diff options
90 files changed, 6490 insertions, 836 deletions
diff --git a/Documentation/admin-guide/bootconfig.rst b/Documentation/admin-guide/bootconfig.rst new file mode 100644 index 000000000000..b342a6796392 --- /dev/null +++ b/Documentation/admin-guide/bootconfig.rst @@ -0,0 +1,190 @@ +.. SPDX-License-Identifier: GPL-2.0 + +.. _bootconfig: + +================== +Boot Configuration +================== + +:Author: Masami Hiramatsu <mhiramat@kernel.org> + +Overview +======== + +The boot configuration expands the current kernel command line to support +additional key-value data when booting the kernel in an efficient way. +This allows administrators to pass a structured-Key config file. + +Config File Syntax +================== + +The boot config syntax is a simple structured key-value. Each key consists +of dot-connected-words, and key and value are connected by ``=``. The value +has to be terminated by semi-colon (``;``) or newline (``\n``). +For array value, array entries are separated by comma (``,``). :: + +KEY[.WORD[...]] = VALUE[, VALUE2[...]][;] + +Unlike the kernel command line syntax, spaces are OK around the comma and ``=``. + +Each key word must contain only alphabets, numbers, dash (``-``) or underscore +(``_``). And each value only contains printable characters or spaces except +for delimiters such as semi-colon (``;``), new-line (``\n``), comma (``,``), +hash (``#``) and closing brace (``}``). + +If you want to use those delimiters in a value, you can use either double- +quotes (``"VALUE"``) or single-quotes (``'VALUE'``) to quote it. Note that +you can not escape these quotes. + +There can be a key which doesn't have value or has an empty value. Those keys +are used for checking if the key exists or not (like a boolean). + +Key-Value Syntax +---------------- + +The boot config file syntax allows user to merge partially same word keys +by brace. For example:: + + foo.bar.baz = value1 + foo.bar.qux.quux = value2 + +These can be written also in:: + + foo.bar { + baz = value1 + qux.quux = value2 + } + +Or more shorter, written as following:: + + foo.bar { baz = value1; qux.quux = value2 } + +In both styles, same key words are automatically merged when parsing it +at boot time. So you can append similar trees or key-values. + +Comments +-------- + +The config syntax accepts shell-script style comments. The comments starting +with hash ("#") until newline ("\n") will be ignored. + +:: + + # comment line + foo = value # value is set to foo. + bar = 1, # 1st element + 2, # 2nd element + 3 # 3rd element + +This is parsed as below:: + + foo = value + bar = 1, 2, 3 + +Note that you can not put a comment between value and delimiter(``,`` or +``;``). This means following config has a syntax error :: + + key = 1 # comment + ,2 + + +/proc/bootconfig +================ + +/proc/bootconfig is a user-space interface of the boot config. +Unlike /proc/cmdline, this file shows the key-value style list. +Each key-value pair is shown in each line with following style:: + + KEY[.WORDS...] = "[VALUE]"[,"VALUE2"...] + + +Boot Kernel With a Boot Config +============================== + +Since the boot configuration file is loaded with initrd, it will be added +to the end of the initrd (initramfs) image file. The Linux kernel decodes +the last part of the initrd image in memory to get the boot configuration +data. +Because of this "piggyback" method, there is no need to change or +update the boot loader and the kernel image itself. + +To do this operation, Linux kernel provides "bootconfig" command under +tools/bootconfig, which allows admin to apply or delete the config file +to/from initrd image. You can build it by the following command:: + + # make -C tools/bootconfig + +To add your boot config file to initrd image, run bootconfig as below +(Old data is removed automatically if exists):: + + # tools/bootconfig/bootconfig -a your-config /boot/initrd.img-X.Y.Z + +To remove the config from the image, you can use -d option as below:: + + # tools/bootconfig/bootconfig -d /boot/initrd.img-X.Y.Z + +Then add "bootconfig" on the normal kernel command line to tell the +kernel to look for the bootconfig at the end of the initrd file. + +Config File Limitation +====================== + +Currently the maximum config size size is 32KB and the total key-words (not +key-value entries) must be under 1024 nodes. +Note: this is not the number of entries but nodes, an entry must consume +more than 2 nodes (a key-word and a value). So theoretically, it will be +up to 512 key-value pairs. If keys contains 3 words in average, it can +contain 256 key-value pairs. In most cases, the number of config items +will be under 100 entries and smaller than 8KB, so it would be enough. +If the node number exceeds 1024, parser returns an error even if the file +size is smaller than 32KB. +Anyway, since bootconfig command verifies it when appending a boot config +to initrd image, user can notice it before boot. + + +Bootconfig APIs +=============== + +User can query or loop on key-value pairs, also it is possible to find +a root (prefix) key node and find key-values under that node. + +If you have a key