Paul Walmsley [Sun, 23 Sep 2012 23:28:23 +0000 (17:28 -0600)]
ARM: OMAP3: hwmod data: add mmu data for iva and isp
Add mmu hwmod data for iva and isp.
Due to compatibility an ifdef CONFIG_OMAP_IOMMU_IVA2 needs to be
propagated (previously on iommu resource info) to hwmod data in OMAP3,
so users of iommu and tidspbridge can avoid issues of two modules
managing mmu data/irqs/resets; this until tidspbridge can be migrated
to iommu framework.
Cc: Benoit Cousson <b-cousson@ti.com> Signed-off-by: Omar Ramirez Luna <omar.luna@linaro.org>
[paul@pwsan.com: fixed some kerneldoc and whitespace; ISP MMUs not present
on AM35xx so restricted these hwmods to 34xx/36xx] Signed-off-by: Paul Walmsley <paul@pwsan.com>
ARM: OMAP: iommu: fix including iommu.h without IOMMU_API selected
If included without IOMMU_API being selected it will break
compilation:
arch/arm/plat-omap/include/plat/iommu.h:
In function 'dev_to_omap_iommu':
arch/arm/plat-omap/include/plat/iommu.h:148:
error: 'struct dev_archdata' has no member named 'iommu'
This will be seen when hwmod includes iommu.h to get the
structure for attributes. Also needed for tidspbridge
incremental migration to use iommu code.
Cc: Tony Lindgren <tony@atomide.com> Signed-off-by: Omar Ramirez Luna <omar.luna@linaro.org> Signed-off-by: Paul Walmsley <paul@pwsan.com>
ARM: OMAP4: hwmod data: make *phy_48m* as the main_clk of ocp2scp
Made *ocp2scp_usb_phy_phy_48m* as the main_clk for ocp2scp.
Since this ocp2scp module does not have any fck but does have a
single opt_clock, it is added as the main_clk for ocp2scp. Also
removed phy_48m as the optional clock since it is now made as the
main clock. By this the driver need not enable/disable phy_48m clk
separately and runtime_get/runtime_put will take care of that.
Cc: Benoît Cousson <b-cousson@ti.com> Reviewed-by: Felipe Balbi <balbi@ti.com> Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com> Acked-by: Benoît Cousson <b-cousson@ti.com> Signed-off-by: Paul Walmsley <paul@pwsan.com>
Tero Kristo [Sun, 23 Sep 2012 23:28:21 +0000 (17:28 -0600)]
ARM: OMAP3: hwmod data: add sad2d hwmod
SAD2D stands for the die to die interface, and is used for communicating
with the optional stacked modem. This hwmod is added in preparation for
the d2d_idle move from pm34xx.c to hwmod data.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
[paul@pwsan.com: SAD2D presumably doesn't exist on non-OMAP34xx/OMAP36xx,
so only add it to the OMAP34xx/OMAP36xx lists] Signed-off-by: Paul Walmsley <paul@pwsan.com>
For a reset sequence to complete cleanly, a module needs its
associated clocks to be enabled, otherwise the timeout check
in prcm code can print a false failure (failed to hardreset)
that occurs because the clocks aren't powered ON and the status
bit checked can't transition without them.
Signed-off-by: Omar Ramirez Luna <omar.luna@linaro.org> Signed-off-by: Paul Walmsley <paul@pwsan.com>
ARM: OMAP: hwmod: partially un-reset hwmods might not be properly enabled
Some IP blocks might not be using/controlling more than one
reset line, this check loosens the restriction to fully use
hwmod framework for those drivers.
E.g.: ipu has reset lines: mmu_cache, cpu0 and cpu1.
- As of now cpu1 is not used and hence (with previous check) the
IP block isn't fully enabled by hwmod code.
- Usually ipu and dsp processors configure their mmu module first
and then enable the processors, this involves:
* Deasserting mmu reset line, and enabling the module.
* Deasserting cpu0 reset line, and enabling the processor.
The ones portrayed in this example are controlled through
rproc_fw_boot in drivers/remoteproc/remoteproc_core.c
While at it, prevent _omap4_module_disable if all the hardreset
lines on an IP block are not under reset.
This will allow the driver to:
a. Deassert the reset line.
b. Enable the hwmod through runtime PM default callbacks.
c. Do its usecase.
d. Disable hwmod through runtime PM.
e. Assert the reset line.
Signed-off-by: Omar Ramirez Luna <omar.luna@linaro.org>
[paul@pwsan.com: updated to apply] Signed-off-by: Paul Walmsley <paul@pwsan.com>
Tero Kristo [Sun, 23 Sep 2012 23:28:20 +0000 (17:28 -0600)]
ARM: OMAP4: hwmod: flag hwmods/modules not supporting module level context status
On OMAP4 most modules/hwmods support module level context status. On
OMAP3 and earlier, we relied on the power domain level context status.
Identify all modules that don't support 'context_offs' by adding a
flag bit, HWMOD_OMAP4_NO_CONTEXT_LOSS_BIT. Rest have a valid
'context_offs' populated in .prcm structure already.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
[paul@pwsan.com: add flag bit rather than overloading .context_offs;
update changelog message] Signed-off-by: Paul Walmsley <paul@pwsan.com>
Tero Kristo [Sun, 23 Sep 2012 23:28:19 +0000 (17:28 -0600)]
ARM: OMAP4: hwmod data: add support for lostcontext_mask
Currently hwmod only provides the offset for the context lose
register, and if we attempt to share the same register between two or
more hwmods, the resulting context loss counts get wrong. Thus, we
need a way to specify which bits are used for the context loss
information for each. This is accomplished by adding a new field to
the omap4 prcm struct, 'lostcontext_mask', which specifies a bit-mask
to use for filtering the register.
Mark the affected hwmods appropriately. 'l4_abe' hwmod uses the
LOSTMEM_AESSMEM bit of RM_ABE_AESS_CONTEXT register, as l4_abe doesn't
have its own dedicated register for this purpose. This register is
shared with 'aess' hwmod, thus both hwmods must also specify which
bits of the register are used for them.
