Pull perf update from Ingo Molnar:
"Lots of changes in this cycle as well, with hundreds of commits from
over 30 contributors. Most of the activity was on the tooling side.
Higher level changes:
- New 'perf kvm' analysis tool, from Xiao Guangrong.
- New 'perf trace' system-wide tracing tool
- uprobes fixes + cleanups from Oleg Nesterov.
- Lots of patches to make perf build on Android out of box, from
Irina Tirdea
- Extend ftrace function tracing utility to be more dynamic for its
users. It allows for data passing to the callback functions, as
well as reading regs as if a breakpoint were to trigger at function
entry.
The main goal of this patch series was to allow kprobes to use
ftrace as an optimized probe point when a probe is placed on an
ftrace nop. With lots of help from Masami Hiramatsu, and going
through lots of iterations, we finally came up with a good
solution.
- Add cpumask for uncore pmu, use it in 'stat', from Yan, Zheng.
- Various tracing updates from Steve Rostedt
- Clean up and improve 'perf sched' performance by elliminating lots
of needless calls to libtraceevent.
- Event group parsing support, from Jiri Olsa
- UI/gtk refactorings and improvements from Namhyung Kim
- Add support for non-tracepoint events in perf script python, from
Feng Tang
- Add --symbols to 'script', similar to the one in 'report', from
Feng Tang.
Infrastructure enhancements and fixes:
- Convert the trace builtins to use the growing evsel/evlist
tracepoint infrastructure, removing several open coded constructs
like switch like series of strcmp to dispatch events, etc.
Basically what had already been showcased in 'perf sched'.
- Add evsel constructor for tracepoints, that uses libtraceevent just
to parse the /format events file, use it in a new 'perf test' to
make sure the libtraceevent format parsing regressions can be more
readily caught.
- Some strange errors were happening in some builds, but not on the
next, reported by several people, problem was some parser related
files, generated during the build, didn't had proper make deps, fix
from Eric Sandeen.
- Introduce struct and cache information about the environment where
a perf.data file was captured, from Namhyung Kim.
- Fix handling of unresolved samples when --symbols is used in
'report', from Feng Tang.
- Add union member access support to 'probe', from Hyeoncheol Lee.
- Fixups to die() removal, from Namhyung Kim.
- Render fixes for the TUI, from Namhyung Kim.
- Don't enable annotation in non symbolic view, from Namhyung Kim.
- Fix pipe mode in 'report', from Namhyung Kim.
- Move related stats code from stat to util/, will be used by the
'stat' kvm tool, from Xiao Guangrong.
- Remove die()/exit() calls from several tools.
- Resolve vdso callchains, from Jiri Olsa
- Don't pass const char pointers to basename, so that we can
unconditionally use libgen.h and thus avoid ifdef BIONIC lines,
from David Ahern
- Refactor hist formatting so that it can be reused with the GTK
browser, From Namhyung Kim
- Fix build for another rbtree.c change, from Adrian Hunter.
- Make 'perf diff' command work with evsel hists, from Jiri Olsa.
- Use the only field_sep var that is set up: symbol_conf.field_sep,
fix from Jiri Olsa.
- .gitignore compiled python binaries, from Namhyung Kim.
- Get rid of die() in more libtraceevent places, from Namhyung Kim.
- Rename libtraceevent 'private' struct member to 'priv' so that it
works in C++, from Steven Rostedt
- Remove lots of exit()/die() calls from tools so that the main perf
exit routine can take place, from David Ahern
- Fix x86 build on x86-64, from David Ahern.
- {int,str,rb}list fixes from Suzuki K Poulose
- perf.data header fixes from Namhyung Kim
- Allow user to indicate objdump path, needed in cross environments,
from Maciek Borzecki
- Fix hardware cache event name generation, fix from Jiri Olsa
- Add round trip test for sw, hw and cache event names, catching the
problem Jiri fixed, after Jiri's patch, the test passes
successfully.
