Steven Rostedt [Wed, 22 Feb 2012 20:50:28 +0000 (15:50 -0500)]
tracing/ring-buffer: Only have tracing_on disable tracing buffers
As the ring-buffer code is being used by other facilities in the
kernel, having tracing_on file disable *all* buffers is not a desired
affect. It should only disable the ftrace buffers that are being used.
Move the code into the trace.c file and use the buffer disabling
for tracing_on() and tracing_off(). This way only the ftrace buffers
will be affected by them and other kernel utilities will not be
confused to why their output suddenly stopped.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Jiri Olsa [Wed, 15 Feb 2012 14:51:54 +0000 (15:51 +0100)]
ftrace, perf: Add filter support for function trace event
Adding support to filter function trace event via perf
interface. It is now possible to use filter interface
in the perf tool like:
perf record -e ftrace:function --filter="(ip == mm_*)" ls
The filter syntax is restricted to the the 'ip' field only,
and following operators are accepted '==' '!=' '||', ending
up with the filter strings like:
ip == f1[, ]f2 ... || ip != f3[, ]f4 ...
with comma ',' or space ' ' as a function separator. If the
space ' ' is used as a separator, the right side of the
assignment needs to be enclosed in double quotes '"', e.g.:
perf record -e ftrace:function --filter '(ip == do_execve,sys_*,ext*)' ls
perf record -e ftrace:function --filter '(ip == "do_execve,sys_*,ext*")' ls
perf record -e ftrace:function --filter '(ip == "do_execve sys_* ext*")' ls
The '==' operator adds trace filter with same effect as would
be added via set_ftrace_filter file.
The '!=' operator adds trace filter with same effect as would
be added via set_ftrace_notrace file.
The right side of the '!=', '==' operators is list of functions
or regexp. to be added to filter separated by space.
The '||' operator is used for connecting multiple filter definitions
together. It is possible to have more than one '==' and '!='
operators within one filter string.
Jiri Olsa [Wed, 15 Feb 2012 14:51:53 +0000 (15:51 +0100)]
ftrace: Allow to specify filter field type for ftrace events
Adding FILTER_TRACE_FN event field type for function tracepoint
event, so it can be properly recognized within filtering code.
Currently all fields of ftrace subsystem events share the common
field type FILTER_OTHER. Since the function trace fields need
special care within the filtering code we need to recognize it
properly, hence adding the FILTER_TRACE_FN event type.
Adding filter parameter to the FTRACE_ENTRY macro, to specify the
filter field type for the event.
Jiri Olsa [Wed, 15 Feb 2012 14:51:52 +0000 (15:51 +0100)]
ftrace, perf: Add support to use function tracepoint in perf
Adding perf registration support for the ftrace function event,
so it is now possible to register it via perf interface.
The perf_event struct statically contains ftrace_ops as a handle
for function tracer. The function tracer is registered/unregistered
in open/close actions.
To be efficient, we enable/disable ftrace_ops each time the traced
process is scheduled in/out (via TRACE_REG_PERF_(ADD|DELL) handlers).
This way tracing is enabled only when the process is running.
Intentionally using this way instead of the event's hw state
PERF_HES_STOPPED, which would not disable the ftrace_ops.
It is now possible to use function trace within perf commands
like:
perf record -e ftrace:function ls
perf stat -e ftrace:function ls
Jiri Olsa [Wed, 15 Feb 2012 14:51:48 +0000 (15:51 +0100)]
ftrace: Add enable/disable ftrace_ops control interface
Adding a way to temporarily enable/disable ftrace_ops. The change
follows the same way as 'global' ftrace_ops are done.
Introducing 2 global ftrace_ops - control_ops and ftrace_control_list
which take over all ftrace_ops registered with FTRACE_OPS_FL_CONTROL
flag. In addition new per cpu flag called 'disabled' is also added to
ftrace_ops to provide the control information for each cpu.
When ftrace_ops with FTRACE_OPS_FL_CONTROL is registered, it is
set as disabled for all cpus.
The ftrace_control_list contains all the registered 'control' ftrace_ops.
The control_ops provides function which iterates ftrace_control_list
and does the check for 'disabled' flag on current cpu.
