current code:

 delta = calc_delta_mine(delta_exec, curr->load.weight, lw);
 delta = min((u64)delta, cfs_rq->sleeper_bonus);

Notice that this calc_delta_mine() line is exactly delta_mine, which
gives:

 delta = min((u64)delta_mine, cfs_rq->sleeper_bonus);
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

current code:

 delta = min(cfs_rq->sleeper_bonus, (u64)delta_exec);
 delta = calc_delta_mine(delta, curr->load.weight, lw);
 delta = min((u64)delta, cfs_rq->sleeper_bonus);

drop the first min(), because we clip against sleeper_bonus in the 3rd line
again. That gives:

 delta = calc_delta_mine(delta_exec, curr->load.weight, lw);
 delta = min((u64)delta, cfs_rq->sleeper_bonus);
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

make the bonus balance more consistent: do not hand out a bonus if
there's too much in flight already, and only deduct as much from a
runner as it has the capacity. This makes the bonus engine a zero-sum
game (as intended).

this also simplifies the code:

   text    data     bss     dec     hex filename
  34770    2998      24   37792    93a0 sched.o.before
  34749    2998      24   37771    938b sched.o.after

and it also avoids overscheduling in sleep-happy workloads like
hackbench.c.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Mitchell Erblich suggested a quality-of-implementation change to
not requeue SCHED_RR tasks if there's only a single task on the
runqueue, by checking for rq->nr_running == 1.

provide a more efficient implementation of that, to check that
particular RT priority-queue only.

[ From: mingo@elte.hu ]

Also first requeue the task then set need_resched - results in slightly
better machine-instruction ordering. Also clean up the code a bit.
Signed-off-by: NDmitry Adamushko <dmitry.adamushko@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Remove trivial conditional branch in Linux scheduler's
can_migrate_task() function.

   text    data     bss     dec     hex filename
   34770    2998      24   37792    93a0 sched.o.before
   34757    2998      24   37779    9393 sched.o.after
Signed-off-by: NSven-Thorsten Dietrich <sven@thebigcorporation.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

remove HZ dependency from the granularity default. Use 10 msec for
the base granularity, 1 msec for wakeup granularity and 25 msec for
batch wakeup granularity. (These defaults are close to the values
that the default HZ=250 setting got previously, and thus it's the
most common setting.)
Signed-off-by: NIngo Molnar <mingo@elte.hu>

when I built with CONFIG_FAIR_GROUP_SCHED=y, I need the following change
to make things right.

[ From: mingo@elte.hu ]

this config option is not upstream-configurable right now but lets fix
this for completeness.
Signed-off-by: NBruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  libata: don't check n_sectors during revalidation if zero
  pata_via: Add Arima W730-K8 and other rebadgings
  pata_sis: Add the FSC Amilo and friends
  pata_pdc2027x: PLL detection fixes
  libata: fix n_sectors failure handling during revalidation

* master.kernel.org:/pub/scm/linux/kernel/git/lethal/sh-2.6.23:
  sh: missing symbol fix for sh4-202
  sh: Fix DSP opcode regression for SH3-DSP parts.

kmalloc() hands us a void pointer, we don't need to cast it.
Signed-off-by: NJesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'master' of ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb:
  V4L/DVB (6070): Fix a warning at dvb_net
  V4L/DVB (6042): b2c2-flexcop: fix Airstar HD5000 tuning regression

* git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched:
  sched: tweak the sched_runtime_limit tunable
  sched: skip updating rq's next_balance under null SD
  sched: fix broken SMT/MC optimizations
  sched: accounting regression since rc1
  sched: fix sysctl directory permissions
  sched: sched_clock_idle_[sleep|wakeup]_event()

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs:
  9p: fix bad error path in conversion routines
  9p: remove deprecated v9fs_fid_lookup_remove()
  9p: update maintainers and documentation
  9p: fix use after free

Renumber AUDIT_TTY_[GS]ET to avoid a conflict with netlink message types
already used in the wild.
Signed-off-by: NMiloslav Trmac <mitr@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc:
  sdhci: tell which spurious interrupt we got
  sdhci: handle data interrupts during command
  mmc: ignore bad max block size in sdhci
  sdhci: be more cautious about block count register
  drivers/mmc/core/host.c: kmalloc + memset conversion to kzalloc
  drivers/mmc/core/bus.c: kmalloc + memset conversion to kzalloc

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6:
  PCI: Run k8t_sound_hostbridge quirk only when needed
  PCI: disable MSI on RX790
  PCI: disable MSI on RD580
  PCI: disable MSI on RS690
  PCI: make pcie_get_readrq visible in pci.h
  PCI: lets kill the 'PCI hidden behind bridge' message
  pci/hotplug/cpqphp_ctrl.c: remove stale BKL use
  PCI: Document pci_iomap()
  PCI: quirk_e100_interrupt() called too early
  PCI: Move prototypes for pci_bus_find_capability to include/linux/pci.h