- Clean target should do clean for lib/traceevent too, fix from David
Ahern
- Check the right variable for allocation failure, fix from Namhyung
Kim
- Set up evsel->tp_format regardless of evsel->name being set
already, fix from Namhyung Kim
- Oprofile fixes from Robert Richter.
- Remove perf_event_attr needless version inflation, from Jiri Olsa
- Introduce libtraceevent strerror like error reporting facility,
from Namhyung Kim
- Add pmu mappings to perf.data header and use event names from cmd
line, from Robert Richter
- Fix include order for bison/flex-generated C files, from Ben
Hutchings
- Build fixes and documentation corrections from David Ahern
- Assorted cleanups from Robert Richter
- Let O= makes handle relative paths, from Steven Rostedt
- perf script python fixes, from Feng Tang.
- Initial bash completion support, from Frederic Weisbecker
- Allow building without libelf, from Namhyung Kim.
- Support DWARF CFI based unwind to have callchains when %bp based
unwinding is not possible, from Jiri Olsa.
- Symbol resolution fixes, while fixing support PPC64 files with an
.opt ELF section was the end goal, several fixes for code that
handles all architectures and cleanups are included, from Cody
Schafer.
- Assorted fixes for Documentation and build in 32 bit, from Robert
Richter
- Cache the libtraceevent event_format associated to each evsel
early, so that we avoid relookups, i.e. calling pevent_find_event
repeatedly when processing tracepoint events.
[ This is to reduce the surface contact with libtraceevents and
make clear what is that the perf tools needs from that lib: so
far parsing the common and per event fields. ]
- Don't stop the build if the audit libraries are not installed, fix
from Namhyung Kim.
- Fix bfd.h/libbfd detection with recent binutils, from Markus
Trippelsdorf.
- Improve warning message when libunwind devel packages not present,
from Jiri Olsa"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (282 commits)
perf trace: Add aliases for some syscalls
perf probe: Print an enum type variable in "enum variable-name" format when showing accessible variables
perf tools: Check libaudit availability for perf-trace builtin
perf hists: Add missing period_* fields when collapsing a hist entry
perf trace: New tool
perf evsel: Export the event_format constructor
perf evsel: Introduce rawptr() method
perf tools: Use perf_evsel__newtp in the event parser
perf evsel: The tracepoint constructor should store sys:name
perf evlist: Introduce set_filter() method
perf evlist: Renane set_filters method to apply_filters
perf test: Add test to check we correctly parse and match syscall open parms
perf evsel: Handle endianity in intval method
perf evsel: Know if byte swap is needed
perf tools: Allow handling a NULL cpu_map as meaning "all cpus"
perf evsel: Improve tracepoint constructor setup
tools lib traceevent: Fix error path on pevent_parse_event
perf test: Fix build failure
trace: Move trace event enable from fs_initcall to core_initcall
tracing: Add an option for disabling markers
...
Pull core locking changes from Ingo Molnar:
"It includes a lockdep improvement plus a spinlock inlining Kconfig
cleanup."
* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking: Adjust spin lock inlining Kconfig options
lockdep: Check if nested lock is actually held
Pull core kernel fixes from Ingo Molnar:
"This is a complex task_work series from Oleg that fixes the bug that
this VFS commit tried to fix:
d35abdb288 hold task_lock around checks in keyctl
but solves the problem without the lockup regression that d35abdb288
introduced in v3.6.
This series came late in v3.6 and I did not feel confident about it so
late in the cycle. Might be worth backporting to -stable if it proves
itself upstream."
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
task_work: Simplify the usage in ptrace_notify() and get_signal_to_deliver()
task_work: Revert "hold task_lock around checks in keyctl"
task_work: task_work_add() should not succeed after exit_task_work()
task_work: Make task_work_add() lockless
Make default just return 0. The current default (checking
TIF_POLLING_NRFLAG) is taken to architectures that need it;
ones that don't do polling in their idle threads don't need
to defined TIF_POLLING_NRFLAG at all.
ia64 defined both TS_POLLING (used by its tsk_is_polling())
and TIF_POLLING_NRFLAG (not used at all). Killed the latter...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Let architectures select GENERIC_KERNEL_THREAD and have their copy_thread()
treat NULL regs as "it came from kernel_thread(), sp argument contains
the function new thread will be calling and stack_size - the argument for
that function". Switching the architectures begins shortly...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
After the previous change is_swbp_at_addr() is always called with
current->mm. Remove this check and move it close to its single caller.
