Commit graph

678,699 commits

Author SHA1 Message Date
Ian Abbott
6541964dcd staging: comedi: jr3_pci: fix initial range[8] max value
`jr3_pci_alloc_spriv()` initializes `spriv->range[8]` to use a maximum
value of 65536, but that will be overwritten with 65535 at a later time
by `jr3_pci_poll_subdevice()` once the "set full scales" command is
complete.  The initial setting looks like a mistake.  This range is only
associated with a couple of dummy channels (channels 56 and 57) to read
back the model number and serial number, so no user code should be
attempting to convert those numbers to physical units.  Just change the
initial value to 65535 to match the final value.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
8abde9acdf staging: comedi: jr3_pci: check PCI BAR 0 size
The various supported boards have different numbers of subdevices from 1
to 4.  Each subdevice needs a block of registers in PCI BAR 0.  Check
the region is large enough for the required number of subdevices.
Return an error from `jr3_pci_auto_attach()` if it is too small.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
7cb1c5c2cd staging: comedi: jr3_pci: check sizes at compile time
The driver currently checks the size of `struct jr3_sensor` is correct
when a device is attached, returning an error if it is wrong.  Replace
that with a compile-time check.  We don't care too much about the size
of `struct jr3_sensor` as it is embedded in the larger `struct
jr3_block` and is followed by a lot of padding.  We should care more
that the size of `struct jr3_block` is correct, as it describes the
overall register layout of a block, and there is an array of such blocks
(one per subdevice).  Check its size at compile-time using the
`BUILD_BUG_ON()` macro.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
6fbd1330e5 staging: comedi: jr3_pci: omit pointless debug info
`jr3_pci_open()` outputs several debug log messages containing serial
numbers of the sensors (one per subdevice) along with a pointer to the
subdevice private data structure.  The latter is of no use, so reformat
the debug log to omit it.

`jr3_pci_alloc_spriv()` outputs a debug log message containing more
useless information about the remapped base address of the board
registers, the sensor registers, and the difference between them.  Get
rid of it.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
9d6a94291e staging: comedi: jr3_pci: use struct jr3_block instead of jr3_t
`struct jr3_t` contains a single array member `block` of member type
`struct jr3_block`.  Rather than using pointers to `struct jr3_t`, just
use pointers to `struct jr3_block` instead and treat it as an array.
Replace the local variables `struct jr3_t __iomem *iobase` with `struct
jr3_block __iomem *block`.  Remove the definition of `struct jr3_t` as
it is no longer needed.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
4e861f0fd8 staging: comedi: jr3_pci: separate out block type
`struct jr3_t` contains a single array member `block` of a tag-less
`struct` type.  Rename the tag-less `struct` type to `struct jr3_block`
and move its definition outside of `struct jr3_t`.  This will allow us
to use pointers of this type.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
6c7f58084f staging: comedi: jr3_pci: rename 'channel' to 'block'
The term "channel" is overloaded in this driver.  Rename the `channel`
member of `struct jr3_t` to `block` to reduce confusion.  `block` is an
array of an anonymous `struct` type, with each element covering the
registers for one subdevice.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
ef3ab9f85c staging: comedi: jr3_pci: rename data to sensor
Rename the `channel[x].data` member of `struct jr3_t` to
`channel[x].sensor` to match its type `struct jr3_sensor`.  Also rename
local variable `ch0data` in `jr3_pci_show_copyright()` to `sensor0` for
consistency.  It points to the `struct jr3_sensor` embedded in the
registers for "channel" 0.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
713eab88c6 staging: comedi: jr3_pci: rename channel to sensor
The driver overloads the term "channel" a lot.  To help reduce
confusion, rename the `channel` member of `struct
jr3_pci_subdev_private` to `sensor` as it points to a `struct
jr3_sensor`.  Also rename the various function parameters and local
variables called `channel` that point to a `struct jr3_sensor` to
`sensor`.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Ian Abbott
c5ad0c7b51 staging: comedi: jr3_pci: rename struct jr3_channel to jr3_sensor
The driver overloads the term "channel" a lot.  To help reduce
confusion, rename `struct jr3_channel` to `struct jr3_sensor`.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Colin Ian King
b612a9762b staging: atomisp: clean up return logic, remove redunant code
There is no need to check if ret is non-zero, remove this
redundant check and just return the error status from the call
to mt9m114_write_reg_array.

