Noralf Trønnes [Sat, 10 Nov 2018 14:56:46 +0000 (15:56 +0100)]
drm/cma-helper: Add DRM_GEM_CMA_VMAP_DRIVER_OPS
This adds functionality to the CMA helper which ensures that the kernel
virtual address is set on the CMA GEM object also for imported buffers.
The drivers have been audited to ensure that none set ->vaddr on imported
buffers, making the conditional dma_buf_vunmap() call in
drm_gem_cma_free_object() safe.
Noralf Trønnes [Sat, 10 Nov 2018 14:56:45 +0000 (15:56 +0100)]
drm/gem: Add drm_gem_object_funcs
This adds an optional function table on GEM objects.
The main benefit is for drivers that support more than one type of
memory (shmem,vram,cma) for their buffers depending on the hardware it
runs on. With the callbacks attached to the GEM object itself, it is
easier to have core helpers for the the various buffer types. The driver
only has to make the decision about buffer type on GEM object creation
and all other callbacks can be handled by the chosen helper.
drm_driver->gem_prime_res_obj has not been added since there's a todo to
put a reservation_object into drm_gem_object.
v3: Add todo entry
v2: Drop drm_gem_object_funcs->prime_mmap in favour of
drm_gem_prime_mmap() (Daniel Vetter)
v1:
- drm_gem_object_funcs.map -> .prime_map let it only do PRIME mmap like
the function it superseeds (Daniel Vetter)
- Flip around the if ladders and make obj->funcs the first choice
highlighting the fact that this the new default way of doing it
(Daniel Vetter)
Quite obviously the driver depends on I2C_MUX, but adding a "depends on"
introduces a recursive dependency, therefore this patch selects I2C_MUX
instead.
Damian Kos [Tue, 6 Nov 2018 15:37:05 +0000 (15:37 +0000)]
drm/rockchip: fix for mailbox read size
Some of the functions (like cdn_dp_dpcd_read, cdn_dp_get_edid_block)
allow to read 64KiB, but the cdn_dp_mailbox_read_receive, that is
used by them, can read only up to 255 bytes at once. Normally, it's
not a big issue as DPCD or EDID reads won't (hopefully) exceed that
value.
The real issue here is the revocation list read during the HDCP
authentication process. (problematic use case:
https://chromium.googlesource.com/chromiumos/third_party/kernel/+/chromeos-4.4/drivers/gpu/drm/rockchip/cdn-dp-reg.c#1152)
The list can reach 127*5+4 bytes (num devs * 5 bytes per ID/Bksv +
4 bytes of an additional info).
In other words - CTSes with HDCP Repeater won't pass without this
fix. Oh, and the driver will most likely stop working (best case
scenario).
Fabrizio Castro [Tue, 6 Nov 2018 11:52:36 +0000 (11:52 +0000)]
drm/bridge/sii902x: Fix EDID readback
While adding SiI9022A support to the iwg23s board, it came
up that when the HDMI transmitter is in pass through mode the
device is not compliant with the I2C specification anymore,
as it requires a far bigger tbuf, due to a delay the HDMI
transmitter is adding when relaying the STOP condition on the
monitor i2c side of things.
When not providing an appropriate delay after the STOP condition
the i2c bus would get stuck. Also, any other traffic on the bus
while talking to the monitor may cause the transaction to fail
or even cause issues with the i2c bus as well.
I2c-gates seemed to reach consent as a possible way to address
these issues, and as such this patch is implementing a solution
based on that. Since others are clearly relying on the current
implementation of the driver, this patch won't require any DT
changes.
Since we don't want any interference during the DDC Bus
Request/Grant procedure and while talking to the monitor, we
have to use the adapter locking primitives rather than the
i2c-mux locking primitives.
Linus Walleij [Thu, 1 Nov 2018 21:32:55 +0000 (22:32 +0100)]
drm/panel: Augment the TPO TPG110 bindings
The TPO TPG110 bindings were using the DPI bindings (popular
in the fbdev subsystem) but this misses the finer points
learned in the DRM subsystem.
We need to augment the bindings for proper DRM integration:
the timings are expressed by the hardware, not put into the
device tree. I.e. this hardware is self-describing and can
report the resolutions and timings needed. It should not
be described in the device tree.
Further the device was incorrectly modeled with GPIO lines
instead of an SPI child, even though the device was using
SPI.
No known deployments of the device using device tree
exist, so it should be fine to augment the bindings.
drm: panel-orientation-quirks: Do rotation quirk for new GPD Win2 FW
I just got a new GDP Win2 device with an updated firmware, which still
requires this quirk to get the rotation right, so add the new firmware
date to the quirk matching table.
YueHaibing [Mon, 12 Nov 2018 13:32:36 +0000 (13:32 +0000)]
drm/qxl: remove set but not used variables 'driver, qbo'
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/gpu/drm/qxl/qxl_release.c: In function 'qxl_release_fence_buffer_objects':
drivers/gpu/drm/qxl/qxl_release.c:431:17: warning:
variable 'qbo' set but not used [-Wunused-but-set-variable]
drivers/gpu/drm/qxl/qxl_release.c:430:24: warning:
variable 'driver' set but not used [-Wunused-but-set-variable]
'qbo' not used since commit f2c24b83ae90 ("drm/ttm: flip the switch, and convert
to dma_fence")
And 'driver' never used since introduction in 8002db6336dd ("qxl: convert qxl driver to proper use for reservations")
YueHaibing [Sat, 10 Nov 2018 03:44:46 +0000 (03:44 +0000)]
drm/qxl: remove set but not used variable 'map'
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/gpu/drm/qxl/qxl_object.c: In function 'qxl_bo_kunmap_atomic_page':
drivers/gpu/drm/qxl/qxl_object.c:189:21: warning:
variable 'map' set but not used [-Wunused-but-set-variable]
Robert Foss [Mon, 12 Nov 2018 16:51:56 +0000 (17:51 +0100)]
drm/virtio: add in/out fence support for explicit synchronization
When the execbuf call receives an in-fence it will get the dma_fence
related to that fence fd and wait on it before submitting the draw call.
On the out-fence side we get fence returned by the submitted draw call
and attach it to a sync_file and send the sync_file fd to userspace. On
error -1 is returned to userspace.
