Legacy PCI over virtio uses a 32bit PFN for the queue. If the
queue pfn is too large to fit in 32bits, which we could hit on
arm64 systems with 52bit physical addresses (even with 64K page
size), we simply miss out a proper link to the other side of
the queue.
Add a check to validate the PFN, rather than silently breaking
the devices.
virtio-mmio with virtio-v1 uses a 32bit PFN for the queue.
If the queue pfn is too large to fit in 32bits, which
we could hit on arm64 systems with 52bit physical addresses
(even with 64K page size), we simply miss out a proper link
to the other side of the queue.
Add a check to validate the PFN, rather than silently breaking
the devices.
Wei Wang [Thu, 16 Aug 2018 07:50:58 +0000 (15:50 +0800)]
virtio_balloon: replace oom notifier with shrinker
The OOM notifier is getting deprecated to use for the reasons:
- As a callout from the oom context, it is too subtle and easy to
generate bugs and corner cases which are hard to track;
- It is called too late (after the reclaiming has been performed).
Drivers with large amuont of reclaimable memory is expected to
release them at an early stage of memory pressure;
- The notifier callback isn't aware of oom contrains; Link: https://lkml.org/lkml/2018/7/12/314
This patch replaces the virtio-balloon oom notifier with a shrinker
to release balloon pages on memory pressure. The balloon pages are
given back to mm adaptively by returning the number of pages that the
reclaimer is asking for (i.e. sc->nr_to_scan).
Currently the max possible value of sc->nr_to_scan passed to the balloon
shrinker is SHRINK_BATCH, which is 128. This is smaller than the
limitation that only VIRTIO_BALLOON_ARRAY_PFNS_MAX (256) pages can be
returned via one invocation of leak_balloon. But this patch still
considers the case that SHRINK_BATCH or shrinker->batch could be changed
to a value larger than VIRTIO_BALLOON_ARRAY_PFNS_MAX, which will need to
do multiple invocations of leak_balloon.
Historically, the feature VIRTIO_BALLOON_F_DEFLATE_ON_OOM has been used
to release balloon pages on OOM. We continue to use this feature bit for
the shrinker, so the shrinker is only registered when this feature bit
has been negotiated with host.
Wei Wang [Thu, 16 Aug 2018 07:50:56 +0000 (15:50 +0800)]
virtio-balloon: remove BUG() in init_vqs
It's a bit overkill to use BUG when failing to add an entry to the
stats_vq in init_vqs. So remove it and just return the error to the
caller to bail out nicely.
Linus Torvalds [Tue, 21 Aug 2018 20:47:29 +0000 (13:47 -0700)]
Merge branch 'siginfo-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace
Pull core signal handling updates from Eric Biederman:
"It was observed that a periodic timer in combination with a
sufficiently expensive fork could prevent fork from every completing.
This contains the changes to remove the need for that restart.
This set of changes is split into several parts:
- The first part makes PIDTYPE_TGID a proper pid type instead
something only for very special cases. The part starts using
PIDTYPE_TGID enough so that in __send_signal where signals are
actually delivered we know if the signal is being sent to a a group
of processes or just a single process.
- With that prep work out of the way the logic in fork is modified so
that fork logically makes signals received while it is running
appear to be received after the fork completes"
* 'siginfo-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (22 commits)
signal: Don't send signals to tasks that don't exist
signal: Don't restart fork when signals come in.
fork: Have new threads join on-going signal group stops
fork: Skip setting TIF_SIGPENDING in ptrace_init_task
signal: Add calculate_sigpending()
fork: Unconditionally exit if a fatal signal is pending
fork: Move and describe why the code examines PIDNS_ADDING
signal: Push pid type down into complete_signal.
signal: Push pid type down into __send_signal
signal: Push pid type down into send_signal
signal: Pass pid type into do_send_sig_info
signal: Pass pid type into send_sigio_to_task & send_sigurg_to_task
signal: Pass pid type into group_send_sig_info
signal: Pass pid and pid type into send_sigqueue
posix-timers: Noralize good_sigevent
signal: Use PIDTYPE_TGID to clearly store where file signals will be sent
pid: Implement PIDTYPE_TGID
pids: Move the pgrp and session pid pointers from task_struct to signal_struct
kvm: Don't open code task_pid in kvm_vcpu_ioctl
pids: Compute task_tgid using signal->leader_pid
...
Linus Torvalds [Tue, 21 Aug 2018 20:18:22 +0000 (13:18 -0700)]
Merge tag 'microblaze-v4.19-rc1' of git://git.monstr.eu/linux-2.6-microblaze
Pull arch/microblaze updates from Michal Simek:
- use generic noncoherent direct mapping
- use LDFLAGS instead of LD
- pci error path fix
- remove incorrect comments
* tag 'microblaze-v4.19-rc1' of git://git.monstr.eu/linux-2.6-microblaze:
microblaze/PCI: Remove stale pcibios_align_resource() comment
microblaze: delete wrong comment about machine_early_init
microblaze: add endianness options to LDFLAGS instead of LD
microblaze: remove consistent_sync and consistent_sync_page
microblaze: use generic dma_noncoherent_ops
microblaze: warn if of_iomap() failed
David S. Miller [Tue, 21 Aug 2018 19:40:26 +0000 (12:40 -0700)]
Merge branch 'sparc32-NO_BOOTMEM'
Mike Rapoport says:
====================
sparc32: switch to NO_BOOTMEM
These patches convert sparc32 to use memblock + nobootmem.
I've made the conversion as simple as possible, just enough to allow moving
HAVE_MEMBLOCK and NO_BOOTMEM to the common SPARC configuration.
v2 changes:
* split whitespace changes to a separate patch
* address Sam's comments [1]
Mike Rapoport [Mon, 6 Aug 2018 10:52:35 +0000 (13:52 +0300)]
sparc32: split ramdisk detection and reservation to a helper function
The detection and reservation of ramdisk memory were separated to allow
bootmem bitmap initialization after the ramdisk boundaries are detected.