Detected by CoverityScan, CID#1416577 ("Identical code for
different branches")

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Colin Ian King
b8ff636cd7 staging: atomisp: remove redundant check for client being null
The previous statement checks if client is null, so the null check
when assigning dev is redundant and can be removed.

Detected by CoverityScan, CID#1416555 ("Logically Dead Code")

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
simran singhal
9573878611 staging: atomisp_fops: Clean up tests if NULL returned on failure
Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
simran singhal
7cf51d34af staging: vpfe_mc_capture: Clean up tests if NULL returned on failure
Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
simran singhal
1e8b15d06a staging: lirc_zilog: Clean up tests if NULL returned on failure
Some functions like kmalloc/kzalloc return NULL on failure.
When NULL represents failure, !x is commonly used.

This was done using Coccinelle:
@@
expression *e;
identifier l1;
@@

e = \(kmalloc\|kzalloc\|kcalloc\|devm_kzalloc\)(...);
...
- e == NULL
+ !e

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:57:00 +01:00
Greg Kroah-Hartman
49637a458b staging: media: atomisp: remove '.' from pci Makefile
Remove the odd './' usage in the Makefile to make the build output a bit
more sane looking and match the normal kernel build style.

Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:51:52 +01:00
Alan Cox
c537f916d0 atomisp: remove FPGA defines
These are not relevant to an upstream kernel driver.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
9403fbe396 atomisp: remove pdaf kernel
This is not used on either Baytrail or Cherrytrail so can simply be deleted

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
43da435042 atomisp: tidy firmware loading code a little
The FWNAME define is never used so can be removed. The option to skip firmware
loading isn't really Cherrytrail specific so remove this and complete the
merging of the two driver versions for this file.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
f406d0338e atomisp: eliminate intel_mid_pm.h
We can do this because the only thing it is used for is identifying the
platform for power management purposes. The driver only supports Baytrail
and Cherrytrail and both of those always need the IPU to be power managed
directly not via PCI D3 states.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
5ac0a3447a atomisp: remove C_RUN define and code
We are not going to be building for anything but Linux so the code bracketed
by C_RUN is not used and not needed.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
de3ae0d087 atomisp: remove HIVECC
We are only going to be building for Linux with gcc, so we can lose bits of
material related to other build targets.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
d9b16766fe atomisp: remove unused code and unify a header
KLOCWORK is never defined so we can remove the workarounds for this in the
code.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Alan Cox
5a0ff3683d atomisp: remove dead code for SSSE3
This is another define which is never used and references a header that doesn't
exist, so consider it dead. Our memcpy is pretty smart anyway.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:48:20 +01:00
Gargi Sharma
cbe7023526 staging: rtl8192e: Remove multiple assignments
This patch removes multiple assignments by factorizing them.
This was done with Coccinelle for the if branch. For the else part
the change was done manually. Braces were also added to the
else part to remove the checkpatch warning, "braces should be
on all arms of if-else statements".

@ identifier i1,i2; constant c; @@
- i1=i2=c;
+ i1=c;
+ i2=c;

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
simran singhal
a4a987decb staging: rtl8192u: Remove typedef phy_ofdm_rx_status_rxsc_sgien_exintfflag
Remove typdef phy_ofdm_rx_status_rxsc_sgien_exintfflag and replace its uses
in the code.

Signed-off-by: simran singhal <singhalsimran0@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Gargi Sharma
4d567cef1b staging: lustre: Use min3 macro
Replace comparsions between three expressions using
two min macros with min3 macro. Done using Coccinelle
Script:

@@
expression e1;
expression e2;
expression e3;
@@
(
- min(min(e1, e2), e3)
+ min3(e1, e2, e3)
|
- min(e1, min(e2, e3))
+ min3(e1, e2, e3)
)

Signed-off-by: Gargi Sharma <gs051095@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Tobin C. Harding
2437a9e023 staging: dgnc: remove item from TODO list
TODO file contains task to verify and correct function return
sites. Need to check for and implement correct usage of goto's when
there is work to be done before returning.

