Writing CP0_Compare clears the timer interrupt pending bit
(CP0_Cause.TI), but this wasn't being done atomically. If a timer
interrupt raced with the write of the guest CP0_Compare, the timer
interrupt could end up being pending even though the new CP0_Compare is
nowhere near CP0_Count.
We were already updating the hrtimer expiry with
kvm_mips_update_hrtimer(), which used both kvm_mips_freeze_hrtimer() and
kvm_mips_resume_hrtimer(). Close the race window by expanding out
kvm_mips_update_hrtimer(), and clearing CP0_Cause.TI and setting
CP0_Compare between the freeze and resume. Since the pending timer
interrupt should not be cleared when CP0_Compare is written via the KVM
user API, an ack argument is added to distinguish the source of the
write.
There's a particularly narrow and subtle race condition when the
software emulated guest timer is frozen which can allow a guest timer
interrupt to be missed.
This happens due to the hrtimer expiry being inexact, so very
occasionally the freeze time will be after the moment when the emulated
CP0_Count transitions to the same value as CP0_Compare (so an IRQ should
be generated), but before the moment when the hrtimer is due to expire
(so no IRQ is generated). The IRQ won't be generated when the timer is
resumed either, since the resume CP0_Count will already match CP0_Compare.
With VZ guests in particular this is far more likely to happen, since
the soft timer may be frozen frequently in order to restore the timer
state to the hardware guest timer. This happens after 5-10 hours of
guest soak testing, resulting in an overflow in guest kernel timekeeping
calculations, hanging the guest. A more focussed test case to
intentionally hit the race (with the help of a new hypcall to cause the
timer state to migrated between hardware & software) hits the condition
fairly reliably within around 30 seconds.
Instead of relying purely on the inexact hrtimer expiry to determine
whether an IRQ should be generated, read the guest CP0_Compare and
directly check whether the freeze time is before or after it. Only if
CP0_Count is on or after CP0_Compare do we check the hrtimer expiry to
determine whether the last IRQ has already been generated (which will
have pushed back the expiry by one timer period).
Commit d28bc9dd25ce reversed the order of two lines which initialize cr0,
allowing the current (old) cr0 value to mess up vcpu initialization.
This was observed in the checks for cr0 X86_CR0_WP bit in the context of
kvm_mmu_reset_context(). Besides, setting vcpu->arch.cr0 after vmx_set_cr0()
is completely redundant. Change the order back to ensure proper vcpu
initialization.
The combination of booting with ovmf firmware when guest vcpus > 1 and kvm's
ept=N option being set results in a VM-entry failure. This patch fixes that.
Fixes: d28bc9dd25ce ("KVM: x86: INIT and reset sequences are different") Signed-off-by: Bruce Rogers <brogers@suse.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
MSR 0x2f8 accessed the 124th Variable Range MTRR ever since MTRR support
was introduced by 9ba075a664df ("KVM: MTRR support").
0x2f8 became harmful when 910a6aae4e2e ("KVM: MTRR: exactly define the
size of variable MTRRs") shrinked the array of VR MTRRs from 256 to 8,
which made access to index 124 out of bounds. The surrounding code only
WARNs in this situation, thus the guest gained a limited read/write
access to struct kvm_arch_vcpu.
0x2f8 is not a valid VR MTRR MSR, because KVM has/advertises only 16 VR
MTRR MSRs, 0x200-0x20f. Every VR MTRR is set up using two MSRs, 0x2f8
was treated as a PHYSBASE and 0x2f9 would be its PHYSMASK, but 0x2f9 was
not implemented in KVM, therefore 0x2f8 could never do anything useful
and getting rid of it is safe.
This fixes CVE-2016-3713.
Fixes: 910a6aae4e2e ("KVM: MTRR: exactly define the size of variable MTRRs") Reported-by: David Matlack <dmatlack@google.com> Signed-off-by: Andy Honig <ahonig@google.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
DMA is optional with this driver. If it was not enabled the devpriv->dma
pointer will be NULL.
Fix the possible NULL pointer dereference when trying to disable the DMA
channels in das1800_ai_cancel() and tidy up the comments to fix the
checkpatch.pl issues:
WARNING: line over 80 characters
It's probably harmless in das1800_ai_setup_dma() because the 'desc' pointer
will not be used if DMA is disabled but fix it there also.
Fixes: 99dfc3357e98 ("staging: comedi: das1800: remove depends on ISA_DMA_API limitation") Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Reviewed-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a USB driver is bound to an interface (either through probing or
by claiming it) or is unbound from an interface, the USB core always
disables Link Power Management during the transition and then
re-enables it afterward. The reason is because the driver might want
to prevent hub-initiated link power transitions, in which case the HCD
would have to recalculate the various LPM parameters. This
recalculation takes place when LPM is re-enabled and the new
parameters are sent to the device and its parent hub.
However, if the driver does not want to prevent hub-initiated link
power transitions then none of this work is necessary. The parameters
don't need to be recalculated, and LPM doesn't need to be disabled and
re-enabled.
It turns out that disabling and enabling LPM can be time-consuming,
enough so that it interferes with user programs that want to claim and
release interfaces rapidly via usbfs. Since the usbfs kernel driver
doesn't set the disable_hub_initiated_lpm flag, we can speed things up
and get the user programs to work by leaving LPM alone whenever the
flag isn't set.
And while we're improving the way disable_hub_initiated_lpm gets used,
let's also fix its kerneldoc.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Tested-by: Matthew Giassa <matthew@giassa.net> CC: Mathias Nyman <mathias.nyman@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current implemenentation restart the sent pattern for each entry in
the sg list. The receiving end expects a continuous pattern, and test
will fail unless scatterilst entries happen to be aligned with the
pattern
Fix this by calculating the pattern byte based on total sent size
instead of just the current sg entry.
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Fixes: 8b5249019352 ("[PATCH] USB: usbtest: scatterlist OUT data pattern testing") Acked-by: Felipe Balbi <felipe.balbi@linux.intel.com> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When binding the function to usb_configuration, check whether the thread
is running before starting another one. Without that, when function
instance is added to multiple configurations, fsg_bing starts multiple
threads with all but the latest one being forgotten by the driver. This
leads to obvious thread leaks, possible lockups when trying to halt the
machine and possible more issues.
This fixes issues with legacy/multi¹ gadget as well as configfs gadgets
when mass_storage function is added to multiple configurations.
This change also simplifies API since the legacy gadgets no longer need
to worry about starting the thread by themselves (which was where bug
in legacy/multi was in the first place).
N.B., this patch doesn’t address adding single mass_storage function
instance to a single configuration twice. Thankfully, there’s no
legitimate reason for such setup plus, if I’m not mistaken, configfs
gadget doesn’t even allow it to be expressed.
¹ I have no example failure though. Conclusion that legacy/multi has
a bug is based purely on me reading the code.
Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Michal Nazarewicz <mina86@mina86.com> Tested-by: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com> Cc: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the current implementation functionfs generates a EFAULT for async read
operations if the read buffer size is larger than the URB data size. Since
a application does not necessarily know how much data the host side is
going to send it typically supplies a buffer larger than the actual data,
which will then result in a EFAULT error.
This behaviour was introduced while refactoring the code to use iov_iter
interface in commit c993c39b8639 ("gadget/function/f_fs.c: use put iov_iter
into io_data"). The original code took the minimum over the URB size and
the user buffer size and then attempted to copy that many bytes using
copy_to_user(). If copy_to_user() could not copy all data a EFAULT error
was generated. Restore the original behaviour by only generating a EFAULT
error when the number of bytes copied is not the size of the URB and the
target buffer has not been fully filled.
Commit 342f39a6c8d3 ("usb: gadget: f_fs: fix check in read operation")
already fixed the same problem for the synchronous read path.
Fixes: c993c39b8639 ("gadget/function/f_fs.c: use put iov_iter into io_data") Acked-by: Michal Nazarewicz <mina86@mina86.com> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: lei liu <liu.lei78@zte.com.cn>
[johan: rebase and replace commit message ] Signed-off-by: Johan Hovold <johan@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: lei liu <liu.lei78@zte.com.cn>
[properly sort them - gregkh] Signed-off-by: Johan Hovold <johan@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
URBs and buffers allocated in attach for Epic devices would never be
deallocated in case of a later probe error (e.g. failure to allocate
minor numbers) as disconnect is then never called.