Since the bootmem initialization is removed, the reservation of ramdisk
memory is done immediately after its boundaries are found.
Split the entire block into a separate helper function.
Mike Rapoport [Mon, 6 Aug 2018 10:52:34 +0000 (13:52 +0300)]
sparc32: switch to NO_BOOTMEM
Each populated sparc_phys_bank is added to memblock.memory. The
reserve_bootmem() calls are replaced with memblock_reserve(), and the
bootmem bitmap initialization is droppped.
Switch to the generic noncoherent direct mapping implementation.
This removes the previous sync_single_for_device implementation, which
looks bogus given that no syncing is happening in the similar but more
important map_single case.
Leo (Sunpeng) Li [Thu, 16 Aug 2018 19:44:39 +0000 (15:44 -0400)]
drm/amd/display: Don't build DCN1 when kcov is enabled
DCN1 contains code that utilizes fp math. When
CONFIG_KCOV_INSTRUMENT_ALL and CONFIG_KCOV_ENABLE_COMPARISONS are
enabled, build errors are found. See this earlier patch for details:
As a short term solution, disable CONFIG_DRM_AMD_DC_DCN1_0 when
KCOV_INSTRUMENT_ALL and KCOV_ENABLE_COMPARISONS are enabled. In
addition, make it a fully derived config, taking into account
CONFIG_X86.
Yintian Tao [Mon, 20 Aug 2018 03:11:48 +0000 (11:11 +0800)]
drm/powerplay: enable dpm under pass-through
Repeat enable dpm under pass-through because there is no actually
hardware-fini and real power-off when guest vm shutdown or reboot.
Otherwise, under pass-through it will be failed to populate populate
and upload SCLK MCLK DPM levels due to zero of pcie_speed_table.count.
Samson Tam [Mon, 30 Jul 2018 16:22:35 +0000 (12:22 -0400)]
drm/amd/display: Do not retain link settings
Do not retrain link settings if lane count and link rate are both
unknown. Causes driver to be stuck reading VBIOS register after
removing emulated connection.
Paul Burton [Tue, 21 Aug 2018 19:12:59 +0000 (12:12 -0700)]
MIPS: lib: Provide MIPS64r6 __multi3() for GCC < 7
Some versions of GCC suboptimally generate calls to the __multi3()
intrinsic for MIPS64r6 builds, resulting in link failures due to the
missing function:
LD vmlinux.o
MODPOST vmlinux.o
kernel/bpf/verifier.o: In function `kmalloc_array':
include/linux/slab.h:631: undefined reference to `__multi3'
fs/select.o: In function `kmalloc_array':
include/linux/slab.h:631: undefined reference to `__multi3'
...
We already have a workaround for this in which we provide the
instrinsic, but we do so selectively for GCC 7 only. Unfortunately the
issue occurs with older GCC versions too - it has been observed with
both GCC 5.4.0 & GCC 6.4.0.
MIPSr6 support was introduced in GCC 5, so all major GCC versions prior
to GCC 8 are affected and we extend our workaround accordingly to all
MIPS64r6 builds using GCC versions older than GCC 8.
Trond Myklebust [Tue, 21 Aug 2018 14:25:34 +0000 (10:25 -0400)]
pNFS: Remove unwanted optimisation of layoutget
If we knew that the file was empty, we wouldn't be asking for a layout.
Any optimisation here is already done before calling pnfs_update_layout().
As it stands, we sometimes end up doing an unnecessary inband read to
the MDS even when holding a layout.
Some versions of GCC for the MIPS architecture suffer from a bug which
can lead to instructions from beyond an unreachable statement being
incorrectly reordered into earlier branch delay slots if the unreachable
statement is the only content of a case in a switch statement. This can
lead to seemingly random behaviour, such as invalid memory accesses from
incorrectly reordered loads or stores, and link failures on microMIPS
builds.
Runtime problems resulting from this bug were initially observed using a
maltasmvp_defconfig v4.4 kernel built using GCC 4.9.2 (from a Codescape
SDK 2015.06-05 toolchain), with the result being an address exception
taken after log messages about the L1 caches (during probe of the L2
cache):
This is early enough that the kernel exception vectors are not in use,
so any further output depends upon the bootloader. This is reproducible
in QEMU where no further output occurs - ie. the system hangs here.
Given the nature of the bug it may potentially be hit with differing
symptoms. The bug is known to affect GCC versions as recent as 7.3, and
it is unclear whether GCC 8 fixed it or just happens not to encounter
the bug in the testcase found at the link above due to differing
optimizations.
This bug can be worked around by placing a volatile asm statement, which
GCC is prevented from reordering past, prior to the
__builtin_unreachable call.
That was actually done already for other reasons by commit 173a3efd3edb
("bug.h: work around GCC PR82365 in BUG()"), but creates problems for
microMIPS builds due to the lack of a .insn directive. The microMIPS ISA
allows for interlinking with regular MIPS32 code by repurposing bit 0 of
the program counter as an ISA mode bit. To switch modes one changes the
value of this bit in the PC. However typical branch instructions encode
their offsets as multiples of 2-byte instruction halfwords, which means
they cannot change ISA mode - this must be done using either an indirect
branch (a jump-register in MIPS terminology) or a dedicated jalx
instruction. In order to ensure that regular branches don't attempt to
target code in a different ISA which they can't actually switch to, the
linker will check that branch targets are code in the same ISA as the
branch.
Unfortunately our empty asm volatile statements don't qualify as code,
and the link for microMIPS builds fails with errors such as:
arch/mips/mm/dma-default.s:3265: Error: branch to a symbol in another ISA mode
arch/mips/mm/dma-default.s:5027: Error: branch to a symbol in another ISA mode
Resolve this by adding a .insn directive within the asm statement which
declares that what comes next is code. This may or may not be true,
since we don't really know what comes next, but as this code is in an
unreachable path anyway that doesn't matter since we won't execute it.