Remove task from TODO list after already having completed audit of
directory drivers/staging/dgnc.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Tobin C. Harding
4f22e0fa01 staging: dgnc: audit goto's in dgnc_tty
TODO file requests fix up of error handling.

Audit dgnc_mgmt.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Tobin C. Harding
7029064dd1 staging: dgnc: audit goto's in dgnc_mgmt
TODO file requests fix up of error handling.

Audit dgnc_mgmt.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Tobin C. Harding
c8863b3ff7 staging: dgnc: audit goto's in dgnc_driver
TODO file requests fix up of error handling.

Audit dgnc_driver.c and fix all return paths to be uniform and inline
with kernel coding style.

Signed-off-by: Tobin C. Harding <me@tobin.cc>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-03-12 14:46:34 +01:00
Sagar Arun Kamble
1f3b1fd386 drm/i915/guc: Update rps.pm_intrmsk_mbz in guc_interrupts_capture/release
Different state is to be maintained for rps.pm_intrmsk_mbz for GuC and
Execlists. Updating it inside guc_interrupts_* routines as in those
routines GuC load/submission params are sanitized and it should not be set
based on HAS_GUC_SCHED during intel_irq_init.

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1489199821-6707-3-git-send-email-sagar.a.kamble@intel.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2017-03-12 12:59:11 +00:00
Sagar Arun Kamble
5dd0455667 drm/i915: s/pm_intr_keep/pm_intrmsk_mbz
"pm_intr_keep" is not conveying the intent that it is bitmask
of interrupts that must be zero(mbz) in GEN6_PMINTRMSK.
Name it "pm_intrmsk_mbz".

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1489199821-6707-2-git-send-email-sagar.a.kamble@intel.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2017-03-12 12:59:08 +00:00
Sagar Arun Kamble
7762ebb9a4 drm/i915/guc: Release GuC interrupts in i915_guc_submission_disable
Like capture of GuC interrupts while enabling GuC submission, release
them while disabling GuC submission.

Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Oscar Mateo <oscar.mateo@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1489199821-6707-1-git-send-email-sagar.a.kamble@intel.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2017-03-12 12:59:06 +00:00
Linus Walleij
6dbb708a8f ARM: gemini: convert to ARMv4 multiplatform
This converts the Gemini platform to ARMv4 multiplatform, deleting
the local <mach/*> include directory, moving an idiomatic local
idling function into the .machine_init() call and getting rid of
the Makefile.boot finally.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:27 +01:00
Linus Walleij
8e39061ec8 ARM: gemini: select ARM_PATCH_PHYS_VIRT and AUTO_ZRELADDR
This platform survives physical to virtual patching
without any hickups, and can use AUTO_ZRELADDR.
We still need to keep Makefile.boot but it is now empty.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:24 +01:00
Linus Walleij
46abf938db ARM: gemini: switch to sparse IRQs
There is no boardfiles or anything else using the fixed IRQs
anymore, switch the platform to use sparse IRQs and delete
the <mach/irqs.h> header.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:22 +01:00
Linus Walleij
ee149d66d7 ARM: gemini: delete all boardfiles
Delete the Gemini boardfiles: we have corresponding, fully-featured
device trees for all these boards. Delete the referenced include
files. Delete the local config symbols, especially one for
"swapped memory", as all supported boards have swapped memory, and
would a new board be supported this is likely not the right way
to achieve it anyways. Only the Kconfig options in the central
arch/arm/Kconfig remains.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:19 +01:00
Linus Walleij
41d9830cd0 ARM: gemini: DT for the Cortina Gemini SoC platforms
This adds initial and compulsory device tree support to the
Gemini ARMv4 platform.

We are selecting a bunch of "absolute minimals" for getting a working
system up with just device tree:

- We select USE_OF for natural reasons or nothing works.