VIRTGPU_EXECBUF_FENCE_FD_IN & VIRTGPU_EXECBUF_FENCE_FD_OUT
are supported at the simultaneously and can be flagged
for simultaneously.
Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.com> Signed-off-by: Robert Foss <robert.foss@collabora.com> Reviewed-by: Emil Velikov <emil.velikov@collabora.com> Link: http://patchwork.freedesktop.org/patch/msgid/20181112165157.32765-4-robert.foss@collabora.com Suggested-by: Rob Herring <robh@kernel.org> Reviewed-by: Emil Velikov <emil.velikov@collabora.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Robert Foss [Mon, 12 Nov 2018 16:51:54 +0000 (17:51 +0100)]
drm/virtio: add virtio_gpu_alloc_fence()
Refactor fence creation, add fences to relevant GPU
operations and add cursor helper functions.
This removes the potential for allocation failures from the
cmd_submit and atomic_commit paths.
Now a fence will be allocated first and only after that
will we proceed with the rest of the execution.
Neil Armstrong [Tue, 6 Nov 2018 09:40:02 +0000 (10:40 +0100)]
drm/meson: Add primary plane scaling
This patch adds support for the Primary Plane scaling.
On the Amlogic GX SoCs, the primary plane is used as On-Screen-Display
layer on top of video, and it's needed to keep the OSD layer to a lower
size as the physical display size to :
- lower the memory bandwidth
- lower the OSD rendering
- lower the memory usage
This use-case is used when setting the display mode to 3840x2160 and the
OSD layer is rendered using the GPU. In this case, the GXBB & GXL cannot
work on more than 2000x2000 buffer, thus needing the OSD layer to be kept
at 1920x1080 and upscaled to 3840x2160 in hardware.
The primary plane atomic check still allow 1:1 scaling, allowing native
3840x2160 if needed by user-space applications.
Neil Armstrong [Tue, 6 Nov 2018 09:40:01 +0000 (10:40 +0100)]
drm/meson: move OSD scaler management into plane atomic update
In preparation to support the Primary Plane scaling, move the basic
OSD Interlace-Only scaler setup code into the primary plane atomic
update callback and handle the vsync scaler update like the overlay
plane scaling registers update.
Neil Armstrong [Tue, 6 Nov 2018 09:40:00 +0000 (10:40 +0100)]
drm/meson: Support Overlay plane for video rendering
The Amlogic Meson GX SoCs support an Overlay plane behind the primary
plane for video rendering.
This Overlay plane support various YUV layouts :
- YUYV
- NV12 / NV21
- YUV444 / 422 / 420 / 411 / 410
The scaler supports a wide range of scaling ratios, but for simplicity,
plane atomic check limits the scaling from x5 to /5 in vertical and
horizontal scaling.
The z-order is fixed and always behind the primary plane and cannot be changed.
The scaling parameter algorithm was taken from the Amlogic vendor kernel
code and rewritten to match the atomic universal plane requirements.
The video rendering using this overlay plane support has been tested using
the new Kodi DRM-KMS Prime rendering path along the in-review V4L2 Mem2Mem
Hardware Video Decoder up to 3840x2160 NV12 frames on various display modes.
YueHaibing [Sat, 10 Nov 2018 02:57:40 +0000 (02:57 +0000)]
drm/sti: remove set but not used variable 'priv'
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/gpu/drm/sti/sti_crtc.c: In function 'sti_crtc_vblank_cb':
drivers/gpu/drm/sti/sti_crtc.c:255:22: warning:
variable 'priv' set but not used [-Wunused-but-set-variable]
It never used since introduction in
commit 9e1f05b28009 ("drm/sti: rename files and functions")
Chris Wilson [Mon, 12 Nov 2018 15:21:30 +0000 (15:21 +0000)]
drm/syncobj: Fix compilation following partial revert
We need to include the revert of commit 783195ec1cad ("drm/syncobj:
disable the timeline UAPI for now v2") along with undoing the change to
drm/i915.
Fixes: 131280a162e7 ("drm: Revert syncobj timeline changes.") Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Chunming Zhou <david1.zhou@amd.com> Cc: Eric Anholt <eric@anholt.net> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Maxime Ripard <maxime.ripard@bootlin.com> Cc: Sean Paul <sean@poorly.run> Cc: David Airlie <airlied@linux.ie> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Sean Paul <seanpaul@chromium.org> Link: https://patchwork.freedesktop.org/patch/msgid/20181112152130.12275-1-chris@chris-wilson.co.uk
Eric Anholt [Thu, 8 Nov 2018 16:04:22 +0000 (08:04 -0800)]
drm: Revert syncobj timeline changes.
Daniel suggested I submit this, since we're still seeing regressions
from it. This is a revert to before 48197bc564c7 ("drm: add syncobj
timeline support v9") and its followon fixes.
