Even in failed case of pm_runtime_get_sync(), the usage_count is
incremented. In order to keep the usage_count with correct value call
appropriate pm_runtime_put().
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
Jin Yao reported the issue (and posted first versions of this change)
with groups being defined over events with different cpu mask.
This causes assert aborts in get_group_fd, like:
# perf stat -M "C2_Pkg_Residency" -a -- sleep 1
perf: util/evsel.c:1464: get_group_fd: Assertion `!(fd == -1)' failed.
Aborted
All the events in the group have to be defined over the same cpus so the
group_fd can be found for every leader/member pair.
Adding check to ensure this condition is met and removing the group
(with warning) if we detect mixed cpus, like:
$ sudo perf stat -e '{power/energy-cores/,cycles},{instructions,power/energy-cores/}'
WARNING: event cpu maps do not match, disabling group:
anon group { power/energy-cores/, cycles }
anon group { instructions, power/energy-cores/ }
Ian asked also for cpu maps details, it's displayed in verbose mode:
$ sudo perf stat -e '{cycles,power/energy-cores/}' -v
WARNING: group events cpu maps do not match, disabling group:
anon group { power/energy-cores/, cycles }
power/energy-cores/: 0
cycles: 0-7
anon group { instructions, power/energy-cores/ }
instructions: 0-7
power/energy-cores/: 0
Committer testing:
[root@seventh ~]# perf stat -e '{power/energy-cores/,cycles},{instructions,power/energy-cores/}'
WARNING: grouped events cpus do not match, disabling group:
anon group { power/energy-cores/, cycles }
anon group { instructions, power/energy-cores/ }
^C
Performance counter stats for 'system wide':
12.62 Joules power/energy-cores/
106,920,637 cycles
80,228,899 instructions # 0.75 insn per cycle
12.62 Joules power/energy-cores/
14.514476987 seconds time elapsed
[root@seventh ~]#
But if we put compatible events in each group it works:
[root@seventh ~]# perf stat -e '{power/energy-cores/,power/energy-ram/},{instructions,cycles}' -a sleep 2
Performance counter stats for 'system wide':
1.95 Joules power/energy-cores/
0.92 Joules power/energy-ram/
29,305,715 instructions # 1.03 insn per cycle
28,423,338 cycles
2.001438142 seconds time elapsed
[root@seventh ~]#
This needs improvement tho:
[root@seventh ~]# perf stat -e '{power/energy-cores/,power/energy-ram/},{instructions,cycles}' sleep 2
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (power/energy-cores/).
/bin/dmesg | grep -i perf may provide additional information.
[root@seventh ~]#
We need to emit a better message, one stating that the power/ events
can't be used for a specific workload, instead it is per-cpu or system
wide.
Fixes: 6a4bb04caa ("perf tools: Enable grouping logic for parsed events")
Co-developed-by: Jin Yao <yao.jin@linux.intel.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200602101736.GE1112120@krava
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
fake_lmem_start does not need to be mutable via module param sysfs. It's
only used during driver probe.
Fixes: 1629224324 ("drm/i915/lmem: add the fake lmem region")
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200601215510.18379-2-jani.nikula@intel.com
(cherry picked from commit f322e851f2)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Allow batch buffers to read their own _local_ cumulative HW runtime of
their logical context.
Fixes: 0f2f397583 ("drm/i915: Add gen9 BCS cmdparsing")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.4+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200601161942.30854-1-chris@chris-wilson.co.uk
(cherry picked from commit f9496520df)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
While the current locking/serialization of the global state
suffices for protecting the obj->state access and the actual
hardware reprogramming, we do have a problem with accessing
the old/new states during nonblocking commits.
The state computation and swap will be protected by the crtc
locks, but the commit_tails can finish out of order, thus also
causing the atomic states to be cleaned up out of order. This
would mean the commit that started first but finished last has
had its new state freed as the no-longer-needed old state by the
other commit.
To fix this let's just refcount the states. obj->state amounts
to one reference, and the intel_atomic_state holds extra references
to both its new and old global obj states.
Fixes: 0ef1905ecf ("drm/i915: Introduce better global state handling")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200527200245.13184-1-ville.syrjala@linux.intel.com
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
(cherry picked from commit f8c86ffa28)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Added support for dynamic clock freq configuration in PWM kernel driver.
Earlier the PWM driver used to cache boot time clock rate by PWM clock
parent during probe. Hence dynamically changing PWM frequency was not
possible for all the possible ranges. With this change, dynamic
calculation is enabled and it is able to set the requested period from
sysfs knob provided the value is supported by clock source.
