If a new hardware queue is added at runtime, we don't allocate scheduler
tags for it, leading to a crash. This hooks up the scheduler framework
to blk_mq_{init,exit}_hctx() to make sure everything gets properly
initialized/freed.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Preparation cleanup for the next couple of fixes, push
blk_mq_sched_setup() and e->ops.mq.init_sched() into a helper.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
While dispatching requests, if we fail to get a driver tag, we mark the
hardware queue as waiting for a tag and put the requests on a
hctx->dispatch list to be run later when a driver tag is freed. However,
blk_mq_dispatch_rq_list() may dispatch requests from multiple hardware
queues if using a single-queue scheduler with a multiqueue device. If
blk_mq_get_driver_tag() fails, it doesn't update the hardware queue we
are processing. This means we end up using the hardware queue of the
previous request, which may or may not be the same as that of the
current request. If it isn't, the wrong hardware queue will end up
waiting for a tag, and the requests will be on the wrong dispatch list,
leading to a hang.
The fix is twofold:
1. Make sure we save which hardware queue we were trying to get a
request for in blk_mq_get_driver_tag() regardless of whether it
succeeds or not.
2. Make blk_mq_dispatch_rq_list() take a request_queue instead of a
blk_mq_hw_queue to make it clear that it must handle multiple
hardware queues, since I've already messed this up on a couple of
occasions.
This didn't appear in testing with nvme and mq-deadline because nvme has
more driver tags than the default number of scheduler tags. However,
with the blk_mq_update_nr_hw_queues() fix, it showed up with nbd.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
inet_rtm_getroute synthesizes a skeletal ICMP skb, which is passed to
ip_route_input when iif is given. If a multipath route is present for
the designated destination, fib_multipath_hash ends up being called with
that skb. However, as that skb contains no information beyond the
protocol type, the calculated hash does not match the one we would see
for a real packet.
There is currently no way to fix this for layer 4 hashing, as
RTM_GETROUTE doesn't have the necessary information to create layer 4
headers. To fix this for layer 3 hashing, set appropriate saddr/daddrs
in the skb and also change the protocol to UDP to avoid special
treatment for ICMP.
Signed-off-by: Florian Larysch <fl@n621.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds the cts_i958 clock to control the clock source of the spdif
output block. This mux is not explicitly mentionned in the documentation
but it is critical to the spdif dai. It is used to select whether the clock
source of the spdif output is cts_amclk (when data are taken from i2s
buffer) or the cts_mclk_i958 (when data are taken from the spdif buffer)
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Add the spdif master clock also referred as cts_mclk_i958
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Add the i2s master clock also referred as cts_amclk
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The audio divider needs a specific clock divider driver.
With am mpll parent clock, which is able to provide a fairly precise rate,
the generic divider tends to select low value of the divider. In such case
the quality of the clock is very poor. For the same final rate, maximizing
the audio clock divider value and selecting the corresponding mpll rate
gives better results. This is what this driver aims to acheive. So far, so
good.
Cc: Hendrik v. Raven <hendrik@consetetur.de>
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The clock controller is getting more complex and it might be possible, in
the future, to have holes in the clk_hw_onecell_data array. Just make sure
we skip those holes if it ever happens.
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The userspace exception injection API and code path are entirely
unprepared for exceptions that might cause a VM-exit from L2 to L1, so
the best course of action may be to simply disallow this for now.
1. The API provides no mechanism for userspace to specify the new DR6
bits for a #DB exception or the new CR2 value for a #PF
exception. Presumably, userspace is expected to modify these registers
directly with KVM_SET_SREGS before the next KVM_RUN ioctl. However, in
the event that L1 intercepts the exception, these registers should not
be changed. Instead, the new values should be provided in the
exit_qualification field of vmcs12 (Intel SDM vol 3, section 27.1).
2. In the case of a userspace-injected #DB, inject_pending_event()
clears DR7.GD before calling vmx_queue_exception(). However, in the
event that L1 intercepts the exception, this is too early, because
DR7.GD should not be modified by a #DB that causes a VM-exit directly
(Intel SDM vol 3, section 27.1).
3. If the injected exception is a #PF, nested_vmx_check_exception()
doesn't properly check whether or not L1 is interested in the
associated error code (using the #PF error code mask and match fields
from vmcs12). It may either return 0 when it should call
nested_vmx_vmexit() or vice versa.
4. nested_vmx_check_exception() assumes that it is dealing with a
hardware-generated exception intercept from L2, with some of the
relevant details (the VM-exit interruption-information and the exit
qualification) live in vmcs02. For userspace-injected exceptions, this
is not the case.
5. prepare_vmcs12() assumes that when its exit_intr_info argument
specifies valid information with a valid error code that it can VMREAD
the VM-exit interruption error code from vmcs02. For
userspace-injected exceptions, this is not the case.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
If we already entered/are about to enter SMM, don't allow switching to
INIT/SIPI_RECEIVED, otherwise the next call to kvm_apic_accept_events()
will report a warning.
