There can 3 reasons for the "command reject" reply produced
by the stack. Each such reply should be accompanied by the
relevand data ( as defined in spec. ). Currently there is one
instance of "command reject" reply with reason "invalid cid"
wich is fixed. Also, added clean-up definitions related to the
"command reject" replies.
Signed-off-by: Ilia Kolomisnky <iliak@ti.com>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
at ext4_trim_all_free() comment, there is no longer an @e4b parameter,
instead it is @group.
Reported-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext4, when FITRIM is called every time, we iterate all the
groups and do trim one by one. It is a bit time wasting if the
group has been trimmed and there is no change since the last
trim.
So this patch adds a new flag in ext4_group_info->bb_state to
indicate that the group has been trimmed, and it will be cleared
if some blocks is freed(in release_blocks_on_commit). Another
trim_minlen is added in ext4_sb_info to record the last minlen
we use to trim the volume, so that if the caller provide a small
one, we will go on the trim regardless of the bb_state.
A simple test with my intel x25m ssd:
df -h shows:
/dev/sdb1 40G 21G 17G 56% /mnt/ext4
Block size: 4096
run the FITRIM with the following parameter:
range.start = 0;
range.len = UINT64_MAX;
range.minlen = 1048576;
without the patch:
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m5.505s
user 0m0.000s
sys 0m1.224s
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m5.359s
user 0m0.000s
sys 0m1.178s
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m5.228s
user 0m0.000s
sys 0m1.151s
with the patch:
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m5.625s
user 0m0.000s
sys 0m1.269s
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m0.002s
user 0m0.000s
sys 0m0.001s
[root@boyu-tm linux-2.6]# time ./ftrim /mnt/ext4/a
real 0m0.002s
user 0m0.000s
sys 0m0.001s
A big improvement for the 2nd and 3rd run.
Even after I delete some big image files, it is still much
faster than iterating the whole disk.
[root@boyu-tm test]# time ./ftrim /mnt/ext4/a
real 0m1.217s
user 0m0.000s
sys 0m0.196s
Cc: Lukas Czerner <lczerner@redhat.com>
Reviewed-by: Andreas Dilger <adilger.kernel@dilger.ca>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
When we trim some free blocks in a group of ext4, we need to
calculate the free blocks properly and check whether there are
enough freed blocks left for us to trim. Current solution will
only calculate free spaces if they are large for a trim which
isn't appropriate.
Let us see a small example:
a group has 1.5M free which are 300k, 300k, 300k, 300k, 300k.
And minblocks is 1M. With current solution, we have to iterate
the whole group since these 300k will never be subtracted from
1.5M. But actually we should exit after we find the first 2
free spaces since the left 3 chunks only sum up to 900K if we
subtract the first 600K although they can't be trimed.
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In 0f0a25b, we adjust 'len' with s_first_data_block - start, but
it could underflow in case blocksize=1K, fstrim_range.len=512 and
fstrim_range.start = 0. In this case, when we run the code:
len -= first_data_blk - start; len will be underflow to -1ULL.
In the end, although we are safe that last_group check later will limit
the trim to the whole volume, but that isn't what the user really want.
So this patch fix it. It also adds the check for 'start' like ext3 so that
we can break immediately if the start is invalid.
Cc: Lukas Czerner <lczerner@redhat.com>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Using function calls in TP_printk causes perf heartburn, so print the
MAJOR/MINOR device numbers instead.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
This patch adds support for permission checks using argv[]/envp[] of execve()
request. Hooks are in the last patch of this pathset.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <jmorris@namei.org>
This patch adds support for permission checks using executable file's realpath
upon execve() and symlink's target upon symlink(). Hooks are in the last patch
of this pathset.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <jmorris@namei.org>
This patch adds support for permission checks using file object's DAC
attributes (e.g. owner/group) when checking file's pathnames. Hooks for passing
file object's pointers are in the last patch of this pathset.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <jmorris@namei.org>
This patch adds support for permission checks using current thread's UID/GID
etc. in addition to pathnames.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <jmorris@namei.org>
Upon corrupted inode or disk failures, we may fail after we already
allocate some blocks from the inode or take some blocks from the
inode's preallocation list, but before we successfully insert the
corresponding extent to the extent tree. In this case, we should free
any allocated blocks and discard the inode's preallocated blocks
because the entries in the inode's preallocation list may be in an
inconsistent state.
