Commit graph

534,192 commits

Author SHA1 Message Date
Andy Lutomirski
7911d3f7af perf/x86: Only allow rdpmc if a perf_event is mapped
We currently allow any process to use rdpmc.  This significantly
weakens the protection offered by PR_TSC_DISABLED, and it could be
helpful to users attempting to exploit timing attacks.

Since we can't enable access to individual counters, use a very
coarse heuristic to limit access to rdpmc: allow access only when
a perf_event is mmapped.  This protects seccomp sandboxes.

There is plenty of room to further tighen these restrictions.  For
example, this allows rdpmc for any x86_pmu event, but it's only
useful for self-monitoring tasks.

As a side effect, cap_user_rdpmc will now be false for AMD uncore
events.  This isn't a real regression, since .event_idx is disabled
for these events anyway for the time being.  Whenever that gets
re-added, the cap_user_rdpmc code can be adjusted or refactored
accordingly.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/a2bdb3cf3a1d70c26980d7c6dddfbaa69f3182bf.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:47 +01:00
Andy Lutomirski
c1317ec2b9 perf: Pass the event to arch_perf_update_userpage()
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/0fea9a7fac3c1eea86cb0a5954184e74f4213666.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:46 +01:00
Andy Lutomirski
1e0fb9ec67 perf: Add pmu callbacks to track event mapping and unmapping
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/266afcba1d1f91ea5501e4e16e94bbbc1a9339b6.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:45 +01:00
Andy Lutomirski
22c4bd9fa9 x86: Add a comment clarifying LDT context switching
The code is correct, but only for a rather subtle reason.  This
confused me for quite a while when I read switch_mm, so clarify the
code to avoid confusing other people, too.

TBH, I wouldn't be surprised if this code was only correct by
accident.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/0db86397f968996fb772c443c251415b0b430ddd.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:43 +01:00
Andy Lutomirski
1e02ce4ccc x86: Store a per-cpu shadow copy of CR4
Context switches and TLB flushes can change individual bits of CR4.
CR4 reads take several cycles, so store a shadow copy of CR4 in a
per-cpu variable.

To avoid wasting a cache line, I added the CR4 shadow to
cpu_tlbstate, which is already touched in switch_mm.  The heaviest
users of the cr4 shadow will be switch_mm and __switch_to_xtra, and
__switch_to_xtra is called shortly after switch_mm during context
switch, so the cacheline is likely to be hot.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/3a54dd3353fffbf84804398e00dfdc5b7c1afd7d.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:42 +01:00
Andy Lutomirski
375074cc73 x86: Clean up cr4 manipulation
CR4 manipulation was split, seemingly at random, between direct
(write_cr4) and using a helper (set/clear_in_cr4).  Unfortunately,
the set_in_cr4 and clear_in_cr4 helpers also poke at the boot code,
which only a small subset of users actually wanted.

This patch replaces all cr4 access in functions that don't leave cr4
exactly the way they found it with new helpers cr4_set_bits,
cr4_clear_bits, and cr4_set_bits_and_update_boot.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/495a10bdc9e67016b8fd3945700d46cfd5c12c2f.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 12:10:41 +01:00
Tomi Valkeinen
d6c2152b3e Merge branches '3.20/fbdev' and '3.20/omapdss' into for-next
Merge fbdev topic branches
2015-02-04 12:44:11 +02:00
Peter Meerwald
811fbb1f58 omapfb: Return error code when applying overlay settings fails
the check of the return code is missing, user space does not get notified
about the error condition:

omapdss OVERLAY error: overlay 2 horizontally not inside the display area (403 + 800 >= 800)
omapdss APPLY error: failed to apply settings: illegal configuration.

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:41:53 +02:00
Tomi Valkeinen
a2408154a1 OMAPDSS: DPI: DRA7xx support
Add support for DRA7xx DPI output.

DRA7xx has three DPI outputs, each of which gets its input from a DISPC
channel. However, DRA72x has only one video PLL, and DRA74x has two
video PLLs. In both cases the video PLLs need to be shared between
multiple outputs. The driver doesn't handle this at the moment.

Also, DRA7xx requires configuring CONTROL module bits to route the clock
from the PLL to the used DISPC channel.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:07 +02:00
Tomi Valkeinen
adb5ff835c OMAPDSS: HDMI: Add DRA7xx support
Add support for DRA7xx to the HDMI driver.