Fixes this on first V3D testcase execution:
[ 48.767088] ============================================
[ 48.772410] WARNING: possible recursive locking detected
[ 48.777739] 4.19.0-rc6+ #489 Not tainted
[ 48.781668] --------------------------------------------
[ 48.786993] shader_runner/3284 is trying to acquire lock:
[ 48.792408] ce309d7f (&(&array->lock)->rlock){....}, at: dma_fence_add_callback+0x30/0x23c
[ 48.800714]
[ 48.800714] but task is already holding lock:
[ 48.806559] c5952bd3 (&(&array->lock)->rlock){....}, at: dma_fence_add_callback+0x30/0x23c
[ 48.814862]
[ 48.814862] other info that might help us debug this:
[ 48.821410] Possible unsafe locking scenario:
[ 48.821410]
[ 48.827338] CPU0
[ 48.829788] ----
[ 48.832239] lock(&(&array->lock)->rlock);
[ 48.836434] lock(&(&array->lock)->rlock);
[ 48.840640]
[ 48.840640] *** DEADLOCK ***
[ 48.840640]
[ 48.846582] May be due to missing lock nesting notation
[ 130.763560] 1 lock held by cts-runner/3270:
[ 130.767745] #0: 7834b793 (&(&array->lock)->rlock){-...}, at: dma_fence_add_callback+0x30/0x23c
[ 130.776461]
stack backtrace:
[ 130.780825] CPU: 1 PID: 3270 Comm: cts-runner Not tainted 4.19.0-rc6+ #486
[ 130.787706] Hardware name: Broadcom STB (Flattened Device Tree)
[ 130.793645] [<c021269c>] (unwind_backtrace) from [<c020db1c>] (show_stack+0x10/0x14)
[ 130.801404] [<c020db1c>] (show_stack) from [<c0c2c4b0>] (dump_stack+0xa8/0xd4)
[ 130.808642] [<c0c2c4b0>] (dump_stack) from [<c0281a84>] (__lock_acquire+0x848/0x1a68)
[ 130.816483] [<c0281a84>] (__lock_acquire) from [<c02835d8>] (lock_acquire+0xd8/0x22c)
[ 130.824326] [<c02835d8>] (lock_acquire) from [<c0c49948>] (_raw_spin_lock_irqsave+0x54/0x68)
[ 130.832777] [<c0c49948>] (_raw_spin_lock_irqsave) from [<c086bf54>] (dma_fence_add_callback+0x30/0x23c)
[ 130.842183] [<c086bf54>] (dma_fence_add_callback) from [<c086d4c8>] (dma_fence_array_enable_signaling+0x58/0xec)
[ 130.852371] [<c086d4c8>] (dma_fence_array_enable_signaling) from [<c086c00c>] (dma_fence_add_callback+0xe8/0x23c)
[ 130.862647] [<c086c00c>] (dma_fence_add_callback) from [<c06d8774>] (drm_syncobj_wait_ioctl+0x518/0x614)
[ 130.872143] [<c06d8774>] (drm_syncobj_wait_ioctl) from [<c06b8458>] (drm_ioctl_kernel+0xb0/0xf0)
[ 130.880940] [<c06b8458>] (drm_ioctl_kernel) from [<c06b8818>] (drm_ioctl+0x1d8/0x390)
[ 130.888782] [<c06b8818>] (drm_ioctl) from [<c03a4510>] (do_vfs_ioctl+0xb0/0x8ac)
[ 130.896187] [<c03a4510>] (do_vfs_ioctl) from [<c03a4d40>] (ksys_ioctl+0x34/0x60)
[ 130.903593] [<c03a4d40>] (ksys_ioctl) from [<c0201000>] (ret_fast_syscall+0x0/0x28)
Cc: Chunming Zhou <david1.zhou@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Eric Anholt <eric@anholt.net> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/261044/
Boris Brezillon [Fri, 9 Nov 2018 10:26:32 +0000 (11:26 +0100)]
drm/vc4: ->x_scaling[1] should never be set to VC4_SCALING_NONE
For the YUV conversion to work properly, ->x_scaling[1] should never
be set to VC4_SCALING_NONE, but vc4_get_scaling_mode() might return
VC4_SCALING_NONE if the horizontal scaling ratio exactly matches the
horizontal subsampling factor. Add a test to turn VC4_SCALING_NONE
into VC4_SCALING_PPF when that happens.
The old ->x_scaling[0] adjustment is dropped as I couldn't find any
mention to this constraint in the spec and it's proven to be
unnecessary (I tested various multi-planar YUV formats with scaling
disabled, and all of them worked fine without this adjustment).
drm/sun4i: tcon: Support an active-low DE signal with RGB interface
Some panels need an active-low data enable (DE) signal for the RGB
interface. This requires flipping a bit in the TCON0 polarity register
when setting up the mode for the RGB interface.
Match the associated bus flag and use it to set the polarity inversion
bit for the DE signal when required.
drm/sun4i: tcon: Get the connector from the encoder in RGB setup
Features such as dithering and pixel data edge configuration currently
rely on the panel registered with the TCON driver. However, bridges are
also supported in addition to panels for RGB setup.
Instead of retrieving the connector from the panel, get it from the
encoder with the dedicated helper.
Even in the case of bridges, the connector is registered with the
encoder from our driver and is accessible when iterating connectors.
Ville Syrjälä [Thu, 1 Nov 2018 18:46:44 +0000 (20:46 +0200)]
drm/atomic: Use explicit old crtc state in drm_atomic_add_affected_planes()
Replace 'crtc->state' with the explicit old crtc state.
Actually it shouldn't matter whether we use the old or the new
crtc state here since any plane that has been removed from the
crtc since the crtc state was duplicated will have been added
to the atomic state already. That is, you can't call
drm_atomic_set_crtc_for_plane() without having the new
plane state already in hand.
Ville Syrjälä [Mon, 29 Oct 2018 18:34:52 +0000 (20:34 +0200)]
drm: Add drm_any_plane_has_format()
Add a function to check whether there is at least one plane that
supports a specific format and modifier combination. Drivers can
use this to reject unsupported formats/modifiers in .fb_create().
v2: Accept anyformat if the driver doesn't do planes (Eric)
s/planes_have_format/any_plane_has_format/ (Eric)
Check the modifier as well since we already have a function
that does both
v3: Don't do the check in the core since we may not know the
modifier yet, instead export the function and let drivers
call it themselves
Daniel Vetter [Mon, 5 Nov 2018 10:12:11 +0000 (11:12 +0100)]
drm/lease: look at ->universal_planes only once
It's lockless, and userspace might chance it underneath us. That's not
really a problem, all userspace gets is a slightly dysfunctional
lease with the current code. But this might change, and gcc might
decide to reload a few too many times, and then boom. So better safe
than sorry.
v2: Remove the now unused lessor_priv argument from validate_lease()
(Keith).
We do lockless access to these flags everywhere, and it's kinda not a
great idea to mix lockless and bitfields. Aside from that gcc isn't
generating great code for these.
If this ever becomes an issue size-wise, I think we need atomic_t here
and atomic bitflag ops.
Eric Anholt [Mon, 5 Nov 2018 23:01:10 +0000 (15:01 -0800)]
drm/syncobj: Fix oops on drm_syncobj_find_fence(file_priv, 0, ...).
This broke rendering on V3D, where we almost always have a 0
in-syncobj.
Signed-off-by: Eric Anholt <eric@anholt.net> Fixes: 48197bc564c7 ("drm: add syncobj timeline support v9") Cc: Chunming Zhou <david1.zhou@amd.com> Cc: Christian König <christian.koenig@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.kernel.org/patch/10669317/ Signed-off-by: Christian König <christian.koenig@amd.com>
Imre Deak [Mon, 5 Nov 2018 14:54:28 +0000 (16:54 +0200)]
drm/vgem: Fix typo in driver feature flags
Fix typo in struct field initializer.
