Kernel doc validator is unhappy with the following
.../perf.c:16: warning: Function parameter or member 'latency_lock' not described in 'DEFINE_SPINLOCK'
.../perf.c:16: warning: expecting prototype for perf.c(). Prototype was for DEFINE_SPINLOCK() instead
Drop kernel doc annotation since the top comment is not in the required format.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20210729163538.40101-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210818134852.1847070-8-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The minimum per-IOMMU PRQ queue size is one 4K page, this is more entries
than the hardcoded limit of 32 in the current VT-d code. Some devices can
support up to 512 outstanding PRQs but underutilized by this limit of 32.
Although, 32 gives some rough fairness when multiple devices share the same
IOMMU PRQ queue, but far from optimal for customized use case. This extends
the per-IOMMU PRQ queue size to four 4K pages and let the devices have as
many outstanding page requests as they can.
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210720013856.4143880-1-baolu.lu@linux.intel.com
Link: https://lore.kernel.org/r/20210818134852.1847070-7-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
We preset the access and dirty bits for IOVA over first level usage only
for the kernel DMA (i.e., when domain type is IOMMU_DOMAIN_DMA). We should
also preset the FL A/D for user space DMA usage. The idea is that even the
user space A/D bit memory write is unnecessary. We should avoid it to
minimize the overhead.
Suggested-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210720013856.4143880-1-baolu.lu@linux.intel.com
Link: https://lore.kernel.org/r/20210818134852.1847070-6-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The commit 8950dcd83a ("iommu/vt-d: Leave scalable mode default off")
leaves the scalable mode default off and end users could turn it on with
"intel_iommu=sm_on". Using the Intel IOMMU scalable mode for kernel DMA,
user-level device access and Shared Virtual Address have been enabled.
This enables the scalable mode by default if the hardware advertises the
support and adds kernel options of "intel_iommu=sm_on/sm_off" for end
users to configure it through the kernel parameters.
Suggested-by: Ashok Raj <ashok.raj@intel.com>
Suggested-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Cc: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20210720013856.4143880-1-baolu.lu@linux.intel.com
Link: https://lore.kernel.org/r/20210818134852.1847070-5-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Introduce a test for aarch64 that ensures CPU resets induced by PSCI are
reflected in the target vCPU's state, even if the target is never run
again. This is a regression test for a race between vCPU migration and
PSCI.
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210818202133.1106786-5-oupton@google.com
According to the PSCI specification, ARM DEN 0022D, 5.1.4 "CPU_ON", the
CPU_ON function takes a target_cpu argument that is bit-compatible with
the affinity fields in MPIDR_EL1. All other bits in the argument are
RES0. Note that the same constraints apply to the target_affinity
argument for the AFFINITY_INFO call.
Enforce the spec by returning INVALID_PARAMS if a guest incorrectly sets
a RES0 bit.
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210818202133.1106786-4-oupton@google.com
The CPU_ON PSCI call takes a payload that KVM uses to configure a
destination vCPU to run. This payload is non-architectural state and not
exposed through any existing UAPI. Effectively, we have a race between
CPU_ON and userspace saving/restoring a guest: if the target vCPU isn't
ran again before the VMM saves its state, the requested PC and context
ID are lost. When restored, the target vCPU will be runnable and start
executing at its old PC.
We can avoid this race by making sure the reset payload is serviced
before userspace can access a vCPU's state.
Fixes: 358b28f09f ("arm/arm64: KVM: Allow a VCPU to fully reset itself")
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210818202133.1106786-3-oupton@google.com
KVM correctly serializes writes to a vCPU's reset state, however since
we do not take the KVM lock on the read side it is entirely possible to
read state from two different reset requests.
Cure the race for now by taking the KVM lock when reading the
reset_state structure.
Fixes: 358b28f09f ("arm/arm64: KVM: Allow a VCPU to fully reset itself")
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210818202133.1106786-2-oupton@google.com
Matthias Fuchs's and Thomas Körper's email addresses aren't valid
anymore. Use the newly created role account instead.
Link: https://lore.kernel.org/r/20210809175843.207864-1-mkl@pengutronix.de
Cc: socketcan@esd.eu
Cc: Stefan Mätje <Stefan.Maetje@esd.eu>
Acked-by: Stefan Mätje <stefan.maetje@esd.eu>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
A make defconfig && make savedefconfig was performed for each
configuration.