Fix by moving deallocation to release and making sure that the
URBs are first unlinked.
The interface read URB is submitted in attach, but was only unlinked by
the driver at disconnect.
In case of a late probe error (e.g. due to failed minor allocation),
disconnect is never called and we would end up with active URBs for an
unbound interface. This in turn could lead to deallocated memory being
dereferenced in the completion callback.
Fixes: f7a33e608d9a ("USB: serial: add quatech2 usb to serial driver") Signed-off-by: Johan Hovold <johan@kernel.org> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The interface instat and indat URBs were submitted in attach, but never
unlinked in release before deallocating the corresponding transfer
buffers.
In the case of a late probe error (e.g. due to failed minor allocation),
disconnect would not have been called before release, causing the
buffers to be freed while the URBs are still in use. We'd also end up
with active URBs for an unbound interface.
The interface read and event URBs are submitted in attach, but were
never explicitly unlinked by the driver. Instead the URBs would have
been killed by usb-serial core on disconnect.
In case of a late probe error (e.g. due to failed minor allocation),
disconnect is never called and we could end up with active URBs for an
unbound interface. This in turn could lead to deallocated memory being
dereferenced in the completion callbacks.
When a message is received and amthif client is not in reading state
the message is ignored and left dangling in the queue. This may happen
after one of the amthif host connections is closed w/o completing the
reading. Another client will pick up a wrong message on next read
attempt which will lead to link reset.
To prevent this the driver has to properly discard the message when
amthif client is not in reading state.
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
hci_vhci driver creates a hci device object dynamically upon each
HCI_VENDOR_PKT write. Although it checks the already created object
and returns an error, it's still racy and may build multiple hci_dev
objects concurrently when parallel writes are performed, as the device
tracks only a single hci_dev object.
This patch introduces a mutex to protect against the concurrent device
creations.
The write handler allocates skbs and queues them into data->readq.
Read side should read them, if there is any. If there is none, skbs
should be dropped by hdev->flush. But this happens only if the device
is HCI_UP, i.e. hdev->power_on work was triggered already. When it was
not, skbs stay allocated in the queue when /dev/vhci is closed. So
purge the queue in ->release.
Program to reproduce:
#include <err.h>
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
Both vhci_get_user and vhci_release race with open_timeout work. They
both contain cancel_delayed_work_sync, but do not test whether the
work actually created hdev or not. Since the work can be in progress
and _sync will wait for finishing it, we can have data->hdev allocated
when cancel_delayed_work_sync returns. But the call sites do 'if
(data->hdev)' *before* cancel_delayed_work_sync.
As a result:
* vhci_get_user allocates a second hdev and puts it into
data->hdev. The former is leaked.
* vhci_release does not release data->hdev properly as it thinks there
is none.
Fix both cases by moving the actual test *after* the call to
cancel_delayed_work_sync.
This can be hit by this program:
#include <err.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <unistd.h>
#include <sys/stat.h>
#include <sys/types.h>
int main(int argc, char **argv)
{
int fd;
srand(time(NULL));
while (1) {
const int delta = (rand() % 200 - 100) * 100;
fd = open("/dev/vhci", O_RDWR);
if (fd < 0)
err(1, "open");
And the result is:
BUG: KASAN: use-after-free in skb_queue_tail+0x13e/0x150 at addr ffff88006b0c1228
Read of size 8 by task kworker/u13:1/32068
=============================================================================
BUG kmalloc-192 (Tainted: G E ): kasan: bad access detected
-----------------------------------------------------------------------------
Disabling lock debugging due to kernel taint
INFO: Allocated in vhci_open+0x50/0x330 [hci_vhci] age=260 cpu=3 pid=32040
...
kmem_cache_alloc_trace+0x150/0x190
vhci_open+0x50/0x330 [hci_vhci]
misc_open+0x35b/0x4e0
chrdev_open+0x23b/0x510
...
INFO: Freed in vhci_release+0xa4/0xd0 [hci_vhci] age=9 cpu=2 pid=32040
...
__slab_free+0x204/0x310
vhci_release+0xa4/0xd0 [hci_vhci]
...
INFO: Slab 0xffffea0001ac3000 objects=16 used=13 fp=0xffff88006b0c1e00 flags=0x5fffff80004080
INFO: Object 0xffff88006b0c1200 @offset=4608 fp=0xffff88006b0c0600
Bytes b4 ffff88006b0c11f0: 09 df 00 00 01 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffff88006b0c1200: 00 06 0c 6b 00 88 ff ff 00 00 00 00 00 00 00 00 ...k............
Object ffff88006b0c1210: 10 12 0c 6b 00 88 ff ff 10 12 0c 6b 00 88 ff ff ...k.......k....
Object ffff88006b0c1220: c0 46 c2 6b 00 88 ff ff c0 46 c2 6b 00 88 ff ff .F.k.....F.k....
Object ffff88006b0c1230: 01 00 00 00 01 00 00 00 e0 ff ff ff 0f 00 00 00 ................
Object ffff88006b0c1240: 40 12 0c 6b 00 88 ff ff 40 12 0c 6b 00 88 ff ff @..k....@..k....
Object ffff88006b0c1250: 50 0d 6e a0 ff ff ff ff 00 02 00 00 00 00 ad de P.n.............
Object ffff88006b0c1260: 00 00 00 00 00 00 00 00 ab 62 02 00 01 00 00 00 .........b......
Object ffff88006b0c1270: 90 b9 19 81 ff ff ff ff 38 12 0c 6b 00 88 ff ff ........8..k....
Object ffff88006b0c1280: 03 00 20 00 ff ff ff ff ff ff ff ff 00 00 00 00 .. .............
Object ffff88006b0c1290: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffff88006b0c12a0: 00 00 00 00 00 00 00 00 00 80 cd 3d 00 88 ff ff ...........=....
Object ffff88006b0c12b0: 00 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 . ..............
Redzone ffff88006b0c12c0: bb bb bb bb bb bb bb bb ........
Padding ffff88006b0c13f8: 00 00 00 00 00 00 00 00 ........
CPU: 3 PID: 32068 Comm: kworker/u13:1 Tainted: G B E 4.4.6-0-default #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.1-0-g4adadbd-20151112_172657-sheep25 04/01/2014
Workqueue: hci0 hci_cmd_work [bluetooth] 00000000ffffffffffffffff81926cfaffff88006be37c68ffff88006bc27180 ffff88006b0c1200ffff88006b0c1234ffffffff81577993ffffffff82489320 ffff88006bc242400000000000000046ffff88006a100000000000026e51eb80
Call Trace:
...
[<ffffffff81ec8ebe>] ? skb_queue_tail+0x13e/0x150
[<ffffffffa06e027c>] ? vhci_send_frame+0xac/0x100 [hci_vhci]
[<ffffffffa0c61268>] ? hci_send_frame+0x188/0x320 [bluetooth]
[<ffffffffa0c61515>] ? hci_cmd_work+0x115/0x310 [bluetooth]
[<ffffffff811a1375>] ? process_one_work+0x815/0x1340
[<ffffffff811a1f85>] ? worker_thread+0xe5/0x11f0
[<ffffffff811a1ea0>] ? process_one_work+0x1340/0x1340
[<ffffffff811b3c68>] ? kthread+0x1c8/0x230
...
Memory state around the buggy address: ffff88006b0c1100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff88006b0c1180: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff88006b0c1200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^ ffff88006b0c1280: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc ffff88006b0c1300: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
008GE0 Toshiba mmc in some Intel Baytrail tablets responds to
MMC_SEND_EXT_CSD in 450-600ms.
This patch will...
() Increase the long read time quirk timeout from 300ms to 600ms. Original
author of that quirk says 300ms was only a guess and that the number
may need to be raised in the future.
() Add this specific MMC to the quirk
Signed-off-by: Matt Gumbel <matthew.k.gumbel@intel.com> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some BIOSes unconditionally send an ACPI notification to RBTN when the
system is resuming from suspend. This makes dell-rbtn send an input
event to userspace as if a function key was pressed. Prevent this by
ignoring all the notifications received while the device is suspended.
