linux-toradex.git, branch v3.10.57

Linux 3.10.57

2014-10-09T19:18:54+00:00

cpufreq: ondemand: Change the calculation of target frequency

2014-10-09T19:18:43+00:00

commit dfa5bb622555d9da0df21b50f46ebdeef390041b upstream.

The ondemand governor calculates load in terms of frequency and
increases it only if load_freq is greater than up_threshold
multiplied by the current or average frequency.  This appears to
produce oscillations of frequency between min and max because,
for example, a relatively small load can easily saturate minimum
frequency and lead the CPU to the max.  Then, it will decrease
back to the min due to small load_freq.

Change the calculation method of load and target frequency on the
basis of the following two observations:

 - Load computation should not depend on the current or average
   measured frequency.  For example, absolute load of 80% at 100MHz
   is not necessarily equivalent to 8% at 1000MHz in the next
   sampling interval.

 - It should be possible to increase the target frequency to any
   value present in the frequency table proportional to the absolute
   load, rather than to the max only, so that:

   Target frequency = C * load

   where we take C = policy->cpuinfo.max_freq / 100.

Tested on Intel i7-3770 CPU @ 3.40GHz and on Quad core 1500MHz Krait.
Phoronix benchmark of Linux Kernel Compilation 3.1 test shows an
increase ~1.5% in performance. cpufreq_stats (time_in_state) shows
that middle frequencies are used more, with this patch.  Highest
and lowest frequencies were used less by ~9%.

[rjw: We have run multiple other tests on kernels with this
 change applied and in the vast majority of cases it turns out
 that the resulting performance improvement also leads to reduced
 consumption of energy.  The change is additionally justified by
 the overall simplification of the code in question.]

Signed-off-by: Stratos Karafotis 
Acked-by: Viresh Kumar 
Signed-off-by: Rafael J. Wysocki 
Cc: Mark Brown 
Signed-off-by: Greg Kroah-Hartman

cpufreq: Fix wrong time unit conversion

2014-10-09T19:18:43+00:00

commit a857c0b9e24e39fe5be82451b65377795f9538d8 upstream.

The time spent by a CPU under a given frequency is stored in jiffies unit
in the cpu var cpufreq_stats_table->time_in_state[i], i being the index of
the frequency.

This is what is displayed in the following file on the right column:

     cat /sys/devices/system/cpu/cpuX/cpufreq/stats/time_in_state
     2301000 19835820
     2300000 3172
     [...]

Now cpufreq converts this jiffies unit delta to clock_t before returning it
to the user as in the above file. And that conversion is achieved using the API
cputime64_to_clock_t().

Although it accidentally works on traditional tick based cputime accounting, where
cputime_t maps directly to jiffies, it doesn't work with other types of cputime
accounting such as CONFIG_VIRT_CPU_ACCOUNTING_* where cputime_t can map to nsecs
or any granularity preffered by the architecture.

For example we get a buggy zero delta on full dyntick configurations:

     cat /sys/devices/system/cpu/cpuX/cpufreq/stats/time_in_state
     2301000 0
     2300000 0
     [...]

Fix this with using the proper jiffies_64_t to clock_t conversion.

Reported-and-tested-by: Carsten Emde 
Signed-off-by: Andreas Schwab 
Signed-off-by: Frederic Weisbecker 
Acked-by: Paul E. McKenney 
Signed-off-by: Rafael J. Wysocki 
Cc: Mark Brown 
Signed-off-by: Greg Kroah-Hartman

nl80211: clear skb cb before passing to netlink

2014-10-09T19:18:43+00:00

commit bd8c78e78d5011d8111bc2533ee73b13a3bd6c42 upstream.

In testmode and vendor command reply/event SKBs we use the
skb cb data to store nl80211 parameters between allocation
and sending. This causes the code for CONFIG_NETLINK_MMAP
to get confused, because it takes ownership of the skb cb
data when the SKB is handed off to netlink, and it doesn't
explicitly clear it.

Clear the skb cb explicitly when we're done and before it
gets passed to netlink to avoid this issue.

Reported-by: Assaf Azulay 
Reported-by: David Spinadel 
Signed-off-by: Johannes Berg 
Signed-off-by: Greg Kroah-Hartman

drbd: fix regression 'out of mem, failed to invoke fence-peer helper'

2014-10-09T19:18:43+00:00

commit bbc1c5e8ad6dfebf9d13b8a4ccdf66c92913eac9 upstream.