This patch only adds the hwmod data, but a future patch should add
code support such that only the specified bits are read and cleared by
the context lose counter update code. If a hwmod doesn't specify
'lostcontext_mask' (default behavior), the whole contents of the
context register should be used without any filtering.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
[paul@pwsan.com: updated to apply after conversion to use flag bit for
missing module context-loss register; combined data and code patches;
dropped code change due to serial driver breakage] Signed-off-by: Paul Walmsley <paul@pwsan.com>
Tero Kristo [Sun, 23 Sep 2012 23:28:19 +0000 (17:28 -0600)]
ARM: OMAP4: powerdomain: add support for reading prev logic and mem states
On OMAP4, there is no support to read previous logic state
or previous memory state achieved when a power domain transitions
to RET. Instead there are module level context registers.
In order to support the powerdomain level logic/mem_off_counters
on OMAP4, instead use the previous power state achieved (RET) and
the *programmed* logic/mem RET state to derive if a powerdomain lost
logic or did not.
If the powerdomain is programmed to enter RET state and lose logic
in RET state, knowing that the powerdomain entered RET is good enough
to derive that the logic was lost as well, in such cases.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
[paul@pwsan.com: removed dependency on functional power state series for now;
bumped copyright date] Signed-off-by: Paul Walmsley <paul@pwsan.com>
The decision was made a few months ago to allow struct omap_hwmod
records and struct clk records to omit clockdomain information if the
clockdomain is not software-controllable. See for example commit 868c157df9721675c19729eed2c96bac6c3f1d01 ("ARM: OMAP2+: hwmod: remove
prm_clkdm, cm_clkdm; allow hwmods to have no clockdomain").
So convert an existing pr_warning() to a pr_debug() (regarding missing
clockdomains in clocks), and add a pr_debug() for missing hwmod
clockdomains. It's still useful to enable these messages for
debugging, since missing clockdomains can cause hard-to-debug problems
with power management; see for example commit 6c4a057bffe9823221eab547e11fac181dc18a2b ("ARM: OMAP4: clock data:
Force a DPLL clkdm/pwrdm ON before a relock").
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Benoît Cousson <b-cousson@ti.com>
Paul Walmsley [Sun, 23 Sep 2012 23:28:18 +0000 (17:28 -0600)]
ARM: OMAP4+: hwmod code: remove clkdm requirement in _omap4_wait_target_*()
We're no longer requiring struct omap_hwmod records to contain a
clockdomain. So we shouldn't return -EINVAL any more from
_omap4_wait_target_disable() or _omap4_wait_target_ready() if there's
no clockdomain defined, since that just gets passed back to the
caller. This can result in pointless warnings under the relaxed data
format.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Benoît Cousson <b-cousson@ti.com>
ARM: OMAP2+: omap_device: expose hwmod assert/deassert to omap devices
This API is meant to be an interface to hwmod assert/deassert
function, omap devices can call them through their platform data to
control their reset lines, they are expected to know the name of the
reset line they are trying to control.
Signed-off-by: Omar Ramirez Luna <omar.luna@linaro.org>
[paul@pwsan.com: tweaked some documentation; fixed CodingStyle issue] Signed-off-by: Paul Walmsley <paul@pwsan.com>
Igor Grinberg [Sun, 23 Sep 2012 23:28:17 +0000 (17:28 -0600)]
ARM: OMAP: hwmod code: remove unused hwmod function prototypes
Several hwmod function prototypes appear to not have an implementation
because the corresponding functions were removed or renamed.
Those prototypes are unneeded anymore - remove them.
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
[paul@pwsan.com: tweaked subject] Signed-off-by: Paul Walmsley <paul@pwsan.com>
Paul Walmsley [Sun, 23 Sep 2012 23:15:44 +0000 (17:15 -0600)]
Merge tag 'cleanup-fixes-for-v3.7' into test_v3.6-rc6_ocb3.7_cff3.7_odaf3.7
These fixes are needed to fix non-omap build breakage for
twl-core driver and to fix omap1_defconfig compile when
led driver changes and omap sparse IRQ changes are merged
together. Also fix warnings for omaps not using pinctrl
framework yet.
Paul Walmsley [Sun, 23 Sep 2012 23:15:11 +0000 (17:15 -0600)]
Merge tag 'omap-cleanup-b-for-3.7' into test_v3.6-rc6_ocb3.7_cff3.7_odaf3.7
smatch and string-wrapping cleanups for the OMAP subarch code.
These changes fix some of the more meaningful warnings that smatch
returns for the OMAP subarch code, and unwraps strings that are
wrapped at the 80-column boundary, to conform with the current
practice.
Basic build, boot, and PM logs are available here:
The new common clk framework includes basic definitions for mux and
divider clocks. These definitions depend on shift and width values
instead of the pre-computed masks that the OMAP/AM33XX clk framework
has traditionally used when accessing the register to control the
mux or divisor.
To ease this transition the masks are left intact and
the width field is simply added alongside the shift and mask data.
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com> Cc: Rajendra Nayak <rnayak@ti.com> Cc: Paul Walmsley <paul@pwsan.com> Cc: Mike Turquette <mturquette@ti.com> Signed-off-by: Paul Walmsley <paul@pwsan.com>
Mike Turquette [Thu, 20 Sep 2012 00:04:14 +0000 (18:04 -0600)]
ARM: OMAP4: cm: add bitfield width values
The new common clk framework includes basic definitions for mux and
divider clocks. These definitions depend on shift and width values
instead of the pre-computed masks that the OMAP clk framework has
traditionally used when accessing the register to control the mux or
divisor.
To ease this transition the masks are left intact and the width field is
simply added alongside the shift and mask data.
Signed-off-by: Mike Turquette <mturquette@ti.com> Signed-off-by: Rajendra Nayak <rnayak@ti.com> Signed-off-by: Paul Walmsley <paul@pwsan.com>
ARM: OMAP2+: clock: Remove all direct dereferencing of struct clk
While we move to Common Clk Framework (CCF), direct deferencing of struct
clk wouldn't be possible anymore. Hence get rid of all such instances
in the current clock code and use macros/helpers similar to the ones that
are provided by CCF.