Also, remove the obsolete comment about is_swbp_at_addr() and
uprobe_state.count.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Unlike set_swbp(), set_orig_insn()->is_swbp_at_addr() makes sense,
although it can't prevent all confusions.
But the usage of is_swbp_at_addr() is equally confusing, and it adds
the extra get_user_pages() we can avoid.
This patch removes set_orig_insn()->is_swbp_at_addr() but changes
write_opcode() to do the necessary checks before replace_page().
Perhaps it also makes sense to ensure PAGE_MAPPING_ANON in unregister
case.
find_active_uprobe() becomes the only user of is_swbp_at_addr(),
we can change its semantics.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
No functional changes, preparations.
1. Extract the kmap-and-memcpy code from read_opcode() into the
new trivial helper, copy_opcode(). The next patch will add
another user.
2. read_opcode() becomes really trivial, fold it into its single
caller, is_swbp_at_addr().
3. Remove "auprobe" argument from write_opcode(), it is not used
since f403072c6.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
A separate patch for better documentation.
set_swbp()->is_swbp_at_addr() is not needed for correctness, it is
harmless to do the unnecessary __replace_page(old_page, new_page)
when these 2 pages are identical.
And it can not be counted as optimization. mmap/register races are
very unlikely, while in the likely case is_swbp_at_addr() adds the
extra get_user_pages() even if the caller is uprobe_mmap(current->mm)
and returns false.
Note also that the semantics/usage of is_swbp_at_addr() in uprobe.c
is confusing. set_swbp() uses it to detect the case when this insn
was already modified by uprobes, that is why it should always compare
the opcode with UPROBE_SWBP_INSN even if the hardware (like powerpc)
has other trap insns. It doesn't matter if this breakpoint was in fact
installed by gdb or application itself, we are going to "steal" this
breakpoint anyway and execute the original insn from vm_file even if
it no longer matches the memory.
OTOH, handle_swbp()->find_active_uprobe() uses is_swbp_at_addr() to
figure out whether we need to send SIGTRAP or not if we can not find
uprobe, so in this case it should return true for all trap variants,
not only for UPROBE_SWBP_INSN.
This patch removes set_swbp()->is_swbp_at_addr(), the next patches
will remove it from set_orig_insn() which is similar to set_swbp()
in this respect. So the only caller will be handle_swbp() and we
can make its semantics clear.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
valid_vma(false) ignores ->vm_flags, this is not actually right.
We should never try to write into MAP_SHARED mapping, this can
confuse an apllication which actually writes to ->vm_file.
With this patch valid_vma(false) ignores VM_WRITE only but checks
other (immutable) bits checked by valid_vma(true). This can also
speedup uprobe_munmap() and uprobe_unregister().
Note: even after this patch _unregister can confuse the probed
application if it does mprotect(PROT_WRITE) after _register and
installs "int3", but this is hardly possible to avoid and this
doesn't differ from gdb case.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
uprobe_register() or uprobe_mmap() requires VM_READ | VM_EXEC, this
is not right. An apllication can do mprotect(PROT_EXEC) later and
execute this code.
Change valid_vma(is_register => true) to check VM_MAYEXEC instead.
No need to check VM_MAYREAD, it is always set.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
write_opcode()->get_user_pages() needs FOLL_FORCE to ensure we can
read the page even if the probed task did mprotect(PROT_NONE) after
uprobe_register(). Without FOLL_WRITE, FOLL_FORCE doesn't have any
side effect but allows to read the !VM_READ memory.
Otherwiese the subsequent uprobe_unregister()->set_orig_insn() fails
and we leak "int3". If that task does mprotect(PROT_READ | EXEC) and
execute the probed insn later it will be killed.
