user/sven/linux.git/kernel/trace/trace.h, branch v6.1.67

tracing: Have trace_event_file have ref counters

2023-11-28T17:07:23Z

commit bb32500fb9b78215e4ef6ee8b4345c5f5d7eafb4 upstream. The following can crash the kernel: # cd /sys/kernel/tracing # echo 'p:sched schedule' > kprobe_events # exec 5>>events/kprobes/sched/enable # > kprobe_events # exec 5>&- The above commands: 1. Change directory to the tracefs directory 2. Create a kprobe event (doesn't matter what one) 3. Open bash file descriptor 5 on the enable file of the kprobe event 4. Delete the kprobe event (removes the files too) 5. Close the bash file descriptor 5 The above causes a crash! BUG: kernel NULL pointer dereference, address: 0000000000000028 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 6 PID: 877 Comm: bash Not tainted 6.5.0-rc4-test-00008-g2c6b6b1029d4-dirty #186 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014 RIP: 0010:tracing_release_file_tr+0xc/0x50 What happens here is that the kprobe event creates a trace_event_file "file" descriptor that represents the file in tracefs to the event. It maintains state of the event (is it enabled for the given instance?). Opening the "enable" file gets a reference to the event "file" descriptor via the open file descriptor. When the kprobe event is deleted, the file is also deleted from the tracefs system which also frees the event "file" descriptor. But as the tracefs file is still opened by user space, it will not be totally removed until the final dput() is called on it. But this is not true with the event "file" descriptor that is already freed. If the user does a write to or simply closes the file descriptor it will reference the event "file" descriptor that was just freed, causing a use-after-free bug. To solve this, add a ref count to the event "file" descriptor as well as a new flag called "FREED". The "file" will not be freed until the last reference is released. But the FREE flag will be set when the event is removed to prevent any more modifications to that event from happening, even if there's still a reference to the event "file" descriptor. Link: https://lore.kernel.org/linux-trace-kernel/20231031000031.1e705592@gandalf.local.home/ Link: https://lore.kernel.org/linux-trace-kernel/20231031122453.7a48b923@gandalf.local.home Cc: stable@vger.kernel.org Cc: Mark Rutland Fixes: f5ca233e2e66d ("tracing: Increase trace array ref count on enable and filter files") Reported-by: Beau Belgrave Tested-by: Beau Belgrave Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing: Increase trace array ref count on enable and filter files

2023-09-23T09:11:11Z

commit f5ca233e2e66dc1c249bf07eefa37e34a6c9346a upstream. When the trace event enable and filter files are opened, increment the trace array ref counter, otherwise they can be accessed when the trace array is being deleted. The ref counter keeps the trace array from being deleted while those files are opened. Link: https://lkml.kernel.org/r/20230907024803.456187066@goodmis.org Link: https://lore.kernel.org/all/1cb3aee2-19af-c472-e265-05176fe9bd84@huawei.com/ Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Andrew Morton Fixes: 8530dec63e7b4 ("tracing: Add tracing_check_open_get_tr()") Tested-by: Linux Kernel Functional Testing Tested-by: Naresh Kamboju Reported-by: Zheng Yejian Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing: Introduce pipe_cpumask to avoid race on trace_pipes

2023-09-13T07:42:25Z

[ Upstream commit c2489bb7e6be2e8cdced12c16c42fa128403ac03 ] There is race issue when concurrently splice_read main trace_pipe and per_cpu trace_pipes which will result in data read out being different from what actually writen. As suggested by Steven: > I believe we should add a ref count to trace_pipe and the per_cpu > trace_pipes, where if they are opened, nothing else can read it. > > Opening trace_pipe locks all per_cpu ref counts, if any of them are > open, then the trace_pipe open will fail (and releases any ref counts > it had taken). > > Opening a per_cpu trace_pipe will up the ref count for just that > CPU buffer. This will allow multiple tasks to read different per_cpu > trace_pipe files, but will prevent the main trace_pipe file from > being opened. But because we only need to know whether per_cpu trace_pipe is open or not, using a cpumask instead of using ref count may be easier. After this patch, users will find that: - Main trace_pipe can be opened by only one user, and if it is opened, all per_cpu trace_pipes cannot be opened; - Per_cpu trace_pipes can be opened by multiple users, but each per_cpu trace_pipe can only be opened by one user. And if one of them is opened, main trace_pipe cannot be opened. Link: https://lore.kernel.org/linux-trace-kernel/20230818022645.1948314-1-zhengyejian1@huawei.com Suggested-by: Steven Rostedt (Google) Signed-off-by: Zheng Yejian Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) Signed-off-by: Sasha Levin

Revert "tracing: Add "(fault)" name injection to kernel probes"

