The s_ack_queue is managed by two pointers into the ring:
r_head_ack_queue and s_tail_ack_queue. r_head_ack_queue is the index of
where the next received request is going to be placed and s_tail_ack_queue
is the entry of the request currently being processed. This works
perfectly fine for normal Verbs as the requests are processed one at a
time and the s_tail_ack_queue is not moved until the request that it
points to is fully completed.
In this fashion, s_tail_ack_queue constantly chases r_head_ack_queue and
the two pointers can easily be used to determine "queue full" and "queue
empty" conditions.
The detection of these two conditions are imported in determining when an
old entry can safely be overwritten with a new received request and the
resources associated with the old request be safely released.
When pipelined TID RDMA WRITE is introduced into this mix, things look
very different. r_head_ack_queue is still the point at which a newly
received request will be inserted, s_tail_ack_queue is still the
currently processed request. However, with pipelined TID RDMA WRITE
requests, s_tail_ack_queue moves to the next request once all TID RDMA
WRITE responses for that request have been sent. The rest of the protocol
for a particular request is managed by other pointers specific to TID RDMA
- r_tid_tail and r_tid_ack - which point to the entries for which the next
TID RDMA DATA packets are going to arrive and the request for which
the next TID RDMA ACK packets are to be generated, respectively.
What this means is that entries in the ring, which are "behind"
s_tail_ack_queue (entries which s_tail_ack_queue has gone past) are no
longer considered complete. This is where the problem is - a newly
received request could potentially overwrite a still active TID RDMA WRITE
request.
The reason why the TID RDMA pointers trail s_tail_ack_queue is that the
normal Verbs send engine uses s_tail_ack_queue as the pointer for the next
response. Since TID RDMA WRITE responses are processed by the normal Verbs
send engine, s_tail_ack_queue had to be moved to the next entry once all
TID RDMA WRITE response packets were sent to get the desired pipelining
between requests. Doing otherwise would mean that the normal Verbs send
engine would not be able to send the TID RDMA WRITE responses for the next
TID RDMA request until the current one is fully completed.
This patch introduces the s_acked_ack_queue index to point to the next
request to complete on the responder side. For requests other than TID
RDMA WRITE, s_acked_ack_queue should always be kept in sync with
s_tail_ack_queue. For TID RDMA WRITE request, it may fall behind
s_tail_ack_queue.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The RC retry timeout value is based on the estimated time for the
response packet to come back. However, for TID RDMA READ request, due
to the use of header suppression, the driver is normally not notified
for each incoming response packet until the last TID RDMA READ response
packet. Consequently, the retry timeout value should be extended to
cover the transaction time for the entire length of a segment (default
256K) instead of that for a single packet. This patch addresses the
issue by introducing new retry timer functions to account for multiple
packets and wrapper functions for backward compatibility.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This patch adds the helper functions to build the TID RDMA READ request
on the requester side. The key is to allocate TID resources (TID flow
and TID entries) and send the resource information to the responder side
along with the read request. Since the TID resources are limited, each
TID RDMA READ request has to be split into segments with a default
segment size of 256K. A software flow is allocated to track the data
transaction for each segment. The work request opcode, packet opcode, and
packet formats for TID RDMA READ protocol are also defined in this patch.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
TID entries are used by hfi1 hardware to receive data payload from
incoming packets directly into a user buffer and thus avoid data copying
by software. This patch implements the functions for TID allocation,
freeing, and programming TID RcvArray entries in hardware for kernel
clients. TID entries are managed via lists of TID groups similar to PSM.
Furthermore, to track TID resource allocation for each request, software
flows are also allocated and freed as needed. Since software flows
consume large amount of memory for tracking TID allocation and freeing,
it is generally desirable to allocate them dynamically in the send queue
and only for TID RDMA requests, but pre-allocate them for receive queue
because the send queue could have thousands of entries while the receive
queue has only a limited number of entries.
Signed-off-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This patch moves some RC helper functions into a header file so that
they can be called from both RC and TID RDMA functions. In addition,
a common function for rewinding a request is created in rdmavt so that
it can be shared between qib and hfi1 driver.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The OPFN protocol uses the COMPARE_SWAP request to exchange data
between the requester and the responder and therefore needs to
be stored in the QP's s_ack_queue when the request is received
on the responder side. However, because the user does not know
anything about the OPFN protocol, this extra entry in the
queue cannot be advertised to the user. This patch adds an extra
entry in a QP's s_ack_queue.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Yishai Hadas says:
Enable DEVX asynchronous query commands
This series enables querying a DEVX object in an asynchronous mode.
