Arnd Bergmann [Thu, 12 Nov 2015 14:37:12 +0000 (15:37 +0100)]
arm64: build vdso without libgcov
On a cross-toolchain without glibc support, libgcov may not be
available, and attempting to build an arm64 kernel with GCOV
enabled then results in a build error:
We don't really want to link libgcov into the vdso anyway, so
this patch just disables GCOV in the vdso directory, just as
we do for most other architectures.
Arnd Bergmann [Thu, 12 Nov 2015 14:20:16 +0000 (15:20 +0100)]
arm64: mark cpus_have_hwcap as __maybe_unused
cpus_have_hwcap() is defined as a 'static' function an only used in
one place that is inside of an #ifdef, so we get a warning when
the only user is disabled:
arch/arm64/kernel/cpufeature.c:699:13: warning: 'cpus_have_hwcap' defined but not used [-Wunused-function]
This marks the function as __maybe_unused, so the compiler knows that
it can drop the function definition without warning about it.
Signed-off-by: Arnd Bergmann <[email protected]> Fixes: 37b01d53ceef ("arm64/HWCAP: Use system wide safe values") Signed-off-by: Catalin Marinas <[email protected]>
Takashi Iwai [Thu, 12 Nov 2015 10:52:13 +0000 (11:52 +0100)]
ALSA: hda - Simplify phantom jack handling for HDMI/DP
The HDMI codec parser may create a phantom jack, but the helper
function snd_hda_jack_add_kctl() treats always as a normal jack. This
is superfluous as the jack query is executed at each time the jack
sync is performed.
Since the HDMI codec parser is the only caller of this function, it's
easier to change back this directly calling the original
__snd_hda_jack_add_kctl() with phantom_jack parameter.
While the 5KE processors have never been taped out, they exists though
a CP0.PRId and experimental RTLs or QEMU implementations. Add a case
entry in the idle code, as they can use the standard idle loop like the
5K processors.
Aaro Koskinen [Fri, 11 Sep 2015 14:46:14 +0000 (17:46 +0300)]
MIPS: vmlinux: create a section for appended DTB
For bootloaders that support booting only ELF kernels and load only ELF
segments to memory there is no easy way to supply DTB without kernel
recompilation. For that purpose, create a section called .appended_dtb
that can be later updated with board-specific DTB using binutils e.g. at
kernel installation time.
While mips can't use the generic compat_siginfo_t directly because
its si_code and si_errno are inverted, we can still make it as
close to the generic version as possible. This makes it easier
to update when new members are added to siginfo_t.
The main changes are adding a missing _sigsys union member and
eliminating the unused _irix_sigchld one.
Dan Williams [Sat, 10 Oct 2015 00:56:28 +0000 (20:56 -0400)]
MIPS: Fix PAGE_MASK definition
Make PAGE_MASK an unsigned long, like it is on x86, to avoid:
In file included from arch/mips/kernel/asm-offsets.c:14:0:
include/linux/mm.h: In function '__pfn_to_pfn_t':
include/linux/mm.h:1050:2: warning: left shift count >= width of type
pfn_t pfn_t = { .val = pfn | (flags & PFN_FLAGS_MASK), };
David Howells [Thu, 12 Nov 2015 09:36:40 +0000 (09:36 +0000)]
X.509: Fix the time validation [ver #2]
This fixes CVE-2015-5327. It affects kernels from 4.3-rc1 onwards.
Fix the X.509 time validation to use month number-1 when looking up the
number of days in that month. Also put the month number validation before
doing the lookup so as not to risk overrunning the array.
Paolo Bonzini [Thu, 12 Nov 2015 09:06:29 +0000 (10:06 +0100)]
Merge branch 'kvm-ppc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc into HEAD
"Paolo,
I have two fixes for HV KVM which I would like to have included in
v4.4-rc1. The first one is a fix for a bug identified by Red Hat
which causes occasional guest crashes. The second one fixes a bug
which causes host stalls and timeouts under certain circumstances when
the host is configured for static 2-way micro-threading mode."
Huaitong Han [Fri, 6 Nov 2015 09:00:23 +0000 (17:00 +0800)]
x86/fpu: Fix get_xsave_addr() behavior under virtualization
KVM uses the get_xsave_addr() function in a different fashion from
the native kernel, in that the 'xsave' parameter belongs to guest vcpu,
not the currently running task.
But 'xsave' is replaced with current task's (host) xsave structure, so
get_xsave_addr() will incorrectly return the bad xsave address to KVM.
Fix it so that the passed in 'xsave' address is used - as intended
originally.