Changes mainly have 2 parts:
- Tegra186 and later chips [1]
- Tegra210 and prior chips [2]
For [1] - Changes implemented to set pwm period dynamically and also
checks added to allow only if requested period(ns) is below or
equals to higher range.
For [2] - Only checks if the requested period(ns) is below or equals to
higher range defined by max clock limit. The limitation in
Tegra210 or prior chips are due to the reason of having only
one PWM controller supporting multiple channels. But later
chips have multiple PWM controller instances each having
single channel support.
Signed-off-by: Sandipan Patra <spatra@nvidia.com>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
The PWM hardware in the JZ4725B works the same as in the JZ4740, but has
only six channels available.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
The PWM in Ingenic SoCs starts in inactive state until the internal
timer reaches the duty value, then becomes active until the timer
reaches the period value. In theory, we should then use (period - duty)
as the real duty value, as a high duty value would otherwise result in
the PWM pin being inactive most of the time.
This is the reason why the duty value was inverted in the driver until
now, but it still had the problem that it would not start with the
active part.
To address this remaining issue, the common trick is to invert the
duty, and invert the polarity when the PWM is enabled.
Since the duty was already inverted, and we invert it again, we now
program the hardware for the requested duty, and simply invert the
polarity when the PWM is enabled.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
Calculating the hardware value for the duty from the hardware value of
the period resulted in a precision loss versus calculating it from the
clock rate directly.
(Also remove a cast that doesn't really need to be here)
Fixes: f6b8a57000 ("pwm: Add Ingenic JZ4740 support")
Cc: <stable@vger.kernel.org>
Suggested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
Depending on MACH_INGENIC prevent us from creating a generic kernel that
works on more than one MIPS board. Instead, we just depend on MIPS being
set.
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
Before commit cfc4c189bc ("pwm: Read initial hardware state at request
time"), a driver's get_state callback would get called once per PWM from
pwmchip_add().
pwm-lpss' runtime-pm code was relying on this, getting a runtime-pm ref for
PWMs which are enabled at probe time from within its get_state callback,
before enabling runtime-pm.
The change to calling get_state at request time causes a number of
problems:
1. PWMs enabled at probe time may get runtime suspended before they are
requested, causing e.g. a LCD backlight controlled by the PWM to turn off.
2. When the request happens when the PWM has been runtime suspended, the
ctrl register will read all 1 / 0xffffffff, causing get_state to store
bogus values in the pwm_state.
3. get_state was using an async pm_runtime_get() call, because it assumed
that runtime-pm has not been enabled yet. If shortly after the request an
apply call is made, then the pwm_lpss_is_updating() check may trigger
because the resume triggered by the pm_runtime_get() call is not complete
yet, so the ctrl register still reads all 1 / 0xffffffff.
This commit fixes these issues by moving the initial pm_runtime_get() call
for PWMs which are enabled at probe time to the pwm_lpss_probe() function;
and by making get_state take a runtime-pm ref before reading the ctrl reg.
BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1828927
Fixes: cfc4c189bc ("pwm: Read initial hardware state at request time")
Cc: stable@vger.kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
The vio and ibmebus buses are used for pseries specific
paravirtualised devices and currently they're initialised by the
generic initcall types. This is mostly fine, but it can result in some
nuisance errors in dmesg when booting on PowerNV on some OSes, e.g.
[ 2.984439] synth uevent: /devices/vio: failed to send uevent
[ 2.984442] vio vio: uevent: failed to send synthetic uevent
[ 17.968551] synth uevent: /devices/vio: failed to send uevent
[ 17.968554] vio vio: uevent: failed to send synthetic uevent
We don't see anything similar for the ibmebus because that depends on
!CONFIG_LITTLE_ENDIAN.
This patch squashes those by switching to using machine_*_initcall()
so the bus type is only registered when the kernel is running on a
pseries machine.
Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
Reviewed-by: Tyrel Datwyler <tyreld@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200421081539.7485-1-oohall@gmail.com
Allwinner A64 is capable of a direct clock output on PWM (see A64 User
Manual chapter 3.10). Add support for this in the sun4i PWM driver.
Signed-off-by: Peter Vasil <peter.vasil@gmail.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
The schib region can be used by userspace to get the subchannel-
information block (SCHIB) for the passthrough subchannel.
This can be useful to get information such as channel path
information via the SCHIB.PMCW fields.