We do this in asm/compiler.h & select CONFIG_HAVE_ARCH_COMPILER_H in
order to have this included by linux/compiler_types.h after
linux/compiler-gcc.h. This will result in asm/compiler.h being included
in all C compilations via the -include linux/compiler_types.h argument
in c_flags, which should be harmless.
Paul Burton [Mon, 20 Aug 2018 22:36:17 +0000 (15:36 -0700)]
compiler.h: Allow arch-specific asm/compiler.h
We have a need to override the definition of
barrier_before_unreachable() for MIPS, which means we either need to add
architecture-specific code into linux/compiler-gcc.h or we need to allow
the architecture to provide a header that can define the macro before
the generic definition. The latter seems like the better approach.
A straightforward approach to the per-arch header is to make use of
asm-generic to provide a default empty header & adjust architectures
which don't need anything specific to make use of that by adding the
header to generic-y. Unfortunately this doesn't work so well due to
commit 28128c61e08e ("kconfig.h: Include compiler types to avoid missed
struct attributes") which caused linux/compiler_types.h to be included
in the compilation of every C file via the -include linux/kconfig.h flag
in c_flags.
Because the -include flag is present for all C files we compile, we need
the architecture-provided header to be present before any C files are
compiled. If any C files can be compiled prior to the asm-generic header
wrappers being generated then we hit a build failure due to missing
header. Such cases do exist - one pointed out by the kbuild test robot
is the compilation of arch/ia64/kernel/nr-irqs.c, which occurs as part
of the archprepare target [1].
This leaves us with a few options:
1) Use generic-y & fix any build failures we find by enforcing
ordering such that the asm-generic target occurs before any C
compilation, such that linux/compiler_types.h can always include
the generated asm-generic wrapper which in turn includes the empty
asm-generic header. This would rely on us finding all the
problematic cases - I don't know for sure that the ia64 issue is
the only one.
2) Add an actual empty header to each architecture, so that we don't
need the generated asm-generic wrapper. This seems messy.
3) Give up & add #ifdef CONFIG_MIPS or similar to
linux/compiler_types.h. This seems messy too.
4) Include the arch header only when it's actually needed, removing
the need for the asm-generic wrapper for all other architectures.
This patch allows us to use approach 4, by including an asm/compiler.h
header from linux/compiler_types.h after the inclusion of the
compiler-specific linux/compiler-*.h header(s). We do this
conditionally, only when CONFIG_HAVE_ARCH_COMPILER_H is selected, in
order to avoid the need for asm-generic wrappers & the associated build
ordering issue described above. The asm/compiler.h header is included
after the generic linux/compiler-*.h header(s) for consistency with the
way linux/compiler-intel.h & linux/compiler-clang.h are included after
the linux/compiler-gcc.h header that they override.
Colin Ian King [Tue, 21 Aug 2018 09:12:48 +0000 (10:12 +0100)]
block/DAC960.c: make some arrays static const, shrinks object size
Don't populate the arrays ReadCacheStatus, WriteCacheStatus and
SenseErrors on the stack but instead make them static const. Makes
the object code smaller by 47 bytes:
Before:
text data bss dec hex filename
160974 34628 832 196434 2ff52 drivers/block/DAC960.o
After:
text data bss dec hex filename
160671 34884 832 196387 2ff23 drivers/block/DAC960.o
Jianchao Wang [Tue, 21 Aug 2018 07:15:04 +0000 (15:15 +0800)]
blk-mq: sync the update nr_hw_queues with blk_mq_queue_tag_busy_iter
For blk-mq, part_in_flight/rw will invoke blk_mq_in_flight/rw to
account the inflight requests. It will access the queue_hw_ctx and
nr_hw_queues w/o any protection. When updating nr_hw_queues and
blk_mq_in_flight/rw occur concurrently, panic comes up.
Before update nr_hw_queues, the q will be frozen. So we could use
q_usage_counter to avoid the race. percpu_ref_is_zero is used here
so that we will not miss any in-flight request. The access to
nr_hw_queues and queue_hw_ctx in blk_mq_queue_tag_busy_iter are
under rcu critical section, __blk_mq_update_nr_hw_queues could use
synchronize_rcu to ensure the zeroed q_usage_counter to be globally
visible.
Jianchao Wang [Tue, 21 Aug 2018 07:15:03 +0000 (15:15 +0800)]
blk-mq: init hctx sched after update ctx and hctx mapping
Currently, when update nr_hw_queues, IO scheduler's init_hctx will
be invoked before the mapping between ctx and hctx is adapted
correctly by blk_mq_map_swqueue. The IO scheduler init_hctx (kyber)
may depend on this mapping and get wrong result and panic finally.
A simply way to fix this is that switch the IO scheduler to 'none'
before update the nr_hw_queues, and then switch it back after
update nr_hw_queues. blk_mq_sched_init_/exit_hctx are removed due
to nobody use them any more.
Heiko Carstens [Tue, 21 Aug 2018 11:37:57 +0000 (13:37 +0200)]
s390: remove gcc version check (4.3 or newer)
git commit cafa0010cd51 ("Raise the minimum required gcc version to 4.6")
raised the minimum gcc version to 4.6. Therefore remove the s390 specific
gcc 4.3 version check, which wasn't sufficient anyway.
s390/zcrypt: hex string mask improvements for apmask and aqmask.
The sysfs attributes /sys/bus/ap/apmask and /sys/bus/ap/aqmask
and the kernel command line arguments ap.apm and ap.aqm get
an improvement of the value parsing with this patch:
The mask values are bitmaps in big endian order starting with bit 0.
So adapter number 0 is the leftmost bit, mask is 0x8000... The sysfs
attributes and the kernel command line accept 2 different formats:
- Absolute hex string starting with 0x like "0x12345678" does set
the mask starting from left to right. If the given string is shorter
than the mask it is padded with 0s on the right. If the string is
longer than the mask an error comes back (EINVAL).
