On the Chromecast, the bootloader provides us with an ATAG_MEM of
start=0x01000000 and size=0x3eff8000. This is clearly incorrect, as the
range given encompasses nearly a GiB but the Chromecast only has 512MiB
of RAM! Additionally, this causes the kernel to be decompressed at
0x00008000, below the claimed beginning of RAM, and so the boot fails.
Since the existing ATAG parsing code runs before the kernel is even
decompressed and irrevocably patches the device tree, don't even try
to bypass it. Instead, use the "linux,usable-memory" property instead
of the "reg" property to define the real range. The ATAG code only
overwrites reg, but linux,usable-memory is checked first in the OF
driver, so the fact that reg gets changed makes no difference.
Signed-off-by: Thomas Hebb <tommyhebb@gmail.com>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Valve Steam Link is a consumer device built around the Marvell BG2CD SoC.
This board file enables the UART, USB and Ethernet interfaces as well as
internal I2C and SDIO, and adds SoC voltage regulator and board-specific
GPIO restart method info.
Cc: Sam Lantinga <saml@valvesoftware.com>
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
1. Driver expects specific order of GPIO interrupt banks. For S5Pv220
and Exynos5410 this order was not preserved so fix and document it.
2. Remove support for Exynos5440 (tree-wide, support is dropped because
there are no real users of this platform, it also did not get testing
since long time).
3. Fix lost state of GPF1..5 pins on Exynos5433 during system suspend.
-----BEGIN PGP SIGNATURE-----
iQItBAABCAAXBQJbBb/7EBxrcnprQGtlcm5lbC5vcmcACgkQwTdm5oaLg9dVGg/5
AX0PgNolOWgApCrqu/fdI5aiTk9CXjezl/OnBR3mZ8iqvacJA2aRrQUZKoGnNHhb
qZp7TzT8UukLq3Aye3zgwjeyFE+FcWDCxVNoJ2pifXNhGU5pkkb8InY/EhcnEX4g
q/ASr54aZ0HpqFgBcKlM3IG/+PQJHIrp54HanGTOxLoBe9nbbWXdtgBNbD9zTv0h
XDpHTDg8mLhvBqFPLR9IMs4sxI7XsQL3r5c84BbXZWjJJ08DReFpiqqwea6y3QBu
rS3qbSjfAr3TndUx3znUx3OWWWG+M65zDSlhpWq0+4/osjrCLo+tRxULq2Aqx7+L
MfIRyN0gRHqf40Eh15xCrZFwXbXTS78sc+gtlnezuNk/BgCGpSI9mAOAjDW6I+Dm
zZg9imJBEV68qmWFoYeTZX2Aa829YYsSyE9iX3+qz29FllB5dDj4mfImmTywYsDk
HyYiyXPR0DU4PvlYRDszwltQpFS5sgO9fkaOQKA45FPWJUc67rhUlrdMD+L/zggy
gGjWx+Ou0n3/wvjpxbUcIqVerdMWKW1xIimiA2CAMqrVQ1zf9aykUY14oVCvZMqP
VcTSUcJfvhQr8AdUxm1uByTaLfZ682l/KHubxGLnMr3Pg0Yr71xaCjTdkr4KP3df
ZCg9ECt5am6dhf6TRXPdUsKPOWMTa2vfm036ZysdwOk=
=7nCe
-----END PGP SIGNATURE-----
Merge tag 'samsung-pinctrl-4.18-2' of https://git.kernel.org/pub/scm/linux/kernel/git/pinctrl/samsung into devel
Samsung pinctrl drivers changes for v4.18
1. Driver expects specific order of GPIO interrupt banks. For S5Pv220
and Exynos5410 this order was not preserved so fix and document it.
2. Remove support for Exynos5440 (tree-wide, support is dropped because
there are no real users of this platform, it also did not get testing
since long time).
3. Fix lost state of GPF1..5 pins on Exynos5433 during system suspend.
- Add support for the new R-Car E3 SoC,
- Add I2C pin groups on R-Car M3-N,
- Small fixes and cleanups.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=xeSy
-----END PGP SIGNATURE-----
Merge tag 'sh-pfc-for-v4.18-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into devel
pinctrl: sh-pfc: Updates for v4.18 (take two)
- Add support for the new R-Car E3 SoC,
- Add I2C pin groups on R-Car M3-N,
- Small fixes and cleanups.
