- 31 Mar, 2009 13 commits
-
-
Ingo Molnar authored
Merge branches 'tracing/docs', 'tracing/filters', 'tracing/ftrace', 'tracing/kprobes', 'tracing/blktrace-v2' and 'tracing/textedit' into tracing/core-v2
-
Li Zefan authored
Impact: fix build warning I passed a const value to trace_seq_putmem(), and I got compile warning. Signed-off-by:
Li Zefan <lizf@cn.fujitsu.com> Acked-by:
Jens Axboe <jens.axboe@oracle.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Eduard - Gabriel Munteanu authored
Impact: cleanup Many declarations within trace_output.h are missing the 'extern' keyword in an inconsistent manner. This adds 'extern' where it should be. Signed-off-by:
Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro> Signed-off-by:
-
Eduard - Gabriel Munteanu authored
trace_seq_reserve() allows a caller to reserve space in a trace_seq and write directly into it. This makes it easier to export binary data to userspace via the tracing interface, by simply filling in a struct. Signed-off-by:
-
Li Zefan authored
Impact: improve ftrace plugin output Before this patch: # cat trace make-5383 [001] 741.240059: 8,7 P N [make] __trace_note_message: cfq1074 # echo 1 > options/blk_classic # cat trace 8,7 1 0.692221252 0 C W 130411392 + 1024 [0] Bad pc action 6361 Bad pc action 283d # echo 0 > options/blk_classic # echo bin > trace_options # cat trace_pipe | blkparse -i - (can't parse messages generated by blk_add_trace_msg()) After this patch: # cat trace <idle>-0 [001] 187.600933: 8,7 C W 145220224 + 8 [0] <idle>-0 [001] 187.600946: 8,7 m N cfq1076 complete # echo 1 > options/blk_classic # cat trace 8,7 1 0.256378996 238 I W 113190728 + 8 [pdflush] 8,7 1 0.256378998 238 m N cfq1076 insert_request # echo 0 > options/blk_classic # echo bin > trace_options # cat trace_pipe | blkparse -i - 8,7 1 0 22.973250293 0 C W 102770576 + 8 [0] 8,7 1 0 22.973259213 0 m N cfq1076 complete Signed-off-by: -
Li Zefan authored
Impact: cleanup blk_trace_event_print() and blk_tracer_print_line() share most of the code. text data bss dec hex filename 8605 393 12 9010 2332 kernel/trace/blktrace.o.orig text data bss dec hex filename 8555 393 12 8960 2300 kernel/trace/blktrace.o This patch also prepares for the next patch, that prints out BLK_TN_MESSAGE. Signed-off-by:
-
Li Zefan authored
Impact: fix mixed ioctl and ftrace-plugin blktrace use memory leak When mixing the use of ioctl-based blktrace and ftrace-based blktrace, we can leak memory in this way: # btrace /dev/sda > /dev/null & # echo 0 > /sys/block/sda/sda1/trace/enable now we leak bt->dropped_file, bt->msg_file, bt->rchan... Signed-off-by:
-
Li Zefan authored
Impact: fix mixed ioctl and ftrace-plugin blktrace use refcount bugs ioctl-based blktrace allocates bt and registers tracepoints when ioctl(BLKTRACESETUP), and do all cleanups when ioctl(BLKTRACETEARDOWN). while ftrace-based blktrace allocates/frees bt when: # echo 1/0 > /sys/block/sda/sda1/trace/enable and registers/unregisters tracepoints when: # echo blk/nop > /debugfs/tracing/current_tracer or # echo 1/0 > /debugfs/tracing/tracing_enable The separatation of allocation and registeration causes 2 problems: 1. current user-space blktrace still calls ioctl(TEARDOWN) when ioctl(SETUP) failed: # echo 1 > /sys/block/sda/sda1/trace/enable # blktrace /dev/sda BLKTRACESETUP: Device or resource busy ^C and now blk_probes_ref == -1 2. Another way to make blk_probes_ref == -1: # plugin sdb && mount sdb1 # echo 1 > /sys/block/sdb/sdb1/trace/enable # remove sdb This patch does the allocation and registeration when writing sdaX/trace/enable. Signed-off-by: -
Li Zefan authored
Impact: fix ftrace plugin timestamp output In the classic user-space blktrace, the output timestamp is sec.nsec not sec.usec. Signed-off-by:
-
Li Zefan authored
'what' is used as the index of array what2act, so it can't >= the array size. Signed-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by:
Steven Rostedt <rostedt@goodmis.org> Signed-off-by:
-
Li Zefan authored
Currently the original blktrace, which is using relay and is used via ioctl, is broken. You can use ftrace to see the output of blktrace, but user-space blktrace is unusable. It's broken by "blktrace: add ftrace plugin" (c71a8961) - if (unlikely(bt->trace_state != Blktrace_running)) + if (unlikely(bt->trace_state != Blktrace_running || !blk_tracer_enabled)) return; With this patch, both ioctl and ftrace can be used, but of course you can't use both of them at the same time. Signed-off-by:
-
Li Zefan authored
t1 t2 ------ ------ do_blk_trace_setup() do_blk_trace_setup() if (!blk_tree_root) { if (!blk_tree_root) blk_tree_root = create_dir() blk_tree_root = create_dir(); (now blk_tree_root == NULL) ... dir = create_dir(name, blk_tree_root); Due to this race, t1 will create 'dir' in /debugfs but not /debugfs/block. Signed-off-by: -
Li Zefan authored
I found the timestamp is wrong: # echo bin > trace_option # echo blk > current_tracer # cat trace_pipe | blkparse -i - 8,0 0 0 0.000000000 504 A W ... ... 8,7 1 0 0.008534097 0 C R ... (should be 8.534097xxx) user-space blkparse expects the timestamp to be nanosecond. Signed-off-by:
-
- 30 Mar, 2009 1 commit
-
-
Ingo Molnar authored
Remove incorrectly introduced headers. Signed-off-by:
-
- 26 Mar, 2009 1 commit
-
-
Tom Zanussi authored
Impact: fix crash (hang) when using TRACE_EVENT_FORMAT filter files filters are only hooked up to the tracepoint events defined using TRACE_EVENT but not the tracers that use TRACE_EVENT_FORMAT, such as ftrace. Do not display the filter files at all for TRACE_EVENT_FORMAT events for the time being. Cc: Steven Rostedt <rostedt@goodmis.org> Cc: =?ISO-8859-1?Q?Fr=E9d=E9ric?= Weisbecker <fweisbec@gmail.com> LKML-Reference: <1237878882.8339.61.camel@charm-linux> Signed-off-by:
-
- 25 Mar, 2009 6 commits
-
-
Zhaolei authored
"Because when we call ftrace_free_rec we change the rec->ip to point to the next record in the chain. Something is very wrong if rec->ip >= s && rec->ip < e and the record is already free." "Note, use FTRACE_WARN_ON() macro. This way it shuts down ftrace if it is hit and helps to avoid further damage later." -- Steven Rostedt <rostedt@goodmis.org> Signed-off-by:Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by:
Steven Rostedt <srostedt@redhat.com>
-
Masami Hiramatsu authored
Fix interrupt emulation code in kretprobe-booster according to pt_regs update (es/ds change and gs adding). This issue has been reported on systemtap-bugzilla: http://sources.redhat.com/bugzilla/show_bug.cgi?id=9965 | On a -tip kernel on x86_32, kretprobe_example (from samples) triggers the | following backtrace when its retprobing a class of functions that cause a | copy_from/to_user(). | | BUG: sleeping function called from invalid context at mm/memory.c:3196 | in_atomic(): 0, irqs_disabled(): 1, pid: 2286, name: cat Signed-off-by:
Masami Hiramatsu <mhiramat@redhat.com> Acked-by:
Ananth N Mavinakayanahalli <ananth@in.ibm.com> Tested-by:
Bharata B Rao <bharata@linux.vnet.ibm.com> Cc: systemtap-ml <systemtap@sources.redhat.com> LKML-Reference: <49C7995C.2010601@redhat.com> Signed-off-by:
-
Lai Jiangshan authored
Impact: cpuset_wq should be initialized after init_workqueues() When I read /debugfs/tracing/trace_stat/workqueues, I got this: # CPU INSERTED EXECUTED NAME # | | | | 0 0 0 cpuset 0 285 285 events/0 0 2 2 work_on_cpu/0 0 1115 1115 khelper 0 325 325 kblockd/0 0 0 0 kacpid 0 0 0 kacpi_notify 0 0 0 ata/0 0 0 0 ata_aux 0 0 0 ksuspend_usbd 0 0 0 aio/0 0 0 0 nfsiod 0 0 0 kpsmoused 0 0 0 kstriped 0 0 0 kondemand/0 0 1 1 hid_compat 0 0 0 rpciod/0 1 64 64 events/1 1 2 2 work_on_cpu/1 1 5 5 kblockd/1 1 0 0 ata/1 1 0 0 aio/1 1 0 0 kondemand/1 1 0 0 rpciod/1 I found "cpuset" is at the earliest. I found a create_singlethread_workqueue() is earlier than init_workqueues(): kernel_init() ->cpuset_init_smp() ->create_singlethread_workqueue() ->do_basic_setup() ->init_workqueues() I think it's better that create_singlethread_workqueue() is called after workqueue subsystem has been initialized. Signed-off-by:
Lai Jiangshan <laijs@cn.fujitsu.com> Acked-by:
Frederic Weisbecker <fweisbec@gmail.com> Cc: Paul Menage <menage@google.com> Cc: miaoxie <miaox@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Andrew Morton <akpm@linux-foundation.org> LKML-Reference: <49C9F416.1050707@cn.fujitsu.com> Signed-off-by:
-
Lai Jiangshan authored
Empty lines separate cpus stat. After previous fix(trace_stat: keep original order) applied, the empty lines are displayed at incorrect position. Signed-off-by:
-
Lai Jiangshan authored
Impact: make trace_stat files show items with the original order trace_stat tracer reverse the items, it makes the output looks a little ugly. Example, when we read trace_stat/workqueues, we get cpu#7's stat. at first, and then cpu#6... cpu#0. Signed-off-by:
-
Lai Jiangshan authored
Impact: Fix incorrect way using seq_file's API Use SEQ_START_TOKEN instead of calling ->stat_headers() int seq_operation->start(). Signed-off-by:
-
- 24 Mar, 2009 16 commits
-
-
Jody McIntyre authored
Fix the tracepoint documentation to refer to "tracepoint-sample" instead of "tracepoint-example" to match what actually exists; fix the directory, and clarify how to compile. Change every instance of "example" in the sample tracepoint code to "sample" for consistency. Signed-off-by:
Jody McIntyre <scjody@sun.com> Acked-by:
Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> Cc: torvalds@linux-foundation.org LKML-Reference: <20090324200027.GH8294@clouds> Signed-off-by:
-
Lai Jiangshan authored
Impact: cleanup struct dyn_ftrace::ip has different usages in his lifecycle, we use union for it. And also for struct dyn_ftrace::flags. Signed-off-by:
-
Lai Jiangshan authored
Impact: fix PID output under namespaces When current namespace is not the global namespace, pid read from set_ftrace_pid is no correct. # ~/newpid_namespace_run bash # echo $$ 1 # echo 1 > set_ftrace_pid # cat set_ftrace_pid 3756 Since we write virtual PID to set_ftrace_pid, we need get virtual PID when we read it. Signed-off-by:
-
Steven Rostedt authored
Impact: give user a choice to show times spent while sleeping The user may want to see the time a function spent sleeping. This patch adds the trace option "sleep-time" to allow that. The "sleep-time" option is default on. echo sleep-time > /debug/tracing/trace_options produces: ------------------------------------------ 2) avahi-d-3428 => <idle>-0 ------------------------------------------ 2) | finish_task_switch() { 2) 0.621 us | _spin_unlock_irq(); 2) 2.202 us | } 2) ! 1002.197 us | } 2) ! 1003.521 us | } where as, echo nosleep-time > /debug/tracing/trace_options produces: 0) <idle>-0 => yum-upd-3416 ------------------------------------------ 0) | finish_task_switch() { 0) 0.643 us | _spin_unlock_irq(); 0) 2.342 us | } 0) + 41.302 us | } 0) + 42.453 us | } Signed-off-by: -
Steven Rostedt authored
Impact: more accurate timings The current method of function graph tracing does not take into account the time spent when a task is not running. This shows functions that call schedule have increased costs: 3) + 18.664 us | } ------------------------------------------ 3) <idle>-0 => kblockd-123 ------------------------------------------ 3) | finish_task_switch() { 3) 1.441 us | _spin_unlock_irq(); 3) 3.966 us | } 3) ! 2959.433 us | } 3) ! 2961.465 us | } This patch uses the tracepoint in the scheduling context switch to account for time that has elapsed while a task is scheduled out. Now we see: ------------------------------------------ 3) <idle>-0 => edac-po-1067 ------------------------------------------ 3) | finish_task_switch() { 3) 0.685 us | _spin_unlock_irq(); 3) 2.331 us | } 3) + 41.439 us | } 3) + 42.663 us | } Signed-off-by: -
Steven Rostedt authored
Impact: prevent crash due to multiple function graph tracers The function graph tracer can currently only handle a single tracer being registered. If another tracer registers with the function graph tracer it can crash the system. Signed-off-by: -
Steven Rostedt authored
This patch move the timestamp from happening in the arch specific code into the general code. This allows for better control by the tracer to time manipulation. Signed-off-by: -
Steven Rostedt authored
If the function profiler does not have any items recorded and one were to cat the function stat file, the kernel would take a BUG with a NULL pointer dereference. Looking further into this, I found that returning NULL from stat_start did not stop the stat logic, and would later call stat_next. This breaks from the way seq_file works, so I looked into fixing the stat code. This is where I noticed that the last next_entry is never freed. It is allocated, and if the stat_next returns NULL, the code breaks out of the loop, unlocks the mutex and exits. We never link the next_entry nor do we free it. Thus it is a real memory leak. This patch rearranges the code a bit to not only fix the memory leak, but also to act more like seq_file where nothing is printed if there is nothing to print. That is, stat_start returns NULL. Signed-off-by: -
Li Zefan authored