Fixes: 3a6eb795641c ("drm/vgem: create a render node for vgem") Cc: Emil Velikov <emil.velikov@collabora.com> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: dri-devel@lists.freedesktop.org Signed-off-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Emil Velikov <emil.velikov@collabora.com> Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181105145428.5590-1-imre.deak@intel.com
Emil Velikov [Fri, 26 Oct 2018 12:06:47 +0000 (13:06 +0100)]
drm/vgem: create a render node for vgem
VGEM doesn't do anything modeset specific, so in a way exposing a
primary node is 'wrong'. At the same time, we extensively use if for
creating dumb buffers, fences, prime fd <> handle imports/exports.
To the point that we explicitly annotate the vgem fence ioctls as
DRM_RENDER_ALLOW and have an IGT test which opens the render node.
Jernej Skrabec [Sun, 4 Nov 2018 18:26:59 +0000 (19:26 +0100)]
drm/sun4i: Add support for H6 HDMI PHY
H6 has Synopsys DWC HDMI 2.0 TX PHY.
There is no freely available documentation for it, only code found in
BSP kernel. However, judging by the code, PHY is very similar to older
Synopsys HDMI PHY described in i.MX6 documentation. Most registers seem
to be the same.
According to i.MX6 documentation, mpll settings are based on pixel clock
and are not specific to each SoC. Because of that, mpll table in this
commit is based on that documentation and not on BSP code. Other PHY
settings were derived from BSP PHY driver code.
Jernej Skrabec [Sun, 4 Nov 2018 18:26:58 +0000 (19:26 +0100)]
drm/sun4i: Add support for Synopsys HDMI PHY
Currently sun8i-hdmi-phy driver supports only custom PHYs connected to
DW HDMI controller. Since newest Allwinner SoCs have unmodified Synopsys
PHY, driver has to be reorganized to support them.
Variant structure is expanded to allow differentiation between custom
and Sysnopsys PHYs and to hold Synopsys PHY settings.
Since DW HDMI bridge platform data has different fields for custom and
Sysnopsys PHY, function sun8i_hdmi_phy_get_ops() is replaced with
sun8i_hdmi_phy_set_ops().
The Allwinner H6 SoC uses a v2.12a DesignWare HDMI controller, with
dedicated CEC and HDCP clocks added; the PHY connected is a standard
DesignWare HDMI PHY.
Jernej Skrabec [Sun, 4 Nov 2018 18:26:54 +0000 (19:26 +0100)]
drm/sun4i: dw-hdmi: Add quirk for setting TMDS clock
It turns out that H6 HDMI BSP kernel driver doesn't change TMDS rate at
all. At this point it is not clear whether it is just not necessary or
it would cause some kind of issues.
Jernej Skrabec [Sun, 4 Nov 2018 18:26:53 +0000 (19:26 +0100)]
drm/sun4i: dw-hdmi: Make mode_valid function configurable
Since it is not possible to access sun8i-dw-hdmi driver private data
inside mode_valid function, make it configurable. That way different
versions of HDMI controllers can set different function, depending on
it's limitations.
Jernej Skrabec [Sun, 4 Nov 2018 18:26:52 +0000 (19:26 +0100)]
drm/sun4i: Not all DW HDMI controllers has scrambled addresses
Currently supported Allwinner SoCs with DW HDMI controller have
scrambled addresses and read lock. However, that is not true in general.
For example, A80 and H6 have normal addresses and normal read access.
Move code for unscrambling addresses and unlocking read access to it's
own function and call it from init function.
Linus Torvalds [Sun, 4 Nov 2018 22:46:04 +0000 (14:46 -0800)]
Merge tag 'tags/upstream-4.20-rc1' of git://git.infradead.org/linux-ubifs
Pull UBIFS updates from Richard Weinberger:
- Full filesystem authentication feature, UBIFS is now able to have the
whole filesystem structure authenticated plus user data encrypted and
authenticated.
- Minor cleanups
* tag 'tags/upstream-4.20-rc1' of git://git.infradead.org/linux-ubifs: (26 commits)
ubifs: Remove unneeded semicolon
Documentation: ubifs: Add authentication whitepaper
ubifs: Enable authentication support
ubifs: Do not update inode size in-place in authenticated mode
ubifs: Add hashes and HMACs to default filesystem
ubifs: authentication: Authenticate super block node
ubifs: Create hash for default LPT
ubfis: authentication: Authenticate master node
ubifs: authentication: Authenticate LPT
ubifs: Authenticate replayed journal
ubifs: Add auth nodes to garbage collector journal head
ubifs: Add authentication nodes to journal
ubifs: authentication: Add hashes to index nodes
ubifs: Add hashes to the tree node cache
ubifs: Create functions to embed a HMAC in a node
ubifs: Add helper functions for authentication support
ubifs: Add separate functions to init/crc a node
ubifs: Format changes for authentication support
ubifs: Store read superblock node
ubifs: Drop write_node
...
Linus Torvalds [Sun, 4 Nov 2018 16:20:09 +0000 (08:20 -0800)]
Merge tag 'nfs-for-4.20-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client bugfixes from Trond Myklebust:
"Highlights include:
Bugfix:
- Fix build issues on architectures that don't provide 64-bit cmpxchg
Cleanups:
- Fix a spelling mistake"
* tag 'nfs-for-4.20-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFS: fix spelling mistake, EACCESS -> EACCES
SUNRPC: Use atomic(64)_t for seq_send(64)
Linus Torvalds [Sun, 4 Nov 2018 16:12:44 +0000 (08:12 -0800)]
Merge tag 'ntb-4.20' of git://github.com/jonmason/ntb
Pull NTB updates from Jon Mason:
"Fairly minor changes and bug fixes:
NTB IDT thermal changes and hook into hwmon, ntb_netdev clean-up of
private struct, and a few bug fixes"
* tag 'ntb-4.20' of git://github.com/jonmason/ntb:
ntb: idt: Alter the driver info comments
ntb: idt: Discard temperature sensor IRQ handler
ntb: idt: Add basic hwmon sysfs interface
ntb: idt: Alter temperature read method
ntb_netdev: Simplify remove with client device drvdata
NTB: transport: Try harder to alloc an aligned MW buffer
ntb: ntb_transport: Mark expected switch fall-throughs
ntb: idt: Set PCIe bus address to BARLIMITx
NTB: ntb_hw_idt: replace IS_ERR_OR_NULL with regular NULL checks
ntb: intel: fix return value for ndev_vec_mask()
ntb_netdev: fix sleep time mismatch
Linus Torvalds [Sun, 4 Nov 2018 01:37:09 +0000 (18:37 -0700)]
Merge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"A memory (under-)allocation fix and a comment fix"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/topology: Fix off by one bug
sched/rt: Update comment in pick_next_task_rt()
Linus Torvalds [Sun, 4 Nov 2018 01:25:17 +0000 (18:25 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"A number of fixes and some late updates:
- make in_compat_syscall() behavior on x86-32 similar to other
platforms, this touches a number of generic files but is not
intended to impact non-x86 platforms.