Since linux kernel 3.13, kthread_run() internally uses
wait_for_completion_killable().  We sometimes may use kthread_run()
while we still have a signal pending, which we used to kick our threads
out of potentially blocking network functions, causing kthread_run() to
mistake that as a new fatal signal and fail.

Fix: flush_signals() before kthread_run().

Signed-off-by: Philipp Reisner 
Signed-off-by: Lars Ellenberg 
Signed-off-by: Jens Axboe 
Signed-off-by: Greg Kroah-Hartman

jiffies: Fix timeval conversion to jiffies

2014-10-09T19:18:43+00:00

commit d78c9300c51d6ceed9f6d078d4e9366f259de28c upstream.

timeval_to_jiffies tried to round a timeval up to an integral number
of jiffies, but the logic for doing so was incorrect: intervals
corresponding to exactly N jiffies would become N+1. This manifested
itself particularly repeatedly stopping/starting an itimer:

setitimer(ITIMER_PROF, &val, NULL);
setitimer(ITIMER_PROF, NULL, &val);

would add a full tick to val, _even if it was exactly representable in
terms of jiffies_ (say, the result of a previous rounding.)  Doing
this repeatedly would cause unbounded growth in val.  So fix the math.

Here's what was wrong with the conversion: we essentially computed
(eliding seconds)

jiffies = usec  * (NSEC_PER_USEC/TICK_NSEC)

by using scaling arithmetic, which took the best approximation of
NSEC_PER_USEC/TICK_NSEC with denominator of 2^USEC_JIFFIE_SC =
x/(2^USEC_JIFFIE_SC), and computed:

jiffies = (usec * x) >> USEC_JIFFIE_SC

and rounded this calculation up in the intermediate form (since we
can't necessarily exactly represent TICK_NSEC in usec.) But the
scaling arithmetic is a (very slight) *over*approximation of the true
value; that is, instead of dividing by (1 usec/ 1 jiffie), we
effectively divided by (1 usec/1 jiffie)-epsilon (rounding
down). This would normally be fine, but we want to round timeouts up,
and we did so by adding 2^USEC_JIFFIE_SC - 1 before the shift; this
would be fine if our division was exact, but dividing this by the
slightly smaller factor was equivalent to adding just _over_ 1 to the
final result (instead of just _under_ 1, as desired.)

In particular, with HZ=1000, we consistently computed that 10000 usec
was 11 jiffies; the same was true for any exact multiple of
TICK_NSEC.

We could possibly still round in the intermediate form, adding
something less than 2^USEC_JIFFIE_SC - 1, but easier still is to
convert usec->nsec, round in nanoseconds, and then convert using
time*spec*_to_jiffies.  This adds one constant multiplication, and is
not observably slower in microbenchmarks on recent x86 hardware.

Tested: the following program:

int main() {
  struct itimerval zero = {{0, 0}, {0, 0}};
  /* Initially set to 10 ms. */
  struct itimerval initial = zero;
  initial.it_interval.tv_usec = 10000;
  setitimer(ITIMER_PROF, &initial, NULL);
  /* Save and restore several times. */
  for (size_t i = 0; i < 10; ++i) {
    struct itimerval prev;
    setitimer(ITIMER_PROF, &zero, &prev);
    /* on old kernels, this goes up by TICK_USEC every iteration */
    printf("previous value: %ld %ld %ld %ld\n",
           prev.it_interval.tv_sec, prev.it_interval.tv_usec,
           prev.it_value.tv_sec, prev.it_value.tv_usec);
    setitimer(ITIMER_PROF, &prev, NULL);
  }
    return 0;
}


Cc: Thomas Gleixner 
Cc: Ingo Molnar 
Cc: Paul Turner 
Cc: Richard Cochran 
Cc: Prarit Bhargava 
Reviewed-by: Paul Turner 
Reported-by: Aaron Jacobs 
Signed-off-by: Andrew Hunter 
[jstultz: Tweaked to apply to 3.17-rc]
Signed-off-by: John Stultz 
[bwh: Backported to 3.16: adjust filename]
Signed-off-by: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

md/raid5: disable 'DISCARD' by default due to safety concerns.

2014-10-09T19:18:43+00:00

commit 8e0e99ba64c7ba46133a7c8a3e3f7de01f23bd93 upstream.

It has come to my attention (thanks Martin) that 'discard_zeroes_data'
is only a hint.  Some devices in some cases don't do what it
says on the label.