Signed-off-by: Jiaying Zhang <jiayingz@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
The current implementation of ext4_free_blocks() always calls
dquot_free_block This looks quite sensible in the most cases: blocks
to be freed are associated with inode and were accounted in quota and
i_blocks some time ago.
However, there is a case when blocks to free were not accounted by the
time calling ext4_free_blocks() yet:
1. delalloc is on, write_begin pre-allocated some space in quota
2. write-back happens, ext4 allocates some blocks in ext4_ext_map_blocks()
3. then ext4_ext_map_blocks() gets an error (e.g. ENOSPC) from
ext4_ext_insert_extent() and calls ext4_free_blocks().
In this scenario, ext4_free_blocks() calls dquot_free_block() who, in
turn, decrements i_blocks for blocks which were not accounted yet (due
to delalloc) After clean umount, e2fsck reports something like:
> Inode 21, i_blocks is 5080, should be 5128. Fix<y>?
because i_blocks was erroneously decremented as explained above.
The patch fixes the problem by passing the new flag
EXT4_FREE_BLOCKS_NO_QUOT_UPDATE to ext4_free_blocks(), to request
that the dquot_free_block() call be skipped.
Signed-off-by: Maxim Patlasov <maxim.patlasov@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
This patch enables fetching configuration data, which is normally provided
via platform_data, from the device-tree instead.
If the device is configured from device-tree data, the platform_data struct
is not used, and button data needs to be allocated dynamically. Big part of
this patch deals with confining pdata usage to the probe function, to make
this possible.
Signed-off-by: David Jander <david@protonic.nl>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
I came across a memory leak during a cyclic cpu-online-offline test.
Signed-off-by: Yu Luming <luming.yu@intel.com>
Cc: Len Brown <lenb@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Jones <davej@redhat.com>
There is no consistency among filesystems from what bios (or requests)
are marked as being metadata. It's interesting to expose this in traces,
but we shouldn't schedule the requests differently based on whether or
not they're marked as being metadata.
Signed-off-by: Justin TerAvest <teravest@google.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/staging:
hwmon: (pmbus) Improve auto-detection of temperature status register
hwmon: (lm95241) Fix negative temperature results
hwmon: (lm95241) Fix chip detection code
It is possible that a PMBus device supports the READ_TEMPERATURE2 and/or
READ_TEMPERATURE3 registers but does not support READ_TEMPERATURE1.
Improve temperature status register detection to address this condition.
Reported-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
Cc: stable@kernel.org # 2.6.39+
Negative temperatures were returned in degrees C instead of milli-Degrees C.
Also, negative temperatures were reported for remote temperature sensors even
if the chip was configured for positive-only results.
Fix by detecting temperature modes, and by treating negative temperatures
similar to positive temperatures, with appropriate sign extension.
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
Cc: stable@kernel.org # 2.6.30+
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
ALSA: hda - Fix a copmile warning
ASoC: ak4642: fixup snd_soc_update_bits mask for PW_MGMT2
ALSA: hda - Change all ADCs for dual-adc switching mode for Realtek
ASoC: Manage WM8731 ACTIVE bit as a supply widget
ASoC: Don't set invalid name string to snd_card->driver field
ASoC: Ensure we delay long enough for WM8994 FLL to lock when starting
ASoC: Tegra: I2S: Ensure clock is enabled when writing regs
ASoC: Fix Blackfin I2S _pointer() implementation return in bounds values
ASoC: tlv320aic3x: Do soft reset to codec when going to bias off state
ASoC: tlv320aic3x: Don't sync first two registers from register cache
audio: tlv320aic26: fix PLL register configuration
* 'fixes' of master.kernel.org:/home/rmk/linux-2.6-arm:
ARM: 6994/1: smp_twd: Fix typo in 'twd_timer_rate' printing
ARM: 6987/1: l2x0: fix disabling function to avoid deadlock
ARM: 6966/1: ep93xx: fix inverted RTS/DTR signals on uart1
ARM: 6980/1: mmci: use StartBitErr to detect bad connections
ARM: 6979/1: mach-vt8500: add forgotten irq_data conversion
ARM: move memory layout sanity checking before meminfo initialization
ARM: 6990/1: MAINTAINERS: add entry for ARM PMU profiling and debugging
ARM: 6989/1: perf: do not start the PMU when no events are present
ARM: dmabounce: fix map_single() error return value
Rearrange struct sk_buff members comments to follow their
definition order. Also, add missing comments for ooo_okay
and dropcount members.