- objtool fixes
- PAT preemption fix
- paravirt fixes/cleanups
- cpufeatures updates for new instructions
- earlyprintk quirk
- make microcode version in sysfs world-readable (it is already
world-readable in procfs)
- minor cleanups and fixes"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
compat: Cleanup in_compat_syscall() callers
x86/compat: Adjust in_compat_syscall() to generic code under !COMPAT
objtool: Support GCC 9 cold subfunction naming scheme
x86/numa_emulation: Fix uniform-split numa emulation
x86/paravirt: Remove unused _paravirt_ident_32
x86/mm/pat: Disable preemption around __flush_tlb_all()
x86/paravirt: Remove GPL from pv_ops export
x86/traps: Use format string with panic() call
x86: Clean up 'sizeof x' => 'sizeof(x)'
x86/cpufeatures: Enumerate MOVDIR64B instruction
x86/cpufeatures: Enumerate MOVDIRI instruction
x86/earlyprintk: Add a force option for pciserial device
objtool: Support per-function rodata sections
x86/microcode: Make revision and processor flags world-readable
Linus Torvalds [Sun, 4 Nov 2018 01:13:43 +0000 (18:13 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf updates and fixes from Ingo Molnar:
"These are almost all tooling updates: 'perf top', 'perf trace' and
'perf script' fixes and updates, an UAPI header sync with the merge
window versions, license marker updates, much improved Sparc support
from David Miller, and a number of fixes"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (66 commits)
perf intel-pt/bts: Calculate cpumode for synthesized samples
perf intel-pt: Insert callchain context into synthesized callchains
perf tools: Don't clone maps from parent when synthesizing forks
perf top: Start display thread earlier
tools headers uapi: Update linux/if_link.h header copy
tools headers uapi: Update linux/netlink.h header copy
tools headers: Sync the various kvm.h header copies
tools include uapi: Update linux/mmap.h copy
perf trace beauty: Use the mmap flags table generated from headers
perf beauty: Wire up the mmap flags table generator to the Makefile
perf beauty: Add a generator for MAP_ mmap's flag constants
tools include uapi: Update asound.h copy
tools arch uapi: Update asm-generic/unistd.h and arm64 unistd.h copies
tools include uapi: Update linux/fs.h copy
perf callchain: Honour the ordering of PERF_CONTEXT_{USER,KERNEL,etc}
perf cs-etm: Correct CPU mode for samples
perf unwind: Take pgoff into account when reporting elf to libdwfl
perf top: Do not use overwrite mode by default
perf top: Allow disabling the overwrite mode
perf trace: Beautify mount's first pathname arg
...
Linus Torvalds [Sun, 4 Nov 2018 01:12:09 +0000 (18:12 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fixes from Ingo Molnar:
"An irqchip driver fix and a memory (over-)allocation fix"
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/irq-mvebu-sei: Fix a NULL vs IS_ERR() bug in probe function
irq/matrix: Fix memory overallocation
Linus Torvalds [Sat, 3 Nov 2018 17:55:23 +0000 (10:55 -0700)]
Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull more arm64 updates from Catalin Marinas:
- fix W+X page (mark RO) allocated by the arm64 kprobes code
- Makefile fix for .i files in out of tree modules
* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: kprobe: make page to RO mode when allocate it
arm64: kdump: fix small typo
arm64: makefile fix build of .i file in external module case
Linus Torvalds [Sat, 3 Nov 2018 17:45:55 +0000 (10:45 -0700)]
Merge tag '4.20-rc1-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes and updates from Steve French:
"Three small fixes (one Kerberos related, one for stable, and another
fixes an oops in xfstest 377), two helpful debugging improvements,
three patches for cifs directio and some minor cleanup"
* tag '4.20-rc1-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: fix signed/unsigned mismatch on aio_read patch
cifs: don't dereference smb_file_target before null check
CIFS: Add direct I/O functions to file_operations
CIFS: Add support for direct I/O write
CIFS: Add support for direct I/O read
smb3: missing defines and structs for reparse point handling
smb3: allow more detailed protocol info on open files for debugging
smb3: on kerberos mount if server doesn't specify auth type use krb5
smb3: add trace point for tree connection
cifs: fix spelling mistake, EACCESS -> EACCES
cifs: fix return value for cifs_listxattr
Linus Torvalds [Sat, 3 Nov 2018 17:35:52 +0000 (10:35 -0700)]
Merge branch 'work.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull 9p fix from Al Viro:
"Regression fix for net/9p handling of iov_iter; broken by braino when
switching to iov_iter_is_kvec() et.al., spotted and fixed by Marc"
* 'work.afs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
iov_iter: Fix 9p virtio breakage
Linus Torvalds [Sat, 3 Nov 2018 17:34:03 +0000 (10:34 -0700)]
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull more SCSI updates from James Bottomley:
"This is a set of minor small (and safe changes) that didn't make the
initial pull request plus some bug fixes"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: mvsas: Remove set but not used variable 'id'
scsi: qla2xxx: Remove two arguments from qlafx00_error_entry()
scsi: qla2xxx: Make sure that qlafx00_ioctl_iosb_entry() initializes 'res'
scsi: qla2xxx: Remove a set-but-not-used variable
scsi: qla2xxx: Make qla2x00_sysfs_write_nvram() easier to analyze
scsi: qla2xxx: Declare local functions 'static'
scsi: qla2xxx: Improve several kernel-doc headers
scsi: qla2xxx: Modify fall-through annotations
scsi: 3w-sas: 3w-9xxx: Use unsigned char for cdb
scsi: mvsas: Use dma_pool_zalloc
scsi: target: Don't request modules that aren't even built
scsi: target: Set response length for REPORT TARGET PORT GROUPS
Linus Torvalds [Sat, 3 Nov 2018 17:21:43 +0000 (10:21 -0700)]
Merge branch 'akpm' (patches from Andrew)
Merge more updates from Andrew Morton:
- more ocfs2 work
- various leftovers
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
memory_hotplug: cond_resched in __remove_pages
bfs: add sanity check at bfs_fill_super()
kernel/sysctl.c: remove duplicated include
kernel/kexec_file.c: remove some duplicated includes
mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask
ocfs2: fix clusters leak in ocfs2_defrag_extent()
ocfs2: dlmglue: clean up timestamp handling
ocfs2: don't put and assigning null to bh allocated outside
ocfs2: fix a misuse a of brelse after failing ocfs2_check_dir_entry
ocfs2: don't use iocb when EIOCBQUEUED returns
ocfs2: without quota support, avoid calling quota recovery
ocfs2: remove ocfs2_is_o2cb_active()
mm: thp: relax __GFP_THISNODE for MADV_HUGEPAGE mappings
include/linux/notifier.h: SRCU: fix ctags
mm: handle no memcg case in memcg_kmem_charge() properly
It has been reported on an older (4.12) kernel but the current upstream
code doesn't cond_resched in the hot remove code at all and the given
range to remove might be really large. Fix the issue by calling
cond_resched once per memory section.