2023-08-03T08:23:55Z

[ Upstream commit 4ed8f337dee32df71435689c19d22e4ee846e15a ] This reverts commit 2e9906f84fc7c99388bb7123ade167250d50f1c0. It was turned out that commit 2e9906f84fc7 ("tracing: Add "(fault)" name injection to kernel probes") did not work correctly and probe events still show just '(fault)' (instead of '"(fault)"'). Also, current '(fault)' is more explicit that it faulted. This also moves FAULT_STRING macro to trace.h so that synthetic event can keep using it, and uses it in trace_probe.c too. Link: https://lore.kernel.org/all/168908495772.123124.1250788051922100079.stgit@devnote2/ Link: https://lore.kernel.org/all/20230706230642.3793a593@rorschach.local.home/ Cc: stable@vger.kernel.org Cc: Andrew Morton Cc: Tom Zanussi Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) Stable-dep-of: 797311bce5c2 ("tracing/probes: Fix to record 0-length data_loc in fetch_store_string*() if fails") Signed-off-by: Sasha Levin

tracing: Allow synthetic events to pass around stacktraces

2023-08-03T08:23:55Z

[ Upstream commit 00cf3d672a9dd409418647e9f98784c339c3ff63 ] Allow a stacktrace from one event to be displayed by the end event of a synthetic event. This is very useful when looking for the longest latency of a sleep or something blocked on I/O. # cd /sys/kernel/tracing/ # echo 's:block_lat pid_t pid; u64 delta; unsigned long[] stack;' > dynamic_events # echo 'hist:keys=next_pid:ts=common_timestamp.usecs,st=stacktrace if prev_state == 1||prev_state == 2' > events/sched/sched_switch/trigger # echo 'hist:keys=prev_pid:delta=common_timestamp.usecs-$ts,s=$st:onmax($delta).trace(block_lat,prev_pid,$delta,$s)' >> events/sched/sched_switch/trigger The above creates a "block_lat" synthetic event that take the stacktrace of when a task schedules out in either the interruptible or uninterruptible states, and on a new per process max $delta (the time it was scheduled out), will print the process id and the stacktrace. # echo 1 > events/synthetic/block_lat/enable # cat trace # TASK-PID CPU# ||||| TIMESTAMP FUNCTION # | | | ||||| | | kworker/u16:0-767 [006] d..4. 560.645045: block_lat: pid=767 delta=66 stack=STACK: => __schedule => schedule => pipe_read => vfs_read => ksys_read => do_syscall_64 => 0x966000aa -0 [003] d..4. 561.132117: block_lat: pid=0 delta=413787 stack=STACK: => __schedule => schedule => schedule_hrtimeout_range_clock => do_sys_poll => __x64_sys_poll => do_syscall_64 => 0x966000aa <...>-153 [006] d..4. 562.068407: block_lat: pid=153 delta=54 stack=STACK: => __schedule => schedule => io_schedule => rq_qos_wait => wbt_wait => __rq_qos_throttle => blk_mq_submit_bio => submit_bio_noacct_nocheck => ext4_bio_write_page => mpage_submit_page => mpage_process_page_bufs => mpage_prepare_extent_to_map => ext4_do_writepages => ext4_writepages => do_writepages => __writeback_single_inode Link: https://lkml.kernel.org/r/20230117152236.010941267@goodmis.org Cc: Masami Hiramatsu Cc: Andrew Morton Cc: Tom Zanussi Cc: Ross Zwisler Cc: Ching-lin Yu Signed-off-by: Steven Rostedt (Google) Stable-dep-of: 797311bce5c2 ("tracing/probes: Fix to record 0-length data_loc in fetch_store_string*() if fails") Signed-off-by: Sasha Levin

tracing: Fix TASK_COMM_LEN in trace event format file

2023-02-14T18:11:54Z

commit b6c7abd1c28a63ad633433d037ee15a1bc3023ba upstream. After commit 3087c61ed2c4 ("tools/testing/selftests/bpf: replace open-coded 16 with TASK_COMM_LEN"), the content of the format file under /sys/kernel/tracing/events/task/task_newtask was changed from field:char comm[16]; offset:12; size:16; signed:0; to field:char comm[TASK_COMM_LEN]; offset:12; size:16; signed:0; John reported that this change breaks older versions of perfetto. Then Mathieu pointed out that this behavioral change was caused by the use of __stringify(_len), which happens to work on macros, but not on enum labels. And he also gave the suggestion on how to fix it: :One possible solution to make this more robust would be to extend :struct trace_event_fields with one more field that indicates the length :of an array as an actual integer, without storing it in its stringified :form in the type, and do the formatting in f_show where it belongs. The result as follows after this change, $ cat /sys/kernel/tracing/events/task/task_newtask/format field:char comm[16]; offset:12; size:16; signed:0; Link: https://lore.kernel.org/lkml/Y+QaZtz55LIirsUO@google.com/ Link: https://lore.kernel.org/linux-trace-kernel/20230210155921.4610-1-laoar.shao@gmail.com/ Link: https://lore.kernel.org/linux-trace-kernel/20230212151303.12353-1-laoar.shao@gmail.com Cc: stable@vger.kernel.org Cc: Alexei Starovoitov Cc: Kajetan Puchalski CC: Qais Yousef Fixes: 3087c61ed2c4 ("tools/testing/selftests/bpf: replace open-coded 16 with TASK_COMM_LEN") Reported-by: John Stultz Debugged-by: Mathieu Desnoyers Suggested-by: Mathieu Desnoyers Suggested-by: Steven Rostedt Signed-off-by: Yafang Shao Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing: Make sure trace_printk() can output as soon as it can be used