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6: (35 commits)
  usb: add PRODUCT, TYPE to usb-interface events
  USB: resubmission unusual_devs modification for Nikon D80
  usb quirks: Add Canon EOS 5D (PC Connection mode) to the autosuspend blacklist
  USB: make EHCI initialize properly on PPC SOCs
  UEAGLE: Remove sysfs files on error case
  USB: fsl_usb2_udc: fix bug in processing setup requests
  USB: g_file_storage: fix bug in DMA buffer handling
  USB: update last_busy field correctly
  USB: fix DoS in pwc USB video driver
  USB: allow retry on descriptor fetch errors
  USB: unkill cxacru atm driver
  USB: Adding support for HTC Smartphones to ipaq
  USB: another quirky device
  USB: quirky mass storage device
  USB: ohci, fix oddball gcc warning
  usb-storage: fix bugs in the disconnect pathway
  usb: typo in usb R8A66597 HCD config
  USB: accept 1-byte Device Status replies, fixing some b0rken devices
  USB: blacklist Samsung ML-2010 printer
  usb-serial: fix oti6858.c segfault in termios handling
  ...

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6:
  sysfs: don't warn on removal of a nonexistent binary file
  HOWTO: latest lxr url address changed
  HOWTO: korean translation of Documentation/HOWTO
  Fix Off-by-one in /sys/module/*/refcnt
  sysfs: fix locking in sysfs_lookup() and sysfs_rename_dir()

Include linux/fs.h to get the prototype for do_pipe().
Build fix.
Signed-off-by: NGreg Ungerer <gerg@uclinux.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Include asm-generic/pgtable.h to pick up the lazy_mmu_mode and
lazy_cpu_mode macros. Won't build without them.
Signed-off-by: NGreg Ungerer <gerg@uclinux.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Include linux/fs.h to get the prototype for getname().
Build fix.
Signed-off-by: NGreg Ungerer <gerg@uclinux.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Whoever did the PCI revision patch slipped up on the it821x, and I
didn't spot this at the time either.  They moved the check for the
errata from the 0x10 revision to 0x11.  Put it back

This one is important for 2.6.23 final as in some cases bad things will
occur if 0x10 revision boards don't get the fixups.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It uses get_futex_key().
Signed-off-by: NAlexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

static function dvb_net_sec declares input arg "pkt" as u8. However, the
same argument at dvb_net_sec_callback is defined as "const u8". When
calling dvb_net_sec, this is casted as just "u8".

gcc 4.2.1 generates a warning about that:

  CC [M]  drivers/media/dvb/dvb-core/dvb_net.o
drivers/media/dvb/dvb-core/dvb_net.c: In function "dvb_net_sec_callback":
drivers/media/dvb/dvb-core/dvb_net.c:905: warning: passing argument 2 of
		"dvb_net_sec" discards qualifiers from pointer target type
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>

Git changeset 6bdcc6e6 dropped the
stand-alone lgh06xf module, whose functionality was absorbed into the
dvb-pll module. However, there was a minor difference between the code
in lgh06xf and dvb-pll, which caused a regression in b2c2-flexcop
devices using the LG-H06xF NIM.

dvb-pll will probe for the presence of an i2c pll chip by performing a
single byte read, the lgh06xf driver did not do this. Unfortunately, the
code in flexcop-i2c.c does not currently support 1 byte or 0 byte reads
as a probe.  Such probes with the current code will always fail.

In order to work around this problem, and restore proper functionality
of the Airstar HD5000 device, this hack was created to make the probe
appear to succeed.  The single byte read in dvb_pll_attach is the only
place where such a probe would ever occur, so this change is safe, and
will not affect any other devices.

Of course, if one knew how to actually perform the read operation, it
would be better to go that route.  In the meantime, however, we must
apply this workaround, in order to prevent the regression that causes
tuning to fail on the Airstar HD5000 ATSC device.

Thanks to Jarod Wilson, who had originally reported this regression, and
to Geoffrey Hausheer, whose original workaround patch led us to find the
actual cause of the problem.
Signed-off-by: NTrent Piepho <xyzzy@speakeasy.org>
Cc: Geoffrey Hausheer <inli3epy93n@phracturedblue.com>
Acked-by: NJarod Wilson <jwilson@redhat.com>
Signed-off-by: NMichael Krufky <mkrufky@linuxtv.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>

When buf_check_overflow() returns != 0 we will hit kfree(ERR_PTR(err))
and it will not be happy about it.
Signed-off-by: NMariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

This patch removes the v9fs_fid_lookup_remove which is no longer used.

Based on original patch from Adrian Bunk <bunk@stusta.de> which
used #if 0 to isolate the code.
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

Updates to the MAINTAINERS file and documentation for 9p to point to the
swik wiki versus the outdated sf.net page.  Also updated some email addresses
and added pointers to papers which better describe the implementation and
application of the Linux 9p client.
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

On 7/22/07, Adrian Bunk <bunk@stusta.de> wrote:
     The Coverity checker spotted the following use-after-free
     in net/9p/mux.c:

     <--  snip  -->

     ...
     struct p9_conn *p9_conn_create(struct p9_transport *trans, int msize,
                                         unsigned char *extended)
     {
     ...
             if (!m->tagpool) {
                     kfree(m);
                     return ERR_PTR(PTR_ERR(m->tagpool));
             }
     ...