Dave Hansen [Wed, 11 Nov 2015 00:23:54 +0000 (16:23 -0800)]
x86/fpu: Fix 32-bit signal frame handling
(This should have gone to LKML originally. Sorry for the extra
noise, folks on the cc.)
Background:
Signal frames on x86 have two formats:
1. For 32-bit executables (whether on a real 32-bit kernel or
under 32-bit emulation on a 64-bit kernel) we have a
'fpregset_t' that includes the "FSAVE" registers.
2. For 64-bit executables (on 64-bit kernels obviously), the
'fpregset_t' is smaller and does not contain the "FSAVE"
state.
When creating the signal frame, we have to be aware of whether
we are running a 32 or 64-bit executable so we create the
correct format signal frame.
Problem:
save_xstate_epilog() uses 'fx_sw_reserved_ia32' whenever it is
called for a 32-bit executable. This is for real 32-bit and
ia32 emulation.
But, fpu__init_prepare_fx_sw_frame() only initializes
'fx_sw_reserved_ia32' when emulation is enabled, *NOT* for real
32-bit kernels.
This leads to really wierd situations where 32-bit programs
lose their extended state when returning from a signal handler.
The kernel copies the uninitialized (zero) 'fx_sw_reserved_ia32'
out to userspace in save_xstate_epilog(). But when returning
from the signal, the kernel errors out in check_for_xstate()
when it does not see FP_XSTATE_MAGIC1 present (because it was
zeroed). This leads to the FPU/XSAVE state being initialized.
For MPX, this leads to the most permissive state and means we
silently lose bounds violations. I think this would also mean
that we could lose *ANY* FPU/SSE/AVX state. I'm not sure why
no one has spotted this bug.
I believe this was broken by:
72a671ced66d ("x86, fpu: Unify signal handling code paths for x86 and x86_64 kernels")
Dave Hansen [Wed, 11 Nov 2015 18:19:34 +0000 (10:19 -0800)]
x86/mpx: Fix 32-bit address space calculation
I received a bug report that running 32-bit MPX binaries on
64-bit kernels was broken. I traced it down to this little code
snippet. We were switching our "number of bounds directory
entries" calculation correctly. But, we didn't switch the other
side of the calculation: the virtual space size.
This meant that we were calculating an absurd size for
bd_entry_virt_space() on 32-bit because we used the 64-bit
virt_space.
This was _also_ broken for 32-bit kernels running on 64-bit
hardware since boot_cpu_data.x86_virt_bits=48 even when running
in 32-bit mode.
Correct that and properly handle all 3 possible cases:
1. 32-bit binary on 64-bit kernel
2. 64-bit binary on 64-bit kernel
3. 32-bit binary on 32-bit kernel
This manifested in having bounds tables not properly unmapped.
It "leaked" memory but had no functional impact otherwise.
Dave Hansen [Wed, 11 Nov 2015 18:19:31 +0000 (10:19 -0800)]
x86/mpx: Do proper get_user() when running 32-bit binaries on 64-bit kernels
When you call get_user(foo, bar), you effectively do a
copy_from_user(&foo, bar, sizeof(*bar));
Note that the sizeof() is implicit.
When we reach out to userspace to try to zap an entire "bounds
table" we need to go read a "bounds directory entry" in order to
locate the table's address. The size of a "directory entry"
depends on the binary being run and is always the size of a
pointer.
But, when we have a 64-bit kernel and a 32-bit application, the
directory entry is still only 32-bits long, but we fetch it with
a 64-bit pointer which makes get_user() does a 64-bit fetch.
Reading 4 extra bytes isn't harmful, unless we are at the end of
and run off the table. It might also cause the zero page to get
faulted in unnecessarily even if you are not at the end.
Fix it up by doing a special 32-bit get_user() via a cast when
we have 32-bit userspace.
Ingo Molnar [Thu, 12 Nov 2015 06:37:53 +0000 (07:37 +0100)]
Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent
Pull perf/urgent fixes from Arnaldo Carvalho de Melo:
- Add missing newlines to some pr_err() calls (Arnaldo Carvalho de Melo)
- Print full source file paths when using
'perf annotate --print-line --full-paths' (Michael Petlan)
- Fix 'perf probe -d' when just one out of uprobes and kprobes is
enabled (Wang Nan)
- Add compiler.h to list.h to fix 'make perf-tar-src-pkg' generated
tarballs, i.e. out of tree building (Arnaldo Carvalho de Melo)
- Add the llvm-src-base.c and llvm-src-kbuild.c files, generated by the
'perf test' LLVM entries, when running it in-tree, to .gitignore (Yunlong Song)
Linus Torvalds [Thu, 12 Nov 2015 04:18:48 +0000 (20:18 -0800)]
Merge tag 'xfs-for-linus-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs
Pull xfs updates from Dave Chinner:
"There is nothing really major here - the only significant addition is
the per-mount operation statistics infrastructure. Otherwises there's
various ACL, xattr, DAX, AIO and logging fixes, and a smattering of
small cleanups and fixes elsewhere.