While here also concatenate some strings split across multiple lines
which seem to be needed anyway.
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: simplified some compound expressions; reformatted some
messages] Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Mike Turquette <mturquette@linaro.org>
ARM: OMAP2+: hwmod: get rid of all omap_clk_get_by_name usage
Moving to Common clk framework for OMAP would mean we no longer use
internal lookup mechanism like omap_clk_get_by_name().
get rid of all its usage mostly from hwmod and omap_device
code.
Moving to clk_get() also means the respective platforms
need the clkdev tables updated with an entry for all clocks
used by hwmod to have clock name same as the alias.
Based on original changes from Mike Turquette.
Signed-off-by: Rajendra Nayak <rnayak@ti.com> Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
[paul@pwsan.com: removed IS_ERR_OR_NULL() conversion (rmk comment);
restricted omap_96m_alwon_fck_3630 to OMAP36xx; added missing AM35xx
clock aliases for emac_fck, emac_ick, vpfe_ick, vpfe_fck; added
aliases rng_ick and several emulation clocks] Signed-off-by: Paul Walmsley <paul@pwsan.com>
As part of Common Clk Framework (CCF) the clk_enable() operation
was split into a clk_prepare() which could sleep, and a clk_enable()
which should never sleep. Similarly the clk_disable() was
split into clk_disable() and clk_unprepare(). This was
needed to handle complex cases where in a clk gate/ungate
would require a slow and a fast part to be implemented.
None of the clocks below seem to be in the 'complex' clocks
category and are just simple clocks which are enabled/disabled
through simple register writes.
Most of the instances also seem to be called in non-atomic
context which means its safe to move all of those from
using a clk_enable() to clk_prepare_enable() and clk_disable() to
clk_disable_unprepare().
For some others, mainly the ones handled through the hwmod framework
there is a possibility that they get called in either an atomic
or a non-atomic context.
The way these get handled below work only as long as clk_prepare
is implemented as a no-op (which is the case today) since this gets
called very early at boot while most subsystems are unavailable.
Hence these are marked with a *HACK* comment, which says we need
to re-visit these once we start doing something meaningful with
clk_prepare/clk_unprepare like doing voltage scaling or something
that involves i2c.
This is in preparation of OMAP moving to CCF.
Based on initial changes from Mike Turquette.
Signed-off-by: Rajendra Nayak <rnayak@ti.com> Signed-off-by: Paul Walmsley <paul@pwsan.com>
Paul Walmsley [Sat, 22 Sep 2012 08:24:16 +0000 (02:24 -0600)]
SPI: OMAP: remove unnecessary includes of plat/clock.h
Remove unnecessary includes of plat/clock.h from the OMAP SPI
controller drivers. These need to be removed to build multi-subarch
ARM kernels which include these drivers.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Grant Likely <grant.likely@secretlab.ca> Acked-by: Grant Likely <grant.likely@secretlab.ca>
Jon Hunter [Sat, 22 Sep 2012 08:24:15 +0000 (02:24 -0600)]
ARM: OMAP4: Add timer clock aliases for device-tree
For OMAP4, the dmtimers are located in the Wake-up, ABE and Peripheral (PER)
power domains. Hence, when the dmtimer is configured to use the "timer_sys_ck"
as its functional clock the actual clock used is different depending on whether
the clock is in the Wake-up, ABE or PER domain. So when we look-up the dmtimer's
"timer_sys_ck" we need to specify the timer device name as well as clock alias
to find the right clock.
Currently, the device names for the timers have the format "omap_timer.X" where
X is the timer instance number. When using to device tree, the format of the
device name created by device-tree is different and has the format
"<reg-address>.<device-name>" (this is assuming that the device-tree "reg"
property is specified). This causes the look-up for the OMAP4 "timer_sys_ck" to
fail. To fix this add new timer clock alias for using device-tree.
Please note that adding a 2nd set of clock aliases for the same clocks to only
temporary until device-tree migration is complete. Then we can remove the legacy
aliases. Hence, I have marked the legacy aliases with a "TODO" to remove them.
Signed-off-by: Jon Hunter <jon-hunter@ti.com>
[paul@pwsan.com: updated to apply] Signed-off-by: Paul Walmsley <paul@pwsan.com>
ARM: OMAP2+: AM33XX: clock data: Add clkdev alias for cpu0
Add AM335x cpu0 clock entry to the corresponding clock data file. This
is useful in getting the correct mpu clock pointer to change the cpu
frequency in cpufreq driver.
Paul Walmsley [Sat, 22 Sep 2012 08:24:14 +0000 (02:24 -0600)]
ARM: OMAP2+: clock data: add some aliases for use by CPUFreq only
These clkdev aliases should make it possible to remove the
cpu_is_omap*() calls and the omap_device*() call from
drivers/cpufreq/omap-cpufreq.c during the next merge window. Those
are interfering with multi-subarch ARM kernels.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Kevin Hilman <khilman@ti.com> Cc: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Kevin Hilman <khilman@ti.com>
Jon Hunter [Sat, 22 Sep 2012 08:24:14 +0000 (02:24 -0600)]
ARM: OMAP2420: Cosmetic fix for timer clock aliases
In commit c59b537 (ARM: OMAP2+: Simplify dmtimer clock aliases) new clock
aliases were added for OMAP2+ devices. For OMAP2420, I incorrectly set the
clock flag as CK_243X instead of CK_242X. This did not introduce a regression
as the clock flags are not checked for OMAP2 devices. This also explains why
I did not catch this when testing on OMAP2420.
Fix the clock flags for these aliases for correctness.
Paul Walmsley [Sat, 22 Sep 2012 16:00:11 +0000 (10:00 -0600)]
Merge tag 'cleanup-fixes-for-v3.7' into test_v3.6-rc6_cff3.7_odaf3.7
These fixes are needed to fix non-omap build breakage for
twl-core driver and to fix omap1_defconfig compile when
led driver changes and omap sparse IRQ changes are merged
together. Also fix warnings for omaps not using pinctrl
framework yet.