The userspace application won't block when calling the firmware and it will be
able to get the response back once that it will be ready.
To enable the above functionality:
- DEVX asynchronous command completion FD object was introduced.
- The applicable file operations were implemented to enable using it by
the user application.
- Query asynchronous method was added to the DEVX object, it will call the
firmware asynchronously and manages the response on the given input FD.
- Hot unplug support was added for the FD to work properly upon
unbind/disassociate.
- mlx5 core fence for asynchronous commands was implemented and used to
prevent racing upon unbind/disassociate.
This branch is based on mlx5-next & v5.0-rc2 due to dependencies, from
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
* branch 'devx-async':
IB/mlx5: Implement DEVX hot unplug for async command FD
IB/mlx5: Implement the file ops of DEVX async command FD
IB/mlx5: Introduce async DEVX obj query API
IB/mlx5: Introduce MLX5_IB_OBJECT_DEVX_ASYNC_CMD_FD
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Introduce async DEVX obj query API to get the command response back to
user space once it's ready without blocking when calling the firmware.
The event's data includes a header with some meta data then the firmware
output command data.
The header includes:
- The input 'wr_id' to let application recognizing the response.
The input FD attribute is used to have the event data ready on.
Downstream patches from this series will implement the file ops to let
application read it.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Introduce MLX5_IB_OBJECT_DEVX_ASYNC_CMD_FD and its initial implementation.
This object is from type class FD and will be used to read DEVX async
commands completion.
The core layer should allow the driver to set object from type FD in a
safe mode, this option was added with a matching comment in place.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When CONFIG_INFINIBAND_ON_DEMAND_PAGING is not set there is no caller to
ib_alloc_odp_umem() so let's remove it.
Signed-off-by: Moni Shoua <monis@mellanox.com>
Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
A sub-range in ODP implicit MR should take its write permission from the
MR and not be set always to allow.
Fixes: d07d1d70ce ("IB/umem: Update on demand page (ODP) support")
Signed-off-by: Moni Shoua <monis@mellanox.com>
Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
APIs that have deferred callbacks should have some kind of cleanup
function that callers can use to fence the callbacks. Otherwise things
like module unloading can lead to dangling function pointers, or worse.
The IB MR code is the only place that calls this function and had a
really poor attempt at creating this fence. Provide a good version in
the core code as future patches will add more places that need this
fence.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Manage indirection mkey upon DEVX flow to support ODP.
To support a page fault event on the indirection mkey it needs to be part
of the device mkey radix tree.
Both the creation and the deletion flows for a DEVX object which is
indirection mkey were adapted to handle that.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
RDMA cgroup registration routine always returns success, so simplify
function to be void and run clang formatter over whole CONFIG_CGROUP_RDMA
art of core_priv.h.
This reduces unwinding error path for regular registration and future net
namespace change functionality for rdma device.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Management Datagram Interface (MAD) is applicable
only when physical port is Infiniband. It makes MAD
command logic to be completely unrelated to eth/core
parts of mlx5.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Jason Gunthorpe <jgg@mellanox.com>
Introduce and use rdma_device_to_ibdev() API for those drivers which are
registering one sysfs group and also use in ib_core.
In subsequent patch, device->provider_ibdev one-to-one mapping is no
longer holds true during accessing sysfs entries.
Therefore, introduce an API rdma_device_to_ibdev() that provides such
information.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Most provider routines are callback routines which ib core invokes.
_callback suffix doesn't convey information about when such callback is
invoked. Therefore, rename port_callback to init_port.
Additionally, store the init_port function pointer in ib_device_ops, so
that it can be accessed in subsequent patches when binding rdma device to
net namespace.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Pull ARM SoC fixes from Olof Johansson:
"A bigger batch than I anticipated this week, for two reasons:
- Some fallout on Davinci from board file -> DTB conversion, that
also includes a few longer-standing fixes (i.e. not recent
regressions).