Note: in fact this is also needed for _register, see the next patch.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Kill UTASK_BP_HIT state, it buys nothing but complicates the code.
It is only used in uprobe_notify_resume() to decide who should be
called, we can check utask->active_uprobe != NULL instead. And this
allows us to simplify handle_swbp(), no need to clear utask->state.
Likewise we could kill UTASK_SSTEP, but UTASK_BP_HIT is worse and
imho should die. The problem is, it creates the special case when
task->utask is NULL, we can't distinguish RUNNING and BP_HIT. With
this patch utask == NULL always means RUNNING.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
If handle_swbp()->add_utask() fails but UPROBE_SKIP_SSTEP is set,
cleanup_ret: path do not restart the insn, this is wrong. Remove
this check and add the additional label for can_skip_sstep() = T
case.
Note also that UPROBE_SKIP_SSTEP can be false positive, we simply
can not trust it unless arch_uprobe_skip_sstep() was already called.
Also, move another UPROBE_SKIP_SSTEP check before can_skip_sstep()
into this helper, this looks more clean and understandable.
Note: probably we should rename "skip" to "emulate" and I think
that "clear UPROBE_SKIP_SSTEP" should be moved to arch_can_skip.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
handle_swbp() sets utask->active_uprobe before handler_chain(),
and UTASK_SSTEP before pre_ssout(). This complicates the code
for no reason, arch_ hooks or consumer->handler() should not
(and can't) use this info.
Change handle_swbp() to initialize them after pre_ssout(), and
remove the no longer needed cleanup-utask code.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
cked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
If handle_swbp()->find_active_uprobe() fails we return with
utask->state = UTASK_BP_HIT.
Change handle_swbp() to reset utask->state at the start. Note
that we do this unconditionally, see the next patch(es).
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Conflicts:
drivers/net/team/team.c
drivers/net/usb/qmi_wwan.c
net/batman-adv/bat_iv_ogm.c
net/ipv4/fib_frontend.c
net/ipv4/route.c
net/l2tp/l2tp_netlink.c
The team, fib_frontend, route, and l2tp_netlink conflicts were simply
overlapping changes.
qmi_wwan and bat_iv_ogm were of the "use HEAD" variety.
With help from Antonio Quartulli.
Signed-off-by: David S. Miller <davem@davemloft.net>
Use generic steal operation on pipe buffer to allow stealing
ring buffer's read page from pipe buffer.
Note that this could reduce the performance of splice on the
splice_write side operation without affinity setting.
Since the ring buffer's read pages are allocated on the
tracing-node, but the splice user does not always execute
splice write side operation on the same node. In this case,
the page will be accessed from the another node.
Thus, it is strongly recommended to assign the splicing
thread to corresponding node.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The original module-init-tools module loader used a fnctl lock on the
.ko file to avoid attempts to simultaneously load a module.
Unfortunately, you can't get an exclusive fcntl lock on a read-only
fd, making this not work for read-only mounted filesystems.
module-init-tools has a hacky sleep-and-loop for this now.
It's not that hard to wait in the kernel, and only return -EEXIST once
the first module has finished loading (or continue loading the module
if the first one failed to initialize for some reason). It's also
consistent with what we do for dependent modules which are still loading.
Suggested-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We use resolve_symbol_wait(), which blocks if the module containing
the symbol is still loading. However:
1) The module_wq we use is only woken after calling the modules' init
function, but there are other failure paths after the module is
placed in the linked list where we need to do the same thing.
2) wake_up() only wakes one waiter, and our waitqueue is shared by all
modules, so we need to wake them all.
3) wake_up_all() doesn't imply a memory barrier: I feel happier calling
it after we've grabbed and dropped the module_mutex, not just after
the state assignment.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Use the mapping of Elf_[SPE]hdr, Elf_Addr, Elf_Sym, Elf_Dyn, Elf_Rel/Rela,
ELF_R_TYPE() and ELF_R_SYM() to either the 32-bit version or the 64-bit version
into asm-generic/module.h for all arches bar MIPS.