Adding 3 inline functions:
ftrace_function_local_disable/ftrace_function_local_enable
- enable/disable the ftrace_ops on current cpu
ftrace_function_local_disabled
- get disabled ftrace_ops::disabled value for current cpu
Steven Rostedt [Tue, 21 Feb 2012 01:37:32 +0000 (20:37 -0500)]
tracing: Don't use p->len field to determine output in __print_*() functions
If more than one __print_*() function is used in a tracepoint
(__print_flags(), __print_symbols(), etc), then the temp seq buffer will
not be zero on entry. Using the temp seq buffer's length to know if
data has been printed or not in the current function is incorrect and
may produce incorrect results.
Currently, no in-tree tracepoint causes this bug, but new ones may
be created.
Cc: Andrew Vagin <avagin@openvz.org> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Andrey Vagin [Sun, 19 Feb 2012 11:16:07 +0000 (14:16 +0300)]
tracing: Don't print an extra separator of flags
If __print_flags() is used after another __print_*() function, the
temp seq_file buffer will not be empty on entry, and the delimiter will
be printed even though there's just one field. We get something like:
|S
instead of just:
S
This is because the length of the temp seq buffer is used to determine
if the delimiter is printed or not. But this algorithm fails when
the seq buffer is not empty on entry, and the delimiter will be printed
because it thinks that a previous field was already printed.
Instead of requiring that users of perf_record_opts set
.sample_id_all_avail to true, just invert the logic, using
.sample_id_all_missing, that doesn't need to be explicitely initialized
since gcc will zero members ommitted in a struct initialization.
Just like the newly introduced .exclude_{guest,host} feature test.
Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-ab772uzk78cwybihf0vt7kxw@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf tools: Handle kernels that don't support attr.exclude_{guest,host}
Just fall back to resetting those fields, if set, warning the user that
that feature is not available.
If guest samples appear they will just be discarded because no struct
machine will be found and thus the event will be accounted as not
handled and dropped, see 0c09571.
Reported-by: Namhyung Kim <namhyung@gmail.com> Tested-by: Joerg Roedel <joerg.roedel@amd.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Joerg Roedel <joerg.roedel@amd.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-vuwxig36mzprl5n7nzvnxxsh@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Robert Richter [Fri, 10 Feb 2012 14:41:56 +0000 (15:41 +0100)]
perf tools: Factor out feature op to process header sections
There is individual code for each feature to process header sections.
Adding a function pointer .process to struct feature_ops for keeping the
implementation in separate functions. Code to process header sections is
now a generic function.
Joerg Roedel [Fri, 10 Feb 2012 17:05:05 +0000 (18:05 +0100)]
perf tools: Change perf_guest default back to false
Setting perf_guest to true by default makes no sense because the perf
subcommands can not setup guest symbol information and thus not process
and guest samples. The only exception is perf-kvm which changes the
perf_guest value on its own. So change the default for perf_guest back
to false.
Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jason Wang <jasowang@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1328893505-4115-3-git-send-email-joerg.roedel@amd.com Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Joerg Roedel [Fri, 10 Feb 2012 17:05:04 +0000 (18:05 +0100)]
perf top: Don't process samples with no valid machine object
The perf sample processing code relies on a valid machine object. Make
sure that this path is only entered when such a object exists.
A counter for samples where no machine object exits is also introduced
to give the user a message about these samples.
Reported-by: David Ahern <dsahern@gmail.com> Reported-by: Jason Wang <jasowang@redhat.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jason Wang <jasowang@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1328893505-4115-2-git-send-email-joerg.roedel@amd.com Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Namhyung Kim [Sun, 12 Feb 2012 10:45:24 +0000 (19:45 +0900)]
perf tools: Fix build dependency of perf python extension
The perf python extention (perf.so) file lacks its dependencies in the
Makefile so that it cannot be refreshed if one of source files it depends
is changed. Fix it by putting them in a separate file and processing it in
both of Makefile and setup.py.
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1329043524-12470-1-git-send-email-namhyung@gmail.com Signed-off-by: Namhyung Kim <namhyung@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Thomas Meyer [Tue, 29 Nov 2011 21:08:00 +0000 (22:08 +0100)]
tracing/trivial: Use kcalloc instead of kzalloc to allocate array
The advantage of kcalloc is, that will prevent integer overflows which could
result from the multiplication of number of elements and size and it is also
a bit nicer to read.
The semantic patch that makes this change is available
in https://lkml.org/lkml/2011/11/25/107
Johannes Berg [Thu, 24 Nov 2011 19:03:08 +0000 (20:03 +0100)]
printk/tracing: Add console output tracing
Add a printk.console trace point to record any printk
messages into the trace, regardless of the current
console loglevel. This can help correlate (existing)
printk debugging with other tracing.