This is useful if the board file needs to reference it.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
This adds most of the remaining Designware IP cores under APB trees in
the interest of documenting assignment of interrupts and memory ranges.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
This adds DT nodes for the Cortex-A9 MPCore SCU, local watchdog and
most importantly the global timer.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Cortex-A9 PMU has no associated memory ranges and "make dtbs W=1" warns
about missing reg or ranges property. To avoid the warning, move the PMU
node out of soc subtree to the root.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Use the correct trigger type for Cortex-A9. This was fixed for several
other SoCs since the kernel started issuing a boot-time warning.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
I believe the flush_cache_all() after scu_enable() is to "Ensure that
the data accessed by CPU0 before the SCU was initialised is visible
to the other CPUs." as commented in scu_enable(). So here
flush_cache_all() is a duplication, remove it.
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
BG2CD SoC uses r3p0 Cortex-A9 MPCore single-CPU cluster. Autoselect
pertinent errata, the SCU and the global timer, and allow use of the
local timer on uniprocessor kernels.
PL310 L2 cache controller has revision r3p2; no errata to select.
Signed-off-by: Alexander Monakov <amonakov@ispras.ru>
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
This patch exports tm_enable()/tm_disable/tm_abort() APIs, which
will be used for PR KVM transactional memory logic.
Signed-off-by: Simon Guo <wei.guo.simon@gmail.com>
Reviewed-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patches add some macros for CR0/TEXASR bits so that PR KVM TM
logic (tbegin./treclaim./tabort.) can make use of them later.
Signed-off-by: Simon Guo <wei.guo.simon@gmail.com>
Reviewed-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
PR KVM will need to reuse msr_check_and_set().
This patch exports this API for reuse.
Signed-off-by: Simon Guo <wei.guo.simon@gmail.com>
Reviewed-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
. Create extra kernel maps to help in decoding samples in x86 PTI entry
trampolines (Adrian Hunter)
. Copy x86 PTI entry trampoline sections in the kcore copy used for
annotation and intel_pt CPU traces decoding (Adrian Hunter)
- Support 'perf annotate --group' for non-explicit recorded event
"groups", showing multiple columns, one for each event, just like
when dealing with explicit event groups (those enclosed with {}) (Jin Yao)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----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=9PtG
-----END PGP SIGNATURE-----
Merge tag 'perf-core-for-mingo-4.18-20180523' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements from Arnaldo Carvalho de Melo:
- Create extra kernel maps to help in decoding samples in x86 PTI entry
trampolines (Adrian Hunter)
- Copy x86 PTI entry trampoline sections in the kcore copy used for
annotation and intel_pt CPU traces decoding (Adrian Hunter)
- Support 'perf annotate --group' for non-explicit recorded event
"groups", showing multiple columns, one for each event, just like
when dealing with explicit event groups (those enclosed with {}) (Jin Yao)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Follow the recent trend for the license description, and also fix the
wrongly stated X11 to MIT.
As already pointed on the DT ML, the X11 license text [1] is explicitly
for the X Consortium and has a couple of extra clauses. The MIT
license text [2] is actually what the current DT files claim.
[1] https://spdx.org/licenses/X11.html
[2] https://spdx.org/licenses/MIT.html
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Follow the recent trend for the license description, and also fix the
wrongly stated X11 to MIT.
As already pointed on the DT ML, the X11 license text [1] is explicitly
for the X Consortium and has a couple of extra clauses. The MIT
license text [2] is actually what the current DT files claim.
[1] https://spdx.org/licenses/X11.html
[2] https://spdx.org/licenses/MIT.html
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Add new macro to wait on a register field to be a specific
value.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Define new clock and powergating flags for VCN block.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add the powerplay callback for powergating VCN (same
as UVD and VCE).
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
VCN clockgating is handled manually like VCE and UVD.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We should return unlock on the error path
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
error: uninitialized symbol 'xxxx'
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The vbios mistakenly sets this bit on some boards without ECC.
This can lead to reduced performance in some workloads. Disable
the bit if the board does not have ECC.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Needed for proper memory setup depending on whether ECC is
enabled on a particular board.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The ForceParWrRMW setting needs to be enabled for ECC, but disabled
when ECC is not enabled.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
First of all it's already being called from the display code from amd_ip_funcs.suspend/resume hooks.
Second of all, the place in amdgpu_device_gpu_recover it's being called is wrong for GPU stalls since
it is called BEFORE we cancel and force completion of all in flight jobs which were not yet processed.
So, as Bas pointed in the ticket we will try to wait for fence in amdgpu_pm_compute_clocks but the pipe
is hanged so we end up in deadlock.
v2: remove unused variable
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106500
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If io_destroy() gets to cancelling everything that can be cancelled and
gets to kiocb_cancel() calling the function driver has left in ->ki_cancel,
it becomes vulnerable to a race with IO completion. At that point req
is already taken off the list and aio_complete() does *NOT* spin until
we (in free_ioctx_users()) releases ->ctx_lock. As the result, it proceeds
to kiocb_free(), freing req just it gets passed to ->ki_cancel().