The use of DISCARD in RAID5 depends on reads from discarded regions
being predictably zero.  If a write to a previously discarded region
performs a read-modify-write cycle it assumes that the parity block
was consistent with the data blocks.  If all were zero, this would
be the case.  If some are and some aren't this would not be the case.
This could lead to data corruption after a device failure when
data needs to be reconstructed from the parity.

As we cannot trust 'discard_zeroes_data', ignore it by default
and so disallow DISCARD on all raid4/5/6 arrays.

As many devices are trustworthy, and as there are benefits to using
DISCARD, add a module parameter to over-ride this caution and cause
DISCARD to work if discard_zeroes_data is set.

If a site want to enable DISCARD on some arrays but not on others they
should select DISCARD support at the filesystem level, and set the
raid456 module parameter.
    raid456.devices_handle_discard_safely=Y

As this is a data-safety issue, I believe this patch is suitable for
-stable.
DISCARD support for RAID456 was added in 3.7

Cc: Shaohua Li 
Cc: "Martin K. Petersen" 
Cc: Mike Snitzer 
Cc: Heinz Mauelshagen 
Acked-by: Martin K. Petersen 
Acked-by: Mike Snitzer 
Fixes: 620125f2bf8ff0c4969b79653b54d7bcc9d40637
Signed-off-by: NeilBrown 
[bwh: Backported to 3.10: adjust context]
Signed-off-by: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

media: vb2: fix VBI/poll regression

2014-10-09T19:18:42+00:00

commit 58d75f4b1ce26324b4d809b18f94819843a98731 upstream.

The recent conversion of saa7134 to vb2 unconvered a poll() bug that
broke the teletext applications alevt and mtt. These applications
expect that calling poll() without having called VIDIOC_STREAMON will
cause poll() to return POLLERR. That did not happen in vb2.

This patch fixes that behavior. It also fixes what should happen when
poll() is called when STREAMON is called but no buffers have been
queued. In that case poll() will also return POLLERR, but only for
capture queues since output queues will always return POLLOUT
anyway in that situation.

This brings the vb2 behavior in line with the old videobuf behavior.

Signed-off-by: Hans Verkuil 
Acked-by: Laurent Pinchart 
Signed-off-by: Mauro Carvalho Chehab 
Signed-off-by: Greg Kroah-Hartman

mm: numa: Do not mark PTEs pte_numa when splitting huge pages

2014-10-09T19:18:42+00:00

commit abc40bd2eeb77eb7c2effcaf63154aad929a1d5f upstream.

This patch reverts 1ba6e0b50b ("mm: numa: split_huge_page: transfer the
NUMA type from the pmd to the pte"). If a huge page is being split due
a protection change and the tail will be in a PROT_NONE vma then NUMA
hinting PTEs are temporarily created in the protected VMA.

 VM_RW|VM_PROTNONE
|-----------------|
      ^
      split here

In the specific case above, it should get fixed up by change_pte_range()
but there is a window of opportunity for weirdness to happen. Similarly,
if a huge page is shrunk and split during a protection update but before
pmd_numa is cleared then a pte_numa can be left behind.

Instead of adding complexity trying to deal with the case, this patch
will not mark PTEs NUMA when splitting a huge page. NUMA hinting faults
will not be triggered which is marginal in comparison to the complexity
in dealing with the corner cases during THP split.

Signed-off-by: Mel Gorman 
Acked-by: Rik van Riel 
Acked-by: Kirill A. Shutemov 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, thp: move invariant bug check out of loop in __split_huge_page_map

2014-10-09T19:18:42+00:00

commit f8303c2582b889351e261ff18c4d8eb197a77db2 upstream.

In __split_huge_page_map(), the check for page_mapcount(page) is
invariant within the for loop.  Because of the fact that the macro is
implemented using atomic_read(), the redundant check cannot be optimized
away by the compiler leading to unnecessary read to the page structure.

This patch moves the invariant bug check out of the loop so that it will
be done only once.  On a 3.16-rc1 based kernel, the execution time of a
microbenchmark that broke up 1000 transparent huge pages using munmap()
had an execution time of 38,245us and 38,548us with and without the
patch respectively.  The performance gain is about 1%.

Signed-off-by: Waiman Long 
Acked-by: Kirill A. Shutemov 
Cc: Andrea Arcangeli 
Cc: Mel Gorman 
Cc: Rik van Riel 
Cc: Scott J Norton 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman