When CFQ dispatches requests forcefully due to a barrier or changing iosched,
it runs through all cfqq's dispatching requests and then expires each queue.
However, it does not activate a cfqq before flushing its IOs resulting in
using stale values for computing slice_used.
This patch fixes it by calling activate queue before flushing reuqests from
each queue.

This is useful mostly for barrier requests because when the iosched is changing
it really doesnt matter if we have incorrect accounting since we're going to
break down all structures anyway.

We also now expire the current timeslice before moving on with the dispatch
to accurately account slice used for that cfqq.

Signed-off-by: Divyesh <Shah&lt;dpshah@google.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Update mtime when writing to backing filesystem using the address space
operations write_begin and write_end.

Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Currently, the io statistics for the root cgroup are maintained, but
they are not shown because the device information is not available at
the point that the root blkio cgroup is created. This patch updates
the device information when the statistics are updated so that the
statistics become visible.

Signed-off-by: Ricky Benitez <rickyb@google.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

I hit this when we had a bug in IDR for a few days. Basically sysfs would
fail to create new inodes since it uses an IDR and therefore class_create would
fail.

While we are unlikely to see this fail we may as well handle it instead of
oopsing.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

elevator_get() not check the name length, if the name length > sizeof(elv),
elv will miss the '\0'. And elv buffer will be replace "-iosched" as something
like aaaaaaaaa, then call request_module() can load an not trust module.

Signed-off-by: Zhitong Wang <zhitong.wangzt@alibaba-inc.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

We take the spin_lock again in fail_all_cmds() so we need to unlock
here.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Even if they are found to be co-operating.

The prio_trees do not have any IDLE cfqqs on them. cfq_close_cooperator()
is called from cfq_select_queue() and cfq_completed_request(). The latter
ensures that the close cooperator code does not get invoked if the current
cfqq is of class IDLE but the former doesn't seem to have any such checks.
So an IDLE cfqq may get merged with a BE cfqq from the same group which
should be avoided.

Signed-off-by: Divyesh <Shah&lt;dpshah@google.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

These have helped us debug some issues we've noticed in earlier IO
controller versions and should be useful now as well. The extra logging
covers:
- idling behavior. Since there are so many conditions based on which we decide
to idle or not, this patch adds a log message for some conditions that we've
found useful.
- workload slices and current prio and workload type

Changelog from v1:
o moved log message from cfq_set_active_queue() to __cfq_set_active_queue()
o changed queue_count to st->count

Signed-off-by: Divyesh <Shah&lt;dpshah@google.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

This macro worked only when applied to variables named 'kobj'.
While this could have been fixed by simply renaming the macro argument,
a more type-safe replacement by an inline function would be preferred.
However, nobody uses this macro, so it's simpler to just remove it.

Signed-off-by: Ferenc Wagner <wferi@niif.hu>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Remove alignment padding to shrink struct request from 336 to 320 bytes
so needing one fewer cacheline and therefore removing 48 bytes from
struct request_queue.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Conflicts:
	block/Kconfig

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Alex Shi reported a kbuild regression which is about 10% performance lost.
He bisected to this commit: 3dde36dd.
The reason is cfqq_close() can't find close cooperator. Restoring
cfq_rq_close()'s threshold to original value makes the regression go away.

Since for_preempt parameter isn't used anymore, this patch deletes it.

Reported-by: Alex Shi <alex.shi@intel.com>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Make the config visible, so we can choose from CONFIG_BLK_CGROUP=y
and CONFIG_BLK_CGROUP=m when CONFIG_IOSCHED_CFQ=m.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

This flag is not used, so best discarded.

Signed-off-by: NeilBrown <neilb@suse.de>
--
Hi Jens,
 I came across this recently - these are the only two occurances
 of "GENHD_FL_DRIVERFS" in the kernel, so it cannot be needed.
NeilBrown
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

These two values are useful when debugging issues surrounding maximum
I/O size.  Put them in sysfs with the rest of the queue limits.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Remove compatibility wrappers and update remaining drivers.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

lcm() was defined to take integer-sized arguments.  The supplied
arguments are multiplied, however, causing us to overflow given
sufficiently large input.  That in turn led to incorrect optimal I/O
size reporting in some cases (RAID over RAID).

Switch lcm() over to unsigned long similar to gcd() and move the
function from blk-settings.c to lib.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6: (34 commits)
  ACPI: processor: push file static MADT pointer into internal map_madt_entry()
  ACPI: processor: refactor internal map_lsapic_id()
  ACPI: processor: refactor internal map_x2apic_id()
  ACPI: processor: refactor internal map_lapic_id()
  ACPI: processor: driver doesn't need to evaluate _PDC
  ACPI: processor: remove early _PDC optin quirks
  ACPI: processor: add internal processor_physically_present()
  ACPI: processor: move acpi_get_cpuid into processor_core.c
  ACPI: processor: export acpi_get_cpuid()
  ACPI: processor: mv processor_pdc.c processor_core.c
  ACPI: processor: mv processor_core.c processor_driver.c
  ACPI: plan to delete "acpi=ht" boot option
  ACPI: remove "acpi=ht" DMI blacklist
  PNPACPI: add bus number support
  PNPACPI: add window support
  resource: add window support
  resource: add bus number support
  resource: expand IORESOURCE_TYPE_BITS to make room for bus resource type
  acpiphp: Execute ACPI _REG method for hotadded devices
  ACPI video: Be more liberal in validating _BQC behaviour
  ...

