels.
Note that this problem is hidden when CONFIG_LOCKDEP=y (ps3_defconfig),
as local_irq_enable_in_hardirq() is defined as a null statement for
that config.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Stefan Becker <Stefan.Becker@nokia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
CC drivers/spi/pxa2xx_spi.o
drivers/spi/pxa2xx_spi.c: In function 'map_dma_buffers':
drivers/spi/pxa2xx_spi.c:331: error: invalid operands to binary &
drivers/spi/pxa2xx_spi.c:331: error: invalid operands to binary &
drivers/spi/pxa2xx_spi.c: In function 'pump_transfers':
drivers/spi/pxa2xx_spi.c:897: warning: format '%lu' expects type 'long unsigned int', but argument 4 has type 'unsigned int'
[dbrownell@users.sourceforge.net: fix warning too ]
Signed-off-by: Mike Rapoport <mike@compulab.co.il> Acked-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixes several chipselect bugs in the pxa2xx_spi driver. These bugs are in
all versions of this driver and prevent using it with chips like m25p16
flash.
1. The spi_transfer.cs_change flag is handled too early:
before spi_transfer.delay_usecs applies, thus making the
delay ineffective at holding chip select.
2. spi_transfer.delay_usecs is ignored on the last transfer
of a message (likewise not holding chipselect long enough).
3. If spi_transfer.cs_change is set on the last transfer, the
chip select is always disabled, instead of the intended
meaning: optionally holding chip select enabled for the
next message.
Those first three bugs were fixed with a relocation of delays
and chip select de-assertions.
4. If a message has the cs_change flag set on the last transfer,
and had the chip select stayed enabled as requested (see 3,
above), it would not have been disabled if the next message is
for a different chip. Fixed by dropping chip select regardless
of cs_change at end of a message, if there is no next message
or if the next message is for a different chip.
This patch should apply to all kernels back to and including 2.6.20;
it was test patched against 2.6.20. An additional patch would be
required for older kernels, but those versions are very buggy anyway.
Signed-off-by: Ned Forrester <nforrester@whoi.edu> Cc: Vernon Sauder <vernoninhand@gmail.com> Cc: Eric Miao <eric.y.miao@gmail.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixes two DMA bugs in the pxa2xx_spi driver. The first bug is in all
versions of this driver; the second was introduced in the 2.6.20 kernel,
and prevents using the driver with chips like m25p16 flash (which can
issue large DMA reads).
1. Zero length transfers are permitted for use to insert timing,
but pxa2xx_spi.c will fail if this is requested in DMA mode.
Fixed by using programmed I/O (PIO) mode for such transfers.
2. Transfers larger than 8191 are not permitted in DMA mode. A
test for length rejects all large transfers regardless of DMA
or PIO mode. Worked around by rejecting only large transfers
with DMA mapped buffers, and forcing all other transfers
larger than 8191 to use PIO mode. A rate limited warning is
issued for DMA transfers forced to PIO mode.
This patch should apply to all kernels back to and including 2.6.20;
it was test patched against 2.6.20. An additional patch would be
required for older kernels, but those versions are very buggy anyway.
Signed-off-by: Ned Forrester <nforrester@whoi.edu> Cc: Vernon Sauder <vernoninhand@gmail.com> Cc: Eric Miao <eric.y.miao@gmail.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The iterator for_each_zone_zonelist() uses a struct zoneref *z cursor when
scanning zonelists to keep track of where in the zonelist it is. The
zoneref that is returned corresponds to the the next zone that is to be
scanned, not the current one. It was intended to be treated as an opaque
list.
When the page allocator is scanning a zonelist, it marks elements in the
zonelist corresponding to zones that are temporarily full. As the
zonelist is being updated, it uses the cursor here;
if (NUMA_BUILD)
zlc_mark_zone_full(zonelist, z);
This is intended to prevent rescanning in the near future but the zoneref
cursor does not correspond to the zone that has been found to be full.
This is an easy misunderstanding to make so this patch corrects the
problem by changing zoneref cursor to be the current zone being scanned
instead of the next one.
Should clear the next pointer of the TX if we are sure that the
next TX (say NXT) will be submitted to the channel too. Overwise,
we break the chain of descriptors, because we lose the information
about the next descriptor to run. So next time, when invoke
async_tx_run_dependencies() with TX, it's TX->next will be NULL, and
NXT will be never submitted.
Signed-off-by: Yuri Tikhonov <yur@emcraft.com> Signed-off-by: Ilya Yanok <yanok@emcraft.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
mmc_block_open() increments md->usage although it returns with -EROFS when
default mounting a MMC/SD card with write protect switch on. This
reference counting bug prevents /dev/mmcblkX from being released on card
removal, and situation worsen with reinsertion until the minor number
range runs out.
Reported-by: <sasin@solomon-systech.com> Acked-by: Pierre Ossman <drzeus-list@drzeus.cx> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The following patch adds it back. Without this the default value of 0
causes the performance of this card to be awful. Restoring these to the
default values yields much better performance.
This regression has been around since 2.6.25.
Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch is not needed in 2.6.27 because it has new c1e-aware idle code.
In kernel 2.6.26 the 32-bit x86 timers are started earlier than before.