Fix is simple - remove from the list after the call of kiocb_cancel(). All
instances of ->ki_cancel() already have to cope with the being called with
iocb still on list - that's what happens in io_cancel(2).
Cc: stable@kernel.org
Fixes: 0460fef2a9 "aio: use cancellation list lazily"
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Enable support for 2 more hardware windows. This require enabling a few
more clocks and set proper plane type for all windows. In the new
configuration primary plane uses hardware window no 3 and cursor uses
window no 5. The remaining hardware windows are used for overlays. This
gives us an overlay plane both below and above primary plane for both
Decon and DeconTV (which uses hardware window nr 0 for background).
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Proper support for Decon's hardware window no 4 and 5 require enabling
a few more clocks ("aclk_smmu_decon1x", "aclk_xiu_decon1x",
"pclk_smmu_decon1x"). Add those clocks to required clocks list in Decon's
dt bindings.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Acked-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
- Add IPP v2 framework.
. it is a rewritten version of the Exynos mem-to-mem image processing
framework which supprts color space conversion, image up/down-scaling
and rotation. This new version replaces existing userspace API with
new easy-to-use and simple ones so we have already applied the use of
these API to real user, Tizen Platform[1], and also makes existing
Scaler, FIMC, GScaler and Rotator drivers to use IPP v2 core API.
And below are patch lists we have applied to a real user,
https://git.tizen.org/cgit/platform/adaptation/samsung_exynos/libtdm-exynos/log/?h=tizen&qt=grep&q=ipphttps://git.tizen.org/cgit/platform/adaptation/samsung_exynos/libtdm-exynos/commit/?h=tizen&id=b59be207365d10efd489e6f71c8a045b558c44fehttps://git.tizen.org/cgit/platform/kernel/linux-exynos/log/?h=tizen&qt=grep&q=ipp
TDM(Tizen Display Manager) is a Display HAL for Tizen platform.
Ps. Only real user using IPP API is Tizen.
[1] https://www.tizen.org/
- Two cleanups
. One is to just remove mode_set callback from MIPI-DSI driver
because drm_display_mode data is already available from crtc
atomic state.
. And other is to just use new return type, vm_fault_t
for page fault handler.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=M74Y
-----END PGP SIGNATURE-----
Merge tag 'exynos-drm-next-for-v4.18' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into exynos-drm-next
- Add S5PV210 FIMD variant support.
- Add IPP v2 framework.
. it is a rewritten version of the Exynos mem-to-mem image processing
framework which supprts color space conversion, image up/down-scaling
and rotation. This new version replaces existing userspace API with
new easy-to-use and simple ones so we have already applied the use of
these API to real user, Tizen Platform[1], and also makes existing
Scaler, FIMC, GScaler and Rotator drivers to use IPP v2 core API.
And below are patch lists we have applied to a real user,
https://git.tizen.org/cgit/platform/adaptation/samsung_exynos/libtdm-exynos/log/?h=tizen&qt=grep&q=ipphttps://git.tizen.org/cgit/platform/adaptation/samsung_exynos/libtdm-exynos/commit/?h=tizen&id=b59be207365d10efd489e6f71c8a045b558c44fehttps://git.tizen.org/cgit/platform/kernel/linux-exynos/log/?h=tizen&qt=grep&q=ipp
TDM(Tizen Display Manager) is a Display HAL for Tizen platform.
Ps. Only real user using IPP API is Tizen.
[1] https://www.tizen.org/
- Two cleanups
. One is to just remove mode_set callback from MIPI-DSI driver
because drm_display_mode data is already available from crtc
atomic state.
. And other is to just use new return type, vm_fault_t
for page fault handler.
This seems to be Synaptics specific, as some Elan touchpads are not
correctly switching to SMBus if we call deactivate before switching to
SMBus on cold boot and on resume.
Tested with the T480s
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
New ICs are using a different scheme for the alternate bus parameter.
Given that they are new and are only using either PS2 only or PS2 + SMBus
Host Notify, we force those new ICs to use the SMBus solution for enhanced
reporting.
This allows the touchpad found on the Lenovo T480s to report 5 fingers
every 8 ms, instead of having a limit of 2 every 8 ms.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Many of the Elantech devices are connected through PS/2 and a different
bus (SMBus or plain I2C).
To not break any existing device, we only enable SMBus based
on a module parameter. If some laptops require the quirk to
be set, we will have to rely on a list of PNPIds or MDI matching
to individually expose those hardware over SMBus.
the parameter mentioned above is elantech_smbus from the psmouse
module.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The command ETP_RESOLUTION_QUERY also contains the bus information.