Link: http://lkml.kernel.org/r/1322161388.5366.54.camel@jlt3.sipsolutions.net Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Ingo Molnar <mingo@redhat.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Geunsik Lim [Wed, 8 Feb 2012 10:05:36 +0000 (19:05 +0900)]
ftrace: sched_switch plugin is deprecated
Actually, sched_switch function tracer is merged into wakeup/wakeup_rt
Update 'mini-HOWTO' for ftrace(Kernel function tracer).
If we want to trace "sched:sched_switch" to trace sched_switch func,
We may utilize event option.(e.g: trace-cmd list -e | grep sched)
This patch is based on Linux-3.3.rc2-SMP-PREEMPT
Steven Rostedt [Tue, 7 Feb 2012 14:46:01 +0000 (09:46 -0500)]
cpuidle/tracing: Denote the tracepoints as being in rcu_idle_exit() section
As the tracepoints in the cpuidle code are called when rcu_idle_exit() is in
effect, the _rcuidle() version must be used, otherwise the rcu_read_lock()s
that protect the tracepoint will not be honored.
Cc: Len Brown <len.brown@intel.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Steven Rostedt [Tue, 7 Feb 2012 14:40:30 +0000 (09:40 -0500)]
x86/tracing: Denote the power and cpuidle tracepoints as _rcuidle()
The power and cpuidle tracepoints are called within a rcu_idle_exit()
section, and must be denoted with the _rcuidle() version of the tracepoint.
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Steven Rostedt [Tue, 7 Feb 2012 14:32:43 +0000 (09:32 -0500)]
tracing/rcu: Add trace_##name##__rcuidle() static tracepoint for inside rcu_idle_exit() sections
Added is a new static inline function that lets *any* tracepoint be used
inside a rcu_idle_exit() section. And this also solves the problem where
the same tracepoint may be used inside a rcu_idle_exit() section as well
as outside of one.
I added a new tracepoint function with a "_rcuidle" extension. All
tracepoints can be used with either the normal "trace_foobar()"
function, or the "trace_foobar_rcuidle()" function when inside a
rcu_idle_exit() section.
All tracepoints defined by TRACE_EVENT() or any of the derivatives
will have a "_rcuidle()" function also defined. When a tracepoint is
used within an rcu_idle_exit() section, the "_rcuidle()" version must
be used. This denotes that the tracepoint is within rcu_idle_exit()
and it allows the rcu read locks within the tracepoint to still
be valid, as this version takes us out of rcu_idle_exit().
Another nice aspect about this patch is that "static inline"s are not
compiled into text when not used. So only the tracepoints that actually
use the _rcuidle() version will have them defined in the actual text
that is booted.
Masami Hiramatsu [Fri, 10 Feb 2012 05:33:40 +0000 (14:33 +0900)]
x86: Fix to decode grouped AVX with VEX pp bits
Fix to decode grouped AVX with VEX pp bits which should be
handled as same as last-prefixes. This fixes below warnings
in posttest with CONFIG_CRYPTO_SHA1_SSSE3=y.
Warning: arch/x86/tools/test_get_len found difference at <sha1_transform_avx>:ffffffff810d5fc0
Warning: ffffffff810d6069: c5 f9 73 de 04 vpsrldq $0x4,%xmm6,%xmm0
Warning: objdump says 5 bytes, but insn_get_length() says 4
...
With this change, test_get_len can decode it correctly.
$ arch/x86/tools/test_get_len -v -y ffffffff810d6069: c5 f9 73 de 04 vpsrldq $0x4,%xmm6,%xmm0
Succeed: decoded and checked 1 instructions
Reflect the change in the soft and hard lockup thresholds and
their relation to the frequency of the hrtimer and NMI events in
the code comments. While at it, remove references to files that
do not exist anymore.
The soft and hard lockup thresholds have changed so the
corresponding Kconfig entries need to be updated accordingly.
Add a reference to watchdog_thresh while at it.
The current version of perf detects whether or not the perf.data file is
written in a different endianness using the attr_size field in the
header of the file. This field represents sizeof(struct perf_event_attr)
as known to perf record. If the sizes do not match, then perf tries the
byte-swapped version. If they match, then the tool assumes a different
endianness.
The issue with the approach is that it assumes the size of
perf_event_attr always has to match between perf record and perf report.