Signed-off-by: Farhan Ali <alifm@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-5-farman@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
This is mostly for the purposes of a later patch, since
we'll need to do the same thing later.
While we are at it, move the resulting function call to ahead
of the unregistering of the IOMMU notifier, so that it's done
in the reverse order of how it was created.
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-4-farman@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Register the chp_event callback to receive channel path related
events for the subchannels managed by vfio-ccw.
Signed-off-by: Farhan Ali <alifm@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-3-farman@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Consolidate some of the cleanup code for the regions, so that
as more are added we reduce code duplication.
Signed-off-by: Farhan Ali <alifm@linux.ibm.com>
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200505122745.53208-2-farman@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Interacting with the I/O and the async regions can yield a number
of errors, which had been undocumented so far. These are part of
the api, so remedy that.
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Message-Id: <20200407111605.1795-1-cohuck@redhat.com>
Remove the explicit prefetch check when using vfio-ccw devices.
This check does not trigger in practice as all Linux channel programs
are intended to use prefetch.
It is expected that all ORBs issued by Linux will request prefetch.
Although non-prefetching ORBs are not rejected, they will prefetch
nonetheless. A warning is issued up to once per 5 seconds when a
forced prefetch occurs.
A non-prefetch ORB does not necessarily result in an error, however
frequent encounters with non-prefetch ORBs indicate that channel
programs are being executed in a way that is inconsistent with what
the guest is requesting. While there is currently no known case of an
error caused by forced prefetch, it is possible in theory that forced
prefetch could result in an error if applied to a channel program that
is dependent on non-prefetch.
Signed-off-by: Jared Rossi <jrossi@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Message-Id: <20200506212440.31323-2-jrossi@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
The CXL_AFU_DRIVER_OPS and CXL_LIB Kconfig options were added to coordinate
merging of new features. They no longer serve any purpose, so remove them.
Signed-off-by: Andrew Donnellan <ajd@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200602040341.10152-1-ajd@linux.ibm.com
PVR value of 0x0F000006 means we are arch v3.1 compliant (i.e.
POWER10). This is used by phyp and kvm when booting as a pseries guest
to detect the presence of new P10 features and to enable the
appropriate hwcap and facility bits.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
[mpe: Fall through to __init_FSCR rather than duplicating it, drop
hack to set current->thread.fscr now that is handled elsewhere.]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-8-alistair@popple.id.au
Matrix multiple assist (MMA) is a new feature added to ISAv3.1 and
POWER10. Support on powernv can be selected via a firmware CPU device
tree feature which enables it via a PCR bit.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-7-alistair@popple.id.au
Prefix instructions have their own FSCR bit which needs to be enabled
via a CPU feature. The kernel will save the FSCR for problem state but
it needs to be enabled initially.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-6-alistair@popple.id.au
On powernv hardware support for ISAv3.1 is advertised via a cpu feature
bit in the device tree. This patch enables the associated HWCAP bit if
the device tree indicates ISAv3.1 is available.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-4-alistair@popple.id.au
Newer ISA versions are enabled by clearing all bits in the PCR
associated with previous versions of the ISA. Enable ISA v3.1 support
by updating the PCR mask to include ISA v3.0. This ensures all PCR
bits corresponding to earlier architecture versions get cleared
thereby enabling ISA v3.1 if supported by the hardware.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-3-alistair@popple.id.au
POWER10 introduces two new architectural features - ISAv3.1 and matrix
multiply assist (MMA) instructions. Userspace detects the presence
of these features via two HWCAP bits introduced in this patch. These
bits have been agreed to by the compiler and binutils team.
According to ISAv3.1 MMA is an optional feature and software that makes
use of it should first check for availability via this HWCAP bit and use
alternate code paths if unavailable.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Tested-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200521014341.29095-2-alistair@popple.id.au
Since the previous commit that saves the value of FSCR configured at
boot into init_task.thread.fscr, the static initialisation in
INIT_THREAD now no longer has any effect.
So remove it.
For non DT CPU features, the end result is the same, because
__init_FSCR() is called on all CPUs that have an FSCR (Power8,
Power9), and it sets FSCR_TAR & FSCR_EBB.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200527145843.2761782-4-mpe@ellerman.id.au
At boot the FSCR is initialised via one of two paths. On most systems
it's set to a hard coded value in __init_FSCR().
On newer skiboot systems we use the device tree CPU features binding,
where firmware can tell Linux what bits to set in FSCR (and HFSCR).
