summaryrefslogtreecommitdiffstats
path: root/kernel/trace
AgeCommit message (Collapse)Author
2019-03-13bpf: fix potential deadlock in bpf_prog_registerAlexei Starovoitov
[ Upstream commit e16ec34039c701594d55d08a5aa49ee3e1abc821 ] Lockdep found a potential deadlock between cpu_hotplug_lock, bpf_event_mutex, and cpuctx_mutex: [ 13.007000] WARNING: possible circular locking dependency detected [ 13.007587] 5.0.0-rc3-00018-g2fa53f892422-dirty #477 Not tainted [ 13.008124] ------------------------------------------------------ [ 13.008624] test_progs/246 is trying to acquire lock: [ 13.009030] 0000000094160d1d (tracepoints_mutex){+.+.}, at: tracepoint_probe_register_prio+0x2d/0x300 [ 13.009770] [ 13.009770] but task is already holding lock: [ 13.010239] 00000000d663ef86 (bpf_event_mutex){+.+.}, at: bpf_probe_register+0x1d/0x60 [ 13.010877] [ 13.010877] which lock already depends on the new lock. [ 13.010877] [ 13.011532] [ 13.011532] the existing dependency chain (in reverse order) is: [ 13.012129] [ 13.012129] -> #4 (bpf_event_mutex){+.+.}: [ 13.012582] perf_event_query_prog_array+0x9b/0x130 [ 13.013016] _perf_ioctl+0x3aa/0x830 [ 13.013354] perf_ioctl+0x2e/0x50 [ 13.013668] do_vfs_ioctl+0x8f/0x6a0 [ 13.014003] ksys_ioctl+0x70/0x80 [ 13.014320] __x64_sys_ioctl+0x16/0x20 [ 13.014668] do_syscall_64+0x4a/0x180 [ 13.015007] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 13.015469] [ 13.015469] -> #3 (&cpuctx_mutex){+.+.}: [ 13.015910] perf_event_init_cpu+0x5a/0x90 [ 13.016291] perf_event_init+0x1b2/0x1de [ 13.016654] start_kernel+0x2b8/0x42a [ 13.016995] secondary_startup_64+0xa4/0xb0 [ 13.017382] [ 13.017382] -> #2 (pmus_lock){+.+.}: [ 13.017794] perf_event_init_cpu+0x21/0x90 [ 13.018172] cpuhp_invoke_callback+0xb3/0x960 [ 13.018573] _cpu_up+0xa7/0x140 [ 13.018871] do_cpu_up+0xa4/0xc0 [ 13.019178] smp_init+0xcd/0xd2 [ 13.019483] kernel_init_freeable+0x123/0x24f [ 13.019878] kernel_init+0xa/0x110 [ 13.020201] ret_from_fork+0x24/0x30 [ 13.020541] [ 13.020541] -> #1 (cpu_hotplug_lock.rw_sem){++++}: [ 13.021051] static_key_slow_inc+0xe/0x20 [ 13.021424] tracepoint_probe_register_prio+0x28c/0x300 [ 13.021891] perf_trace_event_init+0x11f/0x250 [ 13.022297] perf_trace_init+0x6b/0xa0 [ 13.022644] perf_tp_event_init+0x25/0x40 [ 13.023011] perf_try_init_event+0x6b/0x90 [ 13.023386] perf_event_alloc+0x9a8/0xc40 [ 13.023754] __do_sys_perf_event_open+0x1dd/0xd30 [ 13.024173] do_syscall_64+0x4a/0x180 [ 13.024519] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 13.024968] [ 13.024968] -> #0 (tracepoints_mutex){+.+.}: [ 13.025434] __mutex_lock+0x86/0x970 [ 13.025764] tracepoint_probe_register_prio+0x2d/0x300 [ 13.026215] bpf_probe_register+0x40/0x60 [ 13.026584] bpf_raw_tracepoint_open.isra.34+0xa4/0x130 [ 13.027042] __do_sys_bpf+0x94f/0x1a90 [ 13.027389] do_syscall_64+0x4a/0x180 [ 13.027727] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 13.028171] [ 13.028171] other info that might help us debug this: [ 13.028171] [ 13.028807] Chain exists of: [ 13.028807] tracepoints_mutex --> &cpuctx_mutex --> bpf_event_mutex [ 13.028807] [ 13.029666] Possible unsafe locking scenario: [ 13.029666] [ 13.030140] CPU0 CPU1 [ 13.030510] ---- ---- [ 13.030875] lock(bpf_event_mutex); [ 13.031166] lock(&cpuctx_mutex); [ 13.031645] lock(bpf_event_mutex); [ 13.032135] lock(tracepoints_mutex); [ 13.032441] [ 13.032441] *** DEADLOCK *** [ 13.032441] [ 13.032911] 1 lock held by test_progs/246: [ 13.033239] #0: 00000000d663ef86 (bpf_event_mutex){+.+.}, at: bpf_probe_register+0x1d/0x60 [ 13.033909] [ 13.033909] stack backtrace: [ 13.034258] CPU: 1 PID: 246 Comm: test_progs Not tainted 5.0.0-rc3-00018-g2fa53f892422-dirty #477 [ 13.034964] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 [ 13.035657] Call Trace: [ 13.035859] dump_stack+0x5f/0x8b [ 13.036130] print_circular_bug.isra.37+0x1ce/0x1db [ 13.036526] __lock_acquire+0x1158/0x1350 [ 13.036852] ? lock_acquire+0x98/0x190 [ 13.037154] lock_acquire+0x98/0x190 [ 13.037447] ? tracepoint_probe_register_prio+0x2d/0x300 [ 13.037876] __mutex_lock+0x86/0x970 [ 13.038167] ? tracepoint_probe_register_prio+0x2d/0x300 [ 13.038600] ? tracepoint_probe_register_prio+0x2d/0x300 [ 13.039028] ? __mutex_lock+0x86/0x970 [ 13.039337] ? __mutex_lock+0x24a/0x970 [ 13.039649] ? bpf_probe_register+0x1d/0x60 [ 13.039992] ? __bpf_trace_sched_wake_idle_without_ipi+0x10/0x10 [ 13.040478] ? tracepoint_probe_register_prio+0x2d/0x300 [ 13.040906] tracepoint_probe_register_prio+0x2d/0x300 [ 13.041325] bpf_probe_register+0x40/0x60 [ 13.041649] bpf_raw_tracepoint_open.isra.34+0xa4/0x130 [ 13.042068] ? __might_fault+0x3e/0x90 [ 13.042374] __do_sys_bpf+0x94f/0x1a90 [ 13.042678] do_syscall_64+0x4a/0x180 [ 13.042975] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 13.043382] RIP: 0033:0x7f23b10a07f9 [ 13.045155] RSP: 002b:00007ffdef42fdd8 EFLAGS: 00000202 ORIG_RAX: 0000000000000141 [ 13.045759] RAX: ffffffffffffffda RBX: 00007ffdef42ff70 RCX: 00007f23b10a07f9 [ 13.046326] RDX: 0000000000000070 RSI: 00007ffdef42fe10 RDI: 0000000000000011 [ 13.046893] RBP: 00007ffdef42fdf0 R08: 0000000000000038 R09: 00007ffdef42fe10 [ 13.047462] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000000 [ 13.048029] R13: 0000000000000016 R14: 00007f23b1db4690 R15: 0000000000000000 Since tracepoints_mutex will be taken in tracepoint_probe_register/unregister() there is no need to take bpf_event_mutex too. bpf_event_mutex is protecting modifications to prog array used in kprobe/perf bpf progs. bpf_raw_tracepoints don't need to take this mutex. Fixes: c4f6699dfcb8 ("bpf: introduce BPF_RAW_TRACEPOINT") Acked-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
2019-03-10tracing: Fix event filters and triggers to handle negative numbersPavel Tikhomirov
commit 6a072128d262d2b98d31626906a96700d1fc11eb upstream. Then tracing syscall exit event it is extremely useful to filter exit codes equal to some negative value, to react only to required errors. But negative numbers does not work: [root@snorch sys_exit_read]# echo "ret == -1" > filter bash: echo: write error: Invalid argument [root@snorch sys_exit_read]# cat filter ret == -1 ^ parse_error: Invalid value (did you forget quotes)? Similar thing happens when setting triggers. These is a regression in v4.17 introduced by the commit mentioned below, testing without these commit shows no problem with negative numbers. Link: http://lkml.kernel.org/r/20180823102534.7642-1-ptikhomirov@virtuozzo.com Cc: stable@vger.kernel.org Fixes: 80765597bc58 ("tracing: Rewrite filter logic to be simpler and faster") Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-02-27tracing: Fix number of entries in trace headerQuentin Perret