Summary:
- per-mount operational statistics in sysfs
- fixes for concurrent aio append write submission
- various logging fixes
- detection of zeroed logs and invalid log sequence numbers on v5 filesystems
- memory allocation failure message improvements
- a bunch of xattr/ACL fixes
- fdatasync optimisation
- miscellaneous other fixes and cleanups"
* tag 'xfs-for-linus-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs: (39 commits)
xfs: give all workqueues rescuer threads
xfs: fix log recovery op header validation assert
xfs: Fix error path in xfs_get_acl
xfs: optimise away log forces on timestamp updates for fdatasync
xfs: don't leak uuid table on rmmod
xfs: invalidate cached acl if set via ioctl
xfs: Plug memory leak in xfs_attrmulti_attr_set
xfs: Validate the length of on-disk ACLs
xfs: invalidate cached acl if set directly via xattr
xfs: xfs_filemap_pmd_fault treats read faults as write faults
xfs: add ->pfn_mkwrite support for DAX
xfs: DAX does not use IO completion callbacks
xfs: Don't use unwritten extents for DAX
xfs: introduce BMAPI_ZERO for allocating zeroed extents
xfs: fix inode size update overflow in xfs_map_direct()
xfs: clear PF_NOFREEZE for xfsaild kthread
xfs: fix an error code in xfs_fs_fill_super()
xfs: stats are no longer dependent on CONFIG_PROC_FS
xfs: simplify /proc teardown & error handling
xfs: per-filesystem stats counter implementation
...
Linus Torvalds [Thu, 12 Nov 2015 04:11:28 +0000 (20:11 -0800)]
Merge tag 'nfsd-4.4' of git://linux-nfs.org/~bfields/linux
Pull nfsd updates from Bruce Fields:
"Apologies for coming a little late in the merge window. Fortunately
this is another fairly quiet one:
Mainly smaller bugfixes and cleanup. We're still finding some bugs
from the breakup of the big NFSv4 state lock in 3.17 -- thanks
especially to Andrew Elble and Jeff Layton for tracking down some of
the remaining races"
* tag 'nfsd-4.4' of git://linux-nfs.org/~bfields/linux:
svcrpc: document lack of some memory barriers
nfsd: fix race with open / open upgrade stateids
nfsd: eliminate sending duplicate and repeated delegations
nfsd: remove recurring workqueue job to clean DRC
SUNRPC: drop stale comment in svc_setup_socket()
nfsd: ensure that seqid morphing operations are atomic wrt to copies
nfsd: serialize layout stateid morphing operations
nfsd: improve client_has_state to check for unused openowners
nfsd: fix clid_inuse on mount with security change
sunrpc/cache: make cache flushing more reliable.
nfsd: move include of state.h from trace.c to trace.h
sunrpc: avoid warning in gss_key_timeout
lockd: get rid of reference-counted NSM RPC clients
SUNRPC: Use MSG_SENDPAGE_NOTLAST when calling sendpage()
lockd: create NSM handles per net namespace
nfsd: switch unsigned char flags in svc_fh to bools
nfsd: move svc_fh->fh_maxsize to just after fh_handle
nfsd: drop null test before destroy functions
nfsd: serialize state seqid morphing operations
Linus Torvalds [Thu, 12 Nov 2015 03:54:05 +0000 (19:54 -0800)]
Merge tag 'devicetree-fixes-for-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux
Pull DeviceTree fixes from Rob Herring:
- Add empty of_translate_address needed for HiSilicon network driver.
- Fix alignment requirements for CMA regions in DT.
- Fix booting on PPC systems which can't do WARN() early.
- Rename ak4554 binding doc from .c to .txt.
* tag 'devicetree-fixes-for-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
of: Provide static inline function for of_translate_address if needed
drivers: of: of_reserved_mem: fixup the alignment with CMA setup
of: Print rather than WARN'ing when overlap check fails
dt-bindings: ak4554: extension should be .txt
Caesar Wang [Thu, 12 Nov 2015 03:43:11 +0000 (19:43 -0800)]
thermal: rockchip: fix compile error
Missing a include file caused compile error.
drivers/thermal/rockchip_thermal.c: In function 'rockchip_thermal_suspend':
drivers/thermal/rockchip_thermal.c:720:2: error: implicit declaration of function 'pinctrl_pm_select_sleep_state' [-Werror=implicit-function-declaration]
...