- Relative format - a concatenation (done with ',') of the terms
+<bitnr>[-<bitnr>] or -<bitnr>[-<bitnr>]. <bitnr> may be any
valid number (hex, decimal or octal) in the range 0...255.
Here are some examples:
"+0-15,+32,-128,-0xFF"
"-0-255,+1-16,+0x128"
Masami Hiramatsu [Tue, 21 Aug 2018 13:04:57 +0000 (22:04 +0900)]
tracing/kprobes: Fix to check notrace function with correct range
Fix within_notrace_func() to check notrace function correctly.
Since the ftrace_location_range(start, end) function checks
the range inclusively (start <= ftrace-loc <= end), the end
address must not include the entry address of next function.
However, within_notrace_func() uses kallsyms_lookup_size_offset()
to get the function size and calculate the end address from
adding the size to the entry address. This means the end address
is the entry address of the next function.
In the result, within_notrace_func() fails to find notrace
function if the next function of the target function is
ftraced.
Let's subtract 1 from the end address so that ftrace_location_range()
can check it correctly.
Masami Hiramatsu [Tue, 21 Aug 2018 07:27:58 +0000 (16:27 +0900)]
tracing: Allow gcov profiling on only ftrace subsystem
Add GCOV_PROFILE_FTRACE to allow gcov profiling on only files in ftrace
subsystem. This config option will be used for checking kselftest/ftrace
coverage.
Helge Deller [Tue, 21 Aug 2018 12:31:32 +0000 (14:31 +0200)]
parisc: Fix boot failure of 64-bit kernel
Commit c8921d72e390 ("parisc: Fix and improve kernel stack unwinding")
broke booting of 64-bit kernels. On 64-bit kernels function pointers are
actually function descriptors which require dereferencing. In this patch
we instead declare functions in assembly code which are referenced from
C-code as external data pointers with the ENTRY() macro and thus can use
a simple external reference to the functions.
Chao Yu [Tue, 14 Aug 2018 14:37:25 +0000 (22:37 +0800)]
f2fs: readahead encrypted block during GC
During GC, for each encrypted block, we will read block synchronously
into meta page, and then submit it into current cold data log area.
So this block read model with 4k granularity can make poor performance,
like migrating non-encrypted block, let's readahead encrypted block
as well to improve migration performance.
To implement this, we choose meta page that its index is old block
address of the encrypted block, and readahead ciphertext into this
page, later, if readaheaded page is still updated, we will load its
data into target meta page, and submit the write IO.
Note that for OPU, truncation, deletion, we need to invalid meta
page after we invalid old block address, to make sure we won't load
invalid data from target meta page during encrypted block migration.
for ((i = 0; i < 1000; i++))
do {
xfs_io -f /mnt/f2fs/dir/$i -c "pwrite 0 128k" -c "fsync";
} done
for ((i = 0; i < 1000; i+=2))
do {
rm /mnt/f2fs/dir/$i;
} done
Before -
read throughput: 185 MB/s
total read requests: 85177 (of these ~80000 are 4KB size requests).
total write requests: 2546 (of these ~2208 requests are written in 512KB).
After -
read throughput: 758 MB/s
total read requests: 2417 (of these ~2042 are 512KB reads).
total write requests: 2701 (of these ~2034 requests are written in 512KB).
powerpc/topology: Get topology for shared processors at boot
On a shared LPAR, Phyp will not update the CPU associativity at boot
time. Just after the boot system does recognize itself as a shared
LPAR and trigger a request for correct CPU associativity. But by then
the scheduler would have already created/destroyed its sched domains.
This causes
- Broken load balance across Nodes causing islands of cores.
- Performance degradation esp if the system is lightly loaded
- dmesg to wrongly report all CPUs to be in Node 0.
- Messages in dmesg saying borken topology.
- With commit 051f3ca02e46 ("sched/topology: Introduce NUMA identity
node sched domain"), can cause rcu stalls at boot up.
The sched_domains_numa_masks table which is used to generate cpumasks
is only created at boot time just before creating sched domains and
never updated. Hence, its better to get the topology correct before
the sched domains are created.
For example on 64 core Power 8 shared LPAR, dmesg reports
Brought up 512 CPUs
Node 0 CPUs: 0-511
Node 1 CPUs:
Node 2 CPUs:
Node 3 CPUs:
Node 4 CPUs:
Node 5 CPUs:
Node 6 CPUs:
Node 7 CPUs:
Node 8 CPUs:
Node 9 CPUs:
Node 10 CPUs:
Node 11 CPUs:
...
BUG: arch topology borken
the DIE domain not a subset of the NUMA domain
BUG: arch topology borken
the DIE domain not a subset of the NUMA domain
numactl/lscpu output will still be correct with cores spreading across
all nodes:
Socket(s): 64
NUMA node(s): 12
Model: 2.0 (pvr 004d 0200)
Model name: POWER8 (architected), altivec supported
Hypervisor vendor: pHyp
Virtualization type: para
L1d cache: 64K
L1i cache: 32K
NUMA node0 CPU(s): 0-7,32-39,64-71,96-103,176-183,272-279,368-375,464-471
NUMA node1 CPU(s): 8-15,40-47,72-79,104-111,184-191,280-287,376-383,472-479
NUMA node2 CPU(s): 16-23,48-55,80-87,112-119,192-199,288-295,384-391,480-487
NUMA node3 CPU(s): 24-31,56-63,88-95,120-127,200-207,296-303,392-399,488-495
NUMA node4 CPU(s): 208-215,304-311,400-407,496-503
NUMA node5 CPU(s): 168-175,264-271,360-367,456-463
NUMA node6 CPU(s): 128-135,224-231,320-327,416-423
NUMA node7 CPU(s): 136-143,232-239,328-335,424-431
NUMA node8 CPU(s): 216-223,312-319,408-415,504-511
NUMA node9 CPU(s): 144-151,240-247,336-343,432-439
NUMA node10 CPU(s): 152-159,248-255,344-351,440-447
NUMA node11 CPU(s): 160-167,256-263,352-359,448-455
Currently on this LPAR, the scheduler detects 2 levels of Numa and
created numa sched domains for all CPUs, but it finds a single DIE
domain consisting of all CPUs. Hence it deletes all numa sched
domains.