commit 9e7382153f80ba45a0bbcd540fb77d4b15f6e966 upstream. The following commit 441dae8f2f29 ("tracing: Add support for display of tgid in trace output") removed the call to print_event_info() from print_func_help_header_irq() which results in the ftrace header not reporting the number of entries written in the buffer. As this wasn't the original intent of the patch, re-introduce the call to print_event_info() to restore the orginal behaviour. Link: http://lkml.kernel.org/r/20190214152950.4179-1-quentin.perret@arm.com Acked-by: Joel Fernandes <joelaf@google.com> Cc: stable@vger.kernel.org Fixes: 441dae8f2f29 ("tracing: Add support for display of tgid in trace output") Signed-off-by: Quentin Perret <quentin.perret@arm.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-02-20tracing/uprobes: Fix output for multiple string argumentsAndreas Ziegler
commit 0722069a5374b904ec1a67f91249f90e1cfae259 upstream. When printing multiple uprobe arguments as strings the output for the earlier arguments would also include all later string arguments. This is best explained in an example: Consider adding a uprobe to a function receiving two strings as parameters which is at offset 0xa0 in strlib.so and we want to print both parameters when the uprobe is hit (on x86_64): $ echo 'p:func /lib/strlib.so:0xa0 +0(%di):string +0(%si):string' > \ /sys/kernel/debug/tracing/uprobe_events When the function is called as func("foo", "bar") and we hit the probe, the trace file shows a line like the following: [...] func: (0x7f7e683706a0) arg1="foobar" arg2="bar" Note the extra "bar" printed as part of arg1. This behaviour stacks up for additional string arguments. The strings are stored in a dynamically growing part of the uprobe buffer by fetch_store_string() after copying them from userspace via strncpy_from_user(). The return value of strncpy_from_user() is then directly used as the required size for the string. However, this does not take the terminating null byte into account as the documentation for strncpy_from_user() cleary states that it "[...] returns the length of the string (not including the trailing NUL)" even though the null byte will be copied to the destination. Therefore, subsequent calls to fetch_store_string() will overwrite the terminating null byte of the most recently fetched string with the first character of the current string, leading to the "accumulation" of strings in earlier arguments in the output. Fix this by incrementing the return value of strncpy_from_user() by one if we did not hit the maximum buffer size. Link: http://lkml.kernel.org/r/20190116141629.5752-1-andreas.ziegler@fau.de Cc: Ingo Molnar <mingo@redhat.com> Cc: stable@vger.kernel.org Fixes: 5baaa59ef09e ("tracing/probes: Implement 'memory' fetch method for uprobes") Acked-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Andreas Ziegler <andreas.ziegler@fau.de> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-02-15tracing: uprobes: Fix typo in pr_fmt stringAndreas Ziegler
commit ea6eb5e7d15e1838de335609994b4546e2abcaaf upstream. The subsystem-specific message prefix for uprobes was also "trace_kprobe: " instead of "trace_uprobe: " as described in the original commit message. Link: http://lkml.kernel.org/r/20190117133023.19292-1-andreas.ziegler@fau.de Cc: Ingo Molnar <mingo@redhat.com> Cc: stable@vger.kernel.org Acked-by: Masami Hiramatsu <mhiramat@kernel.org> Fixes: 7257634135c24 ("tracing/probe: Show subsystem name in messages") Signed-off-by: Andreas Ziegler <andreas.ziegler@fau.de> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-19tracing: Fix memory leak of instance function hash filtersSteven Rostedt (VMware)
commit 2840f84f74035e5a535959d5f17269c69fa6edc5 upstream. The following commands will cause a memory leak: # cd /sys/kernel/tracing # mkdir instances/foo # echo schedule > instance/foo/set_ftrace_filter # rmdir instances/foo The reason is that the hashes that hold the filters to set_ftrace_filter and set_ftrace_notrace are not freed if they contain any data on the instance and the instance is removed. Found by kmemleak detector. Cc: stable@vger.kernel.org Fixes: 591dffdade9f ("ftrace: Allow for function tracing instance to filter functions") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-19tracing: Fix memory leak in set_trigger_filter()Steven Rostedt (VMware)
commit 3cec638b3d793b7cacdec5b8072364b41caeb0e1 upstream. When create_event_filter() fails in set_trigger_filter(), the filter may still be allocated and needs to be freed. The caller expects the data->filter to be updated with the new filter, even if the new filter failed (we could add an error message by setting set_str parameter of create_event_filter(), but that's another update). But because the error would just exit, filter was left hanging and nothing could free it. Found by kmemleak detector. Cc: stable@vger.kernel.org Fixes: bac5fb97a173a ("tracing: Add and use generic set_trigger_filter() implementation") Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-19tracing: Fix memory leak in create_filter()Steven Rostedt (VMware)
commit b61c19209c2c35ea2a2fe502d484703686eba98c upstream. The create_filter() calls create_filter_start() which allocates a "parse_error" descriptor, but fails to call create_filter_finish() that frees it. The op_stack and inverts in predicate_parse() were also not freed. Found by kmemleak detector. Cc: stable@vger.kernel.org Fixes: 80765597bc587 ("tracing: Rewrite filter logic to be simpler and faster") Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-17bpf: fix check of allowed specifiers in bpf_trace_printkMartynas Pumputis
[ Upstream commit 1efb6ee3edea57f57f9fb05dba8dcb3f7333f61f ] A format string consisting of "%p" or "%s" followed by an invalid specifier (e.g. "%p%\n" or "%s%") could pass the check which would make format_decode (lib/vsprintf.c) to warn. Fixes: 9c959c863f82 ("tracing: Allow BPF programs to call bpf_trace_printk()") Reported-by: syzbot+1ec5c5ec949c4adaa0c4@syzkaller.appspotmail.com Signed-off-by: Martynas Pumputis <m@lambda.lt> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
2018-12-08tracing/fgraph: Fix set_graph_function from showing interruptsSteven Rostedt (VMware)
commit 5cf99a0f3161bc3ae2391269d134d6bf7e26f00e upstream. The tracefs file set_graph_function is used to only function graph functions that are listed in that file (or all functions if the file is empty). The way this is implemented is that the function graph tracer looks at every function, and if the current depth is zero and the function matches something in the file then it will trace that function. When other functions are called, the depth will be greater than zero (because the original function will be at depth zero), and all functions will be traced where the depth is greater than zero. The issue is that when a function is first entered, and the handler that checks this logic is called, the depth is set to zero. If an interrupt comes in and a function in the interrupt handler is traced, its depth will be greater than zero and it will automatically be traced, even if the original function was not. But because the logic only looks at depth it may trace interrupts when it should not be. The recent design change of the function graph tracer to fix other bugs caused the depth to be zero while the function graph callback handler is being called for a longer time, widening the race of this happening. This bug was actually there for a longer time, but because the race window was so small it seldom happened. The Fixes tag below is for the commit that widen the race window, because that commit belongs to a series that will also help fix the original bug. Cc: stable@kernel.org Fixes: 39eb456dacb5 ("function_graph: Use new curr_ret_depth to manage depth instead of curr_ret_stack") Reported-by: Joe Lawrence <joe.lawrence@redhat.com> Tested-by: Joe Lawrence <joe.lawrence@redhat.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Reverse the order of pushing the ret_stack and the callbackSteven Rostedt (VMware)