Fixes: 7e38a5b1daa1 ("thermal: rockchip: support the sleep pinctrl state
to avoid glitches")
According to SPC-4, in a Mode Select, the PS bit in Mode Pages is
reserved and must be set to 0 by the driver. In the sd implementation,
function cache_type_store does a Mode Sense, which might set the PS bit
on the read buffer, followed by a Mode Select, which receives the same
buffer, without explicitly clearing the PS bit. So, in cases where
target supports saving the Mode Page to a non-volatile location, we end
up doing a Mode Select with the PS bit set, which could cause an illegal
request error if the target is checking this.
This was observed on a new firmware change, which was subsequently
reverted, but this changes sd.c to be more compliant with SPC-4.
This patch clears the PS bit in the buffer returned by Mode Select,
right before it is used in the Mode Select command.
Dan Carpenter [Tue, 10 Nov 2015 22:15:24 +0000 (01:15 +0300)]
mptfusion: don't allow negative bytes in kbuf_alloc_2_sgl()
There is a static checker warning here because "bytes" is controlled by
the user and we cap the upper bound with min() but allow negatives.
Negative bytes will result in some nasty warning messages but are not
super harmful. Anyway, no one needs negative bytes so let's just check
for it and return NULL.
scsi: pmcraid: replace struct timeval with ktime_get_real_seconds()
Replace the use of struct timeval and do_gettimeofday() with
64 bit ktime_get_real_seconds. Prevents 32-bit type overflow
in year 2038 on 32-bit systems.
Driver was using the seconds portion of struct timeval (.tv_secs)
to pass a millseconds timestamp to the firmware. This change maintains
that same behavior using ktime_get_real_seconds.
The structure used to pass the timestamp to firmware is 48 bits and
works fine as long as the top 16 bits are zero and they will be zero
for a long time..ie. thousands of years.
Alternative Change: Add sub second granularity to timestamp
As noted above, the driver only used the seconds portion of timeval,
ignores the microseconds portion, and by multiplying by 1000 effectively
does a <<10 and always writes zero into timestamp[0].
The alternative change would pass all the bits to the firmware:
Tina Ruchandani [Fri, 30 Oct 2015 09:11:10 +0000 (02:11 -0700)]
mvumi: 64bit value for seconds_since1970
struct mvumi_hs_page2 stores a "seconds_since1970" field which is of
type u64. It is however, written to, using 'struct timeval' which has
a 32-bit seconds field and whose value will overflow in year 2038.
This patch uses ktime_get_real_seconds() instead since it provides a
64-bit seconds value, which is 2038 safe.
Tim Gardner [Fri, 30 Oct 2015 18:22:58 +0000 (12:22 -0600)]
be2iscsi: Fix bogus WARN_ON length check
drivers/scsi/be2iscsi/be_main.c: In function 'be_sgl_create_contiguous':
drivers/scsi/be2iscsi/be_main.c:3187:18: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]
WARN_ON(!length > 0);
Vitaly Kuznetsov [Fri, 30 Oct 2015 11:37:49 +0000 (12:37 +0100)]
scsi_scan: don't dump trace when scsi_prep_async_scan() is called twice
The only user of scsi_prep_async_scan() is scsi_scan_host() and it
handles the situation correctly. Move 'called twice' reporting to debug
level as well.
The issue is observed on Hyper-V: on any device add/remove event storvsc
driver calls scsi_scan_host() and in case previous scan is still running
we get the message and stack dump on console.
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:35 +0000 (17:30 +0530)]
mpt3sas: Single driver module which supports both SAS 2.0 & SAS 3.0 HBAs
Modified the mpt3sas driver to have a single driver module which
supports both SAS 2.0 & SAS 3.0 HBA devices.
* Added SAS 2.0 HBA device IDs to the mpt3sas_pci_table pci table.
* Created two separate SCSI host templates for SAS2 and SAS3 HBAs so
that, during the driver load time driver can use corresponding host
template(based the pci device ID) while registering a scsi host
adapter instance for that pci device.
* Registered two IOCTL devices, mpt2ctl is for SAS2 HBAs & mpt3ctl for
SAS3 HBAs. Also updated the code to make sure that mpt2ctl device
processes only those ioctl cmds issued for the SAS2 HBAs and mpt3ctl
device processes only those ioctl cmds issued for the SAS3 HBAs.
* Added separate indexing for SAS2 and SAS3 HBAs.
* Replaced compile time check 'MPT2SAS_SCSI' to run time check
'hba_mpi_version_belonged' whereever needed.
* Aliased this merged driver to mpt2sas using MODULE_ALIAS.
* Moved global varaible 'driver_name' to per adapter instance variable.