- drivers/reset material that has been in linux-next for a while, but
didn't get sent to us until now for a variety of reasons
(maintainer out sick, holidays, etc). There's a functional
dependency in there such that one platform (Altera's SoCFPGA) won't
boot without one of the patches; instead of reverting the patch
that got merged, I looked at this set and decided it was small
enough that I'll pick it up anyway. If you disagree I can revisit
with a smaller set.
That being said, there's also a handful of the usual stuff:
- Fix for a crash on Armada 7K/8K when the kernel touches
PSCI-reserved memory
- Fix for PCIe reset on Macchiatobin (Armada 8K development board,
what this email is sent from in fact :)
- Enable a few new-merged modules for Amlogic in arm64 defconfig
- Error path fixes on Integrator
- Build fix for Renesas and Qualcomm
- Initialization fix for Renesas RZ/G2E
.. plus a few more fixlets"
* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (28 commits)
ARM: integrator: impd1: use struct_size() in devm_kzalloc()
qcom-scm: Include <linux/err.h> header
gpio: pl061: handle failed allocations
ARM: dts: kirkwood: Fix polarity of GPIO fan lines
arm64: dts: marvell: mcbin: fix PCIe reset signal
arm64: dts: marvell: armada-ap806: reserve PSCI area
ARM: dts: da850-lcdk: Correct the sound card name
ARM: dts: da850-lcdk: Correct the audio codec regulators
ARM: dts: da850-evm: Correct the sound card name
ARM: dts: da850-evm: Correct the audio codec regulators
ARM: davinci: omapl138-hawk: fix label names in GPIO lookup entries
ARM: davinci: dm644x-evm: fix label names in GPIO lookup entries
ARM: davinci: dm355-evm: fix label names in GPIO lookup entries
ARM: davinci: da850-evm: fix label names in GPIO lookup entries
ARM: davinci: da830-evm: fix label names in GPIO lookup entries
arm64: defconfig: enable modules for amlogic s400 sound card
reset: uniphier-glue: Add AHCI reset control support in glue layer
dt-bindings: reset: uniphier: Add AHCI core reset description
reset: uniphier-usb3: Rename to reset-uniphier-glue
dt-bindings: reset: uniphier: Replace the expression of USB3 with generic peripherals
...
Late reset controller changes for v5.0
This adds missing deassert functionality to the ARC HSDK reset driver,
fixes some indentation and grammar issues in the kernel docs, adds a
helper to count the number of resets on a device for the non-DT case
as well, adds an early reset driver for SoCFPGA and simple reset driver
support for Stratix10, and generalizes the uniphier USB3 glue layer
reset to also cover AHCI.
* tag 'reset-for-5.0-rc2' of git://git.pengutronix.de/git/pza/linux:
reset: uniphier-glue: Add AHCI reset control support in glue layer
dt-bindings: reset: uniphier: Add AHCI core reset description
reset: uniphier-usb3: Rename to reset-uniphier-glue
dt-bindings: reset: uniphier: Replace the expression of USB3 with generic peripherals
ARM: socfpga: dts: document "altr,stratix10-rst-mgr" binding
reset: socfpga: add an early reset driver for SoCFPGA
reset: fix null pointer dereference on dev by dev_name
reset: Add reset_control_get_count()
reset: Improve reset controller kernel docs
ARC: HSDK: improve reset driver
Signed-off-by: Olof Johansson <olof@lixom.net>
Commit 49e54187ae ("ata: libahci_platform: comply to PHY framework") uses
the PHY_MODE_SATA, but that enum had not yet been added. This caused a
build failure for me, with today's linux.git.
Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
in the Marvell Berlin SATA PHY driver.
Fix the build by:
1) Renaming Marvell's defined value to a more scoped name,
in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
2) Adding the missing enum, which was going to be added anyway as part
of [1].
[1] https://lkml.kernel.org/r/20190108163124.6409-3-miquel.raynal@bootlin.com
Fixes: 49e54187ae ("ata: libahci_platform: comply to PHY framework")
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Olof Johansson <olof@lixom.net>
Cc: Grzegorz Jaszczyk <jaz@semihalf.com>
Cc: Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull dma_zalloc_coherent() removal from Christoph Hellwig:
"We've always had a weird situation around dma_zalloc_coherent. To
safely support mapping the allocations to userspace major
architectures like x86 and arm have always zeroed allocations from
dma_alloc_coherent, but a couple other architectures were missing that
zeroing either always or in corner cases.