commit 7c6ea35ef50810aa12ab26f21cb858d980881576 upstream. The function graph profiler uses the ret_stack to store the "subtime" and reuse it by nested functions and also on the return. But the current logic has the profiler callback called before the ret_stack is updated, and it is just modifying the ret_stack that will later be allocated (it's just lucky that the "subtime" is not touched when it is allocated). This could also cause a crash if we are at the end of the ret_stack when this happens. By reversing the order of the allocating the ret_stack and then calling the callbacks attached to a function being traced, the ret_stack entry is no longer used before it is allocated. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Move return callback before update of curr_ret_stackSteven Rostedt (VMware)
commit 552701dd0fa7c3d448142e87210590ba424694a0 upstream. In the past, curr_ret_stack had two functions. One was to denote the depth of the call graph, the other is to keep track of where on the ret_stack the data is used. Although they may be slightly related, there are two cases where they need to be used differently. The one case is that it keeps the ret_stack data from being corrupted by an interrupt coming in and overwriting the data still in use. The other is just to know where the depth of the stack currently is. The function profiler uses the ret_stack to save a "subtime" variable that is part of the data on the ret_stack. If curr_ret_stack is modified too early, then this variable can be corrupted. The "max_depth" option, when set to 1, will record the first functions going into the kernel. To see all top functions (when dealing with timings), the depth variable needs to be lowered before calling the return hook. But by lowering the curr_ret_stack, it makes the data on the ret_stack still being used by the return hook susceptible to being overwritten. Now that there's two variables to handle both cases (curr_ret_depth), we can move them to the locations where they can handle both cases. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Have profiler use curr_ret_stack and not depthSteven Rostedt (VMware)
commit b1b35f2e218a5b57d03bbc3b0667d5064570dc60 upstream. The profiler uses trace->depth to find its entry on the ret_stack, but the depth may not match the actual location of where its entry is (if an interrupt were to preempt the processing of the profiler for another function, the depth and the curr_ret_stack will be different). Have it use the curr_ret_stack as the index to find its ret_stack entry instead of using the depth variable, as that is no longer guaranteed to be the same. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Use new curr_ret_depth to manage depth instead of curr_ret_stackSteven Rostedt (VMware)
commit 39eb456dacb543de90d3bc6a8e0ac5cf51ac475e upstream. Currently, the depth of the ret_stack is determined by curr_ret_stack index. The issue is that there's a race between setting of the curr_ret_stack and calling of the callback attached to the return of the function. Commit 03274a3ffb44 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") moved the calling of the callback to after the setting of the curr_ret_stack, even stating that it was safe to do so, when in fact, it was the reason there was a barrier() there (yes, I should have commented that barrier()). Not only does the curr_ret_stack keep track of the current call graph depth, it also keeps the ret_stack content from being overwritten by new data. The function profiler, uses the "subtime" variable of ret_stack structure and by moving the curr_ret_stack, it allows for interrupts to use the same structure it was using, corrupting the data, and breaking the profiler. To fix this, there needs to be two variables to handle the call stack depth and the pointer to where the ret_stack is being used, as they need to change at two different locations. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Make ftrace_push_return_trace() staticSteven Rostedt (VMware)
commit d125f3f866df88da5a85df00291f88f0baa89f7c upstream. As all architectures now call function_graph_enter() to do the entry work, no architecture should ever call ftrace_push_return_trace(). Make it static. This is needed to prepare for a fix of a design bug on how the curr_ret_stack is used. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-12-05function_graph: Create function_graph_enter() to consolidate architecture codeSteven Rostedt (VMware)
commit 8114865ff82e200b383e46821c25cb0625b842b5 upstream. Currently all the architectures do basically the same thing in preparing the function graph tracer on entry to a function. This code can be pulled into a generic location and then this will allow the function graph tracer to be fixed, as well as extended. Create a new function graph helper function_graph_enter() that will call the hook function (ftrace_graph_entry) and the shadow stack operation (ftrace_push_return_trace), and remove the need of the architecture code to manage the shadow stack. This is needed to prepare for a fix of a design bug on how the curr_ret_stack is used. Cc: stable@kernel.org Fixes: 03274a3ffb449 ("tracing/fgraph: Adjust fgraph depth before calling trace return callback") Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-11-21tracing/kprobes: Check the probe on unloaded module correctlyMasami Hiramatsu
[ Upstream commit 59158ec4aef7d44be51a6f3e7e17fc64c32604eb ] Current kprobe event doesn't checks correctly whether the given event is on unloaded module or not. It just checks the event has ":" in the name. That is not enough because if we define a probe on non-exist symbol on loaded module, it allows to define that (with warning message) To ensure it correctly, this searches the module name on loaded module list and only if there is not, it allows to define it. (this event will be available when the target module is loaded) Link: http://lkml.kernel.org/r/153547309528.26502.8300278470528281328.stgit@devbox Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-11-13tracing: Return -ENOENT if there is no target synthetic eventMasami Hiramatsu