The order of the _OSI related functionalities is as follows:
acpi_blacklisted()
acpi_dmi_osi_linux()
acpi_osi_setup()
acpi_osi_setup()
acpi_update_interfaces() if "!*"
<<<<<<<<<<<<<<<<<<<<<<<<
parse_args()
__setup("acpi_osi=")
acpi_osi_setup_linux()
acpi_update_interfaces() if "!*"
<<<<<<<<<<<<<<<<<<<<<<<<
acpi_early_init()
acpi_initialize_subsystem()
acpi_ut_initialize_interfaces()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
acpi_bus_init()
acpi_os_initialize1()
acpi_install_interface_handler(acpi_osi_handler)
acpi_osi_setup_late()
acpi_update_interfaces() for "!"
>>>>>>>>>>>>>>>>>>>>>>>>
acpi_osi_handler()
Since acpi_osi_setup_linux() can override acpi_dmi_osi_linux(), the command
line setting can override the DMI detection. That's why acpi_blacklisted()
is put before __setup("acpi_osi=").
Then we can notice the following wrong invocation order. There are
acpi_update_interfaces() (marked by <<<<) calls invoked before
acpi_ut_initialize_interfaces() (marked by ^^^^). This makes it impossible
to use acpi_osi=!* correctly from OSI DMI table or from the command line.
The use of acpi_osi=!* is meant to disable both ACPICA
(acpi_gbl_supported_interfaces) and Linux specific strings
(osi_setup_entries) while the ACPICA part should have stopped working
because of the order issue.
This patch fixes this issue by moving acpi_update_interfaces() to where
it is invoked for acpi_osi=! (marked by >>>>) as this is ensured to be
invoked after acpi_ut_initialize_interfaces() (marked by ^^^^). Linux
specific strings are still handled in the original place in order to make
the following command line working: acpi_osi=!* acpi_osi="Module Device".
Note that since acpi_osi=!* is meant to further disable linux specific
string comparing to the acpi_osi=!, there is no such use case in our bug
fixing work and hence there is no one using acpi_osi=!* either from the
command line or from the DMI quirks, this issue is just a theoretical
issue.
Fixes: 741d81280ad2 (ACPI: Add facility to remove all _OSI strings) Tested-by: Lukas Wunner <lukas@wunner.de> Tested-by: Chen Yu <yu.c.chen@intel.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some devices connected to the SDHCI controller may have separate enabling
lines that are controlled through GPIO. These devices need to be powered
on and enabled before probing. This is to ensure all devices connected can
be seen by the controller.
Note, for "stable" this patch depends on the following change:
commit 78a898d0e395 ("ACPI / PM: Export acpi_device_fix_up_power()")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Reported-and-tested-by: Laszlo Fiat <laszlo.fiat@gmail.com> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Reported-by: Laszlo Fiat <laszlo.fiat@gmail.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=112571 Link: http://lkml.kernel.org/r/CA+7w51inLtQSr656bJvOjGG9oQWKYPXH+xxDPJKbeJ=CcrkS9Q@mail.gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some eMMCs set the partition switch timeout too low.
Now typically eMMCs are considered a critical component (e.g. because
they store the root file system) and consequently are expected to be
reliable. Thus we can neglect the use case where eMMCs can't switch
reliably and we might want a lower timeout to facilitate speedy
recovery.
Although we could employ a quirk for the cards that are affected (if
we could identify them all), as described above, there is little
benefit to having a low timeout, so instead simply set a minimum
timeout.
The minimum is set to 300ms somewhat arbitrarily - the examples that
have been seen had a timeout of 10ms but were sometimes taking 60-70ms.
As described in 'can: m_can: tag current CAN FD controllers as non-ISO'
(6cfda7fbebe) it is possible to define fixed configuration options by
setting the according bit in 'ctrlmode' and clear it in 'ctrlmode_supported'.
This leads to the incovenience that the fixed configuration bits can not be
passed by netlink even when they have the correct values (e.g. non-ISO, FD).
This patch fixes that issue and not only allows fixed set bit values to be set
again but now requires(!) to provide these fixed values at configuration time.
A valid CAN FD configuration consists of a nominal/arbitration bittiming, a
data bittiming and a control mode with CAN_CTRLMODE_FD set - which is now
enforced by a new can_validate() function. This fix additionally removed the
inconsistency that was prohibiting the support of 'CANFD-only' controller
drivers, like the RCar CAN FD.
For this reason a new helper can_set_static_ctrlmode() has been introduced to
provide a proper interface to handle static enabled CAN controller options.
Reported-by: Ramesh Shanmugasundaram <ramesh.shanmugasundaram@bp.renesas.com> Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net> Reviewed-by: Ramesh Shanmugasundaram <ramesh.shanmugasundaram@bp.renesas.com> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The GICv3 driver wrongly assumes that it runs on the non-secure
side of a secure-enabled system, while it could be on a system
with a single security state, or a GICv3 with GICD_CTLR.DS set.
Either way, it is important to configure this properly, or
interrupts will simply not be delivered on this HW.
Reported-by: Peter Maydell <peter.maydell@linaro.org> Tested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When an IPI is generated by a CPU, the pattern looks roughly like:
<write shared data>
smp_wmb();
<write to GIC to signal SGI>
On the receiving CPU we rely on the fact that, once we've taken the
interrupt, then the freshly written shared data must be visible to us.
Put another way, the CPU isn't going to speculate taking an interrupt.
Unfortunately, this assumption turns out to be broken.
Consider that CPUx wants to send an IPI to CPUy, which will cause CPUy
to read some shared_data. Before CPUx has done anything, a random
peripheral raises an IRQ to the GIC and the IRQ line on CPUy is raised.
CPUy then takes the IRQ and starts executing the entry code, heading
towards gic_handle_irq. Furthermore, let's assume that a bunch of the
previous interrupts handled by CPUy were SGIs, so the branch predictor
kicks in and speculates that irqnr will be <16 and we're likely to
head into handle_IPI. The prefetcher then grabs a speculative copy of
shared_data which contains a stale value.
Meanwhile, CPUx gets round to updating shared_data and asking the GIC
to send an SGI to CPUy. Internally, the GIC decides that the SGI is
more important than the peripheral interrupt (which hasn't yet been
ACKed) but doesn't need to do anything to CPUy, because the IRQ line
is already raised.
CPUy then reads the ACK register on the GIC, sees the SGI value which
confirms the branch prediction and we end up with a stale shared_data
value.
This patch fixes the problem by adding an smp_rmb() to the IPI entry
code in gic_handle_irq. As it turns out, the combination of a control
dependency and an ISB instruction from the EOI in the GICv3 driver is
enough to provide the ordering we need, so we add a comment there
justifying the absence of an explicit smp_rmb().
Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On a Freescale i.MX53 based board we ran into "BUG: scheduling while
atomic" because input_inject_event locks interrupts, but
imx_pwm_config_v2 sleeps.
Reported-by: H. Nikolaus Schaller <hns@goldelico.com> Tested-by: H. Nikolaus Schaller <hns@goldelico.com> Signed-off-by: Roger Quadros <rogerq@ti.com> Signed-off-by: Lee Jones <lee.jones@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Systems show a minimal load average of 0.00, 0.01, 0.05 even when they
have no load at all.
Uptime and /proc/loadavg on all systems with kernels released during the
last five years up until kernel version 4.6-rc5, show a 5- and 15-minute
minimum loadavg of 0.01 and 0.05 respectively. This should be 0.00 on
idle systems, but the way the kernel calculates this value prevents it
from getting lower than the mentioned values.
Likewise but not as obviously noticeable, a fully loaded system with no
processes waiting, shows a maximum 1/5/15 loadavg of 1.00, 0.99, 0.95
(multiplied by number of cores).
Once the (old) load becomes 93 or higher, it mathematically can never
get lower than 93, even when the active (load) remains 0 forever.
This results in the strange 0.00, 0.01, 0.05 uptime values on idle
systems. Note: 93/2048 = 0.0454..., which rounds up to 0.05.
It is not correct to add a 0.5 rounding (=1024/2048) here, since the
result from this function is fed back into the next iteration again,
so the result of that +0.5 rounding value then gets multiplied by
(2048-2037), and then rounded again, so there is a virtual "ghost"
load created, next to the old and active load terms.
By changing the way the internally kept value is rounded, that internal
value equivalent now can reach 0.00 on idle, and 1.00 on full load. Upon
increasing load, the internally kept load value is rounded up, when the
load is decreasing, the load value is rounded down.