Tony Lindgren [Mon, 17 Sep 2012 23:24:20 +0000 (16:24 -0700)]
ARM: OMAP1: Include gpio-omap.h for board-h2 and board-h3
Merge of the LED related changes with omap sparse IRQ and
hardware.h related changes causes a build issue otherwise:
arch/arm/mach-omap1/board-h2.c:319: error: implicit declaration of function ‘OMAP_MPUIO’
arch/arm/mach-omap1/board-h2.c:319: error: initializer element is not constant
arch/arm/mach-omap1/board-h2.c:319: error: (near initialization for ‘h2_gpio_led_pins[1].gpio’)
Matt Porter [Mon, 17 Sep 2012 23:26:11 +0000 (16:26 -0700)]
ARM: OMAP2+: Enable pinctrl dummy states
Enable pinctrl dummy states for all OMAP platforms that don't
populate DT. This allows drivers to be converted to pinctrl
and not generate new warnings on platforms that do not provide
pinctrl data. These platforms already have pinmuxes configured
before the drivers probe.
Signed-off-by: Matt Porter <mporter@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
Tony Lindgren [Mon, 17 Sep 2012 23:26:10 +0000 (16:26 -0700)]
mfd: Fix compile for twl-core.c by removing cpu_is_omap usage
Commit 7d7e1eba (ARM: OMAP2+: Prepare for irqs.h removal) broke
compile for non-omap as include plat/cpu.h was added. This header
was indirectly included earlier when SPARSE_IRQ was not set, but
does not exist on most platforms.
Fix the problem by removing the cpu_is_omap usage that should
not exist in drivers at all. We can do this by adding proper
clock aliases for the twl-core.c drivers, and drop separate
handling for cases when clock framework is not available as
the behaviour will stay the same.
Note that we need to add a platform device to avoid using the
i2c provided names that may be different on various omaps.
Reported-by: Fengguang Wu <fengguang.wu@intel.com> Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Paul Walmsley <paul@pwsan.com> Acked-by: Samuel Ortiz <sameo@linux.intel.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Cc: Olof Johansson <olof@lixom.net> Cc: Felipe Balbi <balbi@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
Merge tag 'mfd-for-linus-3.6-2' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6
Pull mfd fixes from Samuel Ortiz:
"This is the remaining MFD fixes for 3.6, with 5 pending fixes:
- A tps65217 build error fix.
- A lcp_ich regression fix caused by the MFD driver failing to
initialize the watchdog sub device due to ACPI conflicts.
- 2 MAX77693 interrupt handling bug fixes.
- An MFD core fix, adding an IRQ domain argument to the MFD device
addition API in order to prevent silent and potentially harmful
remapping behaviour changes for drivers supporting non-DT
platforms."
* tag 'mfd-for-linus-3.6-2' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6:
mfd: MAX77693: Fix NULL pointer error when initializing irqs
mfd: MAX77693: Fix interrupt handling bug
mfd: core: Push irqdomain mapping out into devices
mfd: lpc_ich: Fix a 3.5 kernel regression for iTCO_wdt driver
mfd: Move tps65217 regulator plat data handling to regulator
Merge tag 'for-3.6-rc6' of git://gitorious.org/linux-pwm/linux-pwm
Pull pwm fixes from Thierry Reding:
"While this comes a bit later than I had wished, both patches are
rather minor and touch only new drivers so I think these are still
safe for merging."
* tag 'for-3.6-rc6' of git://gitorious.org/linux-pwm/linux-pwm:
pwm: pwm-tiehrpwm: Fix conflicting channel period setting
pwm: pwm-tiecap: Disable APWM mode after configure
Pull scsi target fixes from Nicholas Bellinger:
"Here is the current set of target-pending fixes headed for v3.6-final
The main parts of this series include bug-fixes from Paolo Bonzini to
address an use-after-free bug in pSCSI sense exception handling, along
with addressing some long-standing bugs wrt the handling of zero-
length SCSI CDB payloads also specific to pSCSI pass-through device
backends."
* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
target: go through normal processing for zero-length REQUEST_SENSE
target: support zero allocation length in REQUEST SENSE
target: support zero-size allocation lengths in transport_kmap_data_sg
target: fail REPORT LUNS with less than 16 bytes of payload
target: report too-small parameter lists everywhere
target: go through normal processing for zero-length PSCSI commands
target: fix use-after-free with PSCSI sense data
target: simplify code around transport_get_sense_data
target: move transport_get_sense_data
target: Check idr_get_new return value in iscsi_login_zero_tsih_s1
target: Fix ->data_length re-assignment bug with SCSI overflow
Merge tag 'pm-for-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael J. Wysocki:
"Three ACPI device power management fixes related to checking and
setting device power states."
* tag 'pm-for-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI / PM: Use KERN_DEBUG when no power resources are found
ACPI / PM: Fix resource_lock dead lock in acpi_power_on_device
ACPI / PM: Infer parent power state from child if unknown, v2
Merge tag 'sound-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull more sound fixes from Takashi Iwai:
"Yet more (a bunch of) small fixes that slipped from the previous pull
request. Most of commits are pending ASoC fixes, all of which are
fairly trivial commits."
* tag 'sound-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ASoC: wm8904: correct the index
ALSA: hda - Yet another position_fix quirk for ASUS machines
ASoC: tegra: fix maxburst settings in dmaengine code
ASoC: samsung dma - Don't indicate support for pause/resume.
ASoC: mc13783: Remove mono support
ASoC: arizona: Fix typo in 44.1kHz rates
ASoC: spear: correct the check for NULL dma_buffer pointer
sound: tegra_alc5632: remove HP detect GPIO inversion
ASoC: atmel-ssc: include linux/io.h for raw io
ASoC: dapm: Don't force card bias level to be updated
ASoC: dapm: Make sure we update the bias level for CODECs with no op
ASoC: am3517evm: fix error return code
ASoC: ux500_msp_i2s: better use devm functions and fix error return code
ASoC: imx-sgtl5000: fix error return code
Nikolay Ulyanitsky reported thatthe 3.6-rc5 kernel has a 15-20%
performance drop on PostgreSQL 9.2 on his machine (running "pgbench").