commit 18858511fd8a877303cc34c06efa461b26a0e070 upstream. Return -ENOENT error if there is no target synthetic event. This notices an operation failure to user as below; # echo 'wakeup_latency u64 lat; pid_t pid;' > synthetic_events # echo '!wakeup' >> synthetic_events sh: write error: No such file or directory Link: http://lkml.kernel.org/r/154013449986.25576.9487131386597290172.stgit@devbox Acked-by: Tom Zanussi <zanussi@linux.intel.com> Tested-by: Tom Zanussi <zanussi@linux.intel.com> Cc: Shuah Khan <shuah@kernel.org> Cc: Rajvi Jingar <rajvi.jingar@intel.com> Cc: stable@vger.kernel.org Fixes: 4b147936fa50 ('tracing: Add support for 'synthetic' events') Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-10-19tracing: Fix synthetic event to allow semicolon at endMasami Hiramatsu
Fix synthetic event to allow independent semicolon at end. The synthetic_events interface accepts a semicolon after the last word if there is no space. # echo "myevent u64 var;" >> synthetic_events But if there is a space, it returns an error. # echo "myevent u64 var ;" > synthetic_events sh: write error: Invalid argument This behavior is difficult for users to understand. Let's allow the last independent semicolon too. Link: http://lkml.kernel.org/r/153986835420.18251.2191216690677025744.stgit@devbox Cc: Shuah Khan <shuah@kernel.org> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: stable@vger.kernel.org Fixes: commit 4b147936fa50 ("tracing: Add support for 'synthetic' events") Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-10-19tracing: Fix synthetic event to accept unsigned modifierMasami Hiramatsu
Fix synthetic event to accept unsigned modifier for its field type correctly. Currently, synthetic_events interface returns error for "unsigned" modifiers as below; # echo "myevent unsigned long var" >> synthetic_events sh: write error: Invalid argument This is because argv_split() breaks "unsigned long" into "unsigned" and "long", but parse_synth_field() doesn't expected it. With this fix, synthetic_events can handle the "unsigned long" correctly like as below; # echo "myevent unsigned long var" >> synthetic_events # cat synthetic_events myevent unsigned long var Link: http://lkml.kernel.org/r/153986832571.18251.8448135724590496531.stgit@devbox Cc: Shuah Khan <shuah@kernel.org> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: stable@vger.kernel.org Fixes: commit 4b147936fa50 ("tracing: Add support for 'synthetic' events") Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-10-17tracing: Use trace_clock_local() for looping in preemptirq_delay_test.cSteven Rostedt (VMware)
The preemptirq_delay_test module is used for the ftrace selftest code that tests the latency tracers. The problem is that it uses ktime for the delay loop, and then checks the tracer to see if the delay loop is caught, but the tracer uses trace_clock_local() which uses various different other clocks to measure the latency. As ktime uses the clock cycles, and the code then converts that to nanoseconds, it causes rounding errors, and the preemptirq latency tests are failing due to being off by 1 (it expects to see a delay of 500000 us, but the delay is only 499999 us). This is happening due to a rounding error in the ktime (which is totally legit). The purpose of the test is to see if it can catch the delay, not to test the accuracy between trace_clock_local() and ktime_get(). Best to use apples to apples, and have the delay loop use the same clock as the latency tracer does. Cc: stable@vger.kernel.org Fixes: f96e8577da102 ("lib: Add module for testing preemptoff/irqsoff latency tracers") Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-09-17ring-buffer: Allow for rescheduling when removing pagesVaibhav Nagarnaik
When reducing ring buffer size, pages are removed by scheduling a work item on each CPU for the corresponding CPU ring buffer. After the pages are removed from ring buffer linked list, the pages are free()d in a tight loop. The loop does not give up CPU until all pages are removed. In a worst case behavior, when lot of pages are to be freed, it can cause system stall. After the pages are removed from the list, the free() can happen while the work is rescheduled. Call cond_resched() in the loop to prevent the system hangup. Link: http://lkml.kernel.org/r/20180907223129.71994-1-vnagarnaik@google.com Cc: stable@vger.kernel.org Fixes: 83f40318dab00 ("ring-buffer: Make removal of ring buffer pages atomic") Reported-by: Jason Behmer <jbehmer@google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik@google.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-23Merge tag 'trace-v4.19-1' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace Pull tracing fixes from Steven Rostedt: "Masami found an off by one bug in the code that keeps "notrace" functions from being traced by kprobes. During my testing, I found that there's places that we may want to add kprobes to notrace, thus we may end up changing this code before 4.19 is released. The history behind this change is that we found that adding kprobes to various notrace functions caused the kernel to crashed. We took the safe route and decided not to allow kprobes to trace any notrace function. But because notrace is added to functions that just cause weird side effects to the function tracer, but are still safe, preventing kprobes for all notrace functios may be too much of a big hammer. One such place is __schedule() is marked notrace, to keep function tracer from doing strange recursive loops when it gets traced with NEED_RESCHED set. With this change, one can not add kprobes to the scheduler. Masami also added code to use gcov on ftrace" * tag 'trace-v4.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: tracing/kprobes: Fix to check notrace function with correct range tracing: Allow gcov profiling on only ftrace subsystem
2018-08-22Merge tag 'for-4.19/post-20180822' of git://git.kernel.dk/linux-blockLinus Torvalds
Pull more block updates from Jens Axboe: - Set of bcache fixes and changes (Coly) - The flush warn fix (me) - Small series of BFQ fixes (Paolo) - wbt hang fix (Ming) - blktrace fix (Steven) - blk-mq hardware queue count update fix (Jianchao) - Various little fixes * tag 'for-4.19/post-20180822' of git://git.kernel.dk/linux-block: (31 commits) block/DAC960.c: make some arrays static const, shrinks object size blk-mq: sync the update nr_hw_queues with blk_mq_queue_tag_busy_iter blk-mq: init hctx sched after update ctx and hctx mapping block: remove duplicate initialization tracing/blktrace: Fix to allow setting same value pktcdvd: fix setting of 'ret' error return for a few cases block: change return type to bool block, bfq: return nbytes and not zero from struct cftype .write() method block, bfq: improve code of bfq_bfqq_charge_time block, bfq: reduce write overcharge block, bfq: always update the budget of an entity when needed block, bfq: readd missing reset of parent-entity service blk-wbt: fix IO hang in wbt_wait() block: don't warn for flush on read-only device bcache: add the missing comments for smp_mb()/smp_wmb() bcache: remove unnecessary space before ioctl function pointer arguments bcache: add missing SPDX header bcache: move open brace at end of function definitions to next line bcache: add static const prefix to char * array declarations bcache: fix code comments style ...