The modified code was tested on nohz=off and nohz kernels. It was tested
on vanilla kernel 4.6-rc5 and on centos 7.1 kernel 3.10.0-327. It was
tested on single, dual, and octal cores system. It was tested on virtual
hosts and bare hardware. No unwanted effects have been observed, and the
problems that the patch intended to fix were indeed gone.
Tested-by: Damien Wyart <damien.wyart@free.fr> Signed-off-by: Vik Heyndrickx <vik.heyndrickx@veribox.net> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Doug Smythies <dsmythies@telus.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Fixes: 0f004f5a696a ("sched: Cure more NO_HZ load average woes") Link: http://lkml.kernel.org/r/e8d32bff-d544-7748-72b5-3c86cc71f09f@veribox.net Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch adds the CLK_SET_RATE_PARENT flag for the crypto core and
ahb blocks. Without this flag, clk_set_rate can fail for certain
frequency requests.
Signed-off-by: Andy Gross <andy.gross@linaro.org> Fixes: 3966fab8b6ab ("clk: qcom: Add MSM8916 Global Clock Controller support") Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current sun4i-ss driver could generate data corruption when ciphering/deciphering.
It occurs randomly on end of handled data.
No root cause have been found and the only way to remove it is to replace
all spin_lock_bh by their irq counterparts.
caam_jr_alloc() used to return NULL if a JR device could not be
allocated for a session. In turn, every user of this function used
IS_ERR() function to verify if anything went wrong, which does NOT look
for NULL values. This made the kernel crash if the sanity check failed,
because the driver continued to think it had allocated a valid JR dev
instance to the session and at some point it tries to do a caam_jr_free()
on a NULL JR dev pointer.
This patch is a fix for this issue.
Signed-off-by: Catalin Vasile <cata.vasile@nxp.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Thus it overflows and the resulting number is less than 4080, which makes
3823 / 4080 = 0
an nr_pages is set to this. As we already checked against the minimum that
nr_pages may be, this causes the logic to fail as well, and we crash the
kernel.
There's no reason to have the two DIV_ROUND_UP() (that's just result of
historical code changes), clean up the code and fix this bug.
Fixes: 83f40318dab00 ("ring-buffer: Make removal of ring buffer pages atomic") Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The size variable to change the ring buffer in ftrace is a long. The
nr_pages used to update the ring buffer based on the size is int. On 64 bit
machines this can cause an overflow problem.
For example, the following will cause the ring buffer to crash:
WARNING: CPU: 1 PID: 318 at kernel/trace/ring_buffer.c:1527 rb_update_pages+0x22f/0x260
Which is:
RB_WARN_ON(cpu_buffer, nr_removed);
Note each ring buffer page holds 4080 bytes.
This is because:
1) 10 causes the ring buffer to have 3 pages.
(10kb requires 3 * 4080 pages to hold)
2) (2^31 / 2^10 + 1) * 4080 = 8556384240
The value written into buffer_size_kb is shifted by 10 and then passed
to ring_buffer_resize(). 8556384240 * 2^10 = 8761737461760
3) The size passed to ring_buffer_resize() is then divided by BUF_PAGE_SIZE
which is 4080. 8761737461760 / 4080 = 2147484672
4) nr_pages is subtracted from the current nr_pages (3) and we get: 2147484669. This value is saved in a signed integer nr_pages_to_update
5) 2147484669 is greater than 2^31 but smaller than 2^32, a signed int
turns into the value of -2147482627
6) As the value is a negative number, in update_pages_handler() it is
negated and passed to rb_remove_pages() and 2147482627 pages will
be removed, which is much larger than 3 and it causes the warning
because not all the pages asked to be removed were removed.
In testing with HiKey, we found that since
commit 3f30b158eba5 ("asix: On RX avoid creating bad Ethernet
frames"),
we're seeing lots of noise during network transfers:
[ 239.027993] asix 1-1.1:1.0 eth0: asix_rx_fixup() Data Header synchronisation was lost, remaining 988
[ 239.037310] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0x54ebb5ec, offset 4
[ 239.045519] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0xcdffe7a2, offset 4
[ 239.275044] asix 1-1.1:1.0 eth0: asix_rx_fixup() Data Header synchronisation was lost, remaining 988
[ 239.284355] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0x1d36f59d, offset 4
[ 239.292541] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0xaef3c1e9, offset 4
[ 239.518996] asix 1-1.1:1.0 eth0: asix_rx_fixup() Data Header synchronisation was lost, remaining 988
[ 239.528300] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0x2881912, offset 4
[ 239.536413] asix 1-1.1:1.0 eth0: asix_rx_fixup() Bad Header Length 0x5638f7e2, offset 4
And network throughput ends up being pretty bursty and slow with
a overall throughput of at best ~30kB/s (where as previously we
got 1.1MB/s with the slower USB1.1 "full speed" host).
We found the issue also was reproducible on a x86_64 system,
using a "high-speed" USB2.0 port but the throughput did not
measurably drop (possibly due to the scp transfer being cpu
bound on my slow test hardware).
After lots of debugging, I found the check added in the
problematic commit seems to be calculating the offset
incorrectly.
In the normal case, in the main loop of the function, we do:
(where offset is zero, or set to "offset += (copy_length + 1) &
0xfffe" in the previous loop)
rx->header = get_unaligned_le32(skb->data +
offset);
offset += sizeof(u32);
But the problematic patch calculates:
offset = ((rx->remaining + 1) & 0xfffe) + sizeof(u32);
rx->header = get_unaligned_le32(skb->data + offset);
Adding some debug logic to check those offset calculation used
to find rx->header, the one in problematic code is always too
large by sizeof(u32).
Thus, this patch removes the incorrect " + sizeof(u32)" addition
in the problematic calculation, and resolves the issue.
Cc: Dean Jenkins <Dean_Jenkins@mentor.com> Cc: "David B. Robins" <linux@davidrobins.net> Cc: Mark Craske <Mark_Craske@mentor.com> Cc: Emil Goode <emilgoode@gmail.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: YongQin Liu <yongqin.liu@linaro.org> Cc: Guodong Xu <guodong.xu@linaro.org> Cc: Ivan Vecera <ivecera@redhat.com> Cc: linux-usb@vger.kernel.org Cc: netdev@vger.kernel.org Reported-by: Yongqin Liu <yongqin.liu@linaro.org> Signed-off-by: John Stultz <john.stultz@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
See [MS-NLMP] 3.2.5.1.2 Server Receives an AUTHENTICATE_MESSAGE from the Client:
...
Set NullSession to FALSE
If (AUTHENTICATE_MESSAGE.UserNameLen == 0 AND
AUTHENTICATE_MESSAGE.NtChallengeResponse.Length == 0 AND
(AUTHENTICATE_MESSAGE.LmChallengeResponse == Z(1)
OR
AUTHENTICATE_MESSAGE.LmChallengeResponse.Length == 0))
-- Special case: client requested anonymous authentication
Set NullSession to TRUE
...
Only server which map unknown users to guest will allow
access using a non-null NTChallengeResponse.
For Samba it's the "map to guest = bad user" option.
Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Wrong return code was being returned on SMB3 rmdir of
non-empty directory.
For SMB3 (unlike for cifs), we attempt to delete a directory by
set of delete on close flag on the open. Windows clients set
this flag via a set info (SET_FILE_DISPOSITION to set this flag)
which properly checks if the directory is empty.
With this patch on smb3 mounts we correctly return
"DIRECTORY NOT EMPTY"
on attempts to remove a non-empty directory.
Signed-off-by: Steve French <steve.french@primarydata.com> Acked-by: Sachin Prabhu <sprabhu@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The EC field of the constructed ESR is conditionally modified by ORing in
ESR_ELx_EC_DABT_LOW for a data abort. However, ESR_ELx_EC_SHIFT is missing
from this condition.
Signed-off-by: Matt Evans <matt.evans@arm.com> Acked-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ARM architecture mandates that when changing a page table entry
from a valid entry to another valid entry, an invalid entry is first
written, TLB invalidated, and only then the new entry being written.
The current code doesn't respect this, directly writing the new
entry and only then invalidating TLBs. Let's fix it up.
Reported-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The loop that browses the array compat_hwcap_str will stop when a NULL
is encountered, however NULL is missing at the end of array. This will
lead to overrun until a NULL is found somewhere in the following memory.