The HDMI block on DRA7xx is the same as on OMAP5, except we need to
enable and disable the HDMI PLL via the CONTROL module.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:06 +02:00
Tomi Valkeinen
0006fd63d1 OMAPDSS: DISPC: program dispc polarities to control module
On DRA7xx, DISPC needs to write output signal polarities not only to a
DISPC register, like for all earlier DSS versions, but to control
module's CTRL_CORE_SMA_SW_1 register.

This patch adds support to write the polarities to control module.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:06 +02:00
Tomi Valkeinen
935509275c OMAPDSS: DISPC: Add DRA7xx support
Add DRA7xx support to DISPC driver. The DISPC block is the same as on
OMAP5, except the PLL's used for clocking are "videoX", not "dsiX".

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:06 +02:00
Tomi Valkeinen
99767548b1 OMAPDSS: Add Video PLLs for DRA7xx
DRA7xx SoCs have one (DRA72x) or two (DRA74x) video PLLs. They are
basically the same as DSI PLLs on OMAPs, but without the rest of the DSI
hardware. The video PLLs also require some configuration via the CONTROL
module.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:05 +02:00
Tomi Valkeinen
be40eecf8d OMAPDSS: Add functions for external control of PLL
Add functions which configure the control module register
CTRL_CORE_DSS_PLL_CONTROL found in DRA7xx SoCs. This register configures
whether the PLL registers are accessed internally by DSS, or externally
using OCP2SCP interface. They also configure muxes which route the PLL
output to a particular LCD overlay manager within DSS.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:05 +02:00
Tomi Valkeinen
6d817880cd OMAPDSS: DSS: Add DRA7xx base support
Add base support for DRA7xx to DSS core.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:05 +02:00
Tomi Valkeinen
6761a8f606 Doc/DT: Add DT binding doc for DRA7xx DSS
Add device tree binding documentation for DRA7xx display subsystem.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: devicetree@vger.kernel.org
2015-02-04 12:32:04 +02:00
Tomi Valkeinen
472da57b1c OMAPDSS: add define for DRA7xx HW version
Add define for DRA7xx DSS version.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:04 +02:00
Tomi Valkeinen
a87a6d6b09 OMAPDSS: encoder-tpd12s015: Fix race issue with LS_OE
A race issue has been observed with the encoder-tpd12s015 driver, which
leads to errors when trying to read EDID. This has only now been
observed, as OMAP4 and OMAP5 boards used SoC's GPIOs for LS_OE GPIO. On
dra7-evm boards, the LS_OE is behind a i2c controlled GPIO expander,
which increases the time to set the LS_OE.

This patch simplifies the handling of the LS_OE gpio in the driver by
removing the interrupt handling totally. The only time we actually need
to enable LS_OE is when we are reading the EDID, and thus we can just
set and clear the LS_OE gpio inside the read_edid() function.

This also has the additional benefit of very slightly decreasing the
power consumption.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:04 +02:00
Tomi Valkeinen
fa0c52ab23 OMAPDSS: OMAP5: fix digit output's allowed mgrs
The OMAP5 HW supports directing DIGIT channel to DPI output, but the
driver doesn't support that. However, we have marked that configuration
as possible in the dss features, so in certain cases the driver tries to
use that configuration, leading to broken display.

Fix the problem by allowing DIGIT channel to go only to HDMI output.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:03 +02:00
Tomi Valkeinen
234f9a22f4 OMAPDSS: constify port arrays
The port arrays are constant data, so set them as 'const'.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:03 +02:00
Tomi Valkeinen
eb30199b46 OMAPDSS: PLL: add dss_pll_wait_reset_done()
Add a helper function to wait until the PLL's reset is done.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:03 +02:00
Tomi Valkeinen
64e22ffdab OMAPDSS: Add enum dss_pll_id
In some cases we need global identifiers for the DSS PLLs, for example
when configuring clock muxing on DRA7. For this purpose let's add a
'enum dss_pll_id'.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2015-02-04 12:32:02 +02:00
Jyri Sarha
b8255930e0 ASoC: tlv320aic3x: Fix bad comment before intercon_extra_3104 definition
The intercon_extra_3104 is obviously for tlv320aic3104.