* Created two raid function template and used corresponding raid
function templates based on the run time check
'hba_mpi_version_belonged'.
* Moved mpt2sas_warpdrive.c file from mpt2sas to mpt3sas folder and
renamed it as mpt3sas_warpdrive.c.
* Also renamed the functions in mpt3sas_warpdrive.c file to follow
current driver function name convention.
* Updated the Makefile to build mpt3sas_warpdrive.o file for these
WarpDrive-specific functions.
* Also in function mpt3sas_setup_direct_io(), used sector_div() API
instead of division operator (which gives compilation errors on 32 bit
machines).
* Added module parameter 'hbas_to_enumerate' which permits using this
merged driver as a legacy mpt2sas driver or as a legacy mpt3sas
driver.
Here are the available options for this module parameter:
0 - Merged driver which enumerates both SAS 2.0 & SAS 3.0 HBAs
1 - Acts as legacy mpt2sas driver, which enumerates only SAS 2.0 HBAs
2 - Acts as legacy mpt3sas driver, which enumerates only SAS 3.0 HBAs
* Removed mpt2sas entries from SCSI's Kconfig and Makefile files.
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:31 +0000 (17:30 +0530)]
mpt3sas: Refcount fw_events and fix unsafe list usage
The fw_event_work struct is concurrently referenced at shutdown. Add a
refcount to protect it and refactor the code to use it.
Additionally, refactor _scsih_fw_event_cleanup_queue() such that it no
longer iterates over the list without holding the lock since
_firmware_event_work() concurrently deletes items from the list.
This patch is ported from commit 008549f6e8a1 ("mpt2sas: Refcount
fw_events and fix unsafe list usage"). These changes are also required
for mpt3sas.
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:30 +0000 (17:30 +0530)]
mpt3sas: Refcount sas_device objects and fix unsafe list usage
sas_device objects can be referenced concurrently throughout the driver.
We need a way to make sure threads can't delete them out from under each
other. This patch adds the refcount and refactors the code to use it.
Additionally, we cannot iterate over the sas_device_list without holding
the lock or we risk corrupting random memory if items are added or
deleted as we iterate. This patch refactors _scsih_probe_sas() to use
the sas_device_list in a safe way.
This patch is ported from the following mpt2sas driver commit d224fe0d6097 ("mpt2sas: Refcount sas_device objects and fix unsafe list
usage").
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:24 +0000 (17:30 +0530)]
mpt3sas: Build MPI SGL LIST on GEN2 HBAs and IEEE SGL LIST on GEN3 HBAs
Gen2 HBAs use MPI scatter-gather lists whereas Gen3 HBAs use IEEE
scatter-gather lists. Modify the common code part in such a way that it
will build IEEE SGL tables for Gen3 HBAs and MPI SGL tables for Gen2
HBAs.
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:23 +0000 (17:30 +0530)]
mpt2sas, mpt3sas: Remove SCSI_MPTXSAS_LOGGING entry from Kconfig
Currently there is a logging level option provided for each of our
drivers in the kernel configuration utility. Users can enable this
option to get more verbose information. By default it is enabled.
Only when this option is enabled will the functions which display the
required information get compiled in.
As we are merging the both drivers we can no longer provide this
configuration option. Remove the SCSI_MPTXSAS_LOGGING entry from Kconfig
and unconditionally enable logging (by removing the #ifdef
CONFIG_SCSI_MPT3SAS_LOGGING preprocessor check conditions) so that all
functions which are defined to display more verbose information get
compiled in.
* pm-tools:
Creating a common structure initialization pattern for struct option
cpupower: Enable disabled Cstates if they are below max latency
cpupower: Remove debug message when using cpupower idle-set -D switch
cpupower: cpupower monitor reports uninitialized values for offline cpus
tools/power turbostat: bugfix: print MAX_NON_TURBO_RATIO
tools/power turbostat: simplify Bzy_MHz calculation
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:20 +0000 (17:30 +0530)]
mpt2sas: Move Gen2 HBA's device registration to a separate file
1. Create a mpt2sas_module.c file for mpt2sas where GEN2 HBA devices
register with PCI, SML, IOCTL subsystems.
2. Updated the Makefile to use the object files from mpt3sas folder.
3. Defined a compilation flag SCSI_MPT2SAS which can be used to not
include those sections of code from mpt3sas driver which are not
required for mpt2sas driver.