Most changes are due to options moving around, except for the following
which are due to changing defaults:
- SECCOMP is enabled by default as of commit 282a181b1a ("seccomp:
Move config option SECCOMP to arch/Kconfig") in v5.9
- The soc drivers ASPEED_LPC_CTRL, ASPEED_LPC_SNOOP and ASPEED_P2A_CTRL
are enabled by default when the aspeed platform is enabled, as of
commit 592693a1f8 ("soc: aspeed: Improve kconfig") in v5.9
- The ZBOOT_ROM_TEXT/BSS values fall out of the defconfig as of
commit 39c3e30456 ("ARM: 8984/1: Kconfig: set default
ZBOOT_ROM_TEXT/BSS value to 0x0") in v5.8
- I2C_MUX is selected by MEDIA_SUBDRV_AUTOSELECT, probably as of about
v5.8. It was in the config as it is required bt the PCA I2C muxes
that enabled it
Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Link: https://lore.kernel.org/r/20210819065203.2620911-5-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
Switch hexagon to use the generic code for dma_alloc_coherent from
a global pre-filled pool.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Cain <bcain@codeaurora.org>
Only build the code to support the global coherent pool if support for
it is enabled.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Dillon Min <dillon.minfei@gmail.com>
These two devices are part of the AST2400 and the drivers have support
for that version of the soc.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Link: https://lore.kernel.org/r/20210819065203.2620911-4-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
Raw SerIO is used by the OpenBMC debug-trigger application to take
signals from the host that applications are unresponsive on the BMC.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Link: https://lore.kernel.org/r/20210819065203.2620911-3-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
SLAB_FREELIST_HARDENED can protect from freelist overwrite attacks with
really small overhead.
It works best with the SLUB allocator, so make SLUB the default by
removing SLAB=y.
total used free shared buff/cache available
SLAB 425596 44065.3+/-220 311099+/-3800 14864+/-3900 70432+/-3700 352767+/-3900
SLUB 425592 44225.3+/-280 313275+/-600 12132+/-3.3 68092+/-530 355295+/-280
These figures are the average memory usage from three boots of each
option in qemu, running the Romulus userspace. The output is from
free(1), reported in kilobytes.
Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Link: https://lore.kernel.org/r/20210819065203.2620911-2-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
After commit a6d90e9f22 ("bus: ti-sysc: AM3: RNG is GP only"), clang
with -Wimplicit-fallthrough enabled warns:
drivers/bus/ti-sysc.c:2958:3: warning: unannotated fall-through between
switch labels [-Wimplicit-fallthrough]
default:
^
drivers/bus/ti-sysc.c:2958:3: note: insert 'break;' to avoid
fall-through
default:
^
break;
1 warning generated.
Clang's version of this warning is a little bit more pedantic than
GCC's. Add the missing break to satisfy it to match what has been done
all over the kernel tree.
Fixes: a6d90e9f22 ("bus: ti-sysc: AM3: RNG is GP only")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Remove rtw_mfree_xmit_priv_lock() because it is empty. Remove its only
call from within _rtw_free_xmit_priv().
Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Link: https://lore.kernel.org/r/20210819050808.28825-1-fmdefrancesco@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the unused rtw_update_mem_stat macro and the associated flags
from include/osdep_service.h as this is all dead code.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-8-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the function _rtw_zvmalloc from os_dep/osdep_service.c, as this
function is now unused and is just an inline wrapper around vmalloc
which zeroes out the memory. All previous callers have been converted to
use vzalloc. Also remove the declaration from include/osdep_service.h.
It is considered generally bad practice to declare functions as inline in
the majority of cases, as not only can this qualifier be ignored by the
compiler but the compiler generally makes good decisions about inlining
anyway.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-7-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove rtw_zvmalloc preprocessor definition from include/osdep_service.h
as it now has no callers.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-6-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert all rtw_zvmalloc calls within the driver to use the existing
kernel vzalloc function, which has the same semantics. Also rewrite the
two places where it is mentioned in comments to say vzalloc, and remove
the redundant cast to struct adapter * in ./os_dep/usb_intf.c as vzalloc
returns void *.
The reason for the conversion is that rtw_zvmalloc is just a
preprocessor definition for _rtw_zvmalloc which itself is just an inline
wrapper around vmalloc which then zeroes the memory out. As vzalloc does
the same thing via usage of __GFP_ZERO, this code is redundant and can
subsequently be removed.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-5-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the function _rtw_vmalloc from os_dep/osdep_service.c, converting
its only user (also in os_dep/osdep_service.c) to use plain vmalloc.
This function is just an inline wrapper around vmalloc which returns a u8
pointer, which isn't needed. Also remove the declaration from
include/osdep_service.h.
It is considered generally bad practice to declare functions as inline in
the majority of cases, as not only can this qualifier be ignored by the
compiler but the compiler generally makes good decisions about inlining
anyway.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-4-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove rtw_vmalloc preprocessor definition from include/osdep_service.h
as it now has no callers.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-3-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert the only call to rtw_vmalloc in os_dep/ioctl_linux.c to the
kernel's existing vmalloc function, as rtw_malloc is just a preprocessor
definition for _rtw_vmalloc. The _rtw_vmalloc function is defined inline
and returns a u8, wrapping standard vmalloc. This behaviour is not necessary.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234853.208448-2-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove free_xmit_priv function pointer field from struct hal_ops definition
in include/hal_intf.h, as it is now no longer used anywhere in the driver.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-7-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove function rtw_hal_free_xmit_priv in hal/hal_intf.c and its
declaration in include/hal_intf.h, as well as the single call line in
core/rtw_xmit.c. This function now essentially checks a function
pointer which will always be NULL (due to previous patches) and
therefore will never be called.