Reported-by: Benoît Thébaudeau <benoit.thebaudeau.dev@gmail.com>
Signed-off-by: Jyri Sarha <jsarha@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2015-02-04 10:31:03 +00:00
Josh Poimboeuf
12cf89b550 livepatch: rename config to CONFIG_LIVEPATCH
Rename CONFIG_LIVE_PATCHING to CONFIG_LIVEPATCH to make the naming of
the config and the code more consistent.

Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Reviewed-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2015-02-04 11:25:51 +01:00
Kevin Hao
0a4a3529df gpio: kconfig: replace PPC_OF with PPC
The PPC_OF is a ppc specific option which is used to mean that the
firmware device tree access functions are available. Since all the
ppc platforms have a device tree, it is aways set to 'y' for ppc.
So it makes no sense to keep a such option in the current kernel.
Replace it with PPC.

Signed-off-by: Kevin Hao <haokexin@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-02-04 11:06:45 +01:00
Cristian Stoica
96692a7305 crypto: tcrypt - do not allocate iv on stack for aead speed tests
See also: 9bac019dad

Signed-off-by: Cristian Stoica <cristian.stoica@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-02-04 20:59:42 +11:00
Cristian Stoica
424a5da691 crypto: testmgr - limit IV copy length in aead tests
The working copy of IV is the same size as the transformation's IV.
It is not necessary to copy more than that from the template since
iv_len is usually less than MAX_IVLEN and the rest of the copied data
is garbage.

Signed-off-by: Cristian Stoica <cristian.stoica@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-02-04 20:59:41 +11:00
Cristian Stoica
c47689931f crypto: tcrypt - fix buflen reminder calculation
- This fixes the intent of the code to limit the last scatterlist to
  either a full PAGE or a fraction of it, depending on the number of
  pages needed by buflen and the available space advertised by XBUFLEN.

  The original code always sets the last scatterlist to a fraction of a
  PAGE because the first 'if' is never executed.

- Rearrange the second part of the code to remove the conditional from
  the loop

Signed-off-by: Cristian Stoica <cristian.stoica@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2015-02-04 20:59:41 +11:00
Oleg Drokin
7456a37d55 GFS2: use __vmalloc GFP_NOFS for fs-related allocations.
leaf_dealloc uses vzalloc as a fallback to kzalloc(GFP_NOFS), so
it clearly does not want any shrinker activity within the fs itself.
convert vzalloc into __vmalloc(GFP_NOFS|__GFP_ZERO) to better achieve
this goal.

Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2015-02-04 09:58:41 +00:00
Daniel Vetter
083500baef drm: remove DRM_FORMAT_NV12MT
So this has been merged originally in

commit 83052d4d5c
Author: Seung-Woo Kim <sw0312.kim@samsung.com>
Date:   Thu Dec 15 15:40:55 2011 +0900

    drm: Add multi buffer plane pixel formats

which hasn't seen a lot of review really. The problem is that it's not
a real pixel format, but just a different way to lay out NV12 pixels
in macroblocks, i.e. a tiling format.

The new way of doing this is with the soon-to-be-merged fb modifiers.

This was brough up in some long irc discussion around the entire
topic, as an example of where things have gone wrong. Luckily we can
correct the mistake:
- The kms side support for NV12MT is all dead code because
  format_check in drm_crtc.c never accepted NV12MT.
- The gem side for the gsc support doesn't look better: The code
  forgets to set the pixel format and makes a big mess with the tiling
  mode bits, inadvertedly setting them all.

Conclusion: This never really worked (at least not in upstream) and
hence we can safely correct our mistake here.

Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Rob Clark <robclark@freedesktop.org>
Cc: Daniel Stone <daniel@fooishbar.org>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Acked-by: Joonyoung Shim <jy0922.shim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2015-02-04 10:10:05 +01:00
Mika Westerberg
2479c7300e pinctrl: cherryview: Configure HiZ pins to be input when requested as GPIOs
If the pin is in HiZ mode when it is requested as GPIO its value cannot be
read (it always returns 0). In order to cope with the Linux GPIO subsystem
where we do not have such state at all, turn the pin to be input instead.