Borislav Petkov was able to reproduce this, and bisected it to this
commit 970e178985ca ("sched: Improve scalability via 'CPU buddies' ...")
apparently because the new single-idle-buddy model simply doesn't find
idle CPU's to reschedule on aggressively enough.
Mike Galbraith suspects that it is likely due to the user-mode spinlocks
in PostgreSQL not reacting well to preemption, but we don't really know
the details - I'll just revert the commit for now.
There are hopefully other approaches to improve scheduler scalability
without it causing these kinds of downsides.
Reported-by: Nikolay Ulyanitsky <lystor@gmail.com> Bisected-by: Borislav Petkov <bp@alien8.de> Acked-by: Mike Galbraith <efault@gmx.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Chanwoo Choi [Tue, 21 Aug 2012 06:16:23 +0000 (15:16 +0900)]
mfd: MAX77693: Fix NULL pointer error when initializing irqs
This patch initialize register map of MUIC device because mfd driver
of Maxim MAX77693 use regmap-muic instance of MUIC device when irqs of
Maxim MAX77693 is initialized before call max77693-muic probe() function.
Signed-off-by: Chanwoo Choi <cw00.choi@samsung.com> Signed-off-by: Myungjoo Ham <myungjoo.ham@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Reported-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Chanwoo Choi [Tue, 21 Aug 2012 06:15:52 +0000 (15:15 +0900)]
mfd: MAX77693: Fix interrupt handling bug
This patch fix bug related to interrupt handling for MAX77693 devices.
- Unmask interrupt masking bit for charger/flash/muic to revolve
that interrupt isn't happened when external connector is attached.
- Fix wrong regmap instance when muic interrupt is happened.
This patch were discussed and confirm discussion about this patch on below url:
http://lkml.org/lkml/2012/7/16/118
Signed-off-by: Chanwoo Choi <cw00.choi@samsung.com> Signed-off-by: Myungjoo Ham <myungjoo.ham@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Mark Brown [Tue, 11 Sep 2012 07:16:36 +0000 (15:16 +0800)]
mfd: core: Push irqdomain mapping out into devices
Currently the MFD core supports remapping MFD cell interrupts using an
irqdomain but only if the MFD is being instantiated using device tree
and only if the device tree bindings use the pattern of registering IPs
in the device tree with compatible properties. This will be actively
harmful for drivers which support non-DT platforms and use this pattern
for their DT bindings as it will mean that the core will silently change
remapping behaviour and it is also limiting for drivers which don't do
DT with this particular pattern. There is also a potential fragility if
there are interrupts not associated with MFD cells and all the cells are
omitted from the device tree for some reason.
Instead change the code to take an IRQ domain as an optional argument,
allowing drivers to take the decision about the parent domain for their
interrupts. The one current user of this feature is ab8500-core, it has
the domain lookup pushed out into the driver.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Merge tag 'asoc-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus
ASoC: Updates for 3.6
A bigger set of updates than I'm entirely comfortable with - things
backed up a bit due to travel. As ever the majority of these are small,
focused updates for specific drivers though there are a couple of core
changes. There's been good exposure in -next.
Pull GFS2 fixes from Steven Whitehouse:
"Here are three GFS2 fixes for the current kernel tree. These are all
related to the block reservation code which was added at the merge
window. That code will be getting an update at the forthcoming merge
window too. In the mean time though there are a few smaller issues
which should be fixed.
The first patch resolves an issue with write sizes of greater than 32
bits with the size hinting code. The second ensures that the
allocation data structure is initialised when using xattrs and the
third takes into account allocations which may have been made by other
nodes which affect a reservation on the local node."
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-fixes:
GFS2: Take account of blockages when using reserved blocks
GFS2: Fix missing allocation data for set/remove xattr
GFS2: Make write size hinting code common
Merge branch 'for_linus' of git://cavan.codon.org.uk/platform-drivers-x86
Pull x86 platform driver updates from Matthew Garrett:
"A few small updates for 3.6 - a trivial regression fix and a couple of
conformance updates for the gmux driver, plus some tiny fixes for
asus-wmi, eeepc-laptop and thinkpad_acpi."
* 'for_linus' of git://cavan.codon.org.uk/platform-drivers-x86:
thinkpad_acpi: buffer overflow in fan_get_status()
eeepc-laptop: fix device reference count leakage in eeepc_rfkill_hotplug()
platform/x86: fix asus_laptop.wled_type description
asus-laptop: HRWS/HWRS typo
drivers-platform-x86: remove useless #ifdef CONFIG_ACPI_VIDEO
apple-gmux: Fix port address calculation in gmux_pio_write32()
apple-gmux: Fix index read functions
apple-gmux: Obtain version info from indexed gmux
Merge branch 'i2c-embedded/for-current' of git://git.pengutronix.de/git/wsa/linux
Pull i2c embedded fixes from Wolfram Sang:
"The last bunch of (typical) i2c-embedded driver fixes for 3.6.
Also update the MAINTAINERS file to point to my tree since people keep
asking where to find their patches."
* 'i2c-embedded/for-current' of git://git.pengutronix.de/git/wsa/linux:
i2c: algo: pca: Fix mode selection for PCA9665
MAINTAINERS: fix tree for current i2c-embedded development
i2c: mxs: correctly setup speed for non devicetree
i2c: pnx: Fix read transactions of >= 2 bytes
i2c: pnx: Fix bit definitions
Merge tag 'ecryptfs-3.6-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs
Pull ecryptfs fixes from Tyler Hicks:
- Fixes a regression, introduced in 3.6-rc1, when a file is closed
before its shared memory mapping is dirtied and unmapped. The lower
file was being released when the eCryptfs file was closed and the
dirtied pages could not be written out.
- Adds a call to the lower filesystem's ->flush() from
ecryptfs_flush().
- Fixes a regression, introduced in 2.6.39, when a file is renamed on
top of another file. The target file's inode was not being evicted
and the space taken by the file was not reclaimed until eCryptfs was
unmounted.
