linux/kernel/trace/trace_events_hist.c, branch v5.5

tracing: Fix histogram code when expression has same var as value

2020-01-20T21:11:47Z

While working on a tool to convert SQL syntex into the histogram language of the kernel, I discovered the following bug: # echo 'first u64 start_time u64 end_time pid_t pid u64 delta' >> synthetic_events # echo 'hist:keys=pid:start=common_timestamp' > events/sched/sched_waking/trigger # echo 'hist:keys=next_pid:delta=common_timestamp-$start,start2=$start:onmatch(sched.sched_waking).trace(first,$start2,common_timestamp,next_pid,$delta)' > events/sched/sched_switch/trigger Would not display any histograms in the sched_switch histogram side. But if I were to swap the location of "delta=common_timestamp-$start" with "start2=$start" Such that the last line had: # echo 'hist:keys=next_pid:start2=$start,delta=common_timestamp-$start:onmatch(sched.sched_waking).trace(first,$start2,common_timestamp,next_pid,$delta)' > events/sched/sched_switch/trigger The histogram works as expected. What I found out is that the expressions clear out the value once it is resolved. As the variables are resolved in the order listed, when processing: delta=common_timestamp-$start The $start is cleared. When it gets to "start2=$start", it errors out with "unresolved symbol" (which is silent as this happens at the location of the trace), and the histogram is dropped. When processing the histogram for variable references, instead of adding a new reference for a variable used twice, use the same reference. That way, not only is it more efficient, but the order will no longer matter in processing of the variables. From Tom Zanussi: "Just to clarify some more about what the problem was is that without your patch, we would have two separate references to the same variable, and during resolve_var_refs(), they'd both want to be resolved separately, so in this case, since the first reference to start wasn't part of an expression, it wouldn't get the read-once flag set, so would be read normally, and then the second reference would do the read-once read and also be read but using read-once. So everything worked and you didn't see a problem: from: start2=$start,delta=common_timestamp-$start In the second case, when you switched them around, the first reference would be resolved by doing the read-once, and following that the second reference would try to resolve and see that the variable had already been read, so failed as unset, which caused it to short-circuit out and not do the trigger action to generate the synthetic event: to: delta=common_timestamp-$start,start2=$start With your patch, we only have the single resolution which happens correctly the one time it's resolved, so this can't happen." Link: https://lore.kernel.org/r/20200116154216.58ca08eb@gandalf.local.home Cc: stable@vger.kernel.org Fixes: 067fe038e70f6 ("tracing: Add variable reference handling to hist triggers") Reviewed-by: Tom Zanuss Tested-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware)

tracing: trigger: Replace unneeded RCU-list traversals

2020-01-14T22:12:04Z

With CONFIG_PROVE_RCU_LIST, I had many suspicious RCU warnings when I ran ftracetest trigger testcases. ----- # dmesg -c > /dev/null # ./ftracetest test.d/trigger ... # dmesg | grep "RCU-list traversed" | cut -f 2 -d ] | cut -f 2 -d " " kernel/trace/trace_events_hist.c:6070 kernel/trace/trace_events_hist.c:1760 kernel/trace/trace_events_hist.c:5911 kernel/trace/trace_events_trigger.c:504 kernel/trace/trace_events_hist.c:1810 kernel/trace/trace_events_hist.c:3158 kernel/trace/trace_events_hist.c:3105 kernel/trace/trace_events_hist.c:5518 kernel/trace/trace_events_hist.c:5998 kernel/trace/trace_events_hist.c:6019 kernel/trace/trace_events_hist.c:6044 kernel/trace/trace_events_trigger.c:1500 kernel/trace/trace_events_trigger.c:1540 kernel/trace/trace_events_trigger.c:539 kernel/trace/trace_events_trigger.c:584 ----- I investigated those warnings and found that the RCU-list traversals in event trigger and hist didn't need to use RCU version because those were called only under event_mutex. I also checked other RCU-list traversals related to event trigger list, and found that most of them were called from event_hist_trigger_func() or hist_unregister_trigger() or register/unregister functions except for a few cases. Replace these unneeded RCU-list traversals with normal list traversal macro and lockdep_assert_held() to check the event_mutex is held. Link: http://lkml.kernel.org/r/157680910305.11685.15110237954275915782.stgit@devnote2 Cc: stable@vger.kernel.org Fixes: 30350d65ac567 ("tracing: Add variable support to hist triggers") Reviewed-by: Tom Zanussi Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (VMware)