2018-08-21tracing/kprobes: Fix to check notrace function with correct rangeMasami Hiramatsu
Fix within_notrace_func() to check notrace function correctly. Since the ftrace_location_range(start, end) function checks the range inclusively (start <= ftrace-loc <= end), the end address must not include the entry address of next function. However, within_notrace_func() uses kallsyms_lookup_size_offset() to get the function size and calculate the end address from adding the size to the entry address. This means the end address is the entry address of the next function. In the result, within_notrace_func() fails to find notrace function if the next function of the target function is ftraced. Let's subtract 1 from the end address so that ftrace_location_range() can check it correctly. Link: http://lkml.kernel.org/r/153485669706.16611.17726752296213785504.stgit@devbox Fixes: commit 45408c4f9250 ("tracing: kprobes: Prohibit probing on notrace function") Reported-by: Michael Rodin <michael@rodin.online> Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-21tracing: Allow gcov profiling on only ftrace subsystemMasami Hiramatsu
Add GCOV_PROFILE_FTRACE to allow gcov profiling on only files in ftrace subsystem. This config option will be used for checking kselftest/ftrace coverage. Link: http://lkml.kernel.org/r/153483647755.32472.4746349899604275441.stgit@devbox Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20Merge tag 'trace-v4.19' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace Pull tracing updates from Steven Rostedt: - Restructure of lockdep and latency tracers This is the biggest change. Joel Fernandes restructured the hooks from irqs and preemption disabling and enabling. He got rid of a lot of the preprocessor #ifdef mess that they caused. He turned both lockdep and the latency tracers to use trace events inserted in the preempt/irqs disabling paths. But unfortunately, these started to cause issues in corner cases. Thus, parts of the code was reverted back to where lockdep and the latency tracers just get called directly (without using the trace events). But because the original change cleaned up the code very nicely we kept that, as well as the trace events for preempt and irqs disabling, but they are limited to not being called in NMIs. - Have trace events use SRCU for "rcu idle" calls. This was required for the preempt/irqs off trace events. But it also had to not allow them to be called in NMI context. Waiting till Paul makes an NMI safe SRCU API. - New notrace SRCU API to allow trace events to use SRCU. - Addition of mcount-nop option support - SPDX headers replacing GPL templates. - Various other fixes and clean ups. - Some fixes are marked for stable, but were not fully tested before the merge window opened. * tag 'trace-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (44 commits) tracing: Fix SPDX format headers to use C++ style comments tracing: Add SPDX License format tags to tracing files tracing: Add SPDX License format to bpf_trace.c blktrace: Add SPDX License format header s390/ftrace: Add -mfentry and -mnop-mcount support tracing: Add -mcount-nop option support tracing: Avoid calling cc-option -mrecord-mcount for every Makefile tracing: Handle CC_FLAGS_FTRACE more accurately Uprobe: Additional argument arch_uprobe to uprobe_write_opcode() Uprobes: Simplify uprobe_register() body tracepoints: Free early tracepoints after RCU is initialized uprobes: Use synchronize_rcu() not synchronize_sched() tracing: Fix synchronizing to event changes with tracepoint_synchronize_unregister() ftrace: Remove unused pointer ftrace_swapper_pid tracing: More reverting of "tracing: Centralize preemptirq tracepoints and unify their usage" tracing/irqsoff: Handle preempt_count for different configs tracing: Partial revert of "tracing: Centralize preemptirq tracepoints and unify their usage" tracing: irqsoff: Account for additional preempt_disable trace: Use rcu_dereference_raw for hooks from trace-event subsystem tracing/kprobes: Fix within_notrace_func() to check only notrace functions ...