In reality, this works out because the compat_hwcap2_str array tends to
follow immediately in memory, and that *is* terminated correctly.
Furthermore, the unsigned int compat_elf_hwcap is checked before
printing each capability, so we end up doing the right thing because
the size of the two arrays is less than 32. Still, this is an obvious
mistake and should be fixed.
Note for backporting: commit 12d11817eaafa414 ("arm64: Move
/proc/cpuinfo handling code") moved this code in v4.4. Prior to that
commit, the same change should be made in arch/arm64/kernel/setup.c.
Fixes: 44b82b7700d0 "arm64: Fix up /proc/cpuinfo" Signed-off-by: Julien Grall <julien.grall@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The update to the accessed or dirty states for block mappings must be
done atomically on hardware with support for automatic AF/DBM. The
ptep_set_access_flags() function has been fixed as part of commit 66dbd6e61a52 ("arm64: Implement ptep_set_access_flags() for hardware
AF/DBM"). This patch brings pmdp_set_access_flags() in line with the pte
counterpart.
Fixes: 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits") Reviewed-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When hardware updates of the access and dirty states are enabled, the
default ptep_set_access_flags() implementation based on calling
set_pte_at() directly is potentially racy. This triggers the "racy dirty
state clearing" warning in set_pte_at() because an existing writable PTE
is overridden with a clean entry.
There are two main scenarios for this situation:
1. The CPU getting an access fault does not support hardware updates of
the access/dirty flags. However, a different agent in the system
(e.g. SMMU) can do this, therefore overriding a writable entry with a
clean one could potentially lose the automatically updated dirty
status
2. A more complex situation is possible when all CPUs support hardware
AF/DBM:
a) Initial state: shareable + writable vma and pte_none(pte)
b) Read fault taken by two threads of the same process on different
CPUs
c) CPU0 takes the mmap_sem and proceeds to handling the fault. It
eventually reaches do_set_pte() which sets a writable + clean pte.
CPU0 releases the mmap_sem
d) CPU1 acquires the mmap_sem and proceeds to handle_pte_fault(). The
pte entry it reads is present, writable and clean and it continues
to pte_mkyoung()
e) CPU1 calls ptep_set_access_flags()
If between (d) and (e) the hardware (another CPU) updates the dirty
state (clears PTE_RDONLY), CPU1 will override the PTR_RDONLY bit
marking the entry clean again.
This patch implements an arm64-specific ptep_set_access_flags() function
to perform an atomic update of the PTE flags.
Fixes: 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits") Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Reported-by: Ming Lei <tom.leiming@gmail.com> Tested-by: Julien Grall <julien.grall@arm.com> Cc: Will Deacon <will.deacon@arm.com>
[will: reworded comment] Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, pmd_present() only checks for a non-zero value, returning
true even after pmd_mknotpresent() (which only clears the type bits).
This patch converts pmd_present() to using pte_present(), similar to the
other pmd_*() checks. As a side effect, it will return true for
PROT_NONE mappings, though they are not yet used by the kernel with
transparent huge pages.
For consistency, also change pmd_mknotpresent() to only clear the
PMD_SECT_VALID bit, even though the PMD_TABLE_BIT is already 0 for block
mappings (no functional change). The unused PMD_SECT_PROT_NONE
definition is removed as transparent huge pages use the pte page prot
values.
Fixes: 9c7e535fcc17 ("arm64: mm: Route pmd thp functions through pte equivalents") Reviewed-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With hardware AF/DBM support, pmd modifications (transparent huge pages)
should be performed atomically using load/store exclusive. The initial
patches defined the get-and-clear function and __HAVE_ARCH_* macro
without the "huge" word, leaving the pmdp_huge_get_and_clear() to the
default, non-atomic implementation.
Fixes: 2f4b829c625e ("arm64: Add support for hardware updates of the access and dirty pte bits") Reviewed-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jann reported that the ptrace_may_access() check in
find_lively_task_by_vpid() is racy against exec().
Specifically:
perf_event_open() execve()
ptrace_may_access()
commit_creds()
... if (get_dumpable() != SUID_DUMP_USER)
perf_event_exit_task();
perf_install_in_context()
would result in installing a counter across the creds boundary.
Fix this by wrapping lots of perf_event_open() in cred_guard_mutex.
This should be fine as perf_event_exit_task() is already called with
cred_guard_mutex held, so all perf locks already nest inside it.
Reported-by: Jann Horn <jannh@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: He Kuang <hekuang@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, the PT driver always sets the PMI bit one region (page) before
the STOP region so that we can wake up the consumer before we run out of
room in the buffer and have to disable the event. However, we also need
an interrupt in the last output region, so that we actually get to disable
the event (if no more room from new data is available at that point),
otherwise hardware just quietly refuses to start, but the event is
scheduled in and we end up losing trace data till the event gets removed.
For a cpu-wide event it is even worse since there may not be any
re-scheduling at all and no chance for the ring buffer code to notice
that its buffer is filled up and the event needs to be disabled (so that
the consumer can re-enable it when it finishes reading the data out). In
other words, all the trace data will be lost after the buffer gets filled
up.
This patch makes PT also generate a PMI when the last output region is
full.
Reported-by: Markus Metzger <markus.t.metzger@intel.com> Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Cc: vince@deater.net Link: http://lkml.kernel.org/r/1462886313-13660-2-git-send-email-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The slab name ends up being visible in the directory structure under
/sys, and even if you don't have access rights to the file you can see
the filenames.
Just use a 64-bit counter instead of the pointer to the 'net' structure
to generate a unique name.
This code will go away in 4.7 when the conntrack code moves to a single
kmemcache, but this is the backportable simple solution to avoiding
leaking kernel pointers to user space.
Fixes: 5b3501faa874 ("netfilter: nf_conntrack: per netns nf_conntrack_cachep") Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If device replace entry was found on disk at mounting and its num_write_errors
stats counter has non-NULL value, then replace operation will never be
finished and -EIO error will be reported by btrfs_scrub_dev() because
this counter is never reset.
# mount -o degraded /media/a4fb5c0a-21c5-4fe7-8d0e-fdd87d5f71ee/
# btrfs replace status /media/a4fb5c0a-21c5-4fe7-8d0e-fdd87d5f71ee/
Started on 25.Mar 07:28:00, canceled on 25.Mar 07:28:01 at 0.0%, 40 write errs, 0 uncorr. read errs
# btrfs replace start -B 4 /dev/sdg /media/a4fb5c0a-21c5-4fe7-8d0e-fdd87d5f71ee/
ERROR: ioctl(DEV_REPLACE_START) failed on "/media/a4fb5c0a-21c5-4fe7-8d0e-fdd87d5f71ee/": Input/output error, no error
Reset num_write_errors and num_uncorrectable_read_errors counters in the
dev_replace structure before start of replacing.
Signed-off-by: Yauhen Kharuzhy <yauhen.kharuzhy@zavadatar.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
create_pending_snapshot() will go readonly on _any_ error return from
btrfs_qgroup_inherit(). If qgroups are enabled, a user can crash their fs by
just making a snapshot and asking it to inherit from an invalid qgroup. For
example:
$ btrfs sub snap -i 1/10 /btrfs/ /btrfs/foo
Will cause a transaction abort.
Fix this by only throwing errors in btrfs_qgroup_inherit() when we know
going readonly is acceptable.
The following xfstests test case reproduces this bug:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
here=`pwd`
tmp=/tmp/$$
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
cd /
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
# remove previous $seqres.full before test
rm -f $seqres.full
# real QA test starts here
_supported_fs btrfs
_supported_os Linux
_require_scratch
rm -f $seqres.full
_scratch_mkfs
_scratch_mount
_run_btrfs_util_prog quota enable $SCRATCH_MNT
# The qgroup '1/10' does not exist and should be silently ignored
_run_btrfs_util_prog subvolume snapshot -i 1/10 $SCRATCH_MNT $SCRATCH_MNT/snap1
_scratch_unmount
echo "Silence is golden"
status=0
exit
Signed-off-by: Mark Fasheh <mfasheh@suse.de> Reviewed-by: Qu Wenruo <quwenruo@cn.fujitsu.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dan Carpenter's static checker has found this error, it's introduced by
commit 64c043de466d
("Btrfs: fix up read_tree_block to return proper error")
It's really supposed to 'break' the loop on error like others.