2023-02-01T07:34:37Z

commit 3bb06eb6e9acf7c4a3e1b5bc87aed398ff8e2253 upstream. Currently trace_printk() can be used as soon as early_trace_init() is called from start_kernel(). But if a crash happens, and "ftrace_dump_on_oops" is set on the kernel command line, all you get will be: [ 0.456075] -0 0dN.2. 347519us : Unknown type 6 [ 0.456075] -0 0dN.2. 353141us : Unknown type 6 [ 0.456075] -0 0dN.2. 358684us : Unknown type 6 This is because the trace_printk() event (type 6) hasn't been registered yet. That gets done via an early_initcall(), which may be early, but not early enough. Instead of registering the trace_printk() event (and other ftrace events, which are not trace events) via an early_initcall(), have them registered at the same time that trace_printk() can be used. This way, if there is a crash before early_initcall(), then the trace_printk()s will actually be useful. Link: https://lkml.kernel.org/r/20230104161412.019f6c55@gandalf.local.home Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Fixes: e725c731e3bb1 ("tracing: Split tracing initialization into two for early initialization") Reported-by: "Joel Fernandes (Google)" Tested-by: Joel Fernandes (Google) Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing/probes: Handle system names with hyphens

2023-01-07T10:11:48Z

commit 575b76cb885532aae13a9d979fd476bb2b156cb9 upstream. When creating probe names, a check is done to make sure it matches basic C standard variable naming standards. Basically, starts with alphabetic or underline, and then the rest of the characters have alpha-numeric or underline in them. But system names do not have any true naming conventions, as they are created by the TRACE_SYSTEM macro and nothing tests to see what they are. The "xhci-hcd" trace events has a '-' in the system name. When trying to attach a eprobe to one of these trace points, it fails because the system name does not follow the variable naming convention because of the hyphen, and the eprobe checks fail on this. Allow hyphens in the system name so that eprobes can attach to the "xhci-hcd" trace events. Link: https://lore.kernel.org/all/Y3eJ8GiGnEvVd8%2FN@macondo/ Link: https://lore.kernel.org/linux-trace-kernel/20221122122345.160f5077@gandalf.local.home Cc: Masami Hiramatsu Cc: stable@vger.kernel.org Fixes: 5b7a96220900e ("tracing/probe: Check event/group naming rule at parsing") Reported-by: Rafael Mendonca Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing: Fix complicated dependency of CONFIG_TRACER_MAX_TRACE

2023-01-07T10:11:48Z

commit e25e43a4e5d8cb2323553d8b6a7ba08d2ebab21f upstream. Both CONFIG_OSNOISE_TRACER and CONFIG_HWLAT_TRACER partially enables the CONFIG_TRACER_MAX_TRACE code, but that is complicated and has introduced a bug; It declares tracing_max_lat_fops data structure outside of #ifdefs, but since it is defined only when CONFIG_TRACER_MAX_TRACE=y or CONFIG_HWLAT_TRACER=y, if only CONFIG_OSNOISE_TRACER=y, that declaration comes to a definition(!). To fix this issue, and do not repeat the similar problem, makes CONFIG_OSNOISE_TRACER and CONFIG_HWLAT_TRACER enables the CONFIG_TRACER_MAX_TRACE always. It has there benefits; - Fix the tracing_max_lat_fops bug - Simplify the #ifdefs - CONFIG_TRACER_MAX_TRACE code is fully enabled, or not. Link: https://lore.kernel.org/linux-trace-kernel/167033628155.4111793.12185405690820208159.stgit@devnote3 Fixes: 424b650f35c7 ("tracing: Fix missing osnoise tracer on max_latency") Cc: Daniel Bristot de Oliveira Cc: stable@vger.kernel.org Reported-by: David Howells Reported-by: kernel test robot Signed-off-by: Masami Hiramatsu (Google) Link: https://lore.kernel.org/all/166992525941.1716618.13740663757583361463.stgit@warthog.procyon.org.uk/ (original thread and v1) Link: https://lore.kernel.org/all/202212052253.VuhZ2ulJ-lkp@intel.com/T/#u (v1 error report) Signed-off-by: Steven Rostedt (Google) Signed-off-by: Greg Kroah-Hartman

tracing: Add tracing_reset_all_online_cpus_unlocked() function

2022-11-24T00:06:11Z

Currently the tracing_reset_all_online_cpus() requires the trace_types_lock held. But only one caller of this function actually has that lock held before calling it, and the other just takes the lock so that it can call it. More users of this function is needed where the lock is not held. Add a tracing_reset_all_online_cpus_unlocked() function for the one use case that calls it without being held, and also add a lockdep_assert to make sure it is held when called. Then have tracing_reset_all_online_cpus() take the lock internally, such that callers do not need to worry about taking it. Link: https://lkml.kernel.org/r/20221123192741.658273220@goodmis.org Cc: Masami Hiramatsu Cc: Andrew Morton Cc: Zheng Yejian Signed-off-by: Steven Rostedt (Google)