2018-08-16tracing: Fix SPDX format headers to use C++ style commentsSteven Rostedt (VMware)
The Linux kernel adopted the SPDX License format headers to ease license compliance management, and uses the C++ '//' style comments for the SPDX header tags. Some files in the tracing directory used the C style /* */ comments for them. To be consistent across all files, replace the /* */ C style SPDX tags with the C++ // SPDX tags. Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-16tracing: Add SPDX License format tags to tracing filesSteven Rostedt (VMware)
Add the SPDX License header to ease license compliance management. Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-16tracing: Add SPDX License format to bpf_trace.cSteven Rostedt (VMware)
Add the SPDX License header to ease license compliance management. Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-16tracing/blktrace: Fix to allow setting same valueSteven Rostedt (VMware)
Masami Hiramatsu reported: Current trace-enable attribute in sysfs returns an error if user writes the same setting value as current one, e.g. # cat /sys/block/sda/trace/enable 0 # echo 0 > /sys/block/sda/trace/enable bash: echo: write error: Invalid argument # echo 1 > /sys/block/sda/trace/enable # echo 1 > /sys/block/sda/trace/enable bash: echo: write error: Device or resource busy But this is not a preferred behavior, it should ignore if new setting is same as current one. This fixes the problem as below. # cat /sys/block/sda/trace/enable 0 # echo 0 > /sys/block/sda/trace/enable # echo 1 > /sys/block/sda/trace/enable # echo 1 > /sys/block/sda/trace/enable Link: http://lkml.kernel.org/r/20180816103802.08678002@gandalf.local.home Cc: Ingo Molnar <mingo@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: linux-block@vger.kernel.org Cc: stable@vger.kernel.org Fixes: cd649b8bb830d ("blktrace: remove sysfs_blk_trace_enable_show/store()") Reported-by: Masami Hiramatsu <mhiramat@kernel.org> Tested-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-08-16blktrace: Add SPDX License format headerSteven Rostedt (VMware)
Add the SPDX License header to ease license compliance management. Acked-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-15tracing: Add -mcount-nop option supportVasily Gorbik
-mcount-nop gcc option generates the calls to the profiling functions as nops which allows to avoid patching mcount jump with NOP instructions initially. -mcount-nop gcc option will be activated if platform selects HAVE_NOP_MCOUNT and gcc actually supports it. In addition to that CC_USING_NOP_MCOUNT is defined and could be used by architectures to adapt ftrace patching behavior. Link: http://lkml.kernel.org/r/patch-3.thread-aa7b8d.git-e02ed2dc082b.your-ad-here.call-01533557518-ext-9465@work.hours Signed-off-by: Vasily Gorbik <gor@linux.ibm.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-15Merge tag 'printk-for-4.19' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk Pull printk updates from Petr Mladek: - Different vendors have a different expectation about a console quietness. Make it configurable to reduce bike-shedding about the upstream default - Decide about the message visibility when the message is stored. It avoids races caused by a delayed console handling - Always store printk() messages into the per-CPU buffers again in NMI. The only exception is when flushing trace log in panic(). There the risk of loosing messages is worth an eventual reordering - Handle invalid %pO printf modifiers correctly - Better handle %p printf modifier tests before crng is initialized - Some clean up * tag 'printk-for-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk: lib/vsprintf: Do not handle %pO[^F] as %px printk: Fix warning about unused suppress_message_printing printk/nmi: Prevent deadlock when accessing the main log buffer in NMI printk: Create helper function to queue deferred console handling printk: Split the code for storing a message into the log buffer printk: Clean up syslog_print_all() printk: Remove unnecessary kmalloc() from syslog during clear printk: Make CONSOLE_LOGLEVEL_QUIET configurable printk: make sure to print log on console. lib/test_printf.c: accept "ptrval" as valid result for plain 'p' tests
2018-08-14Merge tag 'docs-4.19' of git://git.lwn.net/linuxLinus Torvalds
Pull documentation update from Jonathan Corbet: "This was a moderately busy cycle for docs, with the usual collection of small fixes and updates. We also have new ktime_get_*() docs from Arnd, some kernel-doc fixes, a new set of Italian translations (non so se vale la pena, ma non fa male - speriamo bene), and some extensive early memory-management documentation improvements from Mike Rapoport" * tag 'docs-4.19' of git://git.lwn.net/linux: (52 commits) Documentation: corrections to console/console.txt Documentation: add ioctl number entry for v4l2-subdev.h Remove gendered language from management style documentation scripts/kernel-doc: Escape all literal braces in regexes docs/mm: add description of boot time memory management docs/mm: memblock: add overview documentation docs/mm: memblock: add kernel-doc description for memblock types docs/mm: memblock: add kernel-doc comments for memblock_add[_node] docs/mm: memblock: update kernel-doc comments mm/memblock: add a name for memblock flags enumeration docs/mm: bootmem: add overview documentation docs/mm: bootmem: add kernel-doc description of 'struct bootmem_data' docs/mm: bootmem: fix kernel-doc warnings docs/mm: nobootmem: fixup kernel-doc comments mm/bootmem: drop duplicated kernel-doc comments Documentation: vm.txt: Adding 'nr_hugepages_mempolicy' parameter description. doc:it_IT: translation for kernel-hacking docs: Fix the reference labels in Locking.rst doc: tracing: Fix a typo of trace_stat mm: Introduce new type vm_fault_t ...
2018-08-14Merge tag 'for-4.19/block-20180812' of git://git.kernel.dk/linux-blockLinus Torvalds
Pull block updates from Jens Axboe: "First pull request for this merge window, there will also be a followup request with some stragglers. This pull request contains: - Fix for a thundering heard issue in the wbt block code (Anchal Agarwal) - A few NVMe pull requests: * Improved tracepoints (Keith) * Larger inline data support for RDMA (Steve Wise) * RDMA setup/teardown fixes (Sagi) * Effects log suppor for NVMe target (Chaitanya Kulkarni) * Buffered IO suppor for NVMe target (Chaitanya Kulkarni) * TP4004 (ANA) support (Christoph) * Various NVMe fixes - Block io-latency controller support. Much needed support for properly containing block devices. (Josef) - Series improving how we handle sense information on the stack (Kees) - Lightnvm fixes and updates/improvements (Mathias/Javier et al) - Zoned device support for null_blk (Matias) - AIX partition fixes (Mauricio Faria de Oliveira) - DIF checksum code made generic (Max Gurtovoy) - Add support for discard in iostats (Michael Callahan / Tejun) - Set of updates for BFQ (Paolo) - Removal of async write support for bsg (Christoph) - Bio page dirtying and clone fixups (Christoph) - Set of bcache fix/changes (via Coly) - Series improving blk-mq queue setup/teardown speed (Ming) - Series improving merging performance on blk-mq (Ming) - Lots of other fixes and cleanups from a slew of folks" * tag 'for-4.19/block-20180812' of git://git.kernel.dk/linux-block: (190 commits) blkcg: Make blkg_root_lookup() work for queues in bypass mode bcache: fix error setting writeback_rate through sysfs interface null_blk: add lock drop/acquire annotation Blk-throttle: reduce tail io latency when iops limit is enforced block: paride: pd: mark expected switch fall-throughs block: Ensure that a request queue is dissociated from the cgroup controller block: Introduce blk_exit_queue() blkcg: Introduce blkg_root_lookup() block: Remove two superfluous #include directives blk-mq: count the hctx as active before allocating tag block: bvec_nr_vecs() returns value for wrong slab bcache: trivial - remove tailing backslash in macro BTREE_FLAG bcache: make the pr_err statement used for ENOENT only in sysfs_attatch section bcache: set max writeback rate when I/O request is idle bcache: add code comments for bset.c bcache: fix mistaken comments in request.c bcache: fix mistaken code comments in bcache.h bcache: add a comment in super.c bcache: avoid unncessary cache prefetch bch_btree_node_get() bcache: display rate debug parameters to 0 when writeback is not running ...