* tag 'ecryptfs-3.6-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs:
eCryptfs: Copy up attributes of the lower target inode after rename
eCryptfs: Call lower ->flush() from ecryptfs_flush()
eCryptfs: Write out all dirty pages just before releasing the lower file
Merge branch 'fixes-for-3.6' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping
Pull one more DMA-mapping fix from Marek Szyprowski:
"This patch fixes very subtle bug (typical off-by-one error) which
might appear in very rare circumstances."
* 'fixes-for-3.6' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:
arm: mm: fix DMA pool affiliation check
Merge tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon fixes from Guenter Roeck:
"Fix word size register read and write operations in ina2xx driver, and
initialize uninitialized structure elements in twl4030-madc-hwmon
driver."
* tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (ina2xx) Fix word size register read and write operations
hwmon: (twl4030-madc-hwmon) Initialize uninitialized structure elements
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"I realise this a bit bigger than I would want at this point.
Exynos is a large chunk, I got them to half what they wanted already,
and hey its ARM based, so not going to hurt many people.
Radeon has only two fixes, but the PLL fixes were a bit bigger, but
required for a lot of scenarios, the fence fix is really urgent.
vmwgfx: I've pulled in a dumb ioctl support patch that I was going to
shove in later and cc stable, but we need it asap, its mainly to stop
mesa growing a really ugly dependency in userspace to run stuff on
vmware, and if I don't stick it in the kernel now, everyone will have
to ship ugly userspace libs to workaround it.
nouveau: single urgent fix found in F18 testing, causes X to not start
properly when f18 plymouth is used
i915: smattering of fixes and debug quieting
gma500: single regression fix
So as I said a bit large, but its fairly well scattered and its all
stuff I'll be shipping in F18's 3.6 kernel."
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (26 commits)
drm/nouveau: fix booting with plymouth + dumb support
drm/radeon: make 64bit fences more robust v3
drm/radeon: rework pll selection (v3)
drm: Drop the NV12M and YUV420M formats
drm/exynos: remove DRM_FORMAT_NV12M from plane module
drm/exynos: fix double call of drm_prime_(init/destroy)_file_private
drm/exynos: add dummy support for dmabuf-mmap
drm/exynos: Add missing braces around sizeof in exynos_mixer.c
drm/exynos: Add missing braces around sizeof in exynos_hdmi.c
drm/exynos: Make g2d_pm_ops static
drm/exynos: Add dependency for G2D in Kconfig
drm/exynos: fixed page align bug.
drm/exynos: Use ERR_CAST inlined function instead of ERR_PTR(PTR_ERR(.. [1]
drm/exynos: Use devm_* functions in exynos_drm_g2d.c file
drm/exynos: Use devm_kzalloc in exynos_drm_hdmi.c file
drm/exynos: Use devm_kzalloc in exynos_drm_vidi.c file
drm/exynos: Remove redundant check in exynos_drm_fimd.c file
drm/exynos: Remove redundant check in exynos_hdmi.c file
vmwgfx: add dumb ioctl support
gma500: Fix regression on Oaktrail devices
...
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"This tree includes various fixes"
Ingo really needs to improve on the whole "explain git pull" part.
"Various fixes" indeed.
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/hwpb: Invoke __perf_event_disable() if interrupts are already disabled
perf/x86: Enable Intel Cedarview Atom suppport
perf_event: Switch to internal refcount, fix race with close()
oprofile, s390: Fix uninitialized memory access when writing to oprofilefs
perf/x86: Fix microcode revision check for SNB-PEBS
Merge tag 'usb-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB patches from Greg Kroah-Hartman:
"Here are a number of USB patches, a bit more than I normally like this
late in the -rc series, but given people's vacations (myself
included), and the kernel summit, it seems to have happened this way.
All are tiny, but they add up. A number of gadget and xhci fixes, and
a few new device ids. All have been tested in linux-next.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>"
* tag 'usb-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (33 commits)
usb: chipidea: udc: don't stall endpoint if request list is empty in isr_tr_complete_low
usb: chipidea: cleanup dma_pool if udc_start() fails
usb: chipidea: udc: fix error path in udc_start()
usb: chipidea: udc: add pullup fuction, needed by the uvc gadget
usb: chipidea: udc: fix setup of endpoint maxpacket size
USB: option: replace ZTE K5006-Z entry with vendor class rule
EHCI: Update qTD next pointer in QH overlay region during unlink
USB: cdc-wdm: fix wdm_find_device* return value
USB: ftdi_sio: do not claim CDC ACM function
usb: dwc3: gadget: fix pending isoc handling
usb: renesas_usbhs: fixup DMA transport data alignment
usb: gadget: at91udc: Don't check for ep->ep.desc
usb: gadget: at91udc: don't overwrite driver data
usb: dwc3: core: fix incorrect usage of resource pointer
usb: musb: musbhsdma: fix IRQ check
usb: musb: tusb6010: fix error path in tusb_probe()
usb: musb: host: fix for musb_start_urb Oops
usb: gadget: dummy_hcd: add support for USB_DT_BOS on rh
usb: gadget: dummy_hcd: fixup error probe path
usb: gadget: s3c-hsotg.c: fix error return code
...
Merge tag 'driver-core-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core fix from Greg Kroah-Hartman:
"Here is one fix for 3.6-rc6 for the kobject.h file.
It fixes a reported oops if CONFIG_HOTPLUG is disabled. It's been in
the linux-next tree for a while now.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>"
* tag 'driver-core-3.6-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
kobject: fix oops with "input0: bad kobj_uevent_env content in show_uevent()"
vfs: make O_PATH file descriptors usable for 'fstat()'
We already use them for openat() and friends, but fstat() also wants to
be able to use O_PATH file descriptors. This should make it more
directly comparable to the O_SEARCH of Solaris.
Note that you could already do the same thing with "fstatat()" and an
empty path, but just doing "fstat()" directly is simpler and faster, so
there is no reason not to just allow it directly.
See also commit 332a2e1244bd, which did the same thing for fchdir, for
the same reasons.