In both cases the value that's configured at boot is not propagated
into the init_task.thread.fscr value prior to the initial fork of init
(pid 1), which means the value is not used by any processes other than
swapper (the idle task).
For the __init_FSCR() case this is OK, because the value in
init_task.thread.fscr is initialised to something sensible. However it
does mean that the value set in __init_FSCR() is not used other than
for swapper, which is odd and confusing.
The bigger problem is for the device tree CPU features case it
prevents firmware from setting (or clearing) FSCR bits for use by user
space. This means all existing kernels can not have features
enabled/disabled by firmware if those features require
setting/clearing FSCR bits.
We can handle both cases by saving the FSCR value into
init_task.thread.fscr after we have initialised it at boot. This fixes
the bug for device tree CPU features, and will allow us to simplify
the initialisation for the __init_FSCR() case in a future patch.
Fixes: 5a61ef74f2 ("powerpc/64s: Support new device tree binding for discovering CPU features")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200527145843.2761782-3-mpe@ellerman.id.au
The device tree CPU features binding includes FSCR bit numbers which
Linux is instructed to set by firmware.
Whether that's a good idea or not, in the case of the DSCR the Linux
implementation has a hard requirement that the FSCR_DSCR bit not be
set by default. We use it to track when a process reads/writes to
DSCR, so it must be clear to begin with.
So if firmware tells us to set FSCR_DSCR we must ignore it.
Currently this does not cause a bug in our DSCR handling because the
value of FSCR that the device tree CPU features code establishes is
only used by swapper. All other tasks use the value hard coded in
init_task.thread.fscr.
However we'd like to fix that in a future commit, at which point this
will become necessary.
Fixes: 5a61ef74f2 ("powerpc/64s: Support new device tree binding for discovering CPU features")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200527145843.2761782-2-mpe@ellerman.id.au
__init_FSCR() was added originally in commit 2468dcf641 ("powerpc:
Add support for context switching the TAR register") (Feb 2013), and
only set FSCR_TAR.
At that point FSCR (Facility Status and Control Register) was not
context switched, so the setting was permanent after boot.
Later we added initialisation of FSCR_DSCR to __init_FSCR(), in commit
54c9b2253d ("powerpc: Set DSCR bit in FSCR setup") (Mar 2013), again
that was permanent after boot.
Then commit 2517617e0d ("powerpc: Fix context switch DSCR on
POWER8") (Aug 2013) added a limited context switch of FSCR, just the
FSCR_DSCR bit was context switched based on thread.dscr_inherit. That
commit said "This clears the H/FSCR DSCR bit initially", but it
didn't, it left the initialisation of FSCR_DSCR in __init_FSCR().
However the initial context switch from init_task to pid 1 would clear
FSCR_DSCR because thread.dscr_inherit was 0.
That commit also introduced the requirement that FSCR_DSCR be clear
for user processes, so that we can take the facility unavailable
interrupt in order to manage dscr_inherit.
Then in commit 152d523e63 ("powerpc: Create context switch helpers
save_sprs() and restore_sprs()") (Dec 2015) FSCR was added to
thread_struct. However it still wasn't fully context switched, we just
took the existing value and set FSCR_DSCR if the new thread had
dscr_inherit set. FSCR was still initialised at boot to FSCR_DSCR |
FSCR_TAR, but that value was not propagated into the thread_struct, so
the initial context switch set FSCR_DSCR back to 0.
Finally commit b57bd2de8c ("powerpc: Improve FSCR init and context
switching") (Jun 2016) added a full context switch of the FSCR, and
added an initialisation of init_task.thread.fscr to FSCR_TAR |
FSCR_EBB, but omitted FSCR_DSCR.
The end result is that swapper runs with FSCR_DSCR set because of the
initialisation in __init_FSCR(), but no other processes do, they use
the value from init_task.thread.fscr.
Having FSCR_DSCR set for swapper allows it to access SPR 3 from
userspace, but swapper never runs userspace, so it has no useful
effect. It's also confusing to have the value initialised in two
places to two different values.
So remove FSCR_DSCR from __init_FSCR(), this at least gets us to the
point where there's a single value of FSCR, even if it's still set in
two places.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Tested-by: Alistair Popple <alistair@popple.id.au>
Link: https://lore.kernel.org/r/20200527145843.2761782-1-mpe@ellerman.id.au
Since commit c55d7b5e64 ("powerpc: Remove STRICT_KERNEL_RWX
incompatibility with RELOCATABLE"), powerpc kernels with
-mprofile-kernel can crash in certain scenarios with a trace like below:
BUG: Unable to handle kernel instruction fetch (NULL pointer?)