However, the kernel perf_event ABI is extensible. New fields can be
added to struct perf_event_attr. Consequently, it is not possible to use
attr_size to detect endianness.
This patch takes another approach by using the magic number written at
the beginning of the perf.data file to detect endianness. The magic
number is an eight-byte signature. It's primary purpose is to identify
(signature) a perf.data file. But it could also be used to encode the
endianness.
The patch introduces a new value for this signature. The key difference
is that the signature is written differently in the file depending on
the endianness. Thus, by comparing the signature from the file with the
tool's own signature it is possible to detect endianness. The new
signature is "PERFILE2".
Backward compatiblity with existing perf.data file is ensured.
Tested-by: David Ahern <dsahern@gmail.com> Acked-by: David Ahern <dsahern@gmail.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com> Cc: Arun Sharma <asharma@fb.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Lin Ming <ming.m.lin@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roberto Agostino Vitillo <ravitillo@lbl.gov> Cc: Robert Richter <robert.richter@amd.com> Cc: Vince Weaver <vweaver1@eecs.utk.edu> Link: http://lkml.kernel.org/r/1328187288-24395-15-git-send-email-eranian@google.com Signed-off-by: Stephane Eranian <eranian@google.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Borislav Petkov [Tue, 7 Feb 2012 12:08:52 +0000 (13:08 +0100)]
x86/sched/perf/AMD: Set sched_clock_stable
Stephane Eranian reported that doing a scheduler latency
measurements with perf on AMD doesn't work out as expected due
to the fact that the sched_clock() granularity is too coarse,
i.e. done in jiffies due to the sched_clock_stable not set,
which, if set, would mean that we get to use the TSC as sample
source which would give us much higher precision.
However, there's no reason not to set sched_clock_stable on AMD
because all families from F10h and upwards do have an invariant
TSC and have the CPUID flag to prove (CPUID_8000_0007_EDX[8]).
Make it so, #1.
Signed-off-by: Borislav Petkov <bp@alien8.de> Cc: Borislav Petkov <bp@amd64.org> Cc: Venki Pallipadi <venki@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Arnaldo Carvalho de Melo <acme@infradead.org> Cc: Robert Richter <robert.richter@amd.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Andreas Herrmann <andreas.herrmann3@amd.com> Link: http://lkml.kernel.org/r/20120206132546.GA30854@quad
[ Should any non-standard system break the TSC, we should
mark them so explicitly, in their platform init handler, or
in a DMI quirk. ] Signed-off-by: Ingo Molnar <mingo@elte.hu>
Jiri Olsa [Wed, 1 Feb 2012 12:54:08 +0000 (13:54 +0100)]
perf tool: Fix perf stack to non executable on x86_64
By adding following objects:
bench/mem-memset-x86-64-asm.o
bench/mem-memcpy-x86-64-asm.o
the x86_64 perf binary ended up with executable stack.
The reason was that above objects are assembler sourced and are missing the
GNU-stack note section. In such case the linker assumes that the final binary
should not be restricted at all and mark the stack as RWX.
Adding section ".note.GNU-stack" definition to mentioned objects, with all
flags disabled, thus omiting those objects from linker stack flags decision.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=783570 Reported-by: Clark Williams <williams@redhat.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1328100848-5630-1-git-send-email-jolsa@redhat.com Signed-off-by: Jiri Olsa <jolsa@redhat.com>
[ committer note: Remaining bits after what was already added to perf/urgent ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A git bisect shows commit f3bda2c as the cause. However, looking back
through the git history, I saw commit 640c03c which seems to have
removed the required initialization for perf_sample->period. The problem
only started showing after commit f3bda2c. The below patch re-introduces
the initialization and it fixes the problem for me.
Jiri Olsa [Sun, 5 Feb 2012 17:11:05 +0000 (18:11 +0100)]
perf tools: Fix prefix matching for kernel maps
In some perf ancient versions we used '[kernel.kallsyms._text]' as the
name for the kernel map.
This got changed with commit:
perf: 'perf kvm' tool for monitoring guest performance from host
commit a1645ce12adb6c9cc9e19d7695466204e3f017fe
Author: Zhang, Yanmin <yanmin_zhang@linux.intel.com>
and we started to use following name '[kernel.kallsyms]_text'.
This name change is important for the report code dealing with ancient
perf data. When processing the kernel map event, we need to recognize
the old naming (dont match the last ']') and initialize the kernel map
correctly.
The subsequent call to maps__set_kallsyms_ref_reloc_sym deals with the
superfluous ']' to get correct symbol name.