To address this, detect the shared processor and update topology soon
after CPUs are setup so that correct topology is updated just before
scheduler creates sched domain.
Linus Torvalds [Tue, 21 Aug 2018 02:37:09 +0000 (19:37 -0700)]
Merge tag 'please-pull-noboot' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux
Pull ia64 NO_BOOTMEM conversion from Tony Luck:
"Mike Rapoport kindly fixed up ia64 to work with NO_BOOTMEM"
* tag 'please-pull-noboot' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux:
ia64: switch to NO_BOOTMEM
ia64: use mem_data to detect nodes' minimal and maximal PFNs
ia64: remove unused num_dma_physpages member from 'struct early_node_data'
ia64: contig/paging_init: reduce code duplication
Linus Torvalds [Tue, 21 Aug 2018 02:02:17 +0000 (19:02 -0700)]
Merge tag 'linux-kselftest-4.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest
Pull Kselftest update from Shuah Khan:
- add cgroup core selftests
- fix compile warnings in android ion test
- fix to bugs in exclude and skip paths in vDSO test
- remove obsolete config options
- add missing .gitignore file
* tag 'linux-kselftest-4.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
selftests/ftrace: Fix kprobe string testcase to not probe notrace function
selftests: mount: remove no longer needed config option
selftests: cgroup: add gitignore file
Add cgroup core selftests
selftests: vDSO - fix to return KSFT_SKIP when test couldn't be run
selftests: vDSO - fix to exclude x86 test on non-x86 platforms
selftests/android: initialize heap_type to avoid compiling warning
Linus Torvalds [Tue, 21 Aug 2018 01:32:00 +0000 (18:32 -0700)]
Merge tag 'trace-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
- Restructure of lockdep and latency tracers
This is the biggest change. Joel Fernandes restructured the hooks
from irqs and preemption disabling and enabling. He got rid of a lot
of the preprocessor #ifdef mess that they caused.
He turned both lockdep and the latency tracers to use trace events
inserted in the preempt/irqs disabling paths. But unfortunately,
these started to cause issues in corner cases. Thus, parts of the
code was reverted back to where lockdep and the latency tracers just
get called directly (without using the trace events). But because the
original change cleaned up the code very nicely we kept that, as well
as the trace events for preempt and irqs disabling, but they are
limited to not being called in NMIs.
- Have trace events use SRCU for "rcu idle" calls. This was required
for the preempt/irqs off trace events. But it also had to not allow
them to be called in NMI context. Waiting till Paul makes an NMI safe
SRCU API.
- New notrace SRCU API to allow trace events to use SRCU.
- Addition of mcount-nop option support
- SPDX headers replacing GPL templates.
- Various other fixes and clean ups.
- Some fixes are marked for stable, but were not fully tested before
the merge window opened.
* tag 'trace-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (44 commits)
tracing: Fix SPDX format headers to use C++ style comments
tracing: Add SPDX License format tags to tracing files
tracing: Add SPDX License format to bpf_trace.c
blktrace: Add SPDX License format header
s390/ftrace: Add -mfentry and -mnop-mcount support
tracing: Add -mcount-nop option support
tracing: Avoid calling cc-option -mrecord-mcount for every Makefile
tracing: Handle CC_FLAGS_FTRACE more accurately
Uprobe: Additional argument arch_uprobe to uprobe_write_opcode()
Uprobes: Simplify uprobe_register() body
tracepoints: Free early tracepoints after RCU is initialized
uprobes: Use synchronize_rcu() not synchronize_sched()
tracing: Fix synchronizing to event changes with tracepoint_synchronize_unregister()
ftrace: Remove unused pointer ftrace_swapper_pid
tracing: More reverting of "tracing: Centralize preemptirq tracepoints and unify their usage"
tracing/irqsoff: Handle preempt_count for different configs
tracing: Partial revert of "tracing: Centralize preemptirq tracepoints and unify their usage"
tracing: irqsoff: Account for additional preempt_disable
trace: Use rcu_dereference_raw for hooks from trace-event subsystem
tracing/kprobes: Fix within_notrace_func() to check only notrace functions
...
Linus Torvalds [Tue, 21 Aug 2018 01:26:55 +0000 (18:26 -0700)]
Merge tag 'ceph-for-4.19-rc1' of git://github.com/ceph/ceph-client
Pull ceph updates from Ilya Dryomov:
"The main things are support for cephx v2 authentication protocol and
basic support for rbd images within namespaces (myself).
Also included are y2038 conversion patches from Arnd, a pile of
miscellaneous fixes from Chengguang and Zheng's feature bit
infrastructure for the filesystem"
* tag 'ceph-for-4.19-rc1' of git://github.com/ceph/ceph-client: (40 commits)
ceph: don't drop message if it contains more data than expected
ceph: support cephfs' own feature bits
crush: fix using plain integer as NULL warning
libceph: remove unnecessary non NULL check for request_key
ceph: refactor error handling code in ceph_reserve_caps()
ceph: refactor ceph_unreserve_caps()
ceph: change to void return type for __do_request()
ceph: compare fsc->max_file_size and inode->i_size for max file size limit
ceph: add additional size check in ceph_setattr()
ceph: add additional offset check in ceph_write_iter()
ceph: add additional range check in ceph_fallocate()
ceph: add new field max_file_size in ceph_fs_client
libceph: weaken sizeof check in ceph_x_verify_authorizer_reply()
libceph: check authorizer reply/challenge length before reading
libceph: implement CEPHX_V2 calculation mode
libceph: add authorizer challenge
libceph: factor out encrypt_authorizer()
libceph: factor out __ceph_x_decrypt()
libceph: factor out __prepare_write_connect()
libceph: store ceph_auth_handshake pointer in ceph_connection
...