Reported-by: Jerome Blin <jerome.blin@intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-02-04 09:59:26 +01:00
Marek Szyprowski
726b6324e3 mmc: pwrseq: add driver for emmc hardware reset
This patch provides a simple mmc-pwrseq-emmc driver, which controls
single gpio line. It perform standard eMMC hw reset procedure, as
descibed by Jedec 4.4 specification. This procedure is performed just
after MMC core enabled power to the given mmc host (to fix possible
issues if bootloader has left eMMC card in initialized or unknown
state), and before performing complete system reboot (also in case of
emergency reboot call). The latter is needed on boards, which doesn't
have hardware reset logic connected to emmc card and (limited or broken)
ROM bootloaders are unable to read second stage from the emmc card if
the card is left in unknown or already initialized state.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
2015-02-04 09:45:09 +01:00
Arnd Bergmann
3981c51666 mmc: moxart: fix probe logic
Jonas Jensen wanted to submit a patch for these, but apparently
forgot about it. I stumbled over this symptom first:

drivers/built-in.o: In function `moxart_probe':
:(.text+0x2af128): undefined reference to `of_dma_request_slave_channel'

This is because of_dma_request_slave_channel is an internal helper
and not exported to loadable module. I'm changing the driver to
use dma_request_slave_channel_reason() instead.

Further problems from inspection:

* The remove function must not call kfree on the host pointer,
  because it is allocated together with the mmc_host.

* The clock is never released

* The dma_cap_mask_t is completely unused and can be removed

* deferred probing does not work if the dma driver is loaded
  after the mmc driver.

This patch should fix all of the above.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jonas Jensen <jonas.jensen@gmail.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
2015-02-04 09:42:29 +01:00
Christophe Ricard
fa00e8fed4 NFC: nci: Move NFCEE discovery logic
NFCEE_DISCOVER_CMD is a specified NCI command used to discover
NFCEE IDs.
Move nci_nfcee_discover() call to nci_discover_se() in order to
guarantee:
- NFCEE_DISCOVER_CMD run when the NCI state machine is initialized
- NFCEE_DISCOVER_CMD is not run in case there is not discover_se
  hook defined by a NFC device driver.

Signed-off-by: Christophe Ricard <christophe-h.ricard@st.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
2015-02-04 09:15:18 +01:00
Christophe Ricard
15d4a8da0e NFC: nci: Move logical connection structure allocation
conn_info is currently allocated only after nfcee_discovery_ntf
which is not generic enough for logical connection other than
NFCEE. The corresponding conn_info is now created in
nci_core_conn_create_rsp().

Signed-off-by: Christophe Ricard <christophe-h.ricard@st.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
2015-02-04 09:14:09 +01:00
Christophe Ricard
3ba5c8466b NFC: nci: Change credits field to credits_cnt
For consistency sake change nci_core_conn_create_rsp structure
credits field to credits_cnt.

Signed-off-by: Christophe Ricard <christophe-h.ricard@st.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
2015-02-04 09:13:15 +01:00
Christophe Ricard
b16ae7160a NFC: nci: Support all destinations type when creating a connection
The current implementation limits nci_core_conn_create_req()
to only manage NCI_DESTINATION_NFCEE.
Add new parameters to nci_core_conn_create() to support all
destination types described in the NCI specification.
Because there are some parameters with variable size dynamic
buffer allocation is needed.

Signed-off-by: Christophe Ricard <christophe-h.ricard@st.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
2015-02-04 09:10:50 +01:00
Christophe Ricard
12bdf27d46 NFC: nci: Add reference to the RF logical connection
The NCI_STATIC_RF_CONN_ID logical connection is the most used
connection. Keeping it directly accessible in the nci_dev
structure will simplify and optimize the access.

Signed-off-by: Christophe Ricard <christophe-h.ricard@st.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
2015-02-04 09:09:53 +01:00
Ingo Tuchscherer
ea96f78813 s390/zcrypt: fixed domain scanning problem (again)
Older machines with more then 16 domains need a special check before
PQAP instructions can be processed. With commit 5bc334bff9 this
check was reverted by accident. This patch re-establishes the additional
code needed for checking the extended domains for older machines.