Signed-off-by: Daniel Baluta <dbaluta@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a debugfs node called "summary" to /sys/kernel/debug/clock/
that displays a quick summary of all clocks registered in the
"clocks" structure. The format of the output from this node is:
<clock-name> <parent-name> <rate> <usecount>
This debugfs node was very helpful for taking a quick snapshot of
the linux clock tree for OMAP and ensuring clock frequencies
calculated by the kernel were indeed correct. This patch helped
uncover some bugs in the linux clock tree for OMAP4.
Signed-off-by: Jon Hunter <jon-hunter@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
On OMAP4, the PRCM recommended sequence for enabling
a module after power-on-reset is:
-1- Force clkdm to SW_WKUP
-2- Enabling the clocks
-3- Configure desired module mode to "enable" or "auto"
-4- Wait for the desired module idle status to be FUNC
-5- Program clkdm in HW_AUTO(if supported)
This sequence applies to all older OMAPs' as well,
however since they use autodeps, it makes sure that
no clkdm is in IDLE, and hence not requiring a force
SW_WKUP when a module is being enabled.
OMAP4 does not need to support autodeps, because
of the dyanamic dependency feature, wherein
the HW takes care of waking up a clockdomain from
idle and hence the module, whenever an interconnect
access happens to the given module.
Implementing the sequence for OMAP4 requires
the clockdomain handling that is currently done in
clock framework to be done as part of hwmod framework
since the step -4- above to "Wait for the desired
module idle status to be FUNC" is done as part of
hwmod framework.
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
[b-cousson@ti.com: Adapt it to the new clkdm hwmod attribute and API]
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
[paul@pwsan.com: dropped mach-omap2/clock.c changes; modified to only
call the clockdomain code if oh->clkdm is set; disable clock->clockdomain
interaction on OMAP4]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The OMAP2/3 clock code was written to notify the clockdomain code when
the first clock in a clockdomain is enabled and when the last enabled
clock in a clockdomain is disabled. OMAP4 requires a different
approach: the hwmod code needs to signal the clockdomain code when to
force-enable and auto-idle a clockdomain during the IP block enable
process. The current conjecture is that once that hwmod sequence is
implemented, it will no longer be necessary for the clock code to call
into the clockdomain code for "optional clocks" on OMAP4.
Add a static flag to the OMAP2+ clock code, clkdm_control, that by
default preserves the OMAP2/3 behavior. Also add a function,
omap2_clk_disable_clkdm_control(), intended to be called from OMAP4
and beyond clock initcalls, that disables the old behavior.
Part of this patch was originally based on a patch by Rajendra Nayak
<rnayak@ti.com>.
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Benoît Cousson <b-cousson@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Since the clkdm state programming is now done from within the hwmod
framework (which uses a per-hwmod lock) instead of the being done
from the clock framework (which used a global lock), there is now a
need to have per-clkdm locking to prevent races between different
hwmods/modules belonging to the same clock domain concurrently
programming the clkdm state.
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The omap_set_pwrdm_state function forces clockdomains
to idle, without checking the existing idle state
programmed, instead based solely on the HW capability
of the clockdomain to support idle.
This is wrong and the clockdomains should be idled
post a state_switch *only* if idle transitions on the
clockdomain were already enabled.
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Add a new function, clkdm_in_hwsup(), that returns true if a clockdomain
is configured for hardware-supervised idle. It does not actually read the
hardware; rather, it checks an internal flag in the struct clockdomain, which
is changed when the clockdomain is switched in and out of hardware-supervised
idle. This should be safe, since all changes to the idle mode should
pass through the clockdomain code.
Based on a set of patches by Rajendra Nayak <rnayak@ti.com> which do
the same thing by checking the hardware bits. This approach should be
faster and more compact.
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Todd Poynor <toddpoynor@google.com>
Cc: Benoît Cousson <b-cousson@ti.com>
Duplicate the existing API for clockdomain enable from clock to enable
a clock domain from hwmod framework.
This will be needed when the hwmod framework will move from the current
clock centric approach to the module based approach.
These APIs are returning 0 for the moment for OMAP2 and OMAP3 until
their hwmods are updated with the clksm attribute.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The call to pwrdm_wait_transition() in clkdm_clk_enable()
is redundant since the function pwrdm_clkdm_state_switch()
which is called next also does the same thing.
Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Take advantage of the explicit modulemode control to fix
the way parents clocks are managed.
A module must be disabled before any parents are disabled.
That programming model was not possible with the previous
implementation that was considering a modulemode as a leaf
clock node managed by the clock fmwk.
This was leading to bad crash upon disable when the parent
clock was gated before the module completed its transition
to idle.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
In OMAP4, a new programming model based on module control instead
of clock control was introduced.
Expose two APIs to allow the upper layer (omap_hwmod) to control
the module mode independently of the parent clocks management.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: renamed 'omap4_cm_' fns to 'omap4_cminst_'; cleaned up
kerneldoc]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Add a new field to provide the mode supported by the module.
The mode will control the way mandatory clocks are managed by the PRCM.
0 : Module is temporarily disabled by SW. OCP access to module are stalled.
Can be used to change timing parameter of GPMC module.
1 : Module is managed automatically by HW according to clock domain
transition. A clock domain sleep transition put module into idle.
A wakeup domain transition put it back into function.
If CLKTRCTRL=3, any OCP access to module is always granted.
Module clocks may be gated according to the clock domain state.
2 : Module is explicitly enabled. Interface clock (if not used for
functions) may be gated according to the clock domain state.
Functional clocks are guarantied to stay present. As long as
in this configuration, power domain sleep transition cannot happen.
Some modules will have a modulemode initialized at 1 (HWCTRL) by default.
This is the case for interconnect and simple module like GPIO, WDT, MAILBOX.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Add a 'context_offs' entry in the prcm.omap4 structure to all
IPs when applicable.
The offset will be used to retrieve the per module context lost
information now available on OMAP4.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The new prminst_xxx accessors based on partition and offset
is now used, so removed all the previous prcm_xxx accessors.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: remove fn prototypes also]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The warm reset function was still using the obsolete API.
Replace it by the new one and move the file to the proper c file.
Change the function names to stick to the file convention as
suggested by Paul Walmsley <paul@pwsan.com>:
prm_xxx -> prminst_xxx
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The RSTCTRL register was accessed using an absolute address.
The usage of hardcoded macros to calculate virtual address from physical
one should be avoided as much as possible.
The usage of an offset will allow future improvement like migration from
the current architecture code toward a module driver.
Update prm_xxx accessors, move definition to the proper header file and
update copyrights.
Change the s16 register offset parameter to u16.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: use '_prminst_' in function names that are part of the
prminst44xx.c file]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
It is mandatory to wait for a module to be in disabled state before
potentially disabling source clock or re-asserting a reset.
omap_hwmod_idle and omap_hwmod_shutdown does not wait for
the module to be fully idle.
Add a cm_xxx accessor to wait the clkctrl idle status to be disabled.
Fix hwmod_[idle|shutdown] to use this API.
Based on Rajendra's initial patch.
Please note that most interconnects hwmod will return one timeout because
it is impossible for them to be in idle since the processor is accessing
the registers though the interconnect.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Todd Poynor <toddpoynor@google.com>
[paul@pwsan.com: move cpu_is_*() tests to the top of _wait_target_disable();
incorporate some feedback from Todd]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
The CLKCTRL register was accessed using an absolute address.
The usage of hardcoded macros to calculate virtual address from physical
one should be avoided as much as possible.
The usage of a offset will allow future improvement like migration from
the current architecture code toward a module driver.
Update cm_xxx accessor, move definition to the proper header file and
update copyrights.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Todd Poynor <toddpoynor@google.com>
[paul@pwsan.com: renamed 'omap4_cm_' fns to 'omap4_cminst_'; removed empty
fn prototype section from cm44xx.h; incorporated comments from Todd;
documented some functions]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
At boot time, lookup the clkdm_name to get the clkdm
structure pointer for further usage.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
In OMAP PRCM terminology, the clock domain is defined as a group of IPs
that share some clocks and most of the time an interface clock.
Every IP does belong to a clockdomain.
For the moment the clock domain attribute is affected to a clock node.
The issue with that approach, is that a clock might or not belong to a
clock domain. Moreover during module transition, it is up to a module
to handle properly the clock domain state and not to a clock node.
Create a clkdm_name attribute to provide this information per hwmod.
Populate this attribute for every OMAP4 hwmod entries.
Future cleanup series with remove that information from the OMAP4 clock
when it is relevant.
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: fix the mpuss_clkdm name]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