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1328461865-6127-1-git-send-email-jolsa@redhat.com Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Jiri Olsa [Mon, 6 Feb 2012 20:54:06 +0000 (18:54 -0200)]
perf tools: Fix perf stack to non executable on x86_64
By adding following objects:
bench/mem-memcpy-x86-64-asm.o
the x86_64 perf binary ended up with executable stack.
The reason was that above object are assembler sourced and is missing the
GNU-stack note section. In such case the linker assumes that the final binary
should not be restricted at all and mark the stack as RWX.
Adding section ".note.GNU-stack" definition to mentioned object, with all
flags disabled, thus omiting this object from linker stack flags decision.
Steven Rostedt [Thu, 26 Jan 2012 01:18:55 +0000 (20:18 -0500)]
tracing/softirq: Move __raise_softirq_irqoff() out of header
The __raise_softirq_irqoff() contains a tracepoint. As tracepoints in headers
can cause issues, and not to mention, bloats the kernel when they are
in a static inline, it is best to move the function that contains the
tracepoint out of the header and into softirq.c.
Jiri Olsa [Mon, 2 Jan 2012 09:04:14 +0000 (10:04 +0100)]
ftrace: Change filter/notrace set functions to return exit code
Currently the ftrace_set_filter and ftrace_set_notrace functions
do not return any return code. So there's no way for ftrace_ops
user to tell wether the filter was correctly applied.
The set_ftrace_filter interface returns error in case the filter
did not match:
cpuc->enabled is always on because when we get to
perf_adjust_freq_unthr_context() the PMU is not totally
disabled. Furthermore when we need to adjust a period,
we only stop the event we need to change and not the
entire PMU. Thus, when we re-enable, cpuc->enabled is
already set. Note that when we stop the event, both
pebs and lbr are stopped if necessary (and possible).
Linus Torvalds [Thu, 2 Feb 2012 23:47:33 +0000 (15:47 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
rbd: fix safety of rbd_put_client()
rbd: fix a memory leak in rbd_get_client()
ceph: create a new session lock to avoid lock inversion
ceph: fix length validation in parse_reply_info()
ceph: initialize client debugfs outside of monc->mutex
ceph: change "ceph.layout" xattr to be "ceph.file.layout"
Alex Elder [Sun, 29 Jan 2012 19:57:43 +0000 (13:57 -0600)]
rbd: fix safety of rbd_put_client()
The rbd_client structure uses a kref to arrange for cleaning up and
freeing an instance when its last reference is dropped. The cleanup
routine is rbd_client_release(), and one of the things it does is
delete the rbd_client from rbd_client_list. It acquires node_lock
to do so, but the way it is done is still not safe.
The problem is that when attempting to reuse an existing rbd_client,
the structure found might already be in the process of getting
destroyed and cleaned up.
Here's the scenario, with "CLIENT" representing an existing
rbd_client that's involved in the race:
Thread on CPU A | Thread on CPU B
--------------- | ---------------
rbd_put_client(CLIENT) | rbd_get_client()
kref_put() | (acquires node_lock)
kref->refcount becomes 0 | __rbd_client_find() returns CLIENT
calls rbd_client_release() | kref_get(&CLIENT->kref);
| (releases node_lock)
(acquires node_lock) |
deletes CLIENT from list | ...and starts using CLIENT...
(releases node_lock) |
and frees CLIENT | <-- but CLIENT gets freed here
Fix this by having rbd_put_client() acquire node_lock. The result
could still be improved, but at least it avoids this problem.
Signed-off-by: Alex Elder <elder@dreamhost.com> Signed-off-by: Sage Weil <sage@newdream.net>
This has been updated since I last sent it as the creation of the new
mm_access() function did almost exactly the same thing as parts of the
previous version of this patch did.
In order to use mm_access() even when /proc isn't enabled, we move it to
kernel/fork.c where other related process mm access functions already
are.
Signed-off-by: Chris Yeoh <yeohc@au1.ibm.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alex Elder [Fri, 13 Jan 2012 01:48:10 +0000 (17:48 -0800)]
ceph: create a new session lock to avoid lock inversion
Lockdep was reporting a possible circular lock dependency in
dentry_lease_is_valid(). That function needs to sample the
session's s_cap_gen and and s_cap_ttl fields coherently, but needs
to do so while holding a dentry lock. The s_cap_lock field was
being used to protect the two fields, but that can't be taken while
holding a lock on a dentry within the session.