Cc: Dan Carpenter <dan.carpenter@oracle.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
csum_dirty_buffer was issuing a warning in case the extent buffer
did not look alright, but was still returning success.
Let's return error in this case, and also add an additional sanity
check on the extent buffer header.
The caller up the chain may BUG_ON on this, for example flush_epd_write_bio will,
but it is better than to have a silent metadata corruption on disk.
Signed-off-by: Alex Lyakas <alex@zadarastorage.com> Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When logging that an inode exists, for example as part of a directory
fsync operation, we were collecting any ordered extents for the inode but
we ended up doing nothing with them except tagging them as processed, by
setting the flag BTRFS_ORDERED_LOGGED on them, which prevented a
subsequent fsync of that inode (using the LOG_INODE_ALL mode) from
collecting and processing them. This created a time window where a second
fsync against the inode, using the fast path, ended up not logging the
checksums for the new extents but it logged the extents since they were
part of the list of modified extents. This happened because the ordered
extents were not collected and checksums were not yet added to the csum
tree - the ordered extents have not gone through btrfs_finish_ordered_io()
yet (which is where we add them to the csum tree by calling
inode.c:add_pending_csums()).
So fix this by not collecting an inode's ordered extents if we are logging
it with the LOG_INODE_EXISTS mode.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we're about to do a fast fsync for an inode and btrfs_inode_in_log()
returns false, it's possible that we had an ordered extent in progress
(btrfs_finish_ordered_io() not run yet) when we noticed that the inode's
last_trans field was not greater than the id of the last committed
transaction, but shortly after, before we checked if there were any
ongoing ordered extents, the ordered extent had just completed and
removed itself from the inode's ordered tree, in which case we end up not
logging the inode, losing some data if a power failure or crash happens
after the fsync handler returns and before the transaction is committed.
Fix this by checking first if there are any ongoing ordered extents
before comparing the inode's last_trans with the id of the last committed
transaction - when it completes, an ordered extent always updates the
inode's last_trans before it removes itself from the inode's ordered
tree (at btrfs_finish_ordered_io()).
Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This happened because under specific timings the path for direct IO reads
can deadlock with concurrent buffered writes. The diagram below shows how
this happens for an example file that has the following layout:
[ extent A ] [ extent B ] [ ....
0K 4K 8K
CPU 1 CPU 2 CPU 3
DIO read against range
[0K, 8K[ starts
btrfs_direct_IO()
--> calls btrfs_get_blocks_direct()
which finds the extent map for the
extent A and leaves the range
[0K, 4K[ locked in the inode's
io tree
buffered write against
range [4K, 8K[ starts
__btrfs_buffered_write()
--> dirties page at 4K
a user space
task calls sync
for e.g or
writepages() is
invoked by mm
writepages()
run_delalloc_range()
cow_file_range()
--> ordered extent X
for the buffered
write is created
and
writeback starts
--> calls btrfs_get_blocks_direct()
again, without submitting first
a bio for reading extent A, and
finds the extent map for extent B
--> calls lock_extent_direct()
--> locks range [4K, 8K[
--> finds ordered extent X
covering range [4K, 8K[
--> unlocks range [4K, 8K[
buffered write against
range [0K, 8K[ starts
__btrfs_buffered_write()
prepare_pages()
--> locks pages with
offsets 0 and 4K
lock_and_cleanup_extent_if_need()
--> blocks attempting to
lock range [0K, 8K[ in
the inode's io tree,
because the range [0, 4K[
is already locked by the
direct IO task at CPU 1
--> calls
btrfs_start_ordered_extent(oe X)
btrfs_start_ordered_extent(oe X)
--> At this point writeback for ordered
extent X has not finished yet
filemap_fdatawrite_range()
btrfs_writepages()
extent_writepages()
extent_write_cache_pages()
--> finds page with offset 0
with the writeback tag
(and not dirty)
--> tries to lock it
--> deadlock, task at CPU 2
has the page locked and
is blocked on the io range
[0, 4K[ that was locked
earlier by this task
So fix this by falling back to a buffered read in the direct IO read path
when an ordered extent for a buffered write is found.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When using the same file as the source and destination for a dedup
(extent_same ioctl) operation we were allowing it to dedup to a
destination offset beyond the file's size, which doesn't make sense and
it's not allowed for the case where the source and destination files are
not the same file. This made de deduplication operation successful only
when the source range corresponded to a hole, a prealloc extent or an
extent with all bytes having a value of 0x00. This was also leaving a
file hole (between i_size and destination offset) without the
corresponding file extent items, which can be reproduced with the
following steps for example:
$ mkfs.btrfs -f /dev/sdi
$ mount /dev/sdi /mnt/sdi
$ xfs_io -f -c "pwrite -S 0xab 304457 404990" /mnt/sdi/foobar
wrote 404990/404990 bytes at offset 304457
395 KiB, 99 ops; 0.0000 sec (31.150 MiB/sec and 7984.5149 ops/sec)
$ /git/hub/duperemove/btrfs-extent-same 24576 /mnt/sdi/foobar 28672 /mnt/sdi/foobar 929792
Deduping 2 total files
(28672, 24576): /mnt/sdi/foobar
(929792, 24576): /mnt/sdi/foobar
1 files asked to be deduped
i: 0, status: 0, bytes_deduped: 24576
24576 total bytes deduped in this operation
$ umount /mnt/sdi
$ btrfsck /dev/sdi
Checking filesystem on /dev/sdi
UUID: 98c528aa-0833-427d-9403-b98032ffbf9d
checking extents
checking free space cache
checking fs roots
root 5 inode 257 errors 100, file extent discount
Found file extent holes:
start: 712704, len: 217088
found 540673 bytes used err is 1
total csum bytes: 400
total tree bytes: 131072
total fs tree bytes: 32768
total extent tree bytes: 16384
btree space waste bytes: 123675
file data blocks allocated: 671744
referenced 671744
btrfs-progs v4.2.3
So fix this by not allowing the destination to go beyond the file's size,
just as we do for the same where the source and destination files are not
the same.
A test for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have two cases where we end up deleting a file at log replay time
when we should not. For this to happen the file must have been renamed
and a directory inode must have been fsynced/logged.
Two examples that exercise these two cases are listed below.
Case 1)
$ mkfs.btrfs -f /dev/sdb
$ mount /dev/sdb /mnt
$ mkdir -p /mnt/a/b
$ mkdir /mnt/c
$ touch /mnt/a/b/foo
$ sync
$ mv /mnt/a/b/foo /mnt/c/
# Create file bar just to make sure the fsync on directory a/ does
# something and it's not a no-op.
$ touch /mnt/a/bar
$ xfs_io -c "fsync" /mnt/a
< power fail / crash >
The next time the filesystem is mounted, the log replay procedure
deletes file foo.
The next time the filesystem is mounted, the log replay procedure
deletes file bar.
The reason why the files are deleted is because when we log inodes
other then the fsync target inode, we ignore their last_unlink_trans
value and leave the log without enough information to later replay the
rename operations. So we need to look at the last_unlink_trans values
and fallback to a transaction commit if they are greater than the
id of the last committed transaction.
So fix this by looking at the last_unlink_trans values and fallback to
transaction commits when needed. Also, when logging other inodes (for
case 1 we logged descendants of the fsync target inode while for case 2
we logged ascendants) we need to care about concurrent tasks updating
the last_unlink_trans of inodes we are logging (which was already an
existing problem in check_parent_dirs_for_sync()). Since we can not
acquire their inode mutex (vfs' struct inode ->i_mutex), as that causes
deadlocks with other concurrent operations that acquire the i_mutex of
2 inodes (other fsyncs or renames for example), we need to serialize on
the log_mutex of the inode we are logging. A task setting a new value for
an inode's last_unlink_trans must acquire the inode's log_mutex and it
must do this update before doing the actual unlink operation (which is
already the case except when deleting a snapshot). Conversely the task
logging the inode must first log the inode and then check the inode's
last_unlink_trans value while holding its log_mutex, as if its value is
not greater then the id of the last committed transaction it means it
logged a safe state of the inode's items, while if its value is not
smaller then the id of the last committed transaction it means the inode
state it has logged might not be safe (the concurrent task might have
just updated last_unlink_trans but hasn't done yet the unlink operation)
and therefore a transaction commit must be done.