Reported-by: ольга крыжановская <olga.kryzhanovska@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: stable@kernel.org # O_PATH introduced in 3.0+ Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Aaron Lu [Fri, 14 Sep 2012 18:54:44 +0000 (20:54 +0200)]
ACPI / PM: Use KERN_DEBUG when no power resources are found
commit a606dac368eed5696fb38e16b1394f1d049c09e9 adds support to link
devices which have _PRx, if a device does not have _PRx, a warning
message will be printed.
This commit is for ZPODD on Intel ZPODD capable platforms, on other
platforms, it has no problem if there is no power resource for this
device, so a warning here is not appropriate, change it to debug.
Reported-by: Borislav Petkov <bp@amd64.org> Signed-off-by: Aaron Lu <aaron.lu@intel.com> Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
eCryptfs: Copy up attributes of the lower target inode after rename
After calling into the lower filesystem to do a rename, the lower target
inode's attributes were not copied up to the eCryptfs target inode. This
resulted in the eCryptfs target inode staying around, rather than being
evicted, because i_nlink was not updated for the eCryptfs inode. This
also meant that eCryptfs didn't do the final iput() on the lower target
inode so it stayed around, as well. This would result in a failure to
free up space occupied by the target file in the rename() operation.
Both target inodes would eventually be evicted when the eCryptfs
filesystem was unmounted.
This patch calls fsstack_copy_attr_all() after the lower filesystem
does its ->rename() so that important inode attributes, such as i_nlink,
are updated at the eCryptfs layer. ecryptfs_evict_inode() is now called
and eCryptfs can drop its final reference on the lower inode.
http://launchpad.net/bugs/561129
Signed-off-by: Tyler Hicks <tyhicks@canonical.com> Tested-by: Colin Ian King <colin.king@canonical.com> Cc: <stable@vger.kernel.org> [2.6.39+]
eCryptfs: Call lower ->flush() from ecryptfs_flush()
Since eCryptfs only calls fput() on the lower file in
ecryptfs_release(), eCryptfs should call the lower filesystem's
->flush() from ecryptfs_flush().
If the lower filesystem implements ->flush(), then eCryptfs should try
to flush out any dirty pages prior to calling the lower ->flush(). If
the lower filesystem does not implement ->flush(), then eCryptfs has no
need to do anything in ecryptfs_flush() since dirty pages are now
written out to the lower filesystem in ecryptfs_release().
eCryptfs: Write out all dirty pages just before releasing the lower file
Fixes a regression caused by:
821f749 eCryptfs: Revert to a writethrough cache model
That patch reverted some code (specifically, 32001d6f) that was
necessary to properly handle open() -> mmap() -> close() -> dirty pages
-> munmap(), because the lower file could be closed before the dirty
pages are written out.
Rather than reapplying 32001d6f, this approach is a better way of
ensuring that the lower file is still open in order to handle writing
out the dirty pages. It is called from ecryptfs_release(), while we have
a lock on the lower file pointer, just before the lower file gets the
final fput() and we overwrite the pointer.
Thomas Kavanagh [Thu, 13 Sep 2012 15:16:55 +0000 (08:16 -0700)]
i2c: algo: pca: Fix mode selection for PCA9665
The code currently always selects turbo mode for PCA9665, no matter which
clock frequency is configured. This is because it compares the clock frequency
against constants reflecting (boundary / 100). Compare against real boundary
frequencies to fix the problem.
Signed-off-by: Thomas Kavanagh <tkavanagh@juniper.net> Signed-off-by: Guenter Roeck <groeck@juniper.net> Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Dave Airlie [Fri, 14 Sep 2012 01:20:46 +0000 (11:20 +1000)]
Merge branch 'drm-fixes-3.6' of git://people.freedesktop.org/~agd5f/linux into drm-fixes
Alex writes:
This is the current set of radeon fixes for 3.6. Two small fixes:
- fix the fence issues introduced in 3.5 with 64-bit fences
- PLL fix for multiple DP heads
* 'drm-fixes-3.6' of git://people.freedesktop.org/~agd5f/linux:
drm/radeon: make 64bit fences more robust v3
drm/radeon: rework pll selection (v3)
Lin Ming [Thu, 13 Sep 2012 22:26:33 +0000 (00:26 +0200)]
ACPI / PM: Fix resource_lock dead lock in acpi_power_on_device
Commit 0090def("ACPI: Add interface to register/unregister device
to/from power resources") used resource_lock to protect the devices list
that relies on power resource. It caused a mutex dead lock, as below
This patch adds a new mutex "devices_lock" to protect the devices list
and calls acpi_power_on_device in acpi_power_on, instead of
__acpi_power_on, after the resource_lock is released.
[rjw: Changed data type of a boolean variable to bool.]
Signed-off-by: Lin Ming <ming.m.lin@intel.com> Signed-off-by: Aaron Lu <aaron.lu@intel.com> Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
ACPI / PM: Infer parent power state from child if unknown, v2
It turns out that there are ACPI BIOSes defining device objects with
_PSx and without either _PSC or _PRx. For devices corresponding to
those ACPI objetcs __acpi_bus_get_power() returns ACPI_STATE_UNKNOWN
and their initial power states are regarded as unknown as a result.
If such a device is a parent of another power-manageable device, the
child cannot be put into a low-power state through ACPI, because
__acpi_bus_set_power() refuses to change power states of devices
whose parents' power states are unknown.
To work around this problem, observe that the ACPI power state of
a device cannot be higher-power (lower-number) than the power state
of its parent. Thus, if the device's _PSC method or the
configuration of its power resources indicates that the device is
in D0, the device's parent has to be in D0 as well. Consequently,
if the parent's power state is unknown when we've just learned that
its child's power state is D0, we can safely set the parent's
power.state field to ACPI_STATE_D0.
Tested-by: Aaron Lu <aaron.lu@intel.com> Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
pktgen: fix crash with vlan and packet size less than 46
If vlan option is being specified in the pktgen and packet size
being requested is less than 46 bytes, despite being illogical
request, pktgen should not crash the kernel.