Commit 6992f533 ("sysfs: Use one lockdep
class per sysfs attribute.") introduced this requirement.  First, at25
was fixed manually.  Then, other occurences were found with coccinelle
and the following semantic patch.  Results were reviewed and fixed up:

    @ init @
    identifier struct_name, bin;
    @@

    	struct struct_name {
    		...
    		struct bin_attribute bin;
    		...
    	};

    @ main extends init @
    expression E;
    statement S;
    identifier name, err;
    @@

    (
    	struct struct_name *name;
    |
    -	struct struct_name *name = NULL;
    +	struct struct_name *name;
    )
    	...
    (
    	sysfs_bin_attr_init(&name->bin);
    |
    +	sysfs_bin_attr_init(&name->bin);
    	if (sysfs_create_bin_file(E, &name->bin))
    		S
    |
    +	sysfs_bin_attr_init(&name->bin);
    	err = sysfs_create_bin_file(E, &name->bin);
    )

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Merge branches 'battery-2.6.34', 'bugzilla-10805', 'bugzilla-14668', 'bugzilla-531916-power-state', 'ht-warn-2.6.34', 'pnp', 'processor-rename', 'sony-2.6.34', 'suse-bugzilla-531547', 'tz-check', 'video' and 'misc-2.6.34' into release

There's no real need for a pointer to the MADT to be global. The only
function who uses it is map_madt_entry.

This allows us to remove some more ugly #ifdefs.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Un-nest the if statements for readability.

Remove comments that re-state the obvious.

Change the control flow so that we no longer need a temp variable.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Untangle the nested if conditions to make this function look
more similar to the other map_*apic_id() functions.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Untangle the if() statement a little for readability.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Now that the early _PDC evaluation path knows how to correctly
evaluate _PDC on only physically present processors, there's no
need for the processor driver to evaluate it later when it loads.

To cover the hotplug case, push _PDC evaluation down into the
hotplug paths.

Cc: x86@kernel.org
Cc: Tony Luck <tony.luck@intel.com>
Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Now that we check for physically present processors before blindly
evaluating _PDC, we no longer need to maintain a DMI opt-in table
nor a kernel param.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Detect if a processor is physically present before evaluating _PDC.

We want this because some BIOS will provide a _PDC even for processors
that are not present. These bogus _PDC methods then attempt to load
non-existent tables, which causes problems.

Avoid those bogus landmines.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Enumerating processors (via MADT/_MAT) belongs in the processor core,
which is always built-in, rather than living in the processor driver
which may not be built.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Rename static get_cpu_id() to acpi_get_cpuid() and export it.

This change also gives us an opportunity to remove the
#ifndef CONFIG_SMP from processor_driver.c and into a header file
where it properly belongs.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

We've renamed the old processor_core.c to processor_driver.c, to
convey the idea that it can be built modular and has driver-like
bits.

Now let's re-create a processor_core.c for the bits needed
statically by the rest of the kernel. The contents of processor_pdc.c
are a good starting spot, so let's just rename that file and
complete our three card monte.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

The ACPI processor driver can be built as a module. But it has
pieces of code that should always be built statically into the
kernel.

The plan is for processor_core.c to contain the static bits while
processor_driver.c contains the module-like bits.

Since the bulk of the code in the current processor_core.c is
module-like, first step is to rename the file to processor_driver.c

Next step will re-create processor_core.c and cherry-pick out
the static bits.

Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>

SuSE added these entries when deploying ACPI in Linux-2.4.
I pulled them into Linux-2.6 on 2003-08-09.
Over the last 6+ years, several entries have proven to be
unnecessary and deleted, while no new entries have been added.
Matthew suggests that they now have negative value, and I agree.

Based-on-patch-by: Matthew Garrett <mjg59@srcf.ucam.org>
Signed-off-by: Len Brown <len.brown@intel.com>

Add support for bus number resources.  This is for bridges with a range of
bus numbers behind them.  Previously, PNP ignored bus number resources.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Add support for resource windows.  This is for bridge resources, i.e.,
regions where a bridge forwards transactions from the primary to the
secondary side.  This does not add support for *setting* windows via
the /proc interface.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Add support for resource windows.  This is for bridge resources, i.e.,
regions where a bridge forwards transactions from the primary to the
secondary side.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

Add support for bus number resources.  This is for bridges with a range of
bus numbers behind them.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

No functional change; this just makes room for another resource type.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>

The original code returns a freed pointer.  This function is expected to
return NULL on errors.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <jmorris@namei.org>