Test cases for xfstests follow in separate patches.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we delete a snapshot, fsync its parent directory and crash/power fail
before the next transaction commit, on the next mount when we attempt to
replay the log tree of the root containing the parent directory we will
fail and prevent the filesystem from mounting, which is solvable by wiping
out the log trees with the btrfs-zero-log tool but very inconvenient as
we will lose any data and metadata fsynced before the parent directory
was fsynced.
For example:
$ mkfs.btrfs -f /dev/sdc
$ mount /dev/sdc /mnt
$ mkdir /mnt/testdir
$ btrfs subvolume snapshot /mnt /mnt/testdir/snap
$ btrfs subvolume delete /mnt/testdir/snap
$ xfs_io -c "fsync" /mnt/testdir
< crash / power failure and reboot >
$ mount /dev/sdc /mnt
mount: mount(2) failed: No such file or directory
And in dmesg/syslog we get the following message and trace:
This happens because when we are replaying the log and processing the
directory entry pointing to the snapshot in the subvolume tree, we treat
its btrfs_dir_item item as having a location with a key type matching
BTRFS_INODE_ITEM_KEY, which is wrong because the type matches
BTRFS_ROOT_ITEM_KEY and therefore must be processed differently, as the
object id refers to a root number and not to an inode in the root
containing the parent directory.
So fix this by triggering a transaction commit if an fsync against the
parent directory is requested after deleting a snapshot. This is the
simplest approach for a rare use case. Some alternative that avoids the
transaction commit would require more code to explicitly delete the
snapshot at log replay time (factoring out common code from ioctl.c:
btrfs_ioctl_snap_destroy()), special care at fsync time to remove the
log tree of the snapshot's root from the log root of the root of tree
roots, amongst other steps.
A test case for xfstests that triggers the issue follows.
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
tmp=/tmp/$$
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
_cleanup_flakey
cd /
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
# real QA test starts here
_need_to_be_root
_supported_fs btrfs
_supported_os Linux
_require_scratch
_require_dm_target flakey
_require_metadata_journaling $SCRATCH_DEV
# Create a snapshot at the root of our filesystem (mount point path), delete it,
# fsync the mount point path, crash and mount to replay the log. This should
# succeed and after the filesystem is mounted the snapshot should not be visible
# anymore.
_run_btrfs_util_prog subvolume snapshot $SCRATCH_MNT $SCRATCH_MNT/snap1
_run_btrfs_util_prog subvolume delete $SCRATCH_MNT/snap1
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT
_flakey_drop_and_remount
[ -e $SCRATCH_MNT/snap1 ] && \
echo "Snapshot snap1 still exists after log replay"
# Similar scenario as above, but this time the snapshot is created inside a
# directory and not directly under the root (mount point path).
mkdir $SCRATCH_MNT/testdir
_run_btrfs_util_prog subvolume snapshot $SCRATCH_MNT $SCRATCH_MNT/testdir/snap2
_run_btrfs_util_prog subvolume delete $SCRATCH_MNT/testdir/snap2
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/testdir
_flakey_drop_and_remount
[ -e $SCRATCH_MNT/testdir/snap2 ] && \
echo "Snapshot snap2 still exists after log replay"
_unmount_flakey
echo "Silence is golden"
status=0
exit
Signed-off-by: Filipe Manana <fdmanana@suse.com> Tested-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current practical default is ~4k on x86_64 (the logic is more complex,
simplified for brevity), the inlined files land in the metadata group and
thus consume space that could be needed for the real metadata.
The inlining brings some usability surprises:
1) total space consumption measured on various filesystems and btrfs
with DUP metadata was quite visible because of the duplicated data
within metadata
2) inlined data may exhaust the metadata, which are more precious in case
the entire device space is allocated to chunks (ie. balance cannot
make the space more compact)
3) performance suffers a bit as the inlined blocks are duplicate and
stored far away on the device.
Proposed fix: set the default to 2048
This fixes namely 1), the total filesysystem space consumption will be on
par with other filesystems.
Partially fixes 2), more data are pushed to the data block groups.
The characteristics of 3) are based on actual small file size
distribution.
The change is independent of the metadata blockgroup type (though it's
most visible with DUP) or system page size as these parameters are not
trival to find out, compared to file size.
Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Let's remove the error message that appears when the tree_id is not
present. This can happen with the quota tree and has been observed in
practice. The applications are supposed to handle -ENOENT and we don't
need to report that in the system log as it's not a fatal error.
truncate_space_check is using btrfs_csum_bytes_to_leaves() but forgetting to
multiply by nodesize so we get an actual byte count. We need a tracepoint here
so that we have the matching reserve for the release that will come later. Also
add a comment to make clear what the intent of truncate_space_check is.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is an issue observed when we hotplug a second DP
4K monitor to the system. Sometimes, the link training
fails for the second monitor after HPD interrupt
generation.
The issue happens when some queued or deferred transactions
are already present on the AUX channel when we initiate
a new transcation to (say) get DPCD or during link training.
We set AUX_IGNORE_HPD_DISCON bit in the AUX_CONTROL
register so that we can ignore any such deferred
transactions when a new AUX transaction is initiated.
Signed-off-by: Arindam Nath <arindam.nath@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Right now MST audio is causing too many kernel panics to really keep
around in the kernel. On top of that, even after fixing said panics it's
still basically non-functional (at least on all the setups I've tested
it on). Revert until we have a proper solution for this.
LPT is pch, so might run into the fdi bandwidth constraint (especially
since it has only 2 lanes). But right now we just force pipe_bpp back
to 24, resulting in a nice loop (which we bail out with a loud
WARN_ON). Fix this.
Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
References: https://bugs.freedesktop.org/show_bug.cgi?id=93477 Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Tested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/1462264381-7573-1-git-send-email-daniel.vetter@ffwll.ch
(cherry picked from commit f58a1acc7e4a1f37d26124ce4c875c647fbcc61f) Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On DCE6.1 PPLL2 is exclusively available to UNIPHYA, so it should not
be taken into consideration when looking for an already enabled PLL
to be shared with other outputs.
This fixes the broken VGA port (TRAVIS DP->VGA bridge) on my Richland
based laptop, where the internal display is connected to UNIPHYA through
a TRAVIS DP->LVDS bridge.
Update CDCLK_FREQ on BDW after changing the cdclk frequency. Not sure
if this is a late addition to the spec, or if I simply overlooked this
step when writing the original code.
This is what Bspec has to say about CDCLK_FREQ:
"Program this field to the CD clock frequency minus one. This is used to
generate a divided down clock for miscellaneous timers in display."
And the "Broadwell Sequences for Changing CD Clock Frequency" section
clarifies this further:
"For CD clock 337.5 MHz, program 337 decimal.
For CD clock 450 MHz, program 449 decimal.
For CD clock 540 MHz, program 539 decimal.
For CD clock 675 MHz, program 674 decimal."
NULL pointer derefence happens when booting with DTB because the
platform data for haptic device is not set in supplied data from parent
MFD device.
The MFD device creates only platform data (from Device Tree) for itself,
not for haptic child.
Unable to handle kernel NULL pointer dereference at virtual address 0000009c
pgd = c0004000
[0000009c] *pgd=00000000
Internal error: Oops: 5 [#1] PREEMPT SMP ARM
(max8997_haptic_probe) from [<c03f9cec>] (platform_drv_probe+0x4c/0xb0)
(platform_drv_probe) from [<c03f8440>] (driver_probe_device+0x214/0x2c0)
(driver_probe_device) from [<c03f8598>] (__driver_attach+0xac/0xb0)
(__driver_attach) from [<c03f67ac>] (bus_for_each_dev+0x68/0x9c)
(bus_for_each_dev) from [<c03f7a38>] (bus_add_driver+0x1a0/0x218)
(bus_add_driver) from [<c03f8db0>] (driver_register+0x78/0xf8)
(driver_register) from [<c0101774>] (do_one_initcall+0x90/0x1d8)
(do_one_initcall) from [<c0a00dbc>] (kernel_init_freeable+0x15c/0x1fc)
(kernel_init_freeable) from [<c06bb5b4>] (kernel_init+0x8/0x114)
(kernel_init) from [<c0107938>] (ret_from_fork+0x14/0x3c)
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Fixes: 104594b01ce7 ("Input: add driver support for MAX8997-haptic")
[k.kozlowski: Write commit message, add CC-stable] Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Payloads of NM entries are not supposed to contain NUL. When we run
into such, only the part prior to the first NUL goes into the
concatenation (i.e. the directory entry name being encoded by a bunch
of NM entries). We do stop when the amount collected so far + the
claimed amount in the current NM entry exceed 254. So far, so good,
but what we return as the total length is the sum of *claimed*
sizes, not the actual amount collected. And that can grow pretty
large - not unlimited, since you'd need to put CE entries in
between to be able to get more than the maximum that could be
contained in one isofs directory entry / continuation chunk and
we are stop once we'd encountered 32 CEs, but you can get about 8Kb
easily. And that's what will be passed to readdir callback as the
name length. 8Kb __copy_to_user() from a buffer allocated by
__get_free_page()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
At the end of process_filter(), collapse_tree() was changed to update
the parg parameter, but the reassignment after the call wasn't removed.