Link: http://lkml.kernel.org/r/20181031125840.23982-1-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Thumshirn <jthumshirn@suse.de> Cc: Dan Williams <dan.j.williams@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Tetsuo Handa [Fri, 2 Nov 2018 22:48:42 +0000 (15:48 -0700)]
bfs: add sanity check at bfs_fill_super()
syzbot is reporting too large memory allocation at bfs_fill_super() [1].
Since file system image is corrupted such that bfs_sb->s_start == 0,
bfs_fill_super() is trying to allocate 8MB of continuous memory. Fix
this by adding a sanity check on bfs_sb->s_start, __GFP_NOWARN and
printf().
Michal Hocko [Fri, 2 Nov 2018 22:48:31 +0000 (15:48 -0700)]
mm, thp: consolidate THP gfp handling into alloc_hugepage_direct_gfpmask
THP allocation mode is quite complex and it depends on the defrag mode.
This complexity is hidden in alloc_hugepage_direct_gfpmask from a large
part currently. The NUMA special casing (namely __GFP_THISNODE) is
however independent and placed in alloc_pages_vma currently. This both
adds an unnecessary branch to all vma based page allocation requests and
it makes the code more complex unnecessarily as well. Not to mention
that e.g. shmem THP used to do the node reclaiming unconditionally
regardless of the defrag mode until recently. This was not only
unexpected behavior but it was also hardly a good default behavior and I
strongly suspect it was just a side effect of the code sharing more than
a deliberate decision which suggests that such a layering is wrong.
Get rid of the thp special casing from alloc_pages_vma and move the
logic to alloc_hugepage_direct_gfpmask. __GFP_THISNODE is applied to the
resulting gfp mask only when the direct reclaim is not requested and
when there is no explicit numa binding to preserve the current logic.
Please note that there's also a slight difference wrt MPOL_BIND now. The
previous code would avoid using __GFP_THISNODE if the local node was
outside of policy_nodemask(). After this patch __GFP_THISNODE is avoided
for all MPOL_BIND policies. So there's a difference that if local node
is actually allowed by the bind policy's nodemask, previously
__GFP_THISNODE would be added, but now it won't be. From the behavior
POV this is still correct because the policy nodemask is used.
Link: http://lkml.kernel.org/r/20180925120326.24392-3-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Alex Williamson <alex.williamson@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Stefan Priebe - Profihost AG <s.priebe@profihost.ag> Cc: Zi Yan <zi.yan@cs.rutgers.edu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Larry Chen [Fri, 2 Nov 2018 22:48:27 +0000 (15:48 -0700)]
ocfs2: fix clusters leak in ocfs2_defrag_extent()
ocfs2_defrag_extent() might leak allocated clusters. When the file
system has insufficient space, the number of claimed clusters might be
less than the caller wants. If that happens, the original code might
directly commit the transaction without returning clusters.
This patch is based on code in ocfs2_add_clusters_in_btree().
[akpm@linux-foundation.org: include localalloc.h, reduce scope of data_ac] Link: http://lkml.kernel.org/r/20180904041621.16874-3-lchen@suse.com Signed-off-by: Larry Chen <lchen@suse.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: Changwei Ge <ge.changwei@h3c.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Arnd Bergmann [Fri, 2 Nov 2018 22:48:23 +0000 (15:48 -0700)]
ocfs2: dlmglue: clean up timestamp handling
The handling of timestamps outside of the 1970..2038 range in the dlm
glue is rather inconsistent: on 32-bit architectures, this has always
wrapped around to negative timestamps in the 1902..1969 range, while on
64-bit kernels all timestamps are interpreted as positive 34 bit numbers
in the 1970..2514 year range.
Now that the VFS code handles 64-bit timestamps on all architectures, we
can make the behavior more consistent here, and return the same result
that we had on 64-bit already, making the file system y2038 safe in the
process. Outside of dlmglue, it already uses 64-bit on-disk timestamps
anway, so that part is fine.
For consistency, I'm changing ocfs2_pack_timespec() to clamp anything
outside of the supported range to the minimum and maximum values. This
avoids a possible ambiguity of values before 1970 in particular, which
used to be interpreted as times at the end of the 2514 range previously.
Link: http://lkml.kernel.org/r/20180619155826.4106487-1-arnd@arndb.de Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: Changwei Ge <ge.changwei@h3c.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Changwei Ge [Fri, 2 Nov 2018 22:48:19 +0000 (15:48 -0700)]
ocfs2: don't put and assigning null to bh allocated outside
ocfs2_read_blocks() and ocfs2_read_blocks_sync() are both used to read
several blocks from disk. Currently, the input argument *bhs* can be
NULL or NOT. It depends on the caller's behavior. If the function
fails in reading blocks from disk, the corresponding bh will be assigned
to NULL and put.