Faulting instruction address: 0x00000000
Oops: Kernel access of bad area, sig: 11 [#1]
LE PAGE_SIZE=64K MMU=Radix SMP NR_CPUS=256 DEBUG_PAGEALLOC NUMA PowerNV
<snip>
NIP [0000000000000000] 0x0
LR [c0080000102c0048] ext4_iomap_end+0x8/0x30 [ext4]
Call Trace:
iomap_apply+0x20c/0x920 (unreliable)
iomap_bmap+0xfc/0x160
ext4_bmap+0xa4/0x180 [ext4]
bmap+0x4c/0x80
jbd2_journal_init_inode+0x44/0x1a0 [jbd2]
ext4_load_journal+0x440/0x860 [ext4]
ext4_fill_super+0x342c/0x3ab0 [ext4]
mount_bdev+0x25c/0x290
ext4_mount+0x28/0x50 [ext4]
legacy_get_tree+0x4c/0xb0
vfs_get_tree+0x4c/0x130
do_mount+0xa18/0xc50
sys_mount+0x158/0x180
system_call+0x5c/0x68
The NIP points to NULL, or a random location (data even), while the LR
always points to the LEP of a function (with an offset of 8), indicating
that something went wrong with ftrace. However, ftrace is not
necessarily active when such crashes occur.
The kernel OOPS sometimes follows a warning from ftrace indicating that
some module functions could not be patched with a nop. Other times, if a
module is loaded early during boot, instruction patching can fail due to
a separate bug, but the error is not reported due to missing error
reporting.
In all the above cases when instruction patching fails, ftrace will be
disabled but certain kernel module functions will be left with default
calls to _mcount(). This is not a problem with ELFv1. However, with
-mprofile-kernel, the default stub is problematic since it depends on a
valid module TOC in r2. If the kernel (or a different module) calls into
a function that does not use the TOC, the function won't have a prologue
to setup the module TOC. When that function calls into _mcount(), we
will end up in the relocation stub that will use the previous TOC, and
end up trying to jump into a random location. From the above trace:
iomap_apply+0x20c/0x920 [kernel TOC]
|
V
ext4_iomap_end+0x8/0x30 [no GEP == kernel TOC]
|
V
_mcount() stub
[uses kernel TOC -> random entry]
To address this, let's change over to using the special stub that is
used for ftrace_[regs_]caller() for _mcount(). This ensures that we are
not dependent on a valid module TOC in r2 for default _mcount()
handling.
Reported-by: Qian Cai <cai@lca.pw>
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Tested-by: Qian Cai <cai@lca.pw>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/8affd4298d22099bbd82544fab8185700a6222b1.1587488954.git.naveen.n.rao@linux.vnet.ibm.com
For -mprofile-kernel, we need special handling when generating stubs for
ftrace calls such as _mcount(). To faciliate this, we check if a
R_PPC64_REL24 relocation is for a symbol named "_mcount()" along with
also checking the instruction sequence. The latter is not really
required since "_mcount()" is an exported symbol and kernel modules
cannot use it. As such, drop the additional checking and simplify the
code. This helps unify stub creation for ftrace stubs with
-mprofile-kernel and aids in code reuse.
Also rename is_mprofile_mcount_callsite() to is_mprofile_ftrace_call()
to reflect the checking being done.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/7d9c316adfa1fb787ad268bb4691e7e4059ff2d5.1587488954.git.naveen.n.rao@linux.vnet.ibm.com
module_trampoline_target() is only used by ftrace. Move the prototype
within the appropriate #ifdef in the header. Also, move the function
body to the end of module_64.c so as to consolidate all ftrace code in
one place.
No functional changes.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/2527351f65c53c5866068ae130dc34c5d4ee8ad9.1587488954.git.naveen.n.rao@linux.vnet.ibm.com
Mapping of early shadow area is implemented by using a single static
page table having all entries pointing to the same early shadow page.
The shadow area must therefore occupy full PGD entries.
The shadow area has a size of 128MB starting at 0xf8000000.
With 4k pages, a PGD entry is 4MB
With 16k pages, a PGD entry is 64MB
With 64k pages, a PGD entry is 1GB which is too big.
Until we rework the early shadow mapping, disable KASAN when the page
size is too big.