Impact: new feature, allow symbolic values in /debug/tracing/act_mask Print stringified act_mask instead of hex value: # cat act_mask read,write,barrier,sync,queue,requeue,issue,complete,fs,pc,ahead,meta, discard,drv_data # echo "meta,write" > act_mask # cat act_mask write,meta Also: - make act_mask accept "ahead", "meta", "discard" and "drv_data" - use strsep() instead of strchr() to parse user input - return -EINVAL if a token is not found in the mask map - fix a bug that 'value' is unsigned, so it can < 0 - propagate error value of blk_trace_mask2str() to userspace, but not always return -ENXIO. Signed-off-by:
-
Li Zefan authored
Impact: fix error flag output t_error() should return t->error but not t->sector. Signed-off-by:
-
Li Zefan authored
Impact: fix the output of IO type category characters Trace categories are the upper 16 bits, not the lower 16 bits. Signed-off-by:
-
Li Zefan authored
Impact: cleanup ddir_act and what2act always stay immutable. Signed-off-by:
-
Tom Zanussi authored
Impact: fix filter use boundary condition / crash Make sure filters for string fields don't use integer values and vice versa. Getting it wrong can crash the system or produce bogus results. Signed-off-by:
Tom Zanussi <tzanussi@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: =?ISO-8859-1?Q?Fr=E9d=E9ric?= Weisbecker <fweisbec@gmail.com> LKML-Reference: <1237878882.8339.61.camel@charm-linux> Signed-off-by:
-
Tom Zanussi authored
Impact: cleanup Instead of just using the trace_seq buffer to print the filters, use trace_seq_printf() as it was intended to be used. Reported-by:
-
Tom Zanussi authored
Impact: fix (small) per trace filter modification memory leak Free the current pred when clearing the filters via the filter files. Signed-off-by:
-
Tom Zanussi authored
Impact: cleanup No need to use the safe version here, so use list_for_each_entry instead of list_for_each_entry_safe in find_event_field(). Signed-off-by:
-
- 23 Mar, 2009 3 commits
-
-
Frederic Weisbecker authored
Impact: fix traces output Sometimes one can observe an imbalance in the traces between function calls and function return traces: func1() { } } The curly brace inside func1() is the return of another function nested inside func1. The return trace have been inserted in the buffer but not the entry. We are storing a return address on the function traces stack while we haven't inserted its entry on the buffer, hence the imbalance on the traces. This is because the tracers doesn't check all failures that can happen on buffer insertion. This patch reports the tracing recursion failures and the ring buffer failures. In such cases, we now restore the original return address for the function, giving up its return trace. Signed-off-by: -
Anton Vorontsov authored
commit 40ada30f ("tracing: clean up menu"), despite the "clean up" in its purpose, introduced a behavioural change for Kconfig symbols: we no longer able to select tracing support on PPC32 (because IRQFLAGS_SUPPORT isn't yet implemented). The IRQFLAGS_SUPPORT is not mandatory for most tracers, tracing core has a special case for platforms w/o irqflags (which, by the way, has become useless as of the commit above). Though according to Ingo Molnar, there was periodic build failures on weird, unmaintained architectures that had no irqflags-tracing support and hence didn't know the raw_irqs_save/restore primitives. Thus we'd better not enable irqflags-less tracing for all architectures. This patch restores the old behaviour for PPC32, and thus brings the tracing back. Other architectures can either add themselves to the exception list or (better) implement TRACE_IRQFLAGS_SUPPORT. Signed-off-by:
Anton Vorontsov <avorontsov@ru.mvista.com> Acked-b: Steven Rostedt <rostedt@goodmis.org> Cc: linuxppc-dev@ozlabs.org LKML-Reference: <20090323220724.GA9851@oksana.dev.rtsoft.ru> Signed-off-by:
-
Frederic Weisbecker authored
Impact: fix a crash with ftrace={nop,boot} parameter If the nop or initcall tracers are launched as boot tracers, they will attempt to create their option directory and files. But these tracers are registered very early and then assigned as "boot tracers" very early if asked to. Since they do this before debugfs has been registered (core initcall), a crash is triggered. Another early tracers could also come later. So we fix it by checking if debugfs is initialized before creating the root tracing directory. Signed-off-by:
-