In most cases, the s_cap_gen and s_cap_ttl fields only get operated
on separately. But in three cases they need to be updated together.
Implement a new lock to protect the spots updating both fields
atomically is required.
Signed-off-by: Alex Elder <elder@dreamhost.com> Reviewed-by: Sage Weil <sage@newdream.net>
Xi Wang [Sun, 15 Jan 2012 03:20:59 +0000 (22:20 -0500)]
ceph: fix length validation in parse_reply_info()
"len" is read from network and thus needs validation. Otherwise, given
a bogus "len" value, p+len could be an out-of-bounds pointer, which is
used in further parsing.
Signed-off-by: Xi Wang <xi.wang@gmail.com> Signed-off-by: Sage Weil <sage@newdream.net>
Sage Weil [Sat, 14 Jan 2012 06:22:03 +0000 (22:22 -0800)]
ceph: initialize client debugfs outside of monc->mutex
Initializing debufs under monc->mutex introduces a lock dependency for
sb->s_type->i_mutex_key, which (combined with several other dependencies)
leads to an annoying lockdep warning. There's no particular reason to do
the debugfs setup under this lock, so move it out.
It used to be the case that our first monmap could come from the OSD; that
is no longer the case with recent servers, so we will reliably set up the
client entry during the initial authentication.
We don't have to worry about racing with debugfs teardown by
ceph_debugfs_client_cleanup() because ceph_destroy_client() calls
ceph_msgr_flush() first, which will wait for the message dispatch work
to complete (and the debugfs init to complete).
Fixes: #1940 Signed-off-by: Sage Weil <sage@newdream.net>
Alex Elder [Thu, 12 Jan 2012 01:41:01 +0000 (17:41 -0800)]
ceph: change "ceph.layout" xattr to be "ceph.file.layout"
The virtual extended attribute named "ceph.layout" is meaningful
only for regular files. Change its name to be "ceph.file.layout" to
more directly reflect that in the ceph xattr namespace. Preserve
the old "ceph.layout" name for the time being (until we decide it's
safe to get rid of it entirely).
Add a missing initializer for "readonly" in the terminating entry.
Signed-off-by: Alex Elder <elder@dreamhost.com> Reviewed-by: Sage Weil <sage@newdream.net>
Robert Richter [Thu, 15 Dec 2011 16:32:40 +0000 (17:32 +0100)]
perf record: Make feature initialization generic
Loop over all features to enable it instead of explicitly enabling every
single feature. Reducing duplicate code and making it more robust to
later changes e.g. when adding more features.
Linus Torvalds [Thu, 2 Feb 2012 19:19:03 +0000 (11:19 -0800)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/radeon/kms/blit: fix blit copy for very large buffers
drm/radeon/kms: fix TRAVIS panel setup
drm/radeon: fix use after free in ATRM bios reading code.
drm/radeon/kms: Fix device tree linkage of DP i2c buses too
drm/radeon: Set DESKTOP_HEIGHT register to the framebuffer (not mode) height.
drm/radeon/kms: disable output polling when suspended
drm/nv50/pm: signedness bug in nv50_pm_clocks_pre()
drm/nouveau/gem: fix fence_sync race / oops
drm/nouveau: fix typo on mxmdcb option
drm/nouveau/mxm: pretend to succeed, even if we can't shadow the MXM-SIS
drm/nouveau/disp: check that panel power gpio is enabled at init time
Linus Torvalds [Thu, 2 Feb 2012 19:11:13 +0000 (11:11 -0800)]
Merge branches 'core-urgent-for-linus', 'perf-urgent-for-linus', 'sched-urgent-for-linus' and 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
bugs, x86: Fix printk levels for panic, softlockups and stack dumps
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf top: Fix number of samples displayed
perf tools: Fix strlen() bug in perf_event__synthesize_event_type()
perf tools: Fix broken build by defining _GNU_SOURCE in Makefile
x86/dumpstack: Remove unneeded check in dump_trace()
perf: Fix broken interrupt rate throttling
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/rt: Fix task stack corruption under __ARCH_WANT_INTERRUPTS_ON_CTXSW
sched: Fix ancient race in do_exit()
sched/nohz: Fix nohz cpu idle load balancing state with cpu hotplug
sched/s390: Fix compile error in sched/core.c
sched: Fix rq->nr_uninterruptible update race
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/reboot: Remove VersaLogic Menlow reboot quirk
x86/reboot: Skip DMI checks if reboot set by user
x86: Properly parenthesize cmpxchg() macro arguments
Ilija Hadzic [Thu, 2 Feb 2012 15:26:24 +0000 (10:26 -0500)]
drm/radeon/kms/blit: fix blit copy for very large buffers
Evergreen and NI blit copy was broken if the buffer maps to a rectangle
whose one dimension is 16384 (max dimension allowed by these chips).