Fixes: 2edb16efc8 ("powerpc/32: Add KASAN support")
Cc: stable@vger.kernel.org # v5.2+
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/7195fcde7314ccbf7a081b356084a69d421b10d4.1590660977.git.christophe.leroy@csgroup.eu
On book3s/32, KUEP is an heavy process as it requires to
set/unset the NX bit in each of the 12 user segments
everytime the kernel is entered/exited from/to user space.
Don't select KUEP by default on book3s/32.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1492bb150c1aaa53d99a604b49992e60ea20cd5f.1586962582.git.christophe.leroy@c-s.fr
On book3s/32, KUAP is an heavy process as it requires to
determine which segments are impacted and unlock/lock
each of them.
And since the implementation of user_access_begin/end, it
is even worth for the time being because unlike __get_user(),
user_access_begin doesn't make difference between read and write
and unlocks access also for read allthought that's unneeded
on book3s/32.
As shown by the size of a kernel built with KUAP and one without,
the overhead is 64k bytes of code. As a comparison a similar
build on an 8xx has an overhead of only 8k bytes of code.
text data bss dec hex filename
7230416 1425868 837376 9493660 90dc9c vmlinux.kuap6xx
7165012 1425548 837376 9427936 8fdbe0 vmlinux.nokuap6xx
6519796 1960028 477464 8957288 88ad68 vmlinux.kuap8xx
6511664 1959864 477464 8948992 888d00 vmlinux.nokuap8xx
Until a more optimised KUAP is implemented on book3s/32,
don't select it by default.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/154a99399317b096ac1f04827b9f8d7a9179ddc1.1586962586.git.christophe.leroy@c-s.fr
To enable/disable kernel access to user space, the 8xx has to
modify the properties of access group 1. This is done by writing
predefined values into SPRN_Mx_AP registers.
As of today, a __put_user() gives:
00000d64 <my_test>:
d64: 3d 20 4f ff lis r9,20479
d68: 61 29 ff ff ori r9,r9,65535
d6c: 7d 3a c3 a6 mtspr 794,r9
d70: 39 20 00 00 li r9,0
d74: 90 83 00 00 stw r4,0(r3)
d78: 3d 20 6f ff lis r9,28671
d7c: 61 29 ff ff ori r9,r9,65535
d80: 7d 3a c3 a6 mtspr 794,r9
d84: 4e 80 00 20 blr
Because only groups 0 and 1 are used, the definition of
groups 2 to 15 doesn't matter.
By setting unused bits to 0 instead on 1, one instruction is
removed for each lock and unlock action:
00000d5c <my_test>:
d5c: 3d 20 40 00 lis r9,16384
d60: 7d 3a c3 a6 mtspr 794,r9
d64: 39 20 00 00 li r9,0
d68: 90 83 00 00 stw r4,0(r3)
d6c: 3d 20 60 00 lis r9,24576
d70: 7d 3a c3 a6 mtspr 794,r9
d74: 4e 80 00 20 blr
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/57425c33dd72f292b1a23570244b81419072a7aa.1586945153.git.christophe.leroy@c-s.fr
kprobe does not handle events happening in real mode.
The very last part of exception exits cannot support a trap.
Blacklist them from kprobe.
While we are at it, remove exc_exit_start symbol which is not
used to avoid having to blacklist it.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/098b0fd3f6299aa1bd692bd576bd7012c84608de.1585670437.git.christophe.leroy@c-s.fr
kprobe does not handle events happening in real mode.
The very last part of syscall cannot support a trap.
Add a symbol syscall_exit_finish to identify that part and
blacklist it from kprobe.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/23eddf49abb03d1359fa0be4206998eb3800f42c.1585670437.git.christophe.leroy@c-s.fr
kprobe does not handle events happening in real mode.
As exception entry points are running with MMU disabled,
blacklist them.
The handling of TLF_NAPPING and TLF_SLEEPING is moved before the
CONFIG_TRACE_IRQFLAGS which contains 'reenable_mmu' because from there
kprobe will be possible as the kernel will run with MMU enabled.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/f61ac599855e674ebb592464d0ea32a3ba9c6644.1585670437.git.christophe.leroy@c-s.fr
In hash_low.S, a lot of named local symbols are used instead of
numbers to ease code readability. However, they don't need to be
visible.
In order to ease blacklisting of functions running with MMU
disabled for kprobe, rename the symbols to .Lsymbols in order
to hide them as if they were numbered labels.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/90c430d9e0f7af772a58aaeaf17bcc6321265340.1585670437.git.christophe.leroy@c-s.fr
Navid Emamdoost