2018-08-10uprobes: Use synchronize_rcu() not synchronize_sched()Steven Rostedt (VMware)
While debugging another bug, I was looking at all the synchronize*() functions being used in kernel/trace, and noticed that trace_uprobes was using synchronize_sched(), with a comment to synchronize with {u,ret}_probe_trace_func(). When looking at those functions, the data is protected with "rcu_read_lock()" and not with "rcu_read_lock_sched()". This is using the wrong synchronize_*() function. Link: http://lkml.kernel.org/r/20180809160553.469e1e32@gandalf.local.home Cc: stable@vger.kernel.org Fixes: 70ed91c6ec7f8 ("tracing/uprobes: Support ftrace_event_file base multibuffer") Acked-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-10tracing: Fix synchronizing to event changes with ↵Steven Rostedt (VMware)
tracepoint_synchronize_unregister() Now that some trace events can be protected by srcu_read_lock(tracepoint_srcu), we need to make sure all locations that depend on this are also protected. There were many places that did a synchronize_sched() thinking that it was enough to protect againts access to trace events. This use to be the case, but now that we use SRCU for _rcuidle() trace events, they may not be protected by synchronize_sched(), as they may be called in paths that RCU is not watching for preempt disable. Fixes: e6753f23d961d ("tracepoint: Make rcuidle tracepoint callers use SRCU") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-10ftrace: Remove unused pointer ftrace_swapper_pidColin Ian King
Pointer ftrace_swapper_pid is defined but is never used hence it is redundant and can be removed. The use of this variable was removed in commit 345ddcc882d8 ("ftrace: Have set_ftrace_pid use the bitmap like events do"). Cleans up clang warning: warning: 'ftrace_swapper_pid' defined but not used [-Wunused-const-variable=] Link: http://lkml.kernel.org/r/20180809125609.13142-1-colin.king@canonical.com Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-10tracing: More reverting of "tracing: Centralize preemptirq tracepoints and ↵Steven Rostedt (VMware)
unify their usage" Joel Fernandes created a nice patch that cleaned up the duplicate hooks used by lockdep and irqsoff latency tracer. It made both use tracepoints. But the latency tracer is triggering warnings when using tracepoints to call into the latency tracer's routines. Mainly, they can be called from NMI context. If that happens, then the SRCU may not work properly because on some architectures, SRCU is not safe to be called in both NMI and non-NMI context. This is a partial revert of the clean up patch c3bc8fd637a9 ("tracing: Centralize preemptirq tracepoints and unify their usage") that adds back the direct calls into the latency tracer. It also only calls the trace events when not in NMI. Link: http://lkml.kernel.org/r/20180809210654.622445925@goodmis.org Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Fixes: c3bc8fd637a9 ("tracing: Centralize preemptirq tracepoints and unify their usage") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-10tracing/irqsoff: Handle preempt_count for different configsSteven Rostedt (VMware)
I was hitting the following warning: WARNING: CPU: 0 PID: 1 at kernel/trace/trace_irqsoff.c:631 tracer_hardirqs_off+0x15/0x2a Modules linked in: CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.18.0-rc6-test+ #13 Hardware name: MSI MS-7823/CSM-H87M-G43 (MS-7823), BIOS V1.6 02/22/2014 EIP: tracer_hardirqs_off+0x15/0x2a Code: ff 85 c0 74 0e 8b 45 00 8b 50 04 8b 45 04 e8 35 ff ff ff 5d c3 55 64 a1 cc 37 51 c1 a9 ff ff ff 7f 89 e5 53 89 d3 89 ca 75 02 <0f> 0b e8 90 fc ff ff 85 c0 74 07 89 d8 e8 0c ff ff ff 5b 5d c3 55 EAX: 80000000 EBX: c04337f0 ECX: c04338e3 EDX: c04338e3 ESI: c04337f0 EDI: c04338e3 EBP: f2aa1d68 ESP: f2aa1d64 DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 EFLAGS: 00210046 CR0: 80050033 CR2: 00000000 CR3: 01668000 CR4: 001406f0 Call Trace: trace_irq_disable_rcuidle+0x63/0x6c trace_hardirqs_off+0x26/0x30 default_send_IPI_mask_allbutself_logical+0x31/0x93 default_send_IPI_allbutself+0x37/0x48 native_send_call_func_ipi+0x4d/0x6a smp_call_function_many+0x165/0x19d ? add_nops+0x34/0x34 ? trace_hardirqs_on_caller+0x2d/0x2d ? add_nops+0x34/0x34 smp_call_function+0x1f/0x23 on_each_cpu+0x15/0x43 ? trace_hardirqs_on_caller+0x2d/0x2d ? trace_hardirqs_on_caller+0x2d/0x2d ? trace_irq_disable_rcuidle+0x1/0x6c text_poke_bp+0xa0/0xc2 ? trace_hardirqs_on_caller+0x2d/0x2d arch_jump_label_transform+0xa7/0xcb ? trace_irq_disable_rcuidle+0x5/0x6c __jump_label_update+0x3e/0x6d jump_label_update+0x7d/0x81 static_key_slow_inc_cpuslocked+0x58/0x6d static_key_slow_inc+0x19/0x20 tracepoint_probe_register_prio+0x19e/0x1d1 ? start_critical_timings+0x1c/0x1c tracepoint_probe_register+0xf/0x11 irqsoff_tracer_init+0x21/0xf2 tracer_init+0x16/0x1a trace_selftest_startup_irqsoff+0x25/0xc4 run_tracer_selftest+0xca/0x131 register_tracer+0xd5/0x172 ? trace_event_define_fields_preemptirq_template+0x45/0x45 init_irqsoff_tracer+0xd/0x11 do_one_initcall+0xab/0x1e8 ? rcu_read_lock_sched_held+0x3d/0x44 ? trace_initcall_level+0x52/0x86 kernel_init_freeable+0x195/0x21a ? rest_init+0xb4/0xb4 kernel_init+0xd/0xe4 ret_from_fork+0x2e/0x38 It is due to running a CONFIG_PREEMPT_VOLUNTARY kernel, which would trigger this warning every time: WARN_ON_ONCE(preempt_count()); Because on CONFIG_PREEMPT_VOLUNTARY, preempt_count() is always zero. This warning is to make sure preempt_count is set because tracer_hardirqs_on() does a preempt_enable_notrace() to make the preempt_trace() work properly, as being called by a trace event, the trace event code disables preemption, and the tracer wants to know what the preemption was before it was called. Instead of enabling preemption like this, just record the preempt_count, subtract PREEMPT_DISABLE_OFFSET from it (which is zero with !CONFIG_PREEMPT set), and pass that value to the necessary functions, which should use the passed in parameter instead of calling preempt_count() directly. Fixes: da5b3ebb45277 ("tracing: irqsoff: Account for additional preempt_disable") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-10tracing: Partial revert of "tracing: Centralize preemptirq tracepoints and ↵Steven Rostedt (VMware)