Linus Torvalds [Mon, 20 Aug 2018 23:30:27 +0000 (16:30 -0700)]
Merge tag 'rtc-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux
Pull RTC updates from Alexandre Belloni:
"It is now possible to add custom sysfs attributes while avoiding a
possible race condition. Unused code has been removed resulting in a
nice reduction of the code base. And more drivers have been switched
to SPDX by their maintainers.
Summary:
Subsystem:
- new helpers to add custom sysfs attributes
- struct rtc_task removal along with rtc_irq_[un]register()
- rtc_irq_set_state and rtc_irq_set_freq are not exported anymore
Drivers:
- armada38x: reset after rtc power loss
- ds1307: now supports m41t11
- isl1208: now supports isl1219 and tamper detection
- pcf2127: internal SRAM support"
* tag 'rtc-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux: (34 commits)
rtc: ds1307: simplify hwmon config
rtc: s5m: Add SPDX license identifier
rtc: maxim: Add SPDX license identifiers
rtc: isl1219: add device tree documentation
rtc: isl1208: set ev-evienb bit from device tree
rtc: isl1208: Add "evdet" interrupt source for isl1219
rtc: isl1208: add support for isl1219 with tamper detection
rtc: sysfs: facilitate attribute add to rtc device
rtc: remove struct rtc_task
char: rtc: remove task handling
rtc: pcf85063: preserve control register value between stop and start
rtc: sh: remove unused variable rtc_dev
rtc: unexport rtc_irq_set_*
rtc: simplify rtc_irq_set_state/rtc_irq_set_freq
rtc: remove irq_task and irq_task_lock
rtc: remove rtc_irq_register/rtc_irq_unregister
rtc: sh: remove dead code
rtc: sa1100: don't set PIE frequency
rtc: ds1307: support m41t11 variant
rtc: ds1307: fix data pointer to m41t0
...
Linus Torvalds [Mon, 20 Aug 2018 22:59:01 +0000 (15:59 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Pull HID updates from Jiri Kosina:
- touch_max detection improvements and quirk handling fixes in wacom
driver from Jason Gerecke and Ping Cheng
- Palm rejection from Dmitry Torokhov and _dial support from Benjamin
Tissoires for hid-multitouch driver
- Low voltage support for i2c-hid driver from Stephen Boyd
- Guitar-Hero support from Nicolas Adenis-Lamarre
- other assorted small fixes and device ID additions
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: (40 commits)
HID: intel_ish-hid: tx_buf memory leak on probe/remove
HID: intel-ish-hid: Prevent loading of driver on Mehlow
HID: cougar: Add support for the Cougar 500k Gaming Keyboard
HID: cougar: make compare_device_paths reusable
HID: intel-ish-hid: remove redundant variable num_frags
HID: multitouch: handle palm for touchscreens
HID: multitouch: touchscreens also use confidence reports
HID: multitouch: report MT_TOOL_PALM for non-confident touches
HID: microsoft: support the Surface Dial
HID: core: do not upper bound the collection stack
HID: input: enable Totem on the Dell Canvas 27
HID: multitouch: remove one copy of values
HID: multitouch: ditch mt_report_id
HID: multitouch: store a per application quirks value
HID: multitouch: Store per collection multitouch data
HID: multitouch: make sure the static list of class is not changed
input: add MT_TOOL_DIAL
HID: elan: Add support for touchpad on the Toshiba Click Mini L9W
HID: elan: Add USB-id for HP x2 10-n000nd touchpad
HID: elan: Add a flag for selecting if the touchpad has a LED
...
Linus Torvalds [Mon, 20 Aug 2018 22:41:37 +0000 (15:41 -0700)]
Merge tag 'backlight-next-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/backlight
Pull backlight updates from Lee Jones:
"Core Framework:
- Remove unused/obsolete code/comments
New Functionality:
- Allow less granular brightness specification for high-res PWMs; pwm_bl
- Align brightness {inc,dec}rements with that perceived by the human-eye; pwm_bl
Fix-ups:
- Prepare for the introduction of -Wimplicit-fall-through; adp8860_bl
Bug Fixes:
- Fix uninitialised variable; pwm_bl"
* tag 'backlight-next-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/backlight:
backlight: pwm_bl: Fix uninitialized variable
backlight: adp8860: Mark expected switch fall-through
backlight: Remove obsolete comment for ->state
dt-bindings: pwm-backlight: Move brightness-levels to optional
backlight: pwm_bl: Compute brightness of LED linearly to human eye
dt-bindings: pwm-backlight: Add a num-interpolation-steps property
backlight: pwm_bl: Linear interpolation between brightness-levels
Linus Torvalds [Mon, 20 Aug 2018 22:38:44 +0000 (15:38 -0700)]
Merge tag 'mfd-next-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd
Pull MFD updates from Lee Jones:
"New Drivers:
- Add Cirrus Logic Madera Codec (CS47L35, CS47L85 and CS47L90/91) driver
- Add ChromeOS EC CEC driver
- Add ROHM BD71837 PMIC driver
New Device Support:
- Add support for Dialog Semi DA9063L PMIC variant to DA9063
- Add support for Intel Ice Lake to Intel-PLSS-PCI
- Add support for X-Powers AXP806 to AXP20x
New Functionality:
- Add support for USB Charging to the ChromeOS Embedded Controller
- Add support for HDMI CEC to the ChromeOS Embedded Controller