Also, use the generic definition mod_arch_specific where possible.
To this end, I've defined three new config bools:
(*) HAVE_MOD_ARCH_SPECIFIC
Arches define this if they don't want to use the empty generic
mod_arch_specific struct.
(*) MODULES_USE_ELF_RELA
Arches define this if their modules can contain RELA records. This causes
the Elf_Rela mapping to be emitted and allows apply_relocate_add() to be
defined by the arch rather than have the core emit an error message.
(*) MODULES_USE_ELF_REL
Arches define this if their modules can contain REL records. This causes
the Elf_Rel mapping to be emitted and allows apply_relocate() to be
defined by the arch rather than have the core emit an error message.
Note that it is possible to allow both REL and RELA records: m68k and mips are
two arches that do this.
With this, some arch asm/module.h files can be deleted entirely and replaced
with a generic-y marker in the arch Kbuild file.
Additionally, I have removed the bits from m32r and score that handle the
unsupported type of relocation record as that's now handled centrally.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cloudlinux have a product called lve that includes a kernel module. This
was previously GPLed but is now under a proprietary license, but the
module continues to declare MODULE_LICENSE("GPL") and makes use of some
EXPORT_SYMBOL_GPL symbols. Forcibly taint it in order to avoid this.
Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org>
Cc: Alex Lyashkov <umka@cloudlinux.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)
iQEcBAABAgAGBQJQX7MuAAoJEHm+PkMAQRiG0h0IAJURkrMCAQUxA+Ik66ReH89s
LQcVd0U9uL4UUOi7f5WR64Vf9Cfu6VVGX9ZKSvjpNskvlQaUQPMIt4pMe6g4X4dI
u0bApEy4XZz3nGabUAghIU8jJ8cDmhCG6kPpSiS7pi7KHc0yIa4WFtJRrIpGaIWT
xuK38YOiOHcSDRlLyWZzainMncQp/ixJdxnqVMTonkVLk0q0b84XzOr4/qlLE5lU
i+TsK3PRKdQXgvZ4CebL+srPBwWX1dmgP3VkeBloQbSSenSeELICbFWavn2ml+sF
GXi4dO93oNquL/Oy5SwI666T4uNcrRPaS+5X+xSZgBW/y2aQVJVJuNZg6ZP/uWk=
=0v2l
-----END PGP SIGNATURE-----
Merge tag 'v3.6-rc7' into next
Linux 3.6-rc7
Requested by David Howells so he can merge his key susbsystem work into
my tree with requisite -linus changesets.
descriptor-related parts of daemonize, done right. As the
result we simplify the locking rules for ->files - we
hold task_lock in *all* cases when we modify ->files.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This command disables NMI-entry. If NMI source has been previously shared
with a serial console ("debug port"), this effectively releases the port
from KDB exclusive use, and makes the console available for normal use.
Of course, NMI can be reenabled, enable_nmi modparam is used for that:
echo 1 > /sys/module/kdb/parameters/enable_nmi
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The new arch callback should manage NMIs that usually cause KGDB to
enter. That is, not all NMIs should be enabled/disabled, but only
those that issue kgdb_handle_exception().
We must mask it as serial-line interrupt can be used as an NMI, so
if the original KGDB-entry cause was say a breakpoint, then every
input to KDB console will cause KGDB to reenter, which we don't want.
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Checking "user" before "is_idle_task()" allows better optimizations
in cases where inlining is possible. Also, "bool" should be passed
"true" or "false" rather than "1" or "0". This commit therefore makes
these changes, as noted in Josh's review.
Reported-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Provide a config option that enables the userspace
RCU extended quiescent state on every CPUs by default.
This is for testing purpose.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
When exceptions or irq are about to resume userspace, if
the task needs to be rescheduled, the arch low level code
calls schedule() directly.
If we call it, it is because we have the TIF_RESCHED flag:
- It can be set after random local calls to set_need_resched()
(RCU, drm, ...)
- A wake up happened and the CPU needs preemption. This can
happen in several ways:
* Remotely: the remote waking CPU has set TIF_RESCHED and send the
wakee an IPI to schedule the new task.