The root cause of why pktgen is not able to handle this case is due
to comparison of signed (datalen) and unsigned data (sizeof), which
eventually passes a huge number to skb_put().
Signed-off-by: Nishank Trivedi <nistrive@cisco.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dan Carpenter [Sat, 1 Sep 2012 19:54:07 +0000 (12:54 -0700)]
thinkpad_acpi: buffer overflow in fan_get_status()
The acpi_evalf() function modifies four bytes of data but in
fan_get_status() we pass a pointer to u8. I have modified the
function to use type checking now.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Matthew Garrett <mjg@redhat.com>
Bernhard Froemel [Sat, 25 Aug 2012 08:30:48 +0000 (10:30 +0200)]
apple-gmux: Fix index read functions
Study of Apple's binary driver revealed that the GMUX_READ_PORT should
be written between calls to gmux_index_wait_ready and
gmux_index_wait_complete (i.e., the new index protocol must be
followed). If this is not done correctly, the indexed
gmux device only partially accepts writes which lead to problems
concerning GPU switching. Special thanks to Seth Forshee who helped
greatly with identifying unnecessary changes.
Signed-off-by: Bernhard Froemel <froemel@vmars.tuwien.ac.at> Signed-off-by: Matthew Garrett <mjg@redhat.com>
Commit a334872224a67b614dc888460377862621f3dac7 added afex support but lacked
several logical changes. This lack can cause afex to crash, and also
have a slight effect on other flows (i.e., driver always assumes the Tx ring
has less available buffers than what it actually has).
This patch adds the missing segments, fixing said issues.
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Barak Witkowski <barak@broadcom.com> Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Under traffic, there are several registers that when read (e.g., via
'ethtool -d') may cause the chip to stall.
This patch corrects the registers read in such flows.
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
bnx2x: correct advertisement of pause capabilities
This patch propagates users' requested flow-control into the link layer,
which will later be used to advertise this flow-control for auto-negotiation
(until now these values were ignored).
Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Prevent updating the xmac PFC configuration when using a link speed
slower than 10G -the umac block is responsible for 1G or slower connections,
therefore it is possible the xmac block is reset when connection is slower.
Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
FW needs the driver statistics for management. Current logic is broken
in that the function that gathers the port statistics does not copy
its own statistics to a place where the FW can use it.
This patch causes every function that can pass statistics to the FW to
do so.
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
During traffic when DCB is enabled, it is possible for multiple instances
of statistics queries to be sent to the chip - this may cause the FW to assert.
This patch prevents the sending of an additional instance of statistics query
while the previous query hasn't completed.
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Christian König [Thu, 13 Sep 2012 08:33:47 +0000 (10:33 +0200)]
drm/radeon: make 64bit fences more robust v3
Only increase the higher 32bits if we really detect a wrap around.
v2: instead of increasing the higher 32bits just use the higher
32bits from the last emitted fence.
v3: also use last emitted fence value as upper limit.
The intention of this patch is to make fences as robust as
they where before introducing 64bit fences. This is
necessary because on older systems it looks like the fence
value gets corrupted on initialization.
Should also fix:
https://bugs.freedesktop.org/show_bug.cgi?id=54129
https://bugs.freedesktop.org/show_bug.cgi?id=54662
https://bugzilla.redhat.com/show_bug.cgi?id=846505
https://bugzilla.redhat.com/show_bug.cgi?id=845639
3.5 needs a separate patch due to changes in the
fence code. Will send that out separately.
Signed-off-by: Christian König <deathsimple@vodafone.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 31 Aug 2012 15:56:50 +0000 (11:56 -0400)]
drm/radeon: rework pll selection (v3)
For DP we can use the same PPLL for all active DP
encoders. Take advantage of that to prevent cases
where we may end up sharing a PPLL between DP and
non-DP which won't work. Also clean up the code
a bit.
v2: - fix missing pll_id assignment in crtc init
v3: - fix DP PPLL check
- document functions
- break in main encoder search loop after matching.
no need to keep checking additional encoders.
net: qmi_wwan: call subdriver with control intf only
This fixes a hang on suspend due to calling wdm_suspend on
the unregistered data interface. The hang should have been
a NULL pointer reference had it not been for a logic error
in the cdc_wdm code.
commit 230718bd net: qmi_wwan: bind to both control and data interface
changed qmi_wwan to use cdc_wdm as a subdriver for devices with
a two-interface QMI/wwan function. The commit failed to update
qmi_wwan_suspend and qmi_wwan_resume, which were written to handle
either a single combined interface function, or no subdriver at all.
The result was that we called into the subdriver both when the
control interface was suspended and when the data interface was
suspended. Calling the subdriver suspend function with an
unregistered interface is not supported and will make the
subdriver bug out.
Signed-off-by: Bjørn Mork <bjorn@mork.no> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ward [Thu, 13 Sep 2012 05:22:35 +0000 (05:22 +0000)]
net_sched: gred: actually perform idling in WRED mode
gred_dequeue() and gred_drop() do not seem to get called when the
queue is empty, meaning that we never start idling while in WRED
mode. And since qidlestart is not stored by gred_store_wred_set(),
we would never stop idling while in WRED mode if we ever started.
This messes up the average queue size calculation that influences
packet marking/dropping behavior.
Now, we start WRED mode idling as we are removing the last packet
from the queue. Also we now actually stop WRED mode idling when we
are enqueuing a packet.
Cc: Bruce Osler <brosler@cisco.com> Signed-off-by: David Ward <david.ward@ll.mit.edu> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ward [Thu, 13 Sep 2012 05:22:34 +0000 (05:22 +0000)]
net_sched: gred: fix qave reporting via netlink
q->vars.qavg is a Wlog scaled value, but q->backlog is not. In order
to pass q->vars.qavg as the backlog value, we need to un-scale it.
Additionally, the qave value returned via netlink should not be Wlog
scaled, so we need to un-scale the result of red_calc_qavg().
This caused artificially high values for "Average Queue" to be shown
by 'tc -s -d qdisc', but did not affect the actual operation of GRED.
Signed-off-by: David Ward <david.ward@ll.mit.edu> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / karo-tx-linux.git / log