Then later we grew anothe dma_zalloc_coherent interface to explicitly
request zeroing, but that just added __GFP_ZERO to the allocation
flags, which for some allocators that didn't end up using the page
allocator ended up being a no-op and still not zeroing the
allocations.
So for this merge window I fixed up all remaining architectures to
zero the memory in dma_alloc_coherent, and made dma_zalloc_coherent a
no-op wrapper around dma_alloc_coherent, which fixes all of the above
issues.
dma_zalloc_coherent is now pointless and can go away, and Luis helped
me writing a cocchinelle script and patch series to kill it, which I
think we should apply now just after -rc1 to finally settle these
issue"
* tag 'remove-dma_zalloc_coherent-5.0' of git://git.infradead.org/users/hch/dma-mapping:
dma-mapping: remove dma_zalloc_coherent()
cross-tree: phase out dma_zalloc_coherent() on headers
cross-tree: phase out dma_zalloc_coherent()
Pull more drm fixes from Daniel Vetter:
"Dave sends out his pull, everybody remembers holidays are over :-)
Since Dave's already in weekend mode and it was quite a few patches I
figured better to apply all the pulls and forward them to you. Hence
here 2nd part of bugfixes for -rc2.
nouveau:
- backlight fix
- falcon register access fix
- fan fix.
i915:
- Disable PSR for Apple panels
- Broxton ERR_PTR error state fix
- Kabylake VECS workaround fix
- Unwind failure on pinning the gen7 ppgtt
- GVT workload request allocation fix
core:
- Fix fb-helper to work correctly with SDL 1.2 bugs
- Fix lockdep warning in the atomic ioctl and setproperty"
* tag 'drm-fixes-2019-01-11-1' of git://anongit.freedesktop.org/drm/drm:
drm/nouveau/falcon: avoid touching registers if engine is off
drm/nouveau: Don't disable polling in fallback mode
drm/nouveau: register backlight on pascal and newer
drm: Fix documentation generation for DP_DPCD_QUIRK_NO_PSR
drm/i915: init per-engine WAs for all engines
drm/i915: Unwind failure on pinning the gen7 ppgtt
drm/i915: Skip the ERR_PTR error state
drm/i915: Disable PSR in Apple panels
gpu/drm: Fix lock held when returning to user space.
drm/fb-helper: Ignore the value of fb_var_screeninfo.pixclock
drm/fb-helper: Partially bring back workaround for bugs of SDL 1.2
drm/i915/gvt: Fix workload request allocation before request add
Pull ceph updates from Ilya Dryomov:
"A patch to allow setting abort_on_full and a fix for an old "rbd
unmap" edge case, marked for stable"
* tag 'ceph-for-5.0-rc2' of git://github.com/ceph/ceph-client:
rbd: don't return 0 on unmap if RBD_DEV_FLAG_REMOVING is set
ceph: use vmf_error() in ceph_filemap_fault()
libceph: allow setting abort_on_full for rbd
Pull x86 fixes from Ingo Molnar:
"A 32-bit build fix, CONFIG_RETPOLINE fixes and rename CONFIG_RESCTRL
to CONFIG_X86_RESCTRL"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, modpost: Replace last remnants of RETPOLINE with CONFIG_RETPOLINE
x86/cache: Rename config option to CONFIG_X86_RESCTRL
samples/seccomp: Fix 32-bit build
Pull power management updates from Rafael Wysocki:
"These fix fallout after starting to use hrtimers in the runtime PM
framework, fix a few cpufreq issues, fix a recently broken reference
to cpuidle documentation, update MAINTAINERS entries for cpufreq and
cpuidle and make the recently added system suspend and resume support
in devfreq actually work.
Specifics:
- Prevent integer overflows from occurring on 32-bit when converting
milliseconds to nanoseconds in the runtime PM framework and update
comments that still refer to jiffies in it (Vincent Guittot,
Ladislav Michl).
- Fix the SCMI cpufreq driver to always use the same frequency units
for arch_set_freq_scale() and make the scale-invariant load
tracking acutally work with this driver (Quentin Perret).