- Add support for HDMI CEC to Intel HDMI
- Add support for accessory detection to Madera devices
- Allow individual pins to be configured via DT' wlf,csnaddr-pd
- Provide legacy platform specific EEPROM/Watchdog commands; rave-sp
Fix-upsL
- Trivial renaming/spelling fixes; cros_ec, da9063-*
- Convert to Managed Resources (devm_*); da9063-*, ti_am335x_tscadc
- Transition to helper macros/functions; da9063-*
- Constify; kempld-core
- Improve error path/messages; wm8994-core
- Disable IRQs locally instead of relying on USB subsystem; dln2
- Remove unused code; rave-sp
- New exports; sec-core
Bug Fixes:
- Fix possible false I2C transaction error; arizona-core
- Fix declared memory area size; hi655x-pmic
- Fix checksum type; rave-sp
- Fix incorrect default serial port configuration: rave-sp
- Fix incorrect coherent DMA mask for sub-devices; sm501"
* tag 'mfd-next-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd: (60 commits)
mfd: madera: Add register definitions for accessory detect
mfd: sm501: Set coherent_dma_mask when creating subdevices
mfd: bd71837: Devicetree bindings for ROHM BD71837 PMIC
mfd: bd71837: Core driver for ROHM BD71837 PMIC
media: platform: cros-ec-cec: Fix dependency on MFD_CROS_EC
mfd: sec-core: Export OF module alias table
mfd: as3722: Disable auto-power-on when AC OK
mfd: axp20x: Support AXP806 in I2C mode
mfd: axp20x: Add self-working mode support for AXP806
dt-bindings: mfd: axp20x: Add "self-working" mode for AXP806
mfd: wm8994: Allow to configure CS/ADDR Pulldown from dts
mfd: wm8994: Allow to configure Speaker Mode Pullup from dts
mfd: rave-sp: Emulate CMD_GET_STATUS on device that don't support it
mfd: rave-sp: Add legacy watchdog ping command translation
mfd: rave-sp: Add legacy EEPROM access command translation
mfd: rave-sp: Initialize flow control and parity of the port
mfd: rave-sp: Fix incorrectly specified checksum type
mfd: rave-sp: Remove unused defines
mfd: hi655x: Fix regmap area declared size for hi655x
mfd: ti_am335x_tscadc: Fix struct clk memory leak
...
Joe Perches [Mon, 20 Aug 2018 20:15:26 +0000 (13:15 -0700)]
Raise the minimum required gcc version to 4.6
Various architectures fail to build properly with older versions of the
gcc compiler.
An example from Guenter Roeck in thread [1]:
>
> In file included from ./include/linux/mm.h:17:0,
> from ./include/linux/pid_namespace.h:7,
> from ./include/linux/ptrace.h:10,
> from arch/openrisc/kernel/asm-offsets.c:32:
> ./include/linux/mm_types.h:497:16: error: flexible array member in otherwise empty struct
>
> This is just an example with gcc 4.5.1 for or32. I have seen the problem
> with gcc 4.4 (for unicore32) as well.
So update the minimum required version of gcc to 4.6.
Max Filippov [Mon, 20 Aug 2018 20:24:24 +0000 (13:24 -0700)]
Merge branch 'xtensa-dma-fixes' (early part) into xtensa-fixes
This switches xtensa arch to the generic noncoherent direct mapping
operations, adds support for DMA_ATTR_NO_KERNEL_MAPPING attribute and
allows for platform-specific handling of coherent memory.
Max Filippov [Tue, 14 Aug 2018 01:14:14 +0000 (18:14 -0700)]
xtensa: drop unneeded platform/hardware.h headers
platform/hardware.h no longer supply any information for processor.h,
vectors.h, setup.c or vmlinux.lds.S, don't include it.
This header is now empty in the platforms/iss, so remove it altogether.
Drop PLATFORM_DEFAULT_MEM_START from the platform/hardware.h headers.
Provide definition of CONFIG_DEFAULT_MEM_START always, allow changing it
only in noMMU configurations when PLATFORM_WANT_DEFAULT_MEM is selected.
Change prompt and description so that it's clear that it controls
PAGE_OFFSET and PHYS_OFFSET.
Max Filippov [Mon, 13 Aug 2018 01:32:13 +0000 (18:32 -0700)]
xtensa: drop unused {CONFIG,PLATFORM}_DEFAULT_MEM_SIZE
Now that noMMU cache attributes are set up separately drop no longer
used macro PLATFORM_DEFAULT_MEM_SIZE and config symbol
CONFIG_DEFAULT_MEM_SIZE used for setting it.
Tony Luck [Mon, 20 Aug 2018 16:31:04 +0000 (09:31 -0700)]
ia64: Fix kernel BUG at lib/ioremap.c:72!
Commit 0bbf47eab469 ("ia64: use asm-generic/io.h") results in a BUG
while booting ia64. This is because asm-generic/io.h defines
PCI_IOBASE, which results in the function acpi_pci_root_remap_iospace()
doing a lot of unnecessary (and wrong) things.
I'd suggested an #if !CONFIG_IA64 in the functon, but Arnd suggested
keeping the fix inside the arch/ia64 tree.
Juergen Gross [Mon, 20 Aug 2018 15:24:20 +0000 (17:24 +0200)]
x86/xen: enable early use of set_fixmap in 32-bit Xen PV guest
Commit 7b25b9cb0dad83 ("x86/xen/time: Initialize pv xen time in
init_hypervisor_platform()") moved the mapping of the shared info area
before pagetable_init(). This breaks booting as 32-bit PV guest as the
use of set_fixmap isn't possible at this time on 32-bit.
This can be worked around by populating the needed PMD on 32-bit
kernel earlier.
In order not to reimplement populate_extra_pte() using extend_brk()
for allocating new page tables extend alloc_low_pages() to do that in
case the early page table pool is not yet available.