It is better to fetch it once, while we are querying for device
information.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
In preparation for SMBus device support, move static device
information that we query form the touchpad upon initialization into
separate structure. This will allow us to query the device without
allocating memory first.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The Elan touchpads over I2C/SMBus also can handle a trackstick.
Unfortunately, nothing tells us if the device supports trackstick (the
information lies in the PS/2 node), so rely on device properties to
determine whether to enable the trackstick node.
Link: https://bugzilla.redhat.com/show_bug.cgi?id=1313939
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: Rob Herring <robh@kernel.org>
Acked-by: KT Liao <kt.liao@emc.com.tw>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Currently we request control of native PCIe hotplug unconditionally.
Native PCIe hotplug events are handled by the pciehp driver, and if it is
not enabled those events will be lost.
Request control of native PCIe hotplug only if the pciehp driver is
enabled, so we will actually handle native PCIe hotplug events.
Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
After a suspend/resume cycle the Presence Detect or Data Link Layer Status
Changed bits might be set. If we don't clear them those events will not
fire anymore and nothing happens for instance when a device is now
hot-unplugged.
Fix this by clearing those bits in a newly introduced function
pcie_reenable_notification(). This should be fine because immediately
after, we check if the adapter is still present by reading directly from
the status register.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: stable@vger.kernel.org
Every once in a while, we should update the examples
to reflect more recent kernel versions.
Update the tables describing kernel releases, the merge window,
and current longterm maintained kernel, from 2.6-era kernels
to 4.x.
Signed-off-by: Tim Bird <tim.bird@sony.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The device node iterators perform an of_node_get() on each iteration, so a
jump out of the loop requires an of_node_put().
The semantic patch that fixes this problem is as follows
(http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
iterator name for_each_child_of_node;
@@
for_each_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
+ of_node_put(child);
? break;
...
}
... when != child
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
When posted work request, it need to compute the length of
all sges of every wr and fill it into the msg_len field of
send wqe. Thus, While posting multiple wr,
tmp_len should be reinitialized to zero.
Fixes: 8b9b8d143b ("RDMA/hns: Fix the endian problem for hns")
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When use cq record db for kernel, it needs to set the hr_cq->db_en
to 1 and configure the dma address of record cq db of qp context.
Fixes: 86188a8810 ("RDMA/hns: Support cq record doorbell for kernel space")
Signed-off-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
From 0aa2e9b921d6db71150633ff290199554f0842a8 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 23 May 2018 10:29:00 -0700
cgwb_release() punts the actual release to cgwb_release_workfn() on
system_wq. Depending on the number of cgroups or block devices, there
can be a lot of cgwb_release_workfn() in flight at the same time.
We're periodically seeing close to 256 kworkers getting stuck with the
following stack trace and overtime the entire system gets stuck.
[<ffffffff810ee40c>] _synchronize_rcu_expedited.constprop.72+0x2fc/0x330
[<ffffffff810ee634>] synchronize_rcu_expedited+0x24/0x30
[<ffffffff811ccf23>] bdi_unregister+0x53/0x290
[<ffffffff811cd1e9>] release_bdi+0x89/0xc0
[<ffffffff811cd645>] wb_exit+0x85/0xa0
[<ffffffff811cdc84>] cgwb_release_workfn+0x54/0xb0
[<ffffffff810a68d0>] process_one_work+0x150/0x410
[<ffffffff810a71fd>] worker_thread+0x6d/0x520
[<ffffffff810ad3dc>] kthread+0x12c/0x160
[<ffffffff81969019>] ret_from_fork+0x29/0x40
[<ffffffffffffffff>] 0xffffffffffffffff
The events leading to the lockup are...
1. A lot of cgwb_release_workfn() is queued at the same time and all
system_wq kworkers are assigned to execute them.
2. They all end up calling synchronize_rcu_expedited(). One of them
wins and tries to perform the expedited synchronization.
3. However, that invovles queueing rcu_exp_work to system_wq and
waiting for it. Because #1 is holding all available kworkers on
system_wq, rcu_exp_work can't be executed. cgwb_release_workfn()
is waiting for synchronize_rcu_expedited() which in turn is waiting
for cgwb_release_workfn() to free up some of the kworkers.
We shouldn't be scheduling hundreds of cgwb_release_workfn() at the
same time. There's nothing to be gained from that. This patch
updates cgwb release path to use a dedicated percpu workqueue with
@max_active of 1.
While this resolves the problem at hand, it might be a good idea to
isolate rcu_exp_work to its own workqueue too as it can be used from
various paths and is prone to this sort of indirect A-A deadlocks.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For some reason we had discard granularity set to 512 always even when
discards were disabled. Fix this by having the default be 0, and then
if we turn it on set the discard granularity to the blocksize.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>