- Fix freeing of dynamic OPPs in the SCPI and SCMI cpufreq drivers
broken during the 4.20 defelopment cycle (Viresh Kumar).
- Prevent the cpufreq core from attempting to return the current
frequency of offline CPUs (Sudeep Holla).
- Add devfreq suspend and resume hooks (missed previously) to the PM
core to make the recently added system suspend and resume support
in devfreq actually work (Lukasz Luba).
- Update MAINTAINERS entries for cpufreq and cpuidle, mostly to add
references to new/current documentation to them (Rafael Wysocki).
- Fix a recently broken reference to cpuidle documentation (Otto
Sabart)"
* tag 'pm-5.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
PM-runtime: Fix autosuspend_delay on 32bits arch
PM-runtime: Fix 'jiffies' in comments after switch to hrtimers
cpufreq: scmi: Fix frequency invariance in slow path
doc: trace: fix reference to cpuidle documentation file
cpufreq: check if policy is inactive early in __cpufreq_get()
cpufreq: scpi/scmi: Fix freeing of dynamic OPPs
cpuidle / Documentation: Update cpuidle MAINTAINERS entry
cpufreq / Documentation: Update cpufreq MAINTAINERS entry
PM: sleep: call devfreq suspend/resume
Pull drm fixes from Dave Airlie:
"Not a huge amount for rc2, assume the usual quiet period, and rc3 will
be most of it.
amdgpu:
- Powerplay fixes
- Virtual display pinning fixes
- Golden register updates for Vega
- Pitch and gem size validation fixes
- SR-IOV init error fix
- Pagetables in system RAM disable for some Raven system
- DP-MST resume fixes
tc358767 bridge:
- fix to work with displayport connector"
* tag 'drm-fixes-2019-01-11' of git://anongit.freedesktop.org/drm/drm: (26 commits)
drm/amdgpu: disable system memory page tables for now
drm/amdgpu: set WRITE_BURST_LENGTH to 64B to workaround SDMA1 hang
drm/amdgpu: fix CPDMA hang in PRT mode for VEGA20
drm/bridge: tc358767: use DP connector if no panel set
drm/bridge: tc358767: fix output H/V syncs
drm/bridge: tc358767: reject modes which require too much BW
drm/bridge: tc358767: fix initial DP0/1_SRCCTRL value
drm/bridge: tc358767: fix single lane configuration
drm/bridge: tc358767: add defines for DP1_SRCCTRL & PHY_2LANE
drm/bridge: tc358767: add bus flags
drm/dp_mst: Add __must_check to drm_dp_mst_topology_mgr_resume()
drm/amdgpu: Don't fail resume process if resuming atomic state fails
drm/amdgpu: Don't ignore rc from drm_dp_mst_topology_mgr_resume()
drm/amdgpu: validate user GEM object size
drm/amdgpu: validate user pitch alignment
drm/amd/powerplay: drop the unnecessary uclk hard min setting
drm/amd/powerplay: avoid possible buffer overflow
drm/amd/powerplay: create pp_od_clk_voltage device file under OD support
drm/amd/powerplay: update OD support flag for SKU with no OD capabilities
drm/amdgpu: make gfx9 enter into rlc safe mode when set MGCG
...
ib_umem_get() can only be called in a method callback, which always has a
udata parameter. This allows ib_umem_get() to derive the ucontext pointer
directly from the udata without requiring the drivers to find it in some
way or another.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
Since commit e6f6d63ed1 ("drm/msm: add headless gpu device for imx5")
the DRM_MSM symbol can be selected by SOC_IMX5 causing the following
error when building imx_v6_v7_defconfig:
In file included from ../drivers/gpu/drm/msm/adreno/a5xx_gpu.c:17:0:
../include/linux/qcom_scm.h: In function 'qcom_scm_set_cold_boot_addr':
../include/linux/qcom_scm.h:73:10: error: 'ENODEV' undeclared (first use in this function)
return -ENODEV;
Include the <linux/err.h> header file to fix this problem.
Reported-by: kernelci.org bot <bot@kernelci.org>
Fixes: e6f6d63ed1 ("drm/msm: add headless gpu device for imx5")
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Andy Gross <andy.gross@linaro.org>
Pull RISC-V updates from Palmer Dabbelt:
"This tag contains a handful of updates that slipped through the cracks
during the merge window due to the holidays. The fixes are mostly
independent, with the exception of one larger audit-related branch.