Same applies if we are already in MP state INIT_RECEIVED and SMM is
requested to be turned on. Refuse to set the VCPU events in this case.
Fixes: cd7764fe9f ("KVM: x86: latch INITs while in system management mode")
Cc: stable@vger.kernel.org # 4.2+
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Its value has never changed; we might as well make it part of the ABI instead
of using the return value of KVM_CHECK_EXTENSION(KVM_CAP_COALESCED_MMIO).
Because PPC does not always make MMIO available, the code has to be made
dependent on CONFIG_KVM_MMIO rather than KVM_COALESCED_MMIO_PAGE_OFFSET.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Remove code from architecture files that can be moved to virt/kvm, since there
is already common code for coalesced MMIO.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
[Removed a pointless 'break' after 'return'.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
In order to simplify adding exit reasons in the future,
the array of exit reason names is now also sorted by
exit reason code.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Now use bit 6 of EPTP to optionally enable A/D bits for EPTP. Another
thing to change is that, when EPT accessed and dirty bits are not in use,
VMX treats accesses to guest paging structures as data reads. When they
are in use (bit 6 of EPTP is set), they are treated as writes and the
corresponding EPT dirty bit is set. The MMU didn't know this detail,
so this patch adds it.
We also have to fix up the exit qualification. It may be wrong because
KVM sets bit 6 but the guest might not.
L1 emulates EPT A/D bits using write permissions, so in principle it may
be possible for EPT A/D bits to be used by L1 even though not available
in hardware. The problem is that guest page-table walks will be treated
as reads rather than writes, so they would not cause an EPT violation.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[Fixed typo in walk_addr_generic() comment and changed bit clear +
conditional-set pattern in handle_ept_violation() to conditional-clear]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
This prepares the MMU paging code for EPT accessed and dirty bits,
which can be enabled optionally at runtime. Code that updates the
accessed and dirty bits will need a pointer to the struct kvm_mmu.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
handle_ept_violation is checking for "guest-linear-address invalid" +
"not a paging-structure walk". However, _all_ EPT violations without
a valid guest linear address are paging structure walks, because those
EPT violations happen when loading the guest PDPTEs.
Therefore, the check can never be true, and even if it were, KVM doesn't
care about the guest linear address; it only uses the guest *physical*
address VMCS field. So, remove the check altogether.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Large pages at the PDPE level can be emulated by the MMU, so the bit
can be set unconditionally in the EPT capabilities MSR. The same is
true of 2MB EPT pages, though all Intel processors with EPT in practice
support those.
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Legacy device assignment has been deprecated since 4.2 (released
1.5 years ago). VFIO is better and everyone should have switched to it.
If they haven't, this should convince them. :)
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Virtual NMIs are only missing in Prescott and Yonah chips. Both are obsolete
for virtualization usage---Yonah is 32-bit only even---so drop vNMI emulation.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
MCG_CAP[63:9] bits are reserved on AMD. However, on an AMD guest, this
MSR returns 0x100010a. More specifically, bit 24 is set, which is simply
wrong. That bit is MCG_SER_P and is present only on Intel. Thus, clean
up the reserved bits in order not to confuse guests.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's combine it in a single function vmx_switch_vmcs().
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
According to the Intel SDM, volume 3, section 28.3.2: Creating and
Using Cached Translation Information, "No linear mappings are used
while EPT is in use." INVEPT will invalidate both the guest-physical
mappings and the combined mappings in the TLBs and paging-structure
caches, so an INVVPID is superfluous.
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Add and use nfnl_msg_type() function to replace opencoded nfnetlink
message type. I suggested this change, Arushi Singhal made an initial
patch to address this but was missing several spots.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
that will also go via the arm64 tree.
-----BEGIN PGP SIGNATURE-----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=WlYI
-----END PGP SIGNATURE-----
Merge tag 'arch-timer-errata' of git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into clockevents/4.12
arm64 arch timer workaround series, including the base patches
that will also go via the arm64 tree.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
The plain Faraday FTTMR010 timer needs a clock to figure out its
tick rate, and the gemini reads it directly from the system
controller set-up. Split the init function and add two paths for
the two compatible-strings. We only support clocking using PCLK
because of lack of documentation on how EXTCLK works.
The Gemini still works like before, but we can also support a
generic, clock-based version.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
After some research it turns out that the "Gemini" timer is
actually a generic IP block from Faraday Technology named
FTTMR010, so as to not make things too confusing we need to
rename the driver and its symbols to make sense.
The implementation remains the same in this patch but we fix
the copy-paste error in the timer name "nomadik_mtu" as we're
at it.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
It turns out that the Cortina Gemini timer block is just a
standard IP block from Faraday Technology named FTTMR010.