Signed-off-by: Ingo Tuchscherer <ingo.tuchscherer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-02-04 09:03:19 +01:00
Ingo Molnar
0967160ad6 Merge branch 'x86/asm' into perf/x86, to avoid conflicts with upcoming patches
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 09:01:12 +01:00
Michal Kazior
608b8f736b ath10k: enable qca6174 hw3.2
The 3.2 revision has a different target BMI
version so it wasn't recognized by ath10k (despite
the chip_id rev being on the supported list
already).

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:22:01 +02:00
SenthilKumar Jegadeesan
467210a67b ath10k: add log level configuration for fw_dbglog
Introduce an optional log level configuration for the existing debugfs fw_dbglog file. It
allows users to configure the desired log level for firmware dbglog messages.

To configure log level as WARN:

echo 0xffffffff  2 > /sys/kernel/debug/ieee80211/phy0/ath10k/fw_dbglog

The values are:

VERBOSE		0
INFO		1
WARN		2
ERR		3

Signed-off-by: SenthilKumar Jegadeesan <sjegadee@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:20:08 +02:00
Helmut Schaa
75930d1a80 ath10k: Use TX cksum offload only for CHECKSUM_PARTIAL
Otherwise ath10k will just checksum everything even if it did not
go through the TCP/IP stack (for example bridged frames). In the worst
case this could mean recreating the checksum for incorrect data.

Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:19:03 +02:00
Michal Kazior
af21319fcd ath10k: fix beacon deadlock
This should fix a very rare occurrence of the following deadlock:

  [<ffffffffa018265e>] ath10k_wmi_tx_beacons_nowait+0x1e/0x50 [ath10k_core]
  [<ffffffffa01829b6>] ath10k_wmi_op_ep_tx_credits+0x16/0x40 [ath10k_core]
  [<ffffffffa017d685>] ath10k_htc_send+0x285/0x3d0 [ath10k_core]
  [<ffffffffa0184b81>] ath10k_wmi_cmd_send_nowait+0x81/0x110 [ath10k_core]
  [<ffffffffa0184c61>] ath10k_wmi_tx_beacon_nowait.part.33+0x51/0x90 [ath10k_core]
  [<ffffffffa0184cd0>] ath10k_wmi_tx_beacons_iter+0x30/0x40 [ath10k_core]
  [<ffffffff81882246>] __iterate_active_interfaces+0xa6/0x100
  [<ffffffffa0184ca0>] ? ath10k_wmi_tx_beacon_nowait.part.33+0x90/0x90 [ath10k_core]
  [<ffffffff818822ae>] ieee80211_iterate_active_interfaces_atomic+0xe/0x10
  [<ffffffffa0182676>] ath10k_wmi_tx_beacons_nowait+0x36/0x50 [ath10k_core]
  [<ffffffffa01829b6>] ath10k_wmi_op_ep_tx_credits+0x16/0x40 [ath10k_core]
  [<ffffffffa017d140>] ath10k_htc_rx+0x280/0x410 [ath10k_core]
  [<ffffffffa01bcbf0>] ? ath10k_ce_completed_recv_next+0x60/0x80 [ath10k_pci]
  [<ffffffffa01bc6ab>] ath10k_pci_ce_recv_data+0x11b/0x1d0 [ath10k_pci]
  [<ffffffffa01bcf44>] ath10k_ce_per_engine_service+0x64/0xc0 [ath10k_pci]
  [<ffffffffa01bcfc2>] ath10k_ce_per_engine_service_any+0x22/0x50 [ath10k_pci]
  [<ffffffffa01bc4d0>] ath10k_pci_tasklet+0x30/0x90 [ath10k_pci]
  [<ffffffff81055a55>] tasklet_action+0xc5/0x100

To prevent this make sure to release ar->data_lock
while calling to ath10k_wmi_beacon_send_ref_nowait().

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:17:06 +02:00
Michal Kazior
9ad501827b ath10k: change dma beacon cmd prototype
The command logic shouldn't really care about
arvif structure.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:17:00 +02:00
SenthilKumar Jegadeesan
627613f8f0 ath10k: prevent setting wrong key idx for station
Ath10k driver sets wrong default key idx that results in
sending unicast frames with multicast key.

The reason for this behavior is that cached broadcast key
is installed for station MAC address on association. After
dot1x completes, unicast key is installed for station
MAC address. Default key idx is set to broadcast key id when
driver tries to send broadcast frame. This causes firmware
to use broadcast key id to transmit unicast frames to stations.