Obviously, above process for non-NULL input bh is not appropriate.
Because the caller doesn't even know its bhs are put and re-assigned.
If buffer head is managed by caller, ocfs2_read_blocks and
ocfs2_read_blocks_sync() should not evaluate it to NULL. It will cause
caller accessing illegal memory, thus crash.
Link: http://lkml.kernel.org/r/HK2PR06MB045285E0F4FBB561F9F2F9B3D5680@HK2PR06MB0452.apcprd06.prod.outlook.com Signed-off-by: Changwei Ge <ge.changwei@h3c.com> Reviewed-by: Guozhonghua <guozhonghua@h3c.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: Changwei Ge <ge.changwei@h3c.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Changwei Ge [Fri, 2 Nov 2018 22:48:15 +0000 (15:48 -0700)]
ocfs2: fix a misuse a of brelse after failing ocfs2_check_dir_entry
Somehow, file system metadata was corrupted, which causes
ocfs2_check_dir_entry() to fail in function ocfs2_dir_foreach_blk_el().
According to the original design intention, if above happens we should
skip the problematic block and continue to retrieve dir entry. But
there is obviouse misuse of brelse around related code.
After failure of ocfs2_check_dir_entry(), current code just moves to
next position and uses the problematic buffer head again and again
during which the problematic buffer head is released for multiple times.
I suppose, this a serious issue which is long-lived in ocfs2. This may
cause other file systems which is also used in a the same host insane.
So we should also consider about bakcporting this patch into linux
-stable.
Link: http://lkml.kernel.org/r/HK2PR06MB045211675B43EED794E597B6D56E0@HK2PR06MB0452.apcprd06.prod.outlook.com Signed-off-by: Changwei Ge <ge.changwei@h3c.com> Suggested-by: Changkuo Shi <shi.changkuo@h3c.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Changwei Ge [Fri, 2 Nov 2018 22:48:11 +0000 (15:48 -0700)]
ocfs2: don't use iocb when EIOCBQUEUED returns
When -EIOCBQUEUED returns, it means that aio_complete() will be called
from dio_complete(), which is an asynchronous progress against
write_iter. Generally, IO is a very slow progress than executing
instruction, but we still can't take the risk to access a freed iocb.
And we do face a BUG crash issue. Using the crash tool, iocb is
obviously freed already.
And the backtrace shows:
ocfs2_file_write_iter+0xcaa/0xd00 [ocfs2]
aio_run_iocb+0x229/0x2f0
do_io_submit+0x291/0x540
SyS_io_submit+0x10/0x20
system_call_fastpath+0x16/0x75
Link: http://lkml.kernel.org/r/1523361653-14439-1-git-send-email-ge.changwei@h3c.com Signed-off-by: Changwei Ge <ge.changwei@h3c.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Guozhonghua [Fri, 2 Nov 2018 22:48:07 +0000 (15:48 -0700)]
ocfs2: without quota support, avoid calling quota recovery
During one dead node's recovery by other node, quota recovery work will
be queued. We should avoid calling quota when it is not supported, so
check the quota flags.
Link: http://lkml.kernel.org/r/71604351584F6A4EBAE558C676F37CA401071AC9FB@H3CMLB12-EX.srv.huawei-3com.com Signed-off-by: guozhonghua <guozhonghua@h3c.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: Changwei Ge <ge.changwei@h3c.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Gang He [Fri, 2 Nov 2018 22:48:03 +0000 (15:48 -0700)]
ocfs2: remove ocfs2_is_o2cb_active()
Remove ocfs2_is_o2cb_active(). We have similar functions to identify
which cluster stack is being used via osb->osb_cluster_stack.
Secondly, the current implementation of ocfs2_is_o2cb_active() is not
totally safe. Based on the design of stackglue, we need to get
ocfs2_stack_lock before using ocfs2_stack related data structures, and
that active_stack pointer can be NULL in the case of mount failure.
Link: http://lkml.kernel.org/r/1495441079-11708-1-git-send-email-ghe@suse.com Signed-off-by: Gang He <ghe@suse.com> Reviewed-by: Joseph Qi <jiangqi903@gmail.com> Reviewed-by: Eric Ren <zren@suse.com> Acked-by: Changwei Ge <ge.changwei@h3c.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: thp: relax __GFP_THISNODE for MADV_HUGEPAGE mappings
THP allocation might be really disruptive when allocated on NUMA system
with the local node full or hard to reclaim. Stefan has posted an
allocation stall report on 4.12 based SLES kernel which suggests the
same issue:
The defrag mode is "madvise" and from the above report it is clear that
the THP has been allocated for MADV_HUGEPAGA vma.
Andrea has identified that the main source of the problem is
__GFP_THISNODE usage:
: The problem is that direct compaction combined with the NUMA
: __GFP_THISNODE logic in mempolicy.c is telling reclaim to swap very
: hard the local node, instead of failing the allocation if there's no
: THP available in the local node.
:
: Such logic was ok until __GFP_THISNODE was added to the THP allocation
: path even with MPOL_DEFAULT.
:
: The idea behind the __GFP_THISNODE addition, is that it is better to
: provide local memory in PAGE_SIZE units than to use remote NUMA THP
: backed memory. That largely depends on the remote latency though, on
: threadrippers for example the overhead is relatively low in my
: experience.
:
: The combination of __GFP_THISNODE and __GFP_DIRECT_RECLAIM results in
: extremely slow qemu startup with vfio, if the VM is larger than the
: size of one host NUMA node. This is because it will try very hard to
: unsuccessfully swapout get_user_pages pinned pages as result of the
: __GFP_THISNODE being set, instead of falling back to PAGE_SIZE
: allocations and instead of trying to allocate THP on other nodes (it
: would be even worse without vfio type1 GUP pins of course, except it'd
: be swapping heavily instead).
Fix this by removing __GFP_THISNODE for THP requests which are
requesting the direct reclaim. This effectivelly reverts 5265047ac301
on the grounds that the zone/node reclaim was known to be disruptive due
to premature reclaim when there was memory free. While it made sense at
the time for HPC workloads without NUMA awareness on rare machines, it
was ultimately harmful in the majority of cases. The existing behaviour
is similar, if not as widespare as it applies to a corner case but
crucially, it cannot be tuned around like zone_reclaim_mode can. The
default behaviour should always be to cause the least harm for the
common case.