Core RISC-V updates:
- The BSS has been moved, which shrinks flat images.
- A fix to test-bpf so it compiles on RV64I-based systems.
- A fix to respect the kernel commandline when there is no device
tree.
- A fix to prevent CPUs from trying to put themselves to sleep when
bringing down the system.
- Support for MODULE_SECTIONS on RV32I-based systems.
- [new in v2] The addition of an SBI earlycon driver. This is
definately a new feature, but I'd like to include it now because I
dropped this patch when submitting the merge window PR that removed
our EARLY_PRINTK support.
RISC-V audit updates:
- The addition of NR_syscalls into unistd.h, which is necessary for
CONFIG_FTRACE_SYSCALLS.
- The definition of CREATE_TRACE_POINTS so __tracepoint_sys_{enter,exit}
get defined.
- A fix for trace_sys_exit() so we can enable HAVE_SYSCALL_TRACEPOINTS
As usual, I've tested this by booting a Fedora-based image on a recent
QEMU (this time just whatever I had lying around).
* tag 'riscv-for-linus-4.21-rc2-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
tty/serial: Add RISC-V SBI earlycon support
riscv: add HAVE_SYSCALL_TRACEPOINTS to Kconfig
riscv: fix trace_sys_exit hook
riscv: define CREATE_TRACE_POINTS in ptrace.c
riscv: define NR_syscalls in unistd.h
riscv: audit: add audit hook in do_syscall_trace_enter/exit()
riscv: add audit support
RISC-V: Support MODULE_SECTIONS mechanism on RV32
MAINTAINERS: SiFive drivers: add myself as a SiFive driver maintainer
MAINTAINERS: SiFive drivers: change the git tree to a SiFive git tree
riscv: don't stop itself in smp_send_stop
arch: riscv: support kernel command line forcing when no DTB passed
tools uapi: fix RISC-V 64-bit support
RISC-V: Make BSS section as the last section in vmlinux.lds.S
syzbot reported the following regression in the latest merge window and
it was confirmed by Qian Cai that a similar bug was visible from a
different context.
======================================================
WARNING: possible circular locking dependency detected
4.20.0+ #297 Not tainted
------------------------------------------------------
syz-executor0/8529 is trying to acquire lock:
000000005e7fb829 (&pgdat->kswapd_wait){....}, at:
__wake_up_common_lock+0x19e/0x330 kernel/sched/wait.c:120
but task is already holding lock:
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at: spin_lock
include/linux/spinlock.h:329 [inline]
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at: rmqueue_bulk
mm/page_alloc.c:2548 [inline]
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at: __rmqueue_pcplist
mm/page_alloc.c:3021 [inline]
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at: rmqueue_pcplist
mm/page_alloc.c:3050 [inline]
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at: rmqueue
mm/page_alloc.c:3072 [inline]
000000009bb7bae0 (&(&zone->lock)->rlock){-.-.}, at:
get_page_from_freelist+0x1bae/0x52a0 mm/page_alloc.c:3491
It appears to be a false positive in that the only way the lock ordering
should be inverted is if kswapd is waking itself and the wakeup
allocates debugging objects which should already be allocated if it's
kswapd doing the waking. Nevertheless, the possibility exists and so
it's best to avoid the problem.
This patch flags a zone as needing a kswapd using the, surprisingly,
unused zone flag field. The flag is read without the lock held to do
the wakeup. It's possible that the flag setting context is not the same
as the flag clearing context or for small races to occur. However, each
race possibility is harmless and there is no visible degredation in
fragmentation treatment.
While zone->flag could have continued to be unused, there is potential
for moving some existing fields into the flags field instead.
Particularly read-mostly ones like zone->initialized and
zone->contiguous.
Link: http://lkml.kernel.org/r/20190103225712.GJ31517@techsingularity.net
Fixes: 1c30844d2d ("mm: reclaim small amounts of memory when an external fragmentation event occurs")
Reported-by: syzbot+93d94a001cfbce9e60e1@syzkaller.appspotmail.com
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Qian Cai <cai@lca.pw>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The implementation of mlx5_core_page_fault_resume() was removed in commit
d5d284b829 ("{net,IB}/mlx5: Move Page fault EQ and ODP logic to
RDMA"). This patch removes declaration too.