tracing: Fix endianness bug in histogram trigger

2019-12-21T21:08:59Z

At least on PA-RISC and s390 synthetic histogram triggers are failing selftests because trace_event_raw_event_synth() always writes a 64 bit values, but the reader expects a field->size sized value. On little endian machines this doesn't hurt, but on big endian this makes the reader always read zero values. Link: http://lore.kernel.org/linux-trace-devel/20191218074427.96184-4-svens@linux.ibm.com Cc: stable@vger.kernel.org Fixes: 4b147936fa509 ("tracing: Add support for 'synthetic' events") Acked-by: Tom Zanussi Signed-off-by: Sven Schnelle Signed-off-by: Steven Rostedt (VMware)

tracing: Increase SYNTH_FIELDS_MAX for synthetic_events

2019-11-15T16:30:29Z

Increase the maximum allowed count of synthetic event fields from 16 to 32 in order to allow for larger-than-usual events. Link: http://lkml.kernel.org/r/20191115091730.9192-1-dedekind1@gmail.com Reviewed-by: Tom Zanussi Signed-off-by: Artem Bityutskiy Signed-off-by: Steven Rostedt (VMware)

tracing: Fix "gfp_t" format for synthetic events

2019-10-18T18:42:53Z

In the format of synthetic events, the "gfp_t" is shown as "signed:1", but in fact the "gfp_t" is "unsigned", should be shown as "signed:0". The issue can be reproduced by the following commands: echo 'memlatency u64 lat; unsigned int order; gfp_t gfp_flags; int migratetype' > /sys/kernel/debug/tracing/synthetic_events cat /sys/kernel/debug/tracing/events/synthetic/memlatency/format name: memlatency ID: 2233 format: field:unsigned short common_type; offset:0; size:2; signed:0; field:unsigned char common_flags; offset:2; size:1; signed:0; field:unsigned char common_preempt_count; offset:3; size:1; signed:0; field:int common_pid; offset:4; size:4; signed:1; field:u64 lat; offset:8; size:8; signed:0; field:unsigned int order; offset:16; size:4; signed:0; field:gfp_t gfp_flags; offset:24; size:4; signed:1; field:int migratetype; offset:32; size:4; signed:1; print fmt: "lat=%llu, order=%u, gfp_flags=%x, migratetype=%d", REC->lat, REC->order, REC->gfp_flags, REC->migratetype Link: http://lkml.kernel.org/r/20191018012034.6404-1-zhengjun.xing@linux.intel.com Reviewed-by: Tom Zanussi Signed-off-by: Zhengjun Xing Signed-off-by: Steven Rostedt (VMware)

tracing: Add locked_down checks to the open calls of files created for tracefs

2019-10-13T00:48:06Z

Added various checks on open tracefs calls to see if tracefs is in lockdown mode, and if so, to return -EPERM. Note, the event format files (which are basically standard on all machines) as well as the enabled_functions file (which shows what is currently being traced) are not lockde down. Perhaps they should be, but it seems counter intuitive to lockdown information to help you know if the system has been modified. Link: http://lkml.kernel.org/r/CAHk-=wj7fGPKUspr579Cii-w_y60PtRaiDgKuxVtBAMK0VNNkA@mail.gmail.com Suggested-by: Linus Torvalds Signed-off-by: Steven Rostedt (VMware)

tracing: Add tracing_check_open_get_tr()

2019-10-13T00:44:07Z

Currently, most files in the tracefs directory test if tracing_disabled is set. If so, it should return -ENODEV. The tracing_disabled is called when tracing is found to be broken. Originally it was done in case the ring buffer was found to be corrupted, and we wanted to prevent reading it from crashing the kernel. But it's also called if a tracing selftest fails on boot. It's a one way switch. That is, once it is triggered, tracing is disabled until reboot. As most tracefs files can also be used by instances in the tracefs directory, they need to be carefully done. Each instance has a trace_array associated to it, and when the instance is removed, the trace_array is freed. But if an instance is opened with a reference to the trace_array, then it requires looking up the trace_array to get its ref counter (as there could be a race with it being deleted and the open itself). Once it is found, a reference is added to prevent the instance from being removed (and the trace_array associated with it freed). Combine the two checks (tracing_disabled and trace_array_get()) into a single helper function. This will also make it easier to add lockdown to tracefs later. Link: http://lkml.kernel.org/r/20191011135458.7399da44@gandalf.local.home Signed-off-by: Steven Rostedt (VMware)