4. Inherited automatic diag buffer feature from mpt3sas driver.
Sreekanth Reddy [Wed, 11 Nov 2015 12:00:19 +0000 (17:30 +0530)]
mpt3sas: Move Gen3 HBA's device registration to a separate file
Created a mpt3sas_module.c file for mpt3sas driver where it can register
SAS3 HBA devices with PCI, SML, IOCTL subsystems. Also removed the
corresponding interfaces from mpt3sas_scsih.c file.
list.h needs WRITE_ONCE() since 7f5f873c6a07 ("rculist: Use WRITE_ONCE()
when deleting from reader-visible list") add it before including the
kernel's list.h file.
This fixes builds of 'make perf-tar-src-pkg' perf tool tarball builds,
i.e. out of tree builds.
Wang Nan [Fri, 6 Nov 2015 09:50:15 +0000 (09:50 +0000)]
perf probe: Verify parameters in two functions
On kernel with only one out of CONFIG_KPROBE_EVENTS and
CONFIG_UPROBE_EVENTS enabled, 'perf probe -d' causes a segfault because
perf_del_probe_events() calls probe_file__get_events() with a negative
fd.
This patch fixes it by adding parameter validation at the entry of
probe_file__get_events() and probe_file__get_rawlist(). Since they are
both non-static public functions (in .h file), parameter verifying is
required.
v1 -> v2: Verify fd at the head of probe_file__get_rawlist() instead of
checking at call site (suggested by Masami and Arnaldo at [1,2]).
perf session: Add missing newlines to some pr_err() calls
Before:
[acme@zoo linux]$ perf evlist
WARNING: The perf.data file's data size field is 0 which is unexpected.
Was the 'perf record' command properly terminated?
non matching sample_type[acme@zoo linux]$
After:
[acme@zoo linux]$ perf evlist
WARNING: The perf.data file's data size field is 0 which is unexpected.
Was the 'perf record' command properly terminated?
non matching sample_type
[acme@zoo linux]$
Michael Petlan [Mon, 9 Nov 2015 15:33:31 +0000 (16:33 +0100)]
perf annotate: Support full source file paths for srcline fix
The --full-paths option did not show the full source file paths in the 'perf
annotate' tool, because the value of the option was not propagated into the
related functions.
With this patch the value of the --full-paths option is known to the function
that composes the srcline string, so it prints the full path when necessary.
Committer Note:
This affects annotate when the --print-line option is used:
Yunlong Song [Mon, 9 Nov 2015 03:10:03 +0000 (11:10 +0800)]
perf test: Add llvm-src-base.c and llvm-src-kbuild.c to .gitignore
Commit b31de018a6284a25e0fdfeb028e724f8417ec3b1 ("perf test: Enhance the
LLVM test: update basic BPF test program") dynamically creates file
llvm-src-base.c during the perf building.
Similarly, the commit 7af3f3d55b80cce40ad94b6b8e173dccedaf25e6 ("perf
test: Enhance the LLVM tests: add kbuild test") dynamically creates file
llvm-src-kbuild.c during the perf building. Add them to .gitignore.
Linus Torvalds [Wed, 11 Nov 2015 18:21:34 +0000 (10:21 -0800)]
Merge tag 'firewire-update' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394
Pull firewire fix from Stefan Richter:
"Work around JMicron initialization quirk, which ffected isochronous
transmission, e.g. audio via FFADO or ALSA"
* tag 'firewire-update' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
firewire: ohci: fix JMicron JMB38x IT context discovery
Eddie Kovsky [Fri, 6 Nov 2015 06:23:01 +0000 (23:23 -0700)]
Documentation: Add note on sending files directly with Mutt
Like 'git send-email', Mutt can also be used to send patches generated
with 'git format-patch'. This works regardless of the editor the
contributor has set up to use with Mutt.
Linus Torvalds [Wed, 11 Nov 2015 17:23:07 +0000 (09:23 -0800)]
Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto fix from Herbert Xu:
"This fixes a bug in the algif_hash interface that may lead to crashes
when used with certain algorithms such as HMAC"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: algif_hash - Only export and import on sockets with data
Linus Torvalds [Wed, 11 Nov 2015 17:16:10 +0000 (09:16 -0800)]
Merge tag 'pwm/for-4.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm
Pull pwm updates from Thierry Reding:
"This round contains a couple of new drivers for the Marvell Berlin
family of SoCs, various SoCs from Renesas and Broadcom as well as the
backlight PWM present on MediaTek SoCs.
Further existing drivers are extended to support a wider range of
hardware.
The remaining patches are minor fixes and cleanups across the board.
Note that one of the patches included in this pull request is against
arch/unicore32. I've included it here because I couldn't get a
response from Guan Xuetao and I consider the change low-risk.