- We select CLKSRC_OF and GEMINI_TIMER so we get timekeeping from
  the clocksource.

- We select GPIO_GEMINI because these are used as irqchips, and
  for a generic driver it is not reasonable for those to have to
  select every possible irqchip in the world to work, the platform
  should simply provide the available irqchips.

- We select a UART that can be exprected to work with
  SERIAL_OF_PLATFORM which is the name for an 8250 OF-probed
  serial port.

- We select the syscon-based reset controller: it's not fun when
  "reboot" doesn't work because of Kconfig, so we just select
  POWER_RESET and POWER_RESET_SYSCON.

- We perhaps a bit controversiallt select ARM_APPENDED_DTB, because
  this platform is using the ancient RedBoot, and can *NOT* be
  expected to upgrade its bootloaders. Appended device tree is
  simply how these devices have to work with device tree.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:12 +01:00
Linus Walleij
c12ddfe1ed ARM: gemini: convert to MULTI_IRQ_HANDLER
In order to enable device tree boot on this machine we must first
convert it to runtime-manage its irq handler, so do this.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:27:04 +01:00
Linus Walleij
6ae4d211ab ARM: dts: add watchdog to the Gemini
This adds watchdog support to the Gemini SoC DTSI file.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:29 +01:00
Linus Walleij
4317c020ae watchdog: add DT bindings for Cortina Gemini
This adds DT bindings for the Cortina systems Gemini SoC
watchdog timer.

Cc: devicetree@vger.kernel.org
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:23 +01:00
Linus Walleij
c4fa8b272e ARM: dts: add a devicetree for Wiliboard WBD-222
This devicetree is simply based on the board file in
arch/arm/mach-gemini/board-wbd222.c and contain the
equivalent platform data, mainly just moving the GPIOs
from the global numberspace to explicitly reference their
parent GPIO.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:20 +01:00
Linus Walleij
2aeeb18201 ARM: dts: add a devicetree for Wiliboard WBD-111
This devicetree is simply based on the board file in
arch/arm/mach-gemini/board-wbd111.c and contain the
equivalent platform data, mainly just moving the GPIOs
from the global numberspace to explicitly reference their
parent GPIO.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:17 +01:00
Linus Walleij
d08bd6b36c ARM: dts: add a devicetree for Teltonika RUT1xx
This devicetree is simply based on the board file in
arch/arm/mach-gemini/board-rut1xx.c and contain the
equivalent platform data, mainly just moving the GPIOs
from the global numberspace to explicitly reference their
parent GPIO.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:14 +01:00
Linus Walleij
fe7bf9dcff ARM: dts: add a devicetree for Raidsonic NAS IB-4220-B
This devicetree is simply based on the board file in
arch/arm/mach-gemini/board-nas4220b.c and contain the
equivalent platform data, mainly just moving the GPIOs
from the global numberspace to explicitly reference &gpio1.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:10 +01:00
Linus Walleij
9be0d7f87e ARM: dts: add device tree for Gemini SoC and SQ201
This adds a device tree for the Gemini SoC and the ITian
Square One SQ201 board that has been my testing target
for Gemini device tree support.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:18:04 +01:00
Linus Walleij
d565194420 ARM: dts: add top-level DT bindings for Cortina Gemini
This adds the top level SoC bindings for Cortina systems Gemini
platforms.

Cc: Janos Laube <janos.dev@gmail.com>
Cc: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Cc: Hans Ulli Kroll <ulli.kroll@googlemail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: devicetree@vger.kernel.org
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2017-03-12 12:17:53 +01:00
Daniel Borkmann
2c4ea6e28d x86/tlb: Fix tlb flushing when lguest clears PGE
Fengguang reported random corruptions from various locations on x86-32
after commits d2852a2240 ("arch: add ARCH_HAS_SET_MEMORY config") and
9d876e79df ("bpf: fix unlocking of jited image when module ronx not set")
that uses the former. While x86-32 doesn't have a JIT like x86_64, the
bpf_prog_lock_ro() and bpf_prog_unlock_ro() got enabled due to
ARCH_HAS_SET_MEMORY, whereas Fengguang's test kernel doesn't have module
support built in and therefore never had the DEBUG_SET_MODULE_RONX setting
enabled.