     <--  snip  -->

Also spotted was a leak of the same structure further down in the function.
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

Michael Gerdau reported reniced task CPU usage weirdnesses.
Such symptoms can be caused by limit underruns so double the
sched_runtime_limit.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Was playing with sched_smt_power_savings/sched_mc_power_savings and
found out that while the scheduler domains are reconstructed when sysfs
settings change, rebalance_domains() can get triggered with null domain
on other cpus, which is setting next_balance to jiffies + 60*HZ.
Resulting in no idle/busy balancing for 60 seconds.

Fix this.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

On a four package system with HT - HT load balancing optimizations were
broken.  For example, if two tasks end up running on two logical threads
of one of the packages, scheduler is not able to pull one of the tasks
to a completely idle package.

In this scenario, for nice-0 tasks, imbalance calculated by scheduler
will be 512 and find_busiest_queue() will return 0 (as each cpu's load
is 1024 > imbalance and has only one task running).

Similarly MC scheduler optimizations also get fixed with this patch.

[ mingo@elte.hu: restored fair balancing by increasing the fuzz and
                 adding it back to the power decision, without the /2
                 factor. ]
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Fix the accounting regression for CONFIG_VIRT_CPU_ACCOUNTING.  It
reverts parts of commit b27f03d4 by
converting fs/proc/array.c back to cputime_t.  The new functions
task_utime and task_stime now return cputime_t instead of clock_t.  If
CONFIG_VIRT_CPU_ACCOUTING is set, task->utime and task->stime are
returned directly instead of using sum_exec_runtime.

Patch is tested on s390x with and without VIRT_CPU_ACCOUTING as well as
on i386.

[ mingo@elte.hu: cleanups, comments. ]
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There are two remaining gotchas:

- The directories have impossible permissions (writeable).

- The ctl_name for the kernel directory is inconsistent with
  everything else.  It should be CTL_KERN.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

construct a more or less wall-clock time out of sched_clock(), by
using ACPI-idle's existing knowledge about how much time we spent
idling. This allows the rq clock to work around TSC-stops-in-C2,
TSC-gets-corrupted-in-C3 type of problems.

( Besides the scheduler's statistics this also benefits blktrace and
  printk-timestamps as well. )

Furthermore, the precise before-C2/C3-sleep and after-C2/C3-wakeup
callbacks allow the scheduler to get out the most of the period where
the CPU has a reliable TSC. This results in slightly more precise
task statistics.

the ACPI bits were acked by Len.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Acked-by: NLen Brown <len.brown@intel.com>

If the initial configuration fails early, n_sectors is left at zero.
Checking against it during revalidation makes retried configuration
fail due to n_sectors mismatch.  Ignore zero n_sectors during
revalidation.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

More cable funnies
Signed-off-by: NAlan Cox <alan@redhat.com>
Tested-by: NMikael Pettersson <mikpe@it.uu.se>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

More short cables
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

Previously I reported that the pata_pdc2027x PLL detection changes
in kernel 2.6.22 broke the driver on my PowerMac:

>pata_pdc2027x: Invalid PLL input clock 1691742kHz, give up!

This is followed by a number of errors and speed reduction
steps on the affected ports.

There are two bugs in pata_pdc2027x's PLL detection code:

1. The PLL counter's start value is read before the chip is
   put in "test mode". Outside of test mode the counter is
   halted, and on the PowerMac the counter is zero because
   the chip hasn't been initialised by its BIOS.

   The fix is to move the read of the start value to after
   test mode is started, but before the mdelay() in test mode.
   This also improves the precision of the PLL detection.

2. The code to compute the number of PLL decrements during the
   mdelay() in test mode fails to consider that the PLL counter
   only is 30 bits wide. If there is a wraparound, it will compute
   an incorrect and much too large value. On the PowerMac, the
   start count is zero, the end count is a large 30-bit value, so
   wraparound occurs and an out of bounds PLL clock is detected.

   The fix is to mask the (start - end) computation to 30 bits.

While debugging this I also noticed that pdc_read_counter()
reads the two halves of the 30-bit PLL counter as 16-bit values,
and then combines them as if the halves only are 15 bits wide.
To avoid confusion, the halves should be read as 15-bit values.

This patch implements all three changes. It fixes the PLL detection
failure on my PowerMac, and doesn't cause any regressions on an x86
with an identical card.
Signed-off-by: NMikael Pettersson <mikpe@it.uu.se>
Signed-off-by: NJeff Garzik <jeff@garzik.org>

If revalidation fails because device has different n_sectors after
configuration the original n_sectors should be restored before failing
revalidation.  Without this fix, n_sectors difference will incorrectly
and silently pass revalidation when revalidation is retried.
Signed-off-by: NTejun Heo <htejun@gmail.com>
Acked-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>