Equivalent patches have been merged and tested in Samsung and PXA
trees. The goal is to finally get rid of legacy code paths that have
repeatedly been causing headaches"
* tag 'pwm/for-4.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm: (24 commits)
pwm: sunxi: Fix whitespace issue
pwm: sysfs: Make use of the DEVICE_ATTR_[RW][WO] macro's
pwm: sysfs: Remove unnecessary temporary variable
unicore32: nb0916: Use PWM lookup table
pwm: pwm-rcar: Revise the device tree binding document about compatible
pwm: Return -ENODEV if no PWM lookup match is found
pwm: sun4i: Add support for PWM controller on sun5i SoCs
pwm: Set enable state properly on failed call to enable
pwm: lpss: Add support for runtime PM
pwm: lpss: Add more Intel Broxton IDs
pwm: lpss: Support all four PWMs on Intel Broxton
pwm: lpss: Add support for multiple PWMs
pwm-pca9685: enable ACPI device found on Galileo Gen2
pwm: Add MediaTek display PWM driver support
dt-bindings: pwm: Add MediaTek display PWM bindings
pwm: tipwmss: Enable on TI DRA7x and AM437x
pwm: atmel-hlcdc: add sama5d2 SoC support.
pwm: Add Broadcom BCM7038 PWM controller support
Documentation: dt: add Broadcom BCM7038 PWM controller binding
pwm: Add support for R-Car PWM Timer
...
Linus Torvalds [Wed, 11 Nov 2015 17:03:01 +0000 (09:03 -0800)]
Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux
Pull thermal updates from Zhang Rui:
- Implement generic devfreq cooling mechanism through frequency
reduction for devices using devfreq. From Ørjan Eide and Javi
Merino.
- Introduce OMAP3 support on TI SoC thermal driver. From Pavel Mack
and Eduardo Valentin.
- A bounch of small fixes on devfreq_cooling, Exynos, IMX, Armada, and
Rockchip thermal drivers.
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (24 commits)
thermal: exynos: Directly return 0 instead of using local ret variable
thermal: exynos: Remove unneeded semicolon
thermal: exynos: Use IS_ERR() because regulator cannot be NULL
thermal: exynos: Fix first temperature read after registering sensor
thermal: exynos: Fix unbalanced regulator disable on probe failure
devfreq_cooling: return on allocation failure
thermal: rockchip: support the sleep pinctrl state to avoid glitches in s2r
dt-bindings: rockchip-thermal: Add the pinctrl states in this document
thermal: devfreq_cooling: Make power a u64
thermal: devfreq_cooling: use a thermal_cooling_device for register and unregister
thermal: underflow bug in imx_set_trip_temp()
thermal: armada: Fix possible overflow in the Armada 380 thermal sensor formula
thermal: imx: register irq handler later in probe
thermal: rockhip: fix setting thermal shutdown polarity
thermal: rockchip: fix handling of invalid readings
devfreq_cooling: add trace information
thermal: Add devfreq cooling
PM / OPP: get the voltage for all OPPs
tools/thermal: tmon: use pkg-config also for CFLAGS
linux/thermal.h: rename KELVIN_TO_CELSIUS to DECI_KELVIN_TO_CELSIUS
...
Vivek Goyal [Mon, 9 Nov 2015 16:23:40 +0000 (09:23 -0700)]
fs/block_dev.c: Remove WARN_ON() when inode writeback fails
If a block device is hot removed and later last reference to device
is put, we try to writeback the dirty inode. But device is gone and
that writeback fails.
Currently we do a WARN_ON() which does not seem to be the right thing.
Convert it to a ratelimited kernel warning.
Stephan Günther [Sun, 8 Nov 2015 01:07:02 +0000 (18:07 -0700)]
NVMe: use split lo_hi_{read,write}q
Some controllers may require ordered split transfers even on 64bit
machines, e.g. Apple's NVMe controller as found in the MacBook8,1 and
MacBookAir7,1 (256/512GB models).
This patch enforces ordered split transfers on 64bit platforms, which
works around that issue for all controllers. As pointed out by Christoph
[1] there should be no performance impact due to that modification.
Signed-off-by: Stephan Guenther <[email protected]> Signed-off-by: Maurice Leclaire <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]>
Updated by me to explicitly use lo_hi_read/writeq instead of playing
define tricks.
Jens Axboe [Thu, 5 Nov 2015 20:29:25 +0000 (13:29 -0700)]
MAINTAINERS: add reference to new linux-block list
In the past, I've resisted doing a non-lkml related block/storage
list. But we have more activity now than we previously did, and
ain't nobody got time to track and follow lkml.
So now [email protected] exists. Please CC your patches
related to block/storage here, and we'll have an easier time
tracking them.
