linux.git/kernel/trace/trace.c, branch v5.9-rc7

tracing: Make the space reserved for the pid wider

2020-09-18T16:42:11+00:00

For 64bit CONFIG_BASE_SMALL=0 systems PID_MAX_LIMIT is set by default to
4194304. During boot the kernel sets a new value based on number of CPUs
but no lower than 32768. It is 1024 per CPU so with 128 CPUs the default
becomes 131072 which needs six digits.
This value can be increased during run time but must not exceed the
initial upper limit.

Systemd sometime after v241 sets it to the upper limit during boot. The
result is that when the pid exceeds five digits, the trace output is a
little hard to read because it is no longer properly padded (same like
on big iron with 98+ CPUs).

Increase the pid padding to seven digits.

Link: https://lkml.kernel.org/r/20200904082331.dcdkrr3bkn3e4qlg@linutronix.de

Signed-off-by: Sebastian Andrzej Siewior 
Signed-off-by: Steven Rostedt (VMware)

Merge tag 'trace-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

2020-08-08T01:29:15+00:00

Pull tracing updates from Steven Rostedt:

 - The biggest news in that the tracing ring buffer can now time events
   that interrupted other ring buffer events.

   Before this change, if an interrupt came in while recording another
   event, and that interrupt also had an event, those events would all
   have the same time stamp as the event it interrupted.

   Now, with the new design, those events will have a unique time stamp
   and rightfully display the time for those events that were recorded
   while interrupting another event.

 - Bootconfig how has an "override" operator that lets the users have a
   default config, but then add options to override the default.

 - A fix was made to properly filter function graph tracing to the
   ftrace PIDs. This came in at the end of the -rc cycle, and needs to
   be backported.

 - Several clean ups, performance updates, and minor fixes as well.

* tag 'trace-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (39 commits)
  tracing: Add trace_array_init_printk() to initialize instance trace_printk() buffers
  kprobes: Fix compiler warning for !CONFIG_KPROBES_ON_FTRACE
  tracing: Use trace_sched_process_free() instead of exit() for pid tracing
  bootconfig: Fix to find the initargs correctly
  Documentation: bootconfig: Add bootconfig override operator
  tools/bootconfig: Add testcases for value override operator
  lib/bootconfig: Add override operator support
  kprobes: Remove show_registers() function prototype
  tracing/uprobe: Remove dead code in trace_uprobe_register()
  kprobes: Fix NULL pointer dereference at kprobe_ftrace_handler
  ftrace: Fix ftrace_trace_task return value
  tracepoint: Use __used attribute definitions from compiler_attributes.h
  tracepoint: Mark __tracepoint_string's __used
  trace : Have tracing buffer info use kvzalloc instead of kzalloc
  tracing: Remove outdated comment in stack handling
  ftrace: Do not let direct or IPMODIFY ftrace_ops be added to module and set trampolines
  ftrace: Setup correct FTRACE_FL_REGS flags for module
  tracing/hwlat: Honor the tracing_cpumask
  tracing/hwlat: Drop the duplicate assignment in start_kthread()
  tracing: Save one trace_event->type by using __TRACE_LAST_TYPE
  ...

tracing: Add trace_array_init_printk() to initialize instance trace_printk() buffers

2020-08-07T21:05:01+00:00

As trace_array_printk() used with not global instances will not add noise to
the main buffer, they are OK to have in the kernel (unlike trace_printk()).
This require the subsystem to create their own tracing instance, and the
trace_array_printk() only writes into those instances.

Add trace_array_init_printk() to initialize the trace_printk() buffers
without printing out the WARNING message.

Reported-by: Sean Paul 
Reviewed-by: Sean Paul 
Signed-off-by: Steven Rostedt (VMware)

Merge tag 'fsnotify_for_v5.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs

2020-08-07T02:29:51+00:00

Pull fsnotify updates from Jan Kara:

 - fanotify fix for softlockups when there are many queued events

 - performance improvement to reduce fsnotify overhead when not used

 - Amir's implementation of fanotify events with names. With these you
   can now efficiently monitor whole filesystem, eg to mirror changes to
   another machine.

* tag 'fsnotify_for_v5.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs: (37 commits)
  fanotify: compare fsid when merging name event
  fsnotify: create method handle_inode_event() in fsnotify_operations
  fanotify: report parent fid + child fid
  fanotify: report parent fid + name + child fid
  fanotify: add support for FAN_REPORT_NAME
  fanotify: report events with parent dir fid to sb/mount/non-dir marks
  fanotify: add basic support for FAN_REPORT_DIR_FID
  fsnotify: remove check that source dentry is positive
  fsnotify: send event with parent/name info to sb/mount/non-dir marks
  audit: do not set FS_EVENT_ON_CHILD in audit marks mask
  inotify: do not set FS_EVENT_ON_CHILD in non-dir mark mask
  fsnotify: pass dir and inode arguments to fsnotify()
  fsnotify: create helper fsnotify_inode()
  fsnotify: send event to parent and child with single callback
  inotify: report both events on parent and child with single callback
  dnotify: report both events on parent and child with single callback
  fanotify: no external fh buffer in fanotify_name_event
  fanotify: use struct fanotify_info to parcel the variable size buffer
  fsnotify: add object type "child" to object type iterator
  fanotify: use FAN_EVENT_ON_CHILD as implicit flag on sb/mount/non-dir marks
  ...

trace : Have tracing buffer info use kvzalloc instead of kzalloc

2020-08-03T15:52:20+00:00

High order memory stuff within trace could introduce OOM, use kvzalloc instead.