After investigating the crashes further, it turned out that using
set_memory_ro() and set_memory_rw() didn't have the desired effect, for
example, setting the pages as read-only on x86-32 would still let
probe_kernel_write() succeed without error. This behavior would manifest
itself in situations where the vmalloc'ed buffer was accessed prior to
set_memory_*() such as in case of bpf_prog_alloc(). In cases where it
wasn't, the page attribute changes seemed to have taken effect, leading to
the conclusion that a TLB invalidate didn't happen. Moreover, it turned out
that this issue reproduced with qemu in "-cpu kvm64" mode, but not for
"-cpu host". When the issue occurs, change_page_attr_set_clr() did trigger
a TLB flush as expected via __flush_tlb_all() through cpa_flush_range(),
though.

There are 3 variants for issuing a TLB flush: invpcid_flush_all() (depends
on CPU feature bits X86_FEATURE_INVPCID, X86_FEATURE_PGE), cr4 based flush
(depends on X86_FEATURE_PGE), and cr3 based flush.  For "-cpu host" case in
my setup, the flush used invpcid_flush_all() variant, whereas for "-cpu
kvm64", the flush was cr4 based. Switching the kvm64 case to cr3 manually
worked fine, and further investigating the cr4 one turned out that
X86_CR4_PGE bit was not set in cr4 register, meaning the
__native_flush_tlb_global_irq_disabled() wrote cr4 twice with the same
value instead of clearing X86_CR4_PGE in the first write to trigger the
flush.

It turned out that X86_CR4_PGE was cleared from cr4 during init from
lguest_arch_host_init() via adjust_pge(). The X86_FEATURE_PGE bit is also
cleared from there due to concerns of using PGE in guest kernel that can
lead to hard to trace bugs (see bff672e630 ("lguest: documentation V:
Host") in init()). The CPU feature bits are cleared in dynamic
boot_cpu_data, but they never propagated to __flush_tlb_all() as it uses
static_cpu_has() instead of boot_cpu_has() for testing which variant of TLB
flushing to use, meaning they still used the old setting of the host
kernel.

Clearing via setup_clear_cpu_cap(X86_FEATURE_PGE) so this would propagate
to static_cpu_has() checks is too late at this point as sections have been
patched already, so for now, it seems reasonable to switch back to
boot_cpu_has(X86_FEATURE_PGE) as it was prior to commit c109bf9599
("x86/cpufeature: Remove cpu_has_pge"). This lets the TLB flush trigger via
cr3 as originally intended, properly makes the new page attributes visible
and thus fixes the crashes seen by Fengguang.

Fixes: c109bf9599 ("x86/cpufeature: Remove cpu_has_pge")
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: bp@suse.de
Cc: Kees Cook <keescook@chromium.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: lkp@01.org
Cc: Laura Abbott <labbott@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernrl.org/r/20170301125426.l4nf65rx4wahohyl@wfg-t540p.sh.intel.com
Link: http://lkml.kernel.org/r/25c41ad9eca164be4db9ad84f768965b7eb19d9e.1489191673.git.daniel@iogearbox.net
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2017-03-12 11:19:29 +01:00
NeilBrown
f5fe1b5190 blk: Ensure users for current->bio_list can see the full list.
Commit 79bd99596b ("blk: improve order of bio handling in generic_make_request()")
changed current->bio_list so that it did not contain *all* of the
queued bios, but only those submitted by the currently running
make_request_fn.

There are two places which walk the list and requeue selected bios,
and others that check if the list is empty.  These are no longer
correct.

So redefine current->bio_list to point to an array of two lists, which
contain all queued bios, and adjust various code to test or walk both
lists.

Signed-off-by: NeilBrown <neilb@suse.com>
Fixes: 79bd99596b ("blk: improve order of bio handling in generic_make_request()")
Signed-off-by: Jens Axboe <axboe@fb.com>
2017-03-11 15:31:37 -07:00