In the mainline kernel, the problem is exposed only when buffers are
very large (1G), but it's still a problem. The problem could be exposed
for smaller buffers if anyone modifies the algorithm for rectangle
construction in r600_blit_create_rect() (the reason why someone would
modify that algorithm is to tune the performance of buffer moves).
The root cause was in i2f() function which only operated on range between
0 and 16383. Fix this by extending the range of i2f() function to 0 to
32767.
While at it improve the function so that the range can be easily
extended in the future (if it becomes necessary), cleanup lines
over 80 characters, and replace in-line comments with one strategic
comment that explains the crux of the function.
Credits to michel@daenzer.net for pointing out the root cause of
the bug.
v2: Fix I2F_MAX_INPUT constant definition goof and warn only once
if input argument is out of range. Edit the comment a little
bit to avoid some linguistic confusion and make it look better
in general.
Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Michel Dänzer <michel@daenzer.net> Signed-off-by: Dave Airlie <airlied@redhat.com>
Linus Torvalds [Thu, 2 Feb 2012 02:30:15 +0000 (18:30 -0800)]
Merge tag 'firewire-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394
firewire fixes post v3.3-rc1
Add workarounds table entries for hardware bugs in
- FireWire part of Sound Blaster Audigy cards,
- Ricoh PCIe 1394 controllers.
Without these, several protocols, e.g. AV/C, do not work on the
Audigy, and the Ricoh PCIe controllers wouldn't work at all.
This does not concern the older Ricoh PCI controllers.
* tag 'firewire-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
firewire: ohci: disable MSI on Ricoh controllers
firewire: ohci: add reset packet quirk for SB Audigy
Randy Dunlap [Thu, 2 Feb 2012 02:15:49 +0000 (18:15 -0800)]
docbook: fix fatal errors in device-drivers docbook and add DMA Management section
Fix 2 fatal errors in the device-drivers docbook.
Also add some missing files from drivers/base/; since several
of these are DMA-related, add a section for DMA Management.
docproc: drivers/base/sys.c: No such file or directory
docproc: drivers/tty/serial/8250.c: No such file or directory
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Thu, 2 Feb 2012 00:00:38 +0000 (16:00 -0800)]
Merge branch 'for-linus2' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security
* 'for-linus2' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
lib: Fix 32-bit sparc udiv_qrnnd() definition in mpilib's longlong.h
lib: Fix multiple definitions of clz_tab
lib/digsig: checks for NULL return value
lib/mpi: added missing NULL check
lib/mpi: added comment on divide by 0 case
lib/mpi: check for possible zero length
lib/digsig: pkcs_1_v1_5_decode_emsa cleanup
lib/digsig: additional sanity checks against badly formated key payload
lib/mpi: removed unused functions
lib/mpi: checks for zero divisor length
lib/mpi: return error code on dividing by zero
lib/mpi: replaced MPI_NULL with normal NULL
lib/mpi: added missing NULL check
David Miller [Wed, 1 Feb 2012 22:17:55 +0000 (00:17 +0200)]
lib: Fix 32-bit sparc udiv_qrnnd() definition in mpilib's longlong.h
This copy of longlong.h is extremely dated and results in compile
errors on sparc32 when MPILIB is enabled, copy over the more uptodate
implementation from arch/sparc/math/sfp-util_32.h
Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: James Morris <jmorris@namei.org>
Linus Torvalds [Wed, 1 Feb 2012 23:18:39 +0000 (15:18 -0800)]
Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
InfiniBand/RDMA fixes for 3.3:
- Fix a crash due to a regression (uninitialized refcnt) introduced in
3.2 with XRC support.
- Close race in how ucma reports events when connect fails.
- Process vendor-specific MADs in mlx4 so that eg FDR-10 data rate works.
- Fix regression in qib caused by over-aggressive PCIe tuning.
- Other small fixes for hardware drivers (ipath, nes, qib).
* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
RDMA/nes: Copyright update
IB/mlx4: pass SMP vendor-specific attribute MADs to firmware
RDMA/nes: Fix fast memory registration opcode
RDMA/nes: Fix fast memory registration length
RDMA/ucma: Discard all events for new connections until accepted
IB/qib: Roll back PCIe tuning change
IB/qib: Use GFP_ATOMIC when locks are held
RDMA/nes: Add missing rcu_read_unlock() in nes_addr_resolve_neigh()
RDMA/nes: Fix for sending MPA reject frame
IB/ipath: Calling PTR_ERR() on right variable in create_file()
RDMA/core: Fix kernel panic by always initializing qp->usecnt
Oleg Nesterov [Tue, 31 Jan 2012 16:15:11 +0000 (17:15 +0100)]
proc: make sure mem_open() doesn't pin the target's memory
Once /proc/pid/mem is opened, the memory can't be released until
mem_release() even if its owner exits.
Change mem_open() to do atomic_inc(mm_count) + mmput(), this only
pins mm_struct. Change mem_rw() to do atomic_inc_not_zero(mm_count)
before access_remote_vm(), this verifies that this mm is still alive.
I am not sure what should mem_rw() return if atomic_inc_not_zero()
fails. With this patch it returns zero to match the "mm == NULL" case,
may be it should return -EINVAL like it did before e268337d.
Perhaps it makes sense to add the additional fatal_signal_pending()
check into the main loop, to ensure we do not hold this memory if
the target task was oom-killed.
This patch fixes merge conflict resolution breakage introduced by merge d3712b9dfcf4 ("Merge tag 'for-linus' of git://github.com/prasad-joshi/logfs_upstream").
The commit changed 'mtd_can_have_bb()' function and made it always
return zero, which is incorrect. Instead, we need it to return whether
the underlying flash device can have bad eraseblocks or not. UBI needs
this information because it affects how it handles the underlying flash.
E.g., if the underlying flash is NOR, it cannot have bad blocks and any
write or erase error is fatal, and all we can do is to switch to R/O
mode. We do not need to reserve a pool of good eraseblocks for bad
eraseblocks handling, and so on.
This patch also removes 'mtd_can_have_bb()' invocations from Logfs to
ensure correct Logfs behavior.
I've tested that with this patch UBI works on top of NOR and NAND
flashes emulated by mtdram and nandsim correspondingly.
This patch is based on patch from Linus Torvalds.
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com> Acked-by: Jörn Engel <joern@logfs.org> Acked-by: Prasad Joshi <prasadjoshi.linux@gmail.com> Acked-by: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michel Dänzer [Wed, 1 Feb 2012 11:09:55 +0000 (12:09 +0100)]
drm/radeon: Set DESKTOP_HEIGHT register to the framebuffer (not mode) height.
The value of this register is transferred to the V_COUNTER register at the
beginning of vertical blank. V_COUNTER is the reference for VLINE waits and
goes from VIEWPORT_Y_START to VIEWPORT_Y_START+VIEWPORT_HEIGHT during scanout,
so if VIEWPORT_Y_START is not 0, V_COUNTER actually went backwards at the
beginning of vertical blank, and VLINE waits excluding the whole scanout area
could never finish (possibly only if VIEWPORT_Y_START is larger than the length
of vertical blank in scanlines). Setting DESKTOP_HEIGHT to the framebuffer
height should prevent this for any kind of VLINE wait.
CC: stable@vger.kernel.org Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
Seth Forshee [Wed, 1 Feb 2012 01:06:25 +0000 (19:06 -0600)]
drm/radeon/kms: disable output polling when suspended
Polling the outputs when the device is suspended can result in erroneous
status updates. Disable output polling during suspend to prevent this
from happening.
Signed-off-by: Seth Forshee <seth.forshee@canonical.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie <airlied@redhat.com>
Dmitry Kasatkin [Thu, 26 Jan 2012 17:13:26 +0000 (19:13 +0200)]
lib/digsig: pkcs_1_v1_5_decode_emsa cleanup
Removed useless 'is_valid' variable in pkcs_1_v1_5_decode_emsa(),
which was inhereted from original code. Client now uses return value
to check for an error.
Dmitry Kasatkin [Thu, 26 Jan 2012 17:13:23 +0000 (19:13 +0200)]
lib/mpi: removed unused functions
do_encode_md() and mpi_get_keyid() are not parts of mpi library.
They were used early versions of gnupg and in digsig project,
but they are not used neither here nor there anymore.
projects / mv-sheeva.git / log