Fixes: d5d284b829 ("{net,IB}/mlx5: Move Page fault EQ and ODP logic to RDMA")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Convert various places to more readable code, which embeds
CONFIG_INFINIBAND_ON_DEMAND_PAGING into the code flow.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
CONFIG_INFINIBAND_ON_DEMAND_PAGING is used in general structures to
micro-optimize the memory footprint. Remove it, so it will allow us to
simplify various ODP device flows.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
dma_zalloc_coherent() is no longer needed as it has no users because
dma_alloc_coherent() already zeroes out memory for us.
The Coccinelle grammar rule that used to check for dma_alloc_coherent()
+ memset() is modified so that it just tells the user that the memset is
not needed anymore.
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
We already need to zero out memory for dma_alloc_coherent(), as such
using dma_zalloc_coherent() is superflous. Phase it out.
This change was generated with the following Coccinelle SmPL patch:
@ replace_dma_zalloc_coherent @
expression dev, size, data, handle, flags;
@@
-dma_zalloc_coherent(dev, size, handle, flags)
+dma_alloc_coherent(dev, size, handle, flags)
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
[hch: re-ran the script on the latest tree]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Introduce a new option abort_on_full, default to false. Then
we can get -ENOSPC when the pool is full, or reaches quota.
[ Don't show abort_on_full in /proc/mounts. ]
Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
This is sort of a mix between a new feature and a bug fix. I've managed
to screw up merging this patch set a handful of times but I think it's
OK this time around. The main new feature here is audit support for
RISC-V, with some fixes to audit-related bugs that cropped up along the
way:
* The addition of NR_syscalls into unistd.h, which is necessary for
CONFIG_FTRACE_SYSCALLS.
* The definition of CREATE_TRACE_POINTS so
__tracepoint_sys_{enter,exit} get defined.
* A fix for trace_sys_exit() so we can enable
CONFIG_HAVE_SYSCALL_TRACEPOINTS.
On RISC-V (riscv) audit is supported through generic lib/audit.c.
The patch adds required arch specific definitions.
Signed-off-by: David Abdurachmanov <david.abdurachmanov@gmail.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
Currently the reset core has internal support for counting the number of
resets for a device described in DT. Generalize this to devices using
lookup resets, and export it for public use.
This will be used by generic drivers that need to be sure a device is
controlled by a single, dedicated reset line (e.g. vfio-platform).
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
[p.zabel@pengutronix.de: fixed a typo in reset_control_get_count comment]
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Pull more Kbuild updates from Masahiro Yamada:
- improve boolinit.cocci and use_after_iter.cocci semantic patches
- fix alignment for kallsyms
- move 'asm goto' compiler test to Kconfig and clean up jump_label
CONFIG option
- generate asm-generic wrappers automatically if arch does not
implement mandatory UAPI headers
- remove redundant generic-y defines
- misc cleanups
* tag 'kbuild-v4.21-3' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
kconfig: rename generated .*conf-cfg to *conf-cfg
kbuild: remove unnecessary stubs for archheader and archscripts
kbuild: use assignment instead of define ... endef for filechk_* rules
arch: remove redundant UAPI generic-y defines
kbuild: generate asm-generic wrappers if mandatory headers are missing
arch: remove stale comments "UAPI Header export list"
riscv: remove redundant kernel-space generic-y
kbuild: change filechk to surround the given command with { }
kbuild: remove redundant target cleaning on failure
kbuild: clean up rule_dtc_dt_yaml
kbuild: remove UIMAGE_IN and UIMAGE_OUT
jump_label: move 'asm goto' support test to Kconfig
kallsyms: lower alignment on ARM
scripts: coccinelle: boolinit: drop warnings on named constants
scripts: coccinelle: check for redeclaration
kconfig: remove unused "file" field of yylval union
nds32: remove redundant kernel-space generic-y
nios2: remove unneeded HAS_DMA define
Pull fscrypt updates from Ted Ts'o:
"Add Adiantum support for fscrypt"
* tag 'fscrypt_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/fscrypt:
fscrypt: add Adiantum support