unify their usage" Joel Fernandes created a nice patch that cleaned up the duplicate hooks used by lockdep and irqsoff latency tracer. It made both use tracepoints. But it caused lockdep to trigger several false positives. We have not figured out why yet, but removing lockdep from using the trace event hooks and just call its helper functions directly (like it use to), makes the problem go away. This is a partial revert of the clean up patch c3bc8fd637a9 ("tracing: Centralize preemptirq tracepoints and unify their usage") that adds direct calls for lockdep, but also keeps most of the clean up done to get rid of the horrible preprocessor if statements. Link: http://lkml.kernel.org/r/20180806155058.5ee875f4@gandalf.local.home Cc: Peter Zijlstra <peterz@infradead.org> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Fixes: c3bc8fd637a9 ("tracing: Centralize preemptirq tracepoints and unify their usage") Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-06tracing: irqsoff: Account for additional preempt_disableJoel Fernandes (Google)
Recently we tried to make the preemptirqsoff tracer to use irqsoff tracepoint probes. However this causes issues as reported by Masami: [2.271078] Testing tracer preemptirqsoff: .. no entries found ..FAILED! [2.381015] WARNING: CPU: 0 PID: 1 at /home/mhiramat/ksrc/linux/kernel/ trace/trace.c:1512 run_tracer_selftest+0xf3/0x154 This is due to the tracepoint code increasing the preempt nesting count by calling an additional preempt_disable before calling into the preemptoff tracer which messes up the preempt_count() check in tracer_hardirqs_off. To fix this, make the irqsoff tracer probes balance the additional outer preempt_disable with a preempt_enable_notrace. The other way to fix this is to just use SRCU for all tracepoints. However we can't do that because we can't use NMIs from RCU context. Link: http://lkml.kernel.org/r/20180806034049.67949-1-joel@joelfernandes.org Fixes: c3bc8fd637a9 ("tracing: Centralize preemptirq tracepoints and unify their usage") Fixes: e6753f23d961 ("tracepoint: Make rcuidle tracepoint callers use SRCU") Reported-by: Masami Hiramatsu <mhiramat@kernel.org> Tested-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-05Merge tag 'v4.18-rc6' into for-4.19/block2Jens Axboe
Pull in 4.18-rc6 to get the NVMe core AEN change to avoid a merge conflict down the line. Signed-of-by: Jens Axboe <axboe@kernel.dk>
2018-08-03trace: Use rcu_dereference_raw for hooks from trace-event subsystemJoel Fernandes (Google)
Since we switched to using SRCU for tracepoints used in the idle path, we can no longer use rcu_dereference_sched for dereferencing points in trace-event hooks. Since tracepoints can now use either SRCU or sched-RCU, just use rcu_dereference_raw for traceevents just like we're doing when dereferencing the tracepoint table. This prevents an RCU warning reported by Masami: [ 282.060593] WARNING: can't dereference registers at 00000000f3c7f62b [ 282.063200] ============================= [ 282.064082] WARNING: suspicious RCU usage [ 282.064963] 4.18.0-rc6+ #15 Tainted: G W [ 282.066048] ----------------------------- [ 282.066923] /home/mhiramat/ksrc/linux/kernel/trace/trace_events.c:242 suspicious rcu_dereference_check() usage! [ 282.068974] [ 282.068974] other info that might help us debug this: [ 282.068974] [ 282.070770] [ 282.070770] RCU used illegally from idle CPU! [ 282.070770] rcu_scheduler_active = 2, debug_locks = 1 [ 282.072938] RCU used illegally from extended quiescent state! [ 282.074183] no locks held by swapper/0/0. [ 282.075071] [ 282.075071] stack backtrace: [ 282.076121] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W [ 282.077782] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) [ 282.079604] Call Trace: [ 282.080212] <IRQ> [ 282.080755] dump_stack+0x85/0xcb [ 282.081523] trace_event_ignore_this_pid+0x66/0x70 [ 282.082541] trace_event_raw_event_preemptirq_template+0xa2/0xb0 [ 282.083774] ? interrupt_entry+0xc4/0xe0 [ 282.084665] ? trace_hardirqs_off_thunk+0x1a/0x1c [ 282.085669] trace_hardirqs_off_caller+0x90/0xd0 [ 282.086597] trace_hardirqs_off_thunk+0x1a/0x1c [ 282.087433] ? call_function_interrupt+0xa/0x20 [ 282.088201] interrupt_entry+0xc4/0xe0 [ 282.088848] ? call_function_interrupt+0xa/0x20 [ 282.089579] </IRQ> [ 282.090029] ? native_safe_halt+0x2/0x10 [ 282.090695] ? default_idle+0x1f/0x160 [ 282.091330] ? default_idle_call+0x24/0x40 [ 282.091997] ? do_idle+0x210/0x250 [ 282.092658] ? cpu_startup_entry+0x6f/0x80 [ 282.093338] ? start_kernel+0x49d/0x4bd [ 282.093987] ? secondary_startup_64+0xa5/0xb0 Link: http://lkml.kernel.org/r/20180803023407.225852-1-joel@joelfernandes.org Reported-by: Masami Hiramatsu <mhiramat@kernel.org> Tested-by: Masami Hiramatsu <mhiramat@kernel.org> Fixes: e6753f23d961 ("tracepoint: Make rcuidle tracepoint callers use SRCU") Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-02tracing/kprobes: Fix within_notrace_func() to check only notrace functionsMasami Hiramatsu
Fix within_notrace_func() to check only notrace functions and to ignore the kprobe-event which can not solve symbol addresses. within_notrace_func() returns true if the given kprobe events probe point seems to be out-of-range. But that is not the correct place to check for it, it should be done in kprobes afterwards. kprobe-events allow users to define a probe point on "currently unloaded module" so that it can trace the event during module load. In this case, the user will put a probe on a symbol which is not in kallsyms yet and it hits the within_notrace_func(). As a result, kprobe-events always refuses if user defines a probe on a "currenly unloaded module". Fixes: commit 45408c4f9250 ("tracing: kprobes: Prohibit probing on notrace function") Link: http://lkml.kernel.org/r/153319624799.29007.13604430345640129926.stgit@devbox Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-02Merge branch 'perf/urgent' into perf/core, to pick up fixesIngo Molnar
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-08-01ftrace: Use true and false for boolean values in ops_references_rec()Gustavo A. R. Silva
Return statements in functions returning bool should use true or false instead of an integer value. This code was detected with the help of Coccinelle. Link: http://lkml.kernel.org/r/20180802010056.GA31012@embeddedor.com Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-01ring-buffer: Make ring_buffer_record_is_set_on() return boolSteven Rostedt (VMware)
The value of ring_buffer_record_is_set_on() is either true or false, so have its return value be bool. Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-01ring-buffer: Make ring_buffer_record_is_on() return boolSteven Rostedt (VMware)
The value of ring_buffer_record_is_on() is either true or false, so have its return value be bool. Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
generated by cgit 1.2.3-korg (git 2.39.0) at 2024-06-01 20:34:14 +0000