Please find the bellowing for the call stack we run across in an android system.
The scenario happens when traced_probes is woken up to get a large quantity of
trace even if free memory is even higher than watermark_low. 

traced_probes invoked oom-killer: gfp_mask=0x140c0c0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),  order=2, oom_score_adj=-1

traced_probes cpuset=system-background mems_allowed=0
CPU: 3 PID: 588 Comm: traced_probes Tainted: G        W  O    4.14.181 #1
Hardware name: Generic DT based system
(unwind_backtrace) from [] (show_stack+0x20/0x24)
(show_stack) from [] (dump_stack+0xa8/0xec)
(dump_stack) from [] (dump_header+0x9c/0x220)
(dump_header) from [] (oom_kill_process+0xc0/0x5c4)
(oom_kill_process) from [] (out_of_memory+0x220/0x310)
(out_of_memory) from [] (__alloc_pages_nodemask+0xff8/0x13a4)
(__alloc_pages_nodemask) from [] (kmalloc_order+0x30/0x48)
(kmalloc_order) from [] (kmalloc_order_trace+0x30/0x118)
(kmalloc_order_trace) from [] (tracing_buffers_open+0x50/0xfc)
(tracing_buffers_open) from [] (do_dentry_open+0x278/0x34c)
(do_dentry_open) from [] (vfs_open+0x50/0x70)
(vfs_open) from [] (path_openat+0x5fc/0x169c)
(path_openat) from [] (do_filp_open+0x94/0xf8)
(do_filp_open) from [] (do_sys_open+0x168/0x26c)
(do_sys_open) from [] (SyS_openat+0x34/0x38)
(SyS_openat) from [] (ret_fast_syscall+0x0/0x28)

Link: https://lkml.kernel.org/r/1596155265-32365-1-git-send-email-zhaoyang.huang@unisoc.com

Signed-off-by: Zhaoyang Huang 
Signed-off-by: Steven Rostedt (VMware)

tracing: Remove outdated comment in stack handling

2020-07-31T02:54:50+00:00

This comment describes the behaviour before commit 2a820bf74918
("tracing: Use percpu stack trace buffer more intelligently").  Since
that commit, interrupts and NMIs do use the per-cpu stacks so the
comment is no longer correct.  Remove it.

(Note that the FTRACE_STACK_SIZE mentioned in the comment has never
existed, it probably should have said FTRACE_STACK_ENTRIES.)

Link: https://lkml.kernel.org/r/20200727092840.18659-1-vincent.whitchurch@axis.com

Signed-off-by: Vincent Whitchurch 
Signed-off-by: Steven Rostedt (VMware)

fsnotify: create helper fsnotify_inode()

2020-07-27T21:13:51+00:00

Simple helper to consolidate biolerplate code.

Link: https://lore.kernel.org/r/20200722125849.17418-5-amir73il@gmail.com
Signed-off-by: Amir Goldstein 
Signed-off-by: Jan Kara

tracefs: Remove unnecessary debug_fs checks.

2020-07-23T15:10:25+00:00

This is a preparation for debugfs restricted mode.
We don't need debugfs to trace, the removed check stop tracefs to work
if debugfs is not initialised. We instead tries to automount within
debugfs and relay on it's handling. The code path is to create a
backward compatibility from when tracefs was part of debugfs, it is now
standalone and does not need debugfs. When debugfs is in restricted
it is compiled in but not active and return EPERM to clients and
tracefs wont work if it assumes it is active it is compiled in
kernel.

Reported-by: kernel test robot 
Signed-off-by: Peter Enderborg 
Reviewed-by: Greg Kroah-Hartman 
Acked-by: Steven Rostedt (VMware) 
Link: https://lore.kernel.org/r/20200716071511.26864-2-peter.enderborg@sony.com
Signed-off-by: Greg Kroah-Hartman

ring-buffer: speed up buffer resets by avoiding synchronize_rcu for each CPU

2020-06-30T21:18:56+00:00

On a 144 thread system, `perf ftrace` takes about 20 seconds to start
up, due to calling synchronize_rcu() for each CPU.

  cat /proc/108560/stack
    0xc0003e7eb336f470
    __switch_to+0x2e0/0x480
    __wait_rcu_gp+0x20c/0x220
    synchronize_rcu+0x9c/0xc0
    ring_buffer_reset_cpu+0x88/0x2e0
    tracing_reset_online_cpus+0x84/0xe0
    tracing_open+0x1d4/0x1f0

On a system with 10x more threads, it starts to become an annoyance.

Batch these up so we disable all the per-cpu buffers first, then
synchronize_rcu() once, then reset each of the buffers. This brings
the time down to about 0.5s.

Link: https://lkml.kernel.org/r/20200625053403.2386972-1-npiggin@gmail.com

Tested-by: Anton Blanchard 
Acked-by: Paul E. McKenney 
Signed-off-by: Nicholas Piggin 
Signed-off-by: Steven Rostedt (VMware)

tracing: Move pipe reference to trace array instead of current_tracer

2020-06-30T18:29:33+00:00

If a process has the trace_pipe open on a trace_array, the current tracer
for that trace array should not be changed. This was original enforced by a
global lock, but when instances were introduced, it was moved to the
current_trace. But this structure is shared by all instances, and a
trace_pipe is for a single instance. There's no reason that a process that
has trace_pipe open on one instance should prevent another instance from
changing its current tracer. Move the reference counter to the trace_array
instead.

This is marked as "Fixes" but is more of a clean up than a true fix.
Backport if you want, but its not critical.

Fixes: cf6ab6d9143b1 ("tracing: Add ref count to tracer for when they are being read by pipe")
Signed-off-by: Steven Rostedt (VMware)