In order to make things clear and understandable, we rename the
bindings with a Faraday compatible as primary and the Cortina
gemini as a more specific case.
For the plain Faraday timer we require two clock references,
while the Gemini can keep it's syscon lookup pattern.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
The clocksource and the sched_clock provided by the arm_global_timer
are quite unstable because their rates depend on the cpu frequency.
On the other side, the arm_global_timer has a higher rating than the
rockchip_timer, it will be selected by default by the time framework
while we want to use the stable rockchip clocksource.
Let's disable the arm_global_timer in order to have the rockchip
clocksource selected by default.
Signed-off-by: Alexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
The patch add two timers to all rk3188 based boards.
The first timer is from alive subsystem and it act as a backup
for the local timers at sleep time. It act the same as other
SoC rockchip timers already present in kernel.
The second timer is from CPU subsystem and act as replacement
for the arm-global-timer clocksource and sched clock. It run
at stable frequency 24MHz.
Signed-off-by: Alexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
The clock supplying the arm-global-timer on the rk3188 is coming from the
the cpu clock itself and thus changes its rate everytime cpufreq adjusts
the cpu frequency making this timer unsuitable as a stable clocksource
and sched clock.
The rk3188, rk3288 and following socs share a separate timer block already
handled by the rockchip-timer driver. Therefore adapt this driver to also
be able to act as clocksource and sched clock on rk3188.
In order to test clocksource you can run following commands and check
how much time it take in real. On rk3188 it take about ~45 seconds.
cpufreq-set -f 1.6GHZ
date; sleep 60; date
In order to use the patch you need to declare two timers in the dts
file. The first timer will be initialized as clockevent provider
and the second one as clocksource. The clockevent must be from
alive subsystem as it used as backup for the local timers at sleep
time.
The patch does not break compatibility with older device tree files.
The older device tree files contain only one timer. The timer
will be initialized as clockevent, as expected.
rk3288 (and probably anything newer) is irrelevant to this patch,
as it has the arch timer interface. This patch may be useful
for Cortex-A9/A5 based parts.
Signed-off-by: Alexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Property set to '"rockchip,rk3228-timer", "rockchip,rk3288-timer"'
to match devicetree bindings.
Signed-off-by: Alexander Kochetkov <al.kochet@gmail.com>
Suggested-by: Heiko Stübner <heiko@sntech.de>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Make all properties description in form '"rockchip,<chip>-timer",
"rockchip,rk3288-timer"' for all chips found in linux kernel.
Suggested-by: Heiko Stübner <heiko@sntech.de>
Signed-off-by: Alexander Kochetkov <al.kochet@gmail.com>
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Add an implementation for the ARM delay timer, which is used for
udelay(). This provides less CPU dependent and more accurate delays -
the CPU loop on Marvell Dove appears to calibrate to around 6% too
short.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Rather than reading the clock rate three times, read it once - we are
about to add a fourth usage.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Joao Pinto says:
====================
net: stmmac: adding multiple buffers
This patch adds multiple buffers to stmmac in a more fragmented
way, in order to make problem debug easier.
I would kindly request to people to test this patch in their HWs in
order to check if everything's functional. Thank you.
====================
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the napi variable to the stmmac_rx_queue
structure and forces that operations like netif_queue_stopped,
netif_wake_queue, netif_stop_queue, netdev_reset_queue and
netdev_sent_queue be made by queue.
Signed-off-by: Joao Pinto <jpinto@synopsys.com>
Tested-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the structure stmmac_tx_queue which contains
tx queues specific data (previously in stmmac_priv).
Signed-off-by: Joao Pinto <jpinto@synopsys.com>
Tested-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the structure stmmac_rx_queue which contains
rx queues specific data (previously in stmmac_priv).
Signed-off-by: Joao Pinto <jpinto@synopsys.com>
Tested-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch breaks several functions into RX and TX scopes, which
will be useful when adding multiple buffers mechanism.
Signed-off-by: Joao Pinto <jpinto@synopsys.com>
Tested-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tobias Klauser says:
====================
Use net_device_stats from struct net_device
Along the lines of previous patches, switch (almost) all remaining net
drivers to use net_device_stats from net_device instead of including a
copy of it in their netdev_priv struct.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
pegasus, use stats from struct net_device. Also remove the now
unnecessary .ndo_get_stats function.
Cc: Petko Manolov <petkan@nucleusys.com>
Cc: linux-usb@vger.kernel.org
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
kaweth_device, use stats from struct net_device. Also remove the now
unnecessary .ndo_get_stats function.
Cc: linux-usb@vger.kernel.org
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
typhoon, use stats from struct net_device.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
de_private, use stats from struct net_device.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
happy_meal, use stats from struct net_device.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in struct
bigmac, use stats from struct net_device.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of using a private copy of struct net_device_stats in
struct w90p910_ether, use stats from struct net_device. Also remove
the now unnecessary .ndo_get_stats function.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