Used TX_USAGE flag to set default key for stations.

Added callback for setting unicast default idx which will be
invoked on every default key idx configuration.

Signed-off-by: SenthilKumar Jegadeesan <sjegadee@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
2015-02-04 09:14:43 +02:00
Mark Rutland
2fde4f94e0 perf: Decouple unthrottling and rotating
Currently the adjusments made as part of perf_event_task_tick() use the
percpu rotation lists to iterate over any active PMU contexts, but these
are not used by the context rotation code, having been replaced by
separate (per-context) hrtimer callbacks. However, some manipulation of
the rotation lists (i.e. removal of contexts) has remained in
perf_rotate_context(). This leads to the following issues:

* Contexts are not always removed from the rotation lists. Removal of
  PMUs which have been placed in rotation lists, but have not been
  removed by a hrtimer callback can result in corruption of the rotation
  lists (when memory backing the context is freed).

  This has been observed to result in hangs when PMU drivers built as
  modules are inserted and removed around the creation of events for
  said PMUs.

* Contexts which do not require rotation may be removed from the
  rotation lists as a result of a hrtimer, and will not be considered by
  the unthrottling code in perf_event_task_tick.

This patch fixes the issue by updating the rotation ist when events are
scheduled in/out, ensuring that each rotation list stays in sync with
the HW state. As each event holds a refcount on the module of its PMU,
this ensures that when a PMU module is unloaded none of its CPU contexts
can be in a rotation list. By maintaining a list of perf_event_contexts
rather than perf_event_cpu_contexts, we don't need separate paths to
handle the cpu and task contexts, which also makes the code a little
simpler.

As the rotation_list variables are not used for rotation, these are
renamed to active_ctx_list, which better matches their current function.
perf_pmu_rotate_{start,stop} are renamed to
perf_pmu_ctx_{activate,deactivate}.

Reported-by: Johannes Jensen <johannes.jensen@arm.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Will Deacon <Will.Deacon@arm.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20150129134511.GR17721@leverpostej
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 08:07:16 +01:00
Mark Rutland
cc34b98bac perf: Drop module reference on event init failure
When initialising an event, perf_init_event will call try_module_get() to
ensure that the PMU's module cannot be removed for the lifetime of the
event, with __free_event() dropping the reference when the event is
finally destroyed. If something fails after the event has been
initialised, but before the event is installed, perf_event_alloc will
drop the reference on the module.

However, if we fail to initialise an event for some reason (e.g. we ask
an uncore PMU to perform sampling, and it refuses to initialise the
event), we do not drop the refcount. If we try to open such a bogus
event without a precise IDR type, we will loop over each PMU in the pmus
list, incrementing each of their refcounts without decrementing them.

This patch adds a module_put when pmu->event_init(event) fails, ensuring
that the refcounts are balanced in failure cases. As the innards of the
precise and search based initialisation look very similar, this logic is
hoisted out into a new helper function. While the early return for the
failed try_module_get is removed from the search case, this is handled
by the remaining return when ret is not -ENOENT.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1420642611-22667-1-git-send-email-mark.rutland@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 08:07:14 +01:00
Jiri Olsa
7c60fc0e02 perf: Use POLLIN instead of POLL_IN for perf poll data in flag
Currently we flag available data (via poll syscall) on perf fd with
POLL_IN macro, which is normally used for SIGIO interface.

We've been lucky, because POLLIN (0x1) is subset of POLL_IN (0x20001)
and sys_poll (do_pollfd function) cut the extra bit out (0x20000).

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1422467678-22341-1-git-send-email-jolsa@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 08:07:13 +01:00
Peter Zijlstra
a83fe28e2e perf: Fix put_event() ctx lock
So what I suspect; but I'm in zombie mode today it seems; is that while
I initially thought that it was impossible for ctx to change when
refcount dropped to 0, I now suspect its possible.

Note that until perf_remove_from_context() the event is still active and
visible on the lists. So a concurrent sys_perf_event_open() from another
task into this task can race.

Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Stephane Eranian <eranian@gmail.com>
Cc: mark.rutland@arm.com
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20150129134434.GB26304@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-04 08:07:12 +01:00