* Remotely enqueued: the remote waking CPU sends an IPI to the target
and the wake up is made by the target.
* Locally: waking CPU == wakee CPU and the wakeup is done locally.
set_need_resched() is called without IPI.
In the case of local and remotely enqueued wake ups, the tick can
be restarted when we enqueue the new task and RCU can exit the
extended quiescent state at the same time. Then by the time we reach
irq exit path and we call schedule, we are not in RCU user mode.
But if we call schedule() only because something called set_need_resched(),
RCU may still be in user mode when we reach schedule.
Also if a wake up is done remotely, the CPU might see the TIF_RESCHED
flag and call schedule while the IPI has not yet happen to restart the
tick and exit RCU user mode.
We need to manually protect against these corner cases.
Create a new API schedule_user() that calls schedule() inside
rcu_user_exit()-rcu_user_enter() in order to protect it. Archs
will need to rely on it now to implement user preemption safely.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
When an exception or an irq exits, and we are going to resume into
interrupted kernel code, the low level architecture code calls
preempt_schedule_irq() if there is a need to reschedule.
If the interrupt/exception occured between a call to rcu_user_enter()
(from syscall exit, exception exit, do_notify_resume exit, ...) and
a real resume to userspace (iret,...), preempt_schedule_irq() can be
called whereas RCU thinks we are in userspace. But preempt_schedule_irq()
is going to run kernel code and may be some RCU read side critical
section. We must exit the userspace extended quiescent state before
we call it.
To solve this, just call rcu_user_exit() in the beginning of
preempt_schedule_irq().
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Clear the syscalls hook of a task when it's scheduled out so that if
the task migrates, it doesn't run the syscall slow path on a CPU
that might not need it.
Also set the syscalls hook on the next task if needed.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
By default we don't want to enter into RCU extended quiescent
state while in userspace because doing this produces some overhead
(eg: use of syscall slowpath). Set it off by default and ready to
run when some feature like adaptive tickless need it.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Allow calls to rcu_user_enter() even if we are already
in userspace (as seen by RCU) and allow calls to rcu_user_exit()
even if we are already in the kernel.
This makes the APIs more flexible to be called from architectures.
Exception entries for example won't need to know if they come from
userspace before calling rcu_user_exit().
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Create a new config option under the RCU menu that put
CPUs under RCU extended quiescent state (as in dynticks
idle mode) when they run in userspace. This require
some contribution from architectures to hook into kernel
and userspace boundaries.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
The current implementation of RCU_FAST_NO_HZ tries reasonably hard to rid
the current CPU of RCU callbacks. This is appropriate when the CPU is
entering idle, where it doesn't have much useful to do anyway, but is most
definitely not what you want when transitioning to user-mode execution.
This commit therefore detects the adaptive-tick case, and refrains from
burning CPU time getting rid of RCU callbacks in that case.
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
In some cases, it is necessary to enter or exit userspace-RCU-idle mode
from an interrupt handler, for example, if some other CPU sends this
CPU a resched IPI. In this case, the current CPU would enter the IPI
handler in userspace-RCU-idle mode, but would need to exit the IPI handler
after having exited that mode.
To allow this to work, this commit adds two new APIs to TREE_RCU:
- rcu_user_enter_after_irq(). This must be called from an interrupt between
rcu_irq_enter() and rcu_irq_exit(). After the irq calls rcu_irq_exit(),
the irq handler will return into an RCU extended quiescent state.
In theory, this interrupt is never a nested interrupt, but in practice
it might interrupt softirq, which looks to RCU like a nested interrupt.
- rcu_user_exit_after_irq(). This must be called from a non-nesting
interrupt, interrupting an RCU extended quiescent state, also
between rcu_irq_enter() and rcu_irq_exit(). After the irq calls
rcu_irq_exit(), the irq handler will return in an RCU non-quiescent
state.