tracing: Make sure variable reference alias has correct var_ref_idx

2019-09-17T15:21:29Z

Original changelog from Steve Rostedt (except last sentence which explains the problem, and the Fixes: tag): I performed a three way histogram with the following commands: echo 'irq_lat u64 lat pid_t pid' > synthetic_events echo 'wake_lat u64 lat u64 irqlat pid_t pid' >> synthetic_events echo 'hist:keys=common_pid:irqts=common_timestamp.usecs if function == 0xffffffff81200580' > events/timer/hrtimer_start/trigger echo 'hist:keys=common_pid:lat=common_timestamp.usecs-$irqts:onmatch(timer.hrtimer_start).irq_lat($lat,pid) if common_flags & 1' > events/sched/sched_waking/trigger echo 'hist:keys=pid:wakets=common_timestamp.usecs,irqlat=lat' > events/synthetic/irq_lat/trigger echo 'hist:keys=next_pid:lat=common_timestamp.usecs-$wakets,irqlat=$irqlat:onmatch(synthetic.irq_lat).wake_lat($lat,$irqlat,next_pid)' > events/sched/sched_switch/trigger echo 1 > events/synthetic/wake_lat/enable Basically I wanted to see: hrtimer_start (calling function tick_sched_timer) Note: # grep tick_sched_timer /proc/kallsyms ffffffff81200580 t tick_sched_timer And save the time of that, and then record sched_waking if it is called in interrupt context and with the same pid as the hrtimer_start, it will record the latency between that and the waking event. I then look at when the task that is woken is scheduled in, and record the latency between the wakeup and the task running. At the end, the wake_lat synthetic event will show the wakeup to scheduled latency, as well as the irq latency in from hritmer_start to the wakeup. The problem is that I found this: -0 [007] d... 190.485261: wake_lat: lat=27 irqlat=190485230 pid=698 -0 [005] d... 190.485283: wake_lat: lat=40 irqlat=190485239 pid=10 -0 [002] d... 190.488327: wake_lat: lat=56 irqlat=190488266 pid=335 -0 [005] d... 190.489330: wake_lat: lat=64 irqlat=190489262 pid=10 -0 [003] d... 190.490312: wake_lat: lat=43 irqlat=190490265 pid=77 -0 [005] d... 190.493322: wake_lat: lat=54 irqlat=190493262 pid=10 -0 [005] d... 190.497305: wake_lat: lat=35 irqlat=190497267 pid=10 -0 [005] d... 190.501319: wake_lat: lat=50 irqlat=190501264 pid=10 The irqlat seemed quite large! Investigating this further, if I had enabled the irq_lat synthetic event, I noticed this: -0 [002] d.s. 249.429308: irq_lat: lat=164968 pid=335 -0 [002] d... 249.429369: wake_lat: lat=55 irqlat=249429308 pid=335 Notice that the timestamp of the irq_lat "249.429308" is awfully similar to the reported irqlat variable. In fact, all instances were like this. It appeared that: irqlat=$irqlat Wasn't assigning the old $irqlat to the new irqlat variable, but instead was assigning the $irqts to it. The issue is that assigning the old $irqlat to the new irqlat variable creates a variable reference alias, but the alias creation code forgets to make sure the alias uses the same var_ref_idx to access the reference. Link: http://lkml.kernel.org/r/1567375321.5282.12.camel@kernel.org Cc: Linux Trace Devel Cc: linux-rt-users Cc: stable@vger.kernel.org Fixes: 7e8b88a30b085 ("tracing: Add hist trigger support for variable reference aliases") Reported-by: Steven Rostedt (VMware) Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware)

tracing: Add "gfp_t" support in synthetic_events

2019-09-05T15:35:14Z

Add "gfp_t" support in synthetic_events, then the "gfp_t" type parameter in some functions can be traced. Prints the gfp flags as hex in addition to the human-readable flag string. Example output: whoopsie-630 [000] ...1 78.969452: testevent: bar=b20 (GFP_ATOMIC|__GFP_ZERO) rcuc/0-11 [000] ...1 81.097555: testevent: bar=a20 (GFP_ATOMIC) rcuc/0-11 [000] ...1 81.583123: testevent: bar=a20 (GFP_ATOMIC) Link: http://lkml.kernel.org/r/20190712015308.9908-1-zhengjun.xing@linux.intel.com Signed-off-by: Zhengjun Xing [ Added printing of flag names ] Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware)

tracing/dynevent: Pass extra arguments to match operation

2019-08-31T16:19:38Z

Pass extra arguments to match operation for checking exact match. If the event doesn't support exact match, it will be ignored. Link: http://lkml.kernel.org/r/156095685930.28024.10405547027475590975.stgit@devnote2 Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (VMware)