What happens is that the "current_op" gets modified and freed and parg
is assigned to the new allocated argument. But after the call to
collapse_tree(), parg is assigned again to the just freed "current_op",
and this causes the tool to crash.
The current_op variable must also be assigned to NULL in case of error,
otherwise it will cause it to be free()ed twice.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Acked-by: Namhyung Kim <namhyung@kernel.org> Fixes: 42d6194d133c ("tools lib traceevent: Refactor process_filter()") Link: http://lkml.kernel.org/r/20160511150936.678c18a1@gandalf.local.home Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The qla1280 driver sets the scsi_host_template's can_queue field to 0xfffff
which results in an allocation failure when allocating the block layer tags
for the driver's queues. This was introduced with the change for host wide
tags in commit 64d513ac31b - "scsi: use host wide tags by default".
Reduce can_queue to MAX_OUTSTANDING_COMMANDS (512) to solve the allocation
error.
Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Fixes: 64d513ac31b - "scsi: use host wide tags by default" Cc: Laura Abbott <labbott@redhat.com> Cc: Michael Reed <mdr@sgi.com> Reviewed-by: Laurence Oberman <loberman@redhat.com> Reviewed-by: Lee Duncan <lduncan@suse.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <jejb@linux.vnet.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* if we have a hashed negative dentry and either CREAT|EXCL on
r/o filesystem, or CREAT|TRUNC on r/o filesystem, or CREAT|EXCL
with failing may_o_create(), we should fail with EROFS or the
error may_o_create() has returned, but not ENOENT. Which is what
the current code ends up returning.
* if we have CREAT|TRUNC hitting a regular file on a read-only
filesystem, we can't fail with EROFS here. At the very least,
not until we'd done follow_managed() - we might have a writable
file (or a device, for that matter) bound on top of that one.
Moreover, the code downstream will see that O_TRUNC and attempt
to grab the write access (*after* following possible mount), so
if we really should fail with EROFS, it will happen. No need
to do that inside atomic_open().
The real logics is much simpler than what the current code is
trying to do - if we decided to go for simple lookup, ended
up with a negative dentry *and* had create_error set, fail with
create_error. No matter whether we'd got that negative dentry
from lookup_real() or had found it in dcache.
The minium voltage of 1800mV is a copy and paste error from the axp20x
regulator info. The correct minimum voltage for the ldo_io regulators
on the axp22x is 700mV.
Fixes: 1b82b4e4f954 ("regulator: axp20x: Add support for AXP22X regulators") Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Chen-Yu Tsai <wens@csie.org> Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The buck9 regulator of S2MPS11 PMIC had incorrect vsel_mask (0xff
instead of 0x1f) thus reading entire register as buck9's voltage. This
effectively caused regulator core to interpret values as higher voltages
than they were and then to set real voltage much lower than intended.
The buck9 provides power to other regulators, including LDO13
and LDO19 which supply the MMC2 (SD card). On Odroid XU3/XU4 the lower
voltage caused SD card detection errors on Odroid XU3/XU4:
mmc1: card never left busy state
mmc1: error -110 whilst initialising SD card
During driver probe the regulator core was checking whether initial
voltage matches the constraints. With incorrect vsel_mask of 0xff and
default value of 0x50, the core interpreted this as 5 V which is outside
of constraints (3-3.775 V). Then the regulator core was adjusting the
voltage to match the constraints. With incorrect vsel_mask this new
voltage mapped to a vere low voltage in the driver.
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com> Tested-by: Javier Martinez Canillas <javier@osg.samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
| If a down prepare callback fails, then DOWN_FAILED is invoked for all
| callbacks which have successfully executed DOWN_PREPARE.
|
| But, workqueue has actually two notifiers. One which handles
| UP/DOWN_FAILED/ONLINE and one which handles DOWN_PREPARE.
|
| Now look at the priorities of those callbacks:
|
| CPU_PRI_WORKQUEUE_UP = 5
| CPU_PRI_WORKQUEUE_DOWN = -5
|
| So the call order on DOWN_PREPARE is:
|
| CB 1
| CB ...
| CB workqueue_up() -> Ignores DOWN_PREPARE
| CB ...
| CB X ---> Fails
|
| So we call up to CB X with DOWN_FAILED
|
| CB 1
| CB ...
| CB workqueue_up() -> Handles DOWN_FAILED
| CB ...
| CB X-1
|
| So the problem is that the workqueue stuff handles DOWN_FAILED in the up
| callback, while it should do it in the down callback. Which is not a good idea
| either because it wants to be called early on rollback...
|
| Brilliant stuff, isn't it? The hotplug rework will solve this problem because
| the callbacks become symetric, but for the existing mess, we need some
| workaround in the workqueue code.
The boot CPU handles housekeeping duty(unbound timers, workqueues,
timekeeping, ...) on behalf of full dynticks CPUs. It must remain
online when nohz full is enabled. There is a priority set to every
notifier_blocks:
So tick_nohz_cpu_down callback failed when down prepare cpu 0, and
notifier_blocks behind tick_nohz_cpu_down will not be called any
more, which leads to workers are actually not unbound. Then hotplug
state machine will fallback to undo and online cpu 0 again. Workers
will be rebound unconditionally even if they are not unbound and
trigger the warning in this progress.
This patch fix it by catching !DISASSOCIATED to avoid rebind bound
workers.
Cc: Tejun Heo <tj@kernel.org> Cc: Lai Jiangshan <jiangshanlai@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Suggested-by: Lai Jiangshan <jiangshanlai@gmail.com> Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The memory range assigned to the PMC (Power Management Controller) was
not including the PMC_PCR register which are used to control peripheral
clocks.
This was working fine thanks to the page granularity of ioremap(), but
started to fail when we switched to syscon/regmap, because regmap is
making sure that all accesses are falling into the reserved range.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> Reported-by: Richard Genoud <richard.genoud@gmail.com> Tested-by: Richard Genoud <richard.genoud@gmail.com> Fixes: 863a81c3be1d ("clk: at91: make use of syscon to share PMC registers in several drivers") Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If a file is renamed to a hardlink of itself POSIX specifies that rename(2)
should do nothing and return success.
This condition is checked in vfs_rename(). However it won't detect hard
links on overlayfs where these are given separate inodes on the overlayfs
layer.
Overlayfs itself detects this condition and returns success without doing
anything, but then vfs_rename() will proceed as if this was a successful
rename (detach_mounts(), d_move()).
The correct thing to do is to detect this condition before even calling
into overlayfs. This patch does this by calling vfs_select_inode() to get
the underlying inodes.
When the PMU driver reports a truncated AUX record, it effectively means
that there is no more usable room in the event's AUX buffer (even though
there may still be some room, so that perf_aux_output_begin() doesn't take
action). At this point the consumer still has to be woken up and the event
has to be disabled, otherwise the event will just keep spinning between
perf_aux_output_begin() and perf_aux_output_end() until its context gets
unscheduled.
Again, for cpu-wide events this means never, so once in this condition,
they will be forever losing data.
Fix this by disabling the event and waking up the consumer in case of a
truncated AUX record.
Reported-by: Markus Metzger <markus.t.metzger@intel.com> Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Cc: vince@deater.net Link: http://lkml.kernel.org/r/1462886313-13660-3-git-send-email-alexander.shishkin@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / linux / log