[ Combined with "Allow calls to rcu_exit_user_irq from nesting irqs." ]
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
RCU currently insists that only idle tasks can enter RCU idle mode, which
prohibits an adaptive tickless kernel (AKA nohz cpusets), which in turn
would mean that usermode execution would always take scheduling-clock
interrupts, even when there is only one task runnable on the CPU in
question.
This commit therefore adds rcu_user_enter() and rcu_user_exit(), which
allow non-idle tasks to enter RCU idle mode. These are quite similar
to rcu_idle_enter() and rcu_idle_exit(), respectively, except that they
omit the idle-task checks.
[ Updated to use "user" flag rather than separate check functions. ]
[ paulmck: Updated to drop exports of new functions based on Josh's patch
getting rid of the need for them. ]
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
The conflicts between kernel/rcutree.h and kernel/rcutree_plugin.h
were due to adjacent insertions and deletions, which were resolved
by simply accepting the changes on both branches.
Move the code that finds out to which context we account the
cputime into generic layer.
Archs that consider the whole time spent in the idle task as idle
time (ia64, powerpc) can rely on the generic vtime_account()
and implement vtime_account_system() and vtime_account_idle(),
letting the generic code to decide when to call which API.
Archs that have their own meaning of idle time, such as s390
that only considers the time spent in CPU low power mode as idle
time, can just override vtime_account().
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Use a naming based on vtime as a prefix for virtual based
cputime accounting APIs:
- account_system_vtime() -> vtime_account()
- account_switch_vtime() -> vtime_task_switch()
It makes it easier to allow for further declension such
as vtime_account_system(), vtime_account_idle(), ... if we
want to find out the context we account to from generic code.
This also make it better to know on which subsystem these APIs
refer to.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
bigrt.2012.09.23a contains additional commits to reduce scheduling latency
from RCU on huge systems (many hundrends or thousands of CPUs).
doctorture.2012.09.23a contains documentation changes and rcutorture fixes.
fixes.2012.09.23a contains miscellaneous fixes.
hotplug.2012.09.23a contains CPU-hotplug-related changes.
idle.2012.09.23a fixes architectures for which RCU no longer considered
the idle loop to be a quiescent state due to earlier
adaptive-dynticks changes. Affected architectures are alpha,
cris, frv, h8300, m32r, m68k, mn10300, parisc, score, xtensa,
and ia64.
We currently use a per socket order-0 page cache for tcp_sendmsg()
operations.
This page is used to build fragments for skbs.
Its done to increase probability of coalescing small write() into
single segments in skbs still in write queue (not yet sent)
But it wastes a lot of memory for applications handling many mostly
idle sockets, since each socket holds one page in sk->sk_sndmsg_page
Its also quite inefficient to build TSO 64KB packets, because we need
about 16 pages per skb on arches where PAGE_SIZE = 4096, so we hit
page allocator more than wanted.
This patch adds a per task frag allocator and uses bigger pages,
if available. An automatic fallback is done in case of memory pressure.
(up to 32768 bytes per frag, thats order-3 pages on x86)
This increases TCP stream performance by 20% on loopback device,
but also benefits on other network devices, since 8x less frags are
mapped on transmit and unmapped on tx completion. Alexander Duyck
mentioned a probable performance win on systems with IOMMU enabled.
Its possible some SG enabled hardware cant cope with bigger fragments,
but their ndo_start_xmit() should already handle this, splitting a
fragment in sub fragments, since some arches have PAGE_SIZE=65536
Successfully tested on various ethernet devices.
(ixgbe, igb, bnx2x, tg3, mellanox mlx4)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Vijay Subramanian <subramanian.vijay@gmail.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Vijay Subramanian <subramanian.vijay@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch splits trace event initialization in two stages:
* ftrace enable
* sysfs event entry creation
This allows to capture trace events from an earlier point
by using 'trace_event' kernel parameter and is important
to trace boot-up allocations.
Note that, in order to enable events at core_initcall,
it's necessary to move init_ftrace_syscalls() from
core_initcall to early_initcall.
Link: http://lkml.kernel.org/r/1347461277-25302-1-git-send-email-elezegarcia@gmail.com
Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