Sathyavathi M [Thu, 5 Nov 2015 19:52:28 +0000 (12:52 -0700)]
NVMe: Increase the max transfer size when mdts is 0
This patch address the issue when IO with 128KB from FIO is split into
two parts, 124KB and 4KB, due to max transfer size(127KB). This degrades
the device performance.
Jan Kara [Wed, 4 Nov 2015 16:13:39 +0000 (17:13 +0100)]
brd: Refuse improperly aligned discard requests
Currently when improperly aligned discard request is submitted, we just
silently discard more / less data which results in filesystem corruption
in some cases. Refuse such misaligned requests.
Zhou Chengming [Fri, 6 Nov 2015 06:25:00 +0000 (14:25 +0800)]
livepatch: x86: fix relocation computation with kASLR
With kASLR enabled, old_addr provided by patch module is being shifted
accrodingly so that the symbol lookups work. To have module relocations
handled properly as well, the same transformation needs to be perfomed
on relocation address information.
Sascha Silbe [Wed, 28 Oct 2015 10:06:08 +0000 (11:06 +0100)]
s390/zcrypt: Fix initialisation when zcrypt is built-in
ap_bus and zcrypt_api assumed module information to always be present
and initialisation to be done in module loading order (symbol
dependencies). These assumptions don't hold if zcrypt is built-in;
THIS_MODULE will be NULL in this case and init call order is linker
order, i.e. Makefile order.
Fix initialisation order by ordering the object files in the Makefile
according to their dependencies, like the module loader would do.
Fix message type registration by using a dedicated "name" field rather
than piggy-backing on the module ("owner") information. There's no
change to the requirement that module name and msgtype name are
identical. The existing name macros are used.
We don't need any special code for dealing with the drivers being
built-in; the generic module support code already does the right
thing.
Test results:
1. CONFIG_MODULES=y, CONFIG_ZCRYPT=y
KVM: boots, no /sys/bus/ap (expected)
LPAR with CEX5: boots, /sys/bus/ap/devices/card*/type present
When zcrypt is built as a module, the module loader ensures that the
driver modules cannot be loaded if the AP bus module returns an error
during initialisation. But if zcrypt and the driver are built-in, the
driver is getting initialised even if the AP bus initialisation
failed. The driver invokes ap_driver_register() during initialisation,
which then causes operations on uninitialised data structures to be
performed.
Explicitly protect ap_driver_register() by introducing an
"initialised" flag that gets set iff the AP bus initialisation was
successful. When the AP bus initialisation failed,
ap_driver_register() will error out with -ENODEV, causing the driver
initialisation to fail as well.
Test results:
1. Inside KVM (no AP bus), zcrypt built-in
Boots. /sys/bus/ap not present (expected).
2. Inside KVM (no AP bus), zcrypt as module
Boots. Loading zcrypt_cex4 fails because loading ap_bus fails
(expected).
Sebastian Ott [Fri, 6 Nov 2015 08:58:06 +0000 (09:58 +0100)]
s390/ipl: fix out of bounds access in scpdata_write
The input buffer in reipl_fcp_scpdata_write is accessed out of bounds
when an offset is specified. The problem is that the offset refers to
the data we should write to and not to the buffer we read from.
So instead of
memcpy(scp_data, buf + off, count);
we could just do
memcpy(scp_data + off, buf, count);
However we not only modify the data but also store its length. For this to
work we'd need to remember a state per open FH. Since that's not possible
with sysfs callbacks let's just fail when an offset is specified.
Jonas Gorski [Mon, 12 Oct 2015 11:13:03 +0000 (13:13 +0200)]
MIPS: Make MIPS_CMDLINE_DTB default
Seval of-enabled machines (bmips, lantiq, xlp, pistachio, ralink) copied
the arguments from dtb to arcs_command_line to prevent the kernel from
overwriting them.
Since there is now an option to keep the dtb arguments, default to the
new option remove the "backup" to arcs_command_line in case of USE_OF is
enabled, except for those platforms that still take the bootloader
arguments or do not use any at all.
Jonas Gorski [Mon, 12 Oct 2015 11:13:02 +0000 (13:13 +0200)]
MIPS: Make the kernel arguments from dtb available
Similar to how arm allows using selecting between bootloader arguments,
dtb arguments and both, allow to select them on mips. But since we have
less control over the place of the dtb do not modify it but instead use
the boot_command_line for merging them.
The default is "use bootloader arguments" to keep the current behaviour
as default.
Jonas Gorski [Mon, 12 Oct 2015 11:13:01 +0000 (13:13 +0200)]
MIPS: Use USE_OF as the guard for appended dtb
Since OF is now a user selectable symbol, the choice for appended dtb
support should only be visible when USE_OF is selected, as this
indicates actual machine support for device tree in MIPS.
projects / linux.git / log