This breaks AMD c1e detection trying to force timer broadcast for the
local apic timer. Copy the code from the 64-bit kernel to force timer
broadcast late.
sata_mv allowed issuing two DMA commands concurrently which the
hardware allows. Unfortunately, libata core layer isn't ready for
this yet and spews ugly warning message and malfunctions on this.
Don't allow concurrent DMA commands for now.
Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Cc: Martin Michlmayr <tbm@cyrius.com> Cc: Mark Lord <liml@rtr.ca> Cc: Artem Bokhan <aptem@ngs.ru> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Thanks to Johann Dahm and David Richter for bug report and testing.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: David Richter <richterd@citi.umich.edu> Tested-by: Johann Dahm <jdahm@umich.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Vegard Nossum reported
----------------------
> I noticed that something weird is going on with /proc/sys/sunrpc/transports.
> This file is generated in net/sunrpc/sysctl.c, function proc_do_xprt(). When
> I "cat" this file, I get the expected output:
> $ cat /proc/sys/sunrpc/transports
> tcp 1048576
> udp 32768
> But I think that it does not check the length of the buffer supplied by
> userspace to read(). With my original program, I found that the stack was
> being overwritten by the characters above, even when the length given to
> read() was just 1.
David Wagner added (among other things) that copy_to_user could be
probably used here.
Ingo Oeser suggested to use simple_read_from_buffer() here.
The conclusion is that proc_do_xprt doesn't check for userside buffer
size indeed so fix this by using Ingo's suggestion.
Reported-by: Vegard Nossum <vegard.nossum@gmail.com> Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com> CC: Ingo Oeser <ioe-lkml@rameria.de> Cc: Neil Brown <neilb@suse.de> Cc: Chuck Lever <chuck.lever@oracle.com> Cc: Greg Banks <gnb@sgi.com> Cc: Tom Tucker <tom@opengridcomputing.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The git commit 7c5026aa9b81dd45df8d3f4e0be73e485976a8b6 ("tg3: Add
link state reporting to UMP firmware") introduced code that waits for
previous firmware events to be serviced before attempting to submit a
new event. Unfortunately that patch contained a bug that cause the
driver to wait 2.5 seconds, rather than 2.5 milliseconds as intended.
This patch fixes that bug.
This bug revealed that not all firmware versions service driver events
though. Since we do not know which versions of the firmware do and don't
service these events, the driver needs some way to minimize the effects
of the delay. This patch solves the problem by recording a jiffies
timestamp when it submits an event to the hardware. If the jiffies
counter shows that 2.5 milliseconds have already passed, a wait is not
needed and the driver can proceed to submit a new event.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Authenc works in two stages for encryption, it first encrypts and
then computes an ICV. The context memory of the request is used
by both operations. The problem is that when an asynchronous
encryption completes, we will compute the ICV and then reread the
context memory of the encryption to get the original request.
It just happens that we have a buffer of 16 bytes in front of the
request pointer, so ICVs of 16 bytes (such as SHA1) do not trigger
the bug. However, any attempt to uses a larger ICV instantly kills
the machine when the first asynchronous encryption is completed.
This patch fixes this by saving the request pointer before we start
the ICV computation.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
First, lmb_enforce_memory_limit() interprets it's argument
(mostly, heh) as a size limit not an address limit. So pass
the raw cmdline_memory_size value into it. And we don't
need to check it against zero, lmb_enforce_memory_limit() does
that for us.
Next, free_initmem() needs special handling when the kernel
command line trims the available memory. The problem case is
if the trimmed out memory is where the kernel image itself
resides.
When that memory is trimmed out, we don't add those physical
ram areas to the sparsemem active ranges, amongst other things.
Which means that this free_initmem() code will free up invalid
page structs, resulting in either crashes or hangs.
Just quick fix this by not freeing initmem at all if "mem="
was given on the boot command line.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ever since commit 4c563f7669c10a12354b72b518c2287ffc6ebfb3
("[XFRM]: Speed up xfrm_policy and xfrm_state walking") it is
illegal to call __xfrm_state_destroy (and thus xfrm_state_put())
with xfrm_state_lock held. If we do, we'll deadlock since we
have the lock already and __xfrm_state_destroy() tries to take
it again.
Fix this by pushing the xfrm_state_put() calls after the lock
is dropped.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The number of identifiers needs to be checked against the option
length. Also, the identifier index provided needs to be verified
to make sure that it doesn't exceed the bounds of the array.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The bonds check to prevent buffer overlflow was not exactly
right. It still allowed overflow of up to 8 bytes which is
sizeof(struct sctp_authkey).
Since optlen is already checked against the size of that struct,
we are guaranteed not to cause interger overflow either.
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
which broke userpsace, in particular the XORP build as reported by
Jose Calhariz, the debain package maintainer for XORP.
Nothing originally in linux/mroute.h was exported to userspace
ever, but some of this stuff started to be when it was moved into
this new linux/pim.h, and that was wrong. If we didn't provide these
definitions for 10 years we can reasonable expect that applications
defined this stuff locally or used GLIBC headers providing the
protocol definitions. And as such the only result of this can
be conflict and userland build breakage.
The commit #1 had such a short and terse commit message, that we
cannot even know why such a move and set of new userland exports were
even made.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / karo-tx-linux.git / log