Acked-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-6-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove empty function rtl8188eu_free_xmit_priv from hal/rtl8188eu_xmit.c
and its declaration from include/rtl8188e_xmit.h, as well as the one
place where it is stored in the free_xmit_priv function pointer of the
HalFunc struct inside padapter. This is safe, as the wrapper function
checks for NULL before calling this function via the function pointer.
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-5-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove txrpt_ccx_sw_88e and txrpt_ccx_qtime_88e macro definitions from
include/rtl8188e_xmit.h, as these were only called from the now removed
dump_txrpt_ccx_88e function, which was itself not called from anywhere
anyway.
Acked-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-4-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove unused function dump_txrpt_ccx_88e from hal/rtl8188e_xmit.c and
remove its declaration in include/rtl8188e_xmit.h, as this function is
not called from anywhere, and is thus dead code.
Acked-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-3-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove _dbg_dump_tx_info from hal/rtl8188e_xmit.c as it is just a
dumping function that contains a lot of unclear DBG_88E calls, and has
no other external effect, other than calling a function that ultimately
determines whether or not to dump/trigger the DBG_88E calls. Also remove
its declaration and single call site.
Acked-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
Link: https://lore.kernel.org/r/20210818234253.208271-2-phil@philpotter.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the removal of the legacy IDE driver in kb7fb14d3ac63 ("ide: remove
the legacy ide driver"), this crashpoint no longer points to a valid
function.
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kevin Mitchell <kevmitch@arista.com>
Link: https://lore.kernel.org/r/20210819022940.561875-3-kevmitch@arista.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When scsi_dispatch_cmd was moved to scsi_lib.c and made static, some
compilers (i.e., at least gcc 8.4.0) decided to compile this
inline. This is a problem for lkdtm.ko, which inserted a kprobe
on this function for the SCSI_DISPATCH_CMD crashpoint.
Move this crashpoint one function up the call chain to
scsi_queue_rq. Though this is also a static function, it should never be
inlined because it is assigned as a structure entry. Therefore,
kprobe_register should always be able to find it.
Fixes: 82042a2cdb ("scsi: move scsi_dispatch_cmd to scsi_lib.c")
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kevin Mitchell <kevmitch@arista.com>
Link: https://lore.kernel.org/r/20210819022940.561875-2-kevmitch@arista.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Recent builds of the vendor u-boot tree disable features of the
BMC that may allow unwanted access if not correctly configured. This
includes the firmware hub to ahb bridge (FWH2AHB), which is used by this
driver.
The bit to "un-disable" it is in the SCU. Set it only when the ioctl is
called and we are running on the ast2600, as to not open up the
'backdoor' unless there's userspace trying to use it.
Fixes: deb50313ba83 ("soc: aspeed-lpc-ctrl: LPC to AHB mapping on ast2600")
Link: https://lore.kernel.org/r/20210629073520.318514-2-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
Add AST26XX series AST2625-A3 SOC ID, taken from the vendor u-boot SDK:
arch/arm/mach-aspeed/ast2600/scu_info.c
+ SOC_ID("AST2625-A3", 0x0503040305030403),
Reviewed-by: Dylan Hung <dylan_hung@aspeedtech.com>
Link: https://lore.kernel.org/r/20210818010534.2508122-1-joel@jms.id.au
Signed-off-by: Joel Stanley <joel@jms.id.au>
With quotaoff not allowing disabling of accounting there is no need
for untagged lookups in this code, so remove the dead leftovers.
Repoted-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
[djwong: convert to for_each_perag_tag]
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Constify the rest of the btree functions that take structure and union
pointers and are not supposed to modify them.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
This btree function is called when updating a record in the rightmost
block of a btree so that we can update the AGF's longest free extent
length field. Neither parameter is supposed to be updated, so mark them
both const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The @start pointer passed to each per-AG btree type's ->alloc_block
function isn't supposed to be modified, since it's a hint about the
location of the btree block being split that is to be fed to the
allocator, so mark the parameter const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The pointer passed to each per-AG btree type's ->set_root function isn't
supposed to be modified (that function sets an external pointer to the
root block) so mark them const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
xchk_btree calls a user-supplied function to validate each btree record
that it finds. Those functions are not supposed to change the record
data, so mark the parameter const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The inorder functions are simple predicates, which means that they don't
modify the parameters. Mark them all const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
These functions initialize a key from a record, but they aren't supposed
to modify the record. Mark it const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The query_range functions are supposed to call a caller-supplied
function on each record found in the dataset. These functions don't
own the memory storing the record, so don't let them change the record.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Range query functions are not supposed to modify the query keys that are
being passed in, so mark them all const.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>