If there are specialised use cases out there that want zone_reclaim_mode
in specific cases, then it can be built on top. Longterm we should
consider a memory policy which allows for the node reclaim like behavior
for the specific memory ranges which would allow a
: Both patches look correct to me but I'm responding to this one because
: it's the fix. The change makes sense and moves further away from the
: severe stalling behaviour we used to see with both THP and zone reclaim
: mode.
:
: I put together a basic experiment with usemem configured to reference a
: buffer multiple times that is 80% the size of main memory on a 2-socket
: box with symmetric node sizes and defrag set to "always". The defrag
: setting is not the default but it would be functionally similar to
: accessing a buffer with madvise(MADV_HUGEPAGE). Usemem is configured to
: reference the buffer multiple times and while it's not an interesting
: workload, it would be expected to complete reasonably quickly as it fits
: within memory. The results were;
:
: usemem
: vanilla noreclaim-v1
: Amean Elapsd-1 42.78 ( 0.00%) 26.87 ( 37.18%)
: Amean Elapsd-3 27.55 ( 0.00%) 7.44 ( 73.00%)
: Amean Elapsd-4 5.72 ( 0.00%) 5.69 ( 0.45%)
:
: This shows the elapsed time in seconds for 1 thread, 3 threads and 4
: threads referencing buffers 80% the size of memory. With the patches
: applied, it's 37.18% faster for the single thread and 73% faster with two
: threads. Note that 4 threads showing little difference does not indicate
: the problem is related to thread counts. It's simply the case that 4
: threads gets spread so their workload mostly fits in one node.
:
: The overall view from /proc/vmstats is more startling
:
: 4.19.0-rc1 4.19.0-rc1
: vanillanoreclaim-v1r1
: Minor Faults 35593425 708164
: Major Faults 484088 36
: Swap Ins 3772837 0
: Swap Outs 3932295 0
:
: Massive amounts of swap in/out without the patch
:
: Direct pages scanned 6013214 0
: Kswapd pages scanned 0 0
: Kswapd pages reclaimed 0 0
: Direct pages reclaimed 4033009 0
:
: Lots of reclaim activity without the patch
:
: Kswapd efficiency 100% 100%
: Kswapd velocity 0.000 0.000
: Direct efficiency 67% 100%
: Direct velocity 11191.956 0.000
:
: Mostly from direct reclaim context as you'd expect without the patch.
:
: Page writes by reclaim 3932314.000 0.000
: Page writes file 19 0
: Page writes anon 3932295 0
: Page reclaim immediate 42336 0
:
: Writes from reclaim context is never good but the patch eliminates it.
:
: We should never have default behaviour to thrash the system for such a
: basic workload. If zone reclaim mode behaviour is ever desired but on a
: single task instead of a global basis then the sensible option is to build
: a mempolicy that enforces that behaviour.
This was a severe regression compared to previous kernels that made
important workloads unusable and it starts when __GFP_THISNODE was
added to THP allocations under MADV_HUGEPAGE. It is not a significant
risk to go to the previous behavior before __GFP_THISNODE was added, it
worked like that for years.
This was simply an optimization to some lucky workloads that can fit in
a single node, but it ended up breaking the VM for others that can't
possibly fit in a single node, so going back is safe.
[mhocko@suse.com: rewrote the changelog based on the one from Andrea] Link: http://lkml.kernel.org/r/20180925120326.24392-2-mhocko@kernel.org Fixes: 5265047ac301 ("mm, thp: really limit transparent hugepage allocation to local node") Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Stefan Priebe <s.priebe@profihost.ag> Debugged-by: Andrea Arcangeli <aarcange@redhat.com> Reported-by: Alex Williamson <alex.williamson@redhat.com> Reviewed-by: Mel Gorman <mgorman@techsingularity.net> Tested-by: Mel Gorman <mgorman@techsingularity.net> Cc: Zi Yan <zi.yan@cs.rutgers.edu> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: <stable@vger.kernel.org> [4.1+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sam Protsenko [Fri, 2 Nov 2018 22:47:53 +0000 (15:47 -0700)]
include/linux/notifier.h: SRCU: fix ctags
ctags indexing ("make tags" command) throws this warning:
ctags: Warning: include/linux/notifier.h:125:
null expansion of name pattern "\1"
This is the result of DEFINE_PER_CPU() macro expansion. Fix that by
getting rid of line break.
Similar fix was already done in commit 25528213fe9f ("tags: Fix
DEFINE_PER_CPU expansions"), but this one probably wasn't noticed.
Link: http://lkml.kernel.org/r/20181030202808.28027-1-semen.protsenko@linaro.org Fixes: 9c80172b902d ("kernel/SRCU: provide a static initializer") Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Roman Gushchin [Fri, 2 Nov 2018 22:47:49 +0000 (15:47 -0700)]
mm: handle no memcg case in memcg_kmem_charge() properly
Mike Galbraith reported a regression caused by the commit 9b6f7e163cd0
("mm: rework memcg kernel stack accounting") on a system with
"cgroup_disable=memory" boot option: the system panics with the following
stack trace:
The problem occurs because get_mem_cgroup_from_current() returns the NULL
pointer if memory controller is disabled. Let's check if this is a case
at the beginning of memcg_kmem_charge() and just return 0 if
mem_cgroup_disabled() returns true. This is how we handle this case in
many other places in the memory controller code.
Link: http://lkml.kernel.org/r/20181029215123.17830-1-guro@fb.com Fixes: 9b6f7e163cd0 ("mm: rework memcg kernel stack accounting") Signed-off-by: Roman Gushchin <guro@fb.com> Reported-by: Mike Galbraith <efault@gmx.de> Acked-by: Rik van Riel <riel@surriel.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Shakeel Butt <shakeelb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
ARM: dts: stm32: update HASH1 dmas property on stm32mp157c
Remove unused parameter from HASH1 dmas property on stm32mp157c SoC.
Fixes: 1e726a40e067 ("ARM: dts: stm32: Add HASH support on stm32mp157c") Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
[Olof: Bug doesn't cause any harm, so shouldn't need stable backport] Signed-off-by: Olof Johansson <olof@lixom.net>