Fixes: 7b25b9cb0dad83 ("x86/xen/time: Initialize pv xen time in init_hypervisor_platform()") Signed-off-by: Juergen Gross <[email protected]> Reviewed-by: Thomas Gleixner <[email protected]> Signed-off-by: Boris Ostrovsky <[email protected]>
Jan Beulich [Mon, 25 Jun 2018 10:17:35 +0000 (04:17 -0600)]
xen/ACPI: don't upload Px/Cx data for disabled processors
This is unnecessary and triggers a warning in the hypervisor.
Often systems have more processor entries in their ACPI tables than are
actually installed/active. The ACPI_STA_DEVICE_PRESENT bit cannot be
reliably used, but the ACPI_MADT_ENABLED bit can. In order to not
introduce new functions in the main ACPI processor driver code, simply
use acpi_get_phys_id(), which does more than we need, but which checks
the MADT enabled bit in the process. Any CPU for which we can't
determine the APIC ID is unlikely to work properly anyway, so the extra
checks done by acpi_get_phys_id() should do no harm.
Jan Beulich [Mon, 25 Jun 2018 10:45:49 +0000 (04:45 -0600)]
x86/Xen: further refine add_preferred_console() invocations
As the sequence of invocations matters, add "tty" only after "hvc" when
a VGA console is available (which is often the case for Dom0, but hardly
ever for DomU).
Jan Beulich [Mon, 25 Jun 2018 10:34:01 +0000 (04:34 -0600)]
x86/Xen: mark xen_setup_gdt() __init
Its only caller is __init, so to avoid section mismatch warnings when a
compiler decides to not inline the function marke this function so as
well. Take the opportunity and also make the function actually use its
argument: The sole caller passes in zero anyway.
Palmer Dabbelt [Mon, 20 Aug 2018 18:32:10 +0000 (11:32 -0700)]
RISC-V: Fix sys_riscv_flush_icache
This contains a pair of patches that together fix sys_riscv_flush_icache
on all systems:
* The first enables sys_riscv_flush_icache() for non-SMP systems.
* The second fixes a bug in our syscall header that caused
sys_riscv_flush_icache to never get generated.
riscv does not enable CONFIG_COMPAT in default configurations:
defconfig, allmodconfig and allnoconfig.
Remove the asm/compat.h as it does not seem to add any value to
the architecture without CONFIG_COMPAT.
Now that time compat syscalls are being reused in non CONFIG_COMPAT
modes, asm-generic/compat.h provides definitions for riscv 32 bit
mode.
Palmer Dabbelt [Fri, 3 Aug 2018 19:27:19 +0000 (12:27 -0700)]
RISC-V: Don't use a global include guard for uapi/asm/syscalls.h
This file is expected to be included multiple times in the same file in
order to allow the __SYSCALL macro to generate system call tables. With
a global include guard we end up missing __NR_riscv_flush_icache in the
syscall table, which results in icache flushes that escape the vDSO call
to not actually do anything.
The fix is to move to per-#define include guards, which allows the
system call tables to actually be populated. Thanks to Macrus Comstedt
for finding and fixing the bug!
Palmer Dabbelt [Thu, 9 Aug 2018 19:53:57 +0000 (12:53 -0700)]
RISC-V: Define sys_riscv_flush_icache when SMP=n
This would be necessary to make non-SMP builds work, but there is
another error in the implementation of our syscall linkage that actually
just causes sys_riscv_flush_icache to never build. I've build tested
this on allnoconfig and allnoconfig+SMP=y, as well as defconfig like
normal.
If the BIOS is not supplying NUMA information:
- set the default table count to 1 for all possible nodes
- select node 0 (instead of current NUMA) node to get consistent
performance
- generate an error indicating that the BIOS should be upgraded
Dan Williams [Sat, 14 Jul 2018 04:50:37 +0000 (21:50 -0700)]
libnvdimm, pmem: Restore page attributes when clearing errors
Use clear_mce_nospec() to restore WB mode for the kernel linear mapping
of a pmem page that was marked 'HWPoison'. A page with 'HWPoison' set
has also been marked UC in PAT (page attribute table) via
set_mce_nospec() to prevent speculative retrievals of poison.
The 'HWPoison' flag is only cleared when overwriting an entire page.
Currently memory_failure() returns zero if the error was handled. On
that result mce_unmap_kpfn() is called to zap the page out of the kernel
linear mapping to prevent speculative fetches of potentially poisoned
memory. However, in the case of dax mapped devmap pages the page may be
in active permanent use by the device driver, so it cannot be unmapped
from the kernel.
Instead of marking the page not present, marking the page UC should
be sufficient for preventing poison from being pre-fetched into the
cache. Convert mce_unmap_pfn() to set_mce_nospec() remapping the page as
UC, to hide it from speculative accesses.
Given that that persistent memory errors can be cleared by the driver,
include a facility to restore the page to cacheable operation,
clear_mce_nospec().
Dan Williams [Sat, 14 Jul 2018 04:50:27 +0000 (21:50 -0700)]
x86/mm/pat: Prepare {reserve, free}_memtype() for "decoy" addresses
In preparation for using set_memory_uc() instead set_memory_np() for
isolating poison from speculation, teach the memtype code to sanitize
physical addresses vs __PHYSICAL_MASK.
The motivation for using set_memory_uc() for this case is to allow
ongoing access to persistent memory pages via the pmem-driver +
memcpy_mcsafe() until the poison is repaired.
Vlastimil Babka [Mon, 20 Aug 2018 09:58:35 +0000 (11:58 +0200)]
x86/speculation/l1tf: Fix overflow in l1tf_pfn_limit() on 32bit
On 32bit PAE kernels on 64bit hardware with enough physical bits,
l1tf_pfn_limit() will overflow unsigned long. This in turn affects
max_swapfile_size() and can lead to swapon returning -EINVAL. This has been
observed in a 32bit guest with 42 bits physical address size, where
max_swapfile_size() overflows exactly to 1 << 32, thus zero, and produces
the following warning to dmesg:
[ 6.396845] Truncating oversized swap area, only using 0k out of 2047996k
projects / linux.git / log