string, you can query the value directly with the key +using xbc_find_value(). If you want to know what keys exist in the boot +config, you can use xbc_for_each_key_value() to iterate key-value pairs. +Note that you need to use xbc_array_for_each_value() for accessing +each array's value, e.g.:: + + vnode = NULL; + xbc_find_value("key.word", &vnode); + if (vnode && xbc_node_is_array(vnode)) + xbc_array_for_each_value(vnode, value) { + printk("%s ", value); + } + +If you want to focus on keys which have a prefix string, you can use +xbc_find_node() to find a node by the prefix string, and iterate +keys under the prefix node with xbc_node_for_each_key_value(). + +But the most typical usage is to get the named value under prefix +or get the named array under prefix as below:: + + root = xbc_find_node("key.prefix"); + value = xbc_node_find_value(root, "option", &vnode); + ... + xbc_node_for_each_array_value(root, "array-option", value, anode) { + ... + } + +This accesses a value of "key.prefix.option" and an array of +"key.prefix.array-option". + +Locking is not needed, since after initialization, the config becomes +read-only. All data and keys must be copied if you need to modify it. + + +Functions and structures +======================== + +.. kernel-doc:: include/linux/bootconfig.h +.. kernel-doc:: lib/bootconfig.c + diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst index 4433f3929481..f1d0ccffbe72 100644 --- a/Documentation/admin-guide/index.rst +++ b/Documentation/admin-guide/index.rst @@ -64,6 +64,7 @@ configure specific aspects of kernel behavior to your liking. binderfs binfmt-misc blockdev/index + bootconfig braille-console btmrvl cgroup-v1/index diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index ddc5ccdd4cd1..dbc22d684627 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -437,6 +437,12 @@ no delay (0). Format: integer + bootconfig [KNL] + Extended command line options can be added to an initrd + and this will cause the kernel to look for it. + + See Documentation/admin-guide/bootconfig.rst + bert_disable [ACPI] Disable BERT OS support on buggy BIOSes. diff --git a/Documentation/trace/boottime-trace.rst b/Documentation/trace/boottime-trace.rst new file mode 100644 index 000000000000..dcb390075ca1 --- /dev/null +++ b/Documentation/trace/boottime-trace.rst @@ -0,0 +1,184 @@ +.. SPDX-License-Identifier: GPL-2.0 + +================= +Boot-time tracing +================= + +:Author: Masami Hiramatsu <mhiramat@kernel.org> + +Overview +======== + +Boot-time tracing allows users to trace boot-time process including +device initialization with full features of ftrace including per-event +filter and actions, histograms, kprobe-events and synthetic-events, +and trace instances. +Since kernel command line is not enough to control these complex features, +this uses bootconfig file to describe tracing feature programming. + +Options in the Boot Config +========================== + +Here is the list of available options list for boot time tracing in +boot config file [1]_. All options are under "ftrace." or "kernel." +prefix. See kernel parameters for the options which starts +with "kernel." prefix [2]_. + +.. [1] See :ref:`Documentation/admin-guide/bootconfig.rst <bootconfig>` +.. [2] See :ref:`Documentation/admin-guide/kernel-parameters.rst <kernelparameters>` + +Ftrace Global Options +--------------------- + +Ftrace global options have "kernel." prefix in boot config, which means +these options are passed as a part of kernel legacy command line. + +kernel.tp_printk + Output trace-event data on printk buffer too. + +kernel.dump_on_oops [= MODE] + Dump ftrace on Oops. If MODE = 1 or omitted, dump trace buffer + on all CPUs. If MODE = 2, dump a buffer on a CPU which kicks Oops. + +kernel.traceoff_on_warning + Stop tracing if WARN_ON() occurs. + +kernel.fgraph_max_depth = MAX_DEPTH + Set MAX_DEPTH to maximum depth of fgraph tracer. + +kernel.fgraph_filters = FILTER[, FILTER2...] + Add fgraph tracing function filters. + +kernel.fgraph_notraces = FILTER[, FILTER2...] + Add fgraph non-tracing function filters. + + +Ftrace Per-instance Options +--------------------------- + +These options can be used for each instance including global ftrace node. + +ftrace.[instance.INSTANCE.]options = OPT1[, OPT2[...]] + Enable given ftrace options. + +ftrace.[instance.INSTANCE.]trace_clock = CLOCK + Set given CLOCK to ftrace's trace_clock. + +ftrace.[instance.INSTANCE.]buffer_size = SIZE + Configure ftrace buffer size to SIZE. You can use "KB" or "MB" + for that SIZE. + +ftrace.[instance.INSTANCE.]alloc_snapshot + Allocate snapshot buffer. + +ftrace.[instance.INSTANCE.]cpumask = CPUMASK + Set CPUMASK as trace cpu-mask. + +ftrace.[instance.INSTANCE.]events = EVENT[, EVENT2[...]] + Enable given events on boot. You can use a wild card in EVENT. + +ftrace.[instance.INSTANCE.]tracer = TRACER + Set TRACER to current tracer on boot. (e.g. function) + +ftrace.[instance.INSTANCE.]ftrace.filters + This will take an array of tracing function filter rules. + +ftrace.[instance.INSTANCE.]ftrace.notraces + This will take an array of NON-tracing function filter rules. + + +Ftrace Per-Event Options +------------------------ + +These options are setting per-event options. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.enable + Enable GROUP:EVENT tracing. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.filter = FILTER + Set FILTER rule to the GROUP:EVENT. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.actions = ACTION[, ACTION2[...]] + Set ACTIONs to the GROUP:EVENT. + +ftrace.[instance.INSTANCE.]event.kprobes.EVENT.probes = PROBE[, PROBE2[...]] + Defines new kprobe event based on PROBEs. It is able to define + multiple probes on one event, but those must have same type of + arguments. This option is available only for the event which + group name is "kprobes". + +ftrace.[instance.INSTANCE.]event.synthetic.EVENT.fields = FIELD[, FIELD2[...]] + Defines new synthetic event with FIELDs. Each field should be + "type varname". + +Note that kprobe and synthetic event definitions can be written under +instance node, but those are also visible from other instances. So please +take care for event name conflict. + + +Examples +======== + +For example, to add filter and actions for each event, define kprobe +events, and synthetic events with histogram, write a boot config like +below:: + + ftrace.event { + task.task_newtask { + filter = "pid < 128" + enable + } + kprobes.vfs_read { + probes = "vfs_read $arg1 $arg2" + filter = "common_pid < 200" + enable + } + synthetic.initcall_latency { + fields = "unsigned long func", "u64 lat" + actions = "hist:keys=func.sym,lat:vals=lat:sort=lat" + } + initcall.initcall_start { + actions = "hist:keys=func:ts0=common_timestamp.usecs" + } + initcall.initcall_finish { + actions = "hist:keys=func:lat=common_timestamp.usecs-$ts0:onmatch(initcall.initcall_start).initcall_latency(func,$lat)" + } + } + +Also, boot-time tracing supports "instance" node, which allows us to run +several tracers for different purpose at once. For example, one tracer +is for tracing functions starting with "user\_", and others tracing +"kernel\_" functions, you can write boot config as below:: + + ftrace.instance { + foo { + tracer = "function" + ftrace.filters = "user_*" + } + bar { + tracer = "function" + ftrace.filters = "kernel_*" + } + } + +The instance node also accepts event nodes so that each instance +can customize its event tracing. + +This boot-time tracing also supports ftrace kernel parameters via boot +config. +For example, following kernel parameters:: + + trace_options=sym-addr trace_event=initcall:* tp_printk trace_buf_size=1M ftrace=function ftrace_filter="vfs*" + +This can be written in boot config like below:: + + kernel { + trace_options = sym-addr + trace_event = "initcall:*" + tp_printk + trace_buf_size = 1M + ftrace = function + ftrace_filter = "vfs*" + } + +Note that parameters start with "kernel" prefix instead of "ftrace". diff --git a/Documentation/trace/events.rst b/Documentation/trace/events.rst index f7e1fcc0953c..ed79b220bd07 100644 --- a/Documentation/trace/events.rst +++ b/Documentation/trace/events.rst @@ -525,3 +525,518 @@ The following commands are supported: event counts (hitcount). See Documentation/trace/histogram.rst for details and examples. + +6.3 In-kernel trace event API +----------------------------- + +In most cases, the command-line interface to trace events is more than +sufficient. Sometimes, however, applications might find the need for +more complex relationships than can be expressed through a simple +series of linked command-line expressions, or putting together sets of +commands may be simply too cumbersome. An example might be an +application that needs to 'listen' to the trace stream in order to +maintain an in-kernel state machine detecting, for instance, when an +illegal kernel state occurs in the scheduler. + +The trace event subsystem provides an in-kernel API allowing modules +or other kernel code to generate user-defined 'synthetic' events at +will, which can be used to either augment the existing trace stream +and/or signal that a particular important state has occurred. + +A similar in-kernel API is also available for creating kprobe and +kretprobe events. + +Both the synthetic event and k/ret/probe event APIs are built on top +of a lower-level "dynevent_cmd" event command API, which is also +available for more specialized applications, or as the basis of other +higher-level trace event APIs. + +The API provided for these purposes is describe below and allows the +following: + + - dynamically creating synthetic event definitions + - dynamically creating kprobe and kretprobe event definitions + - tracing synthetic events from in-kernel code + - the low-level "dynevent_cmd" API + +6.3.1 Dyamically creating synthetic event definitions +----------------------------------------------------- + +There are a couple ways to create a new synthetic event from a kernel +module or other kernel code. + +The first creates the event in one step, using synth_event_create(). +In this method, the name of the event to create and an array defining +the fields is supplied to synth_event_create(). If successful, a +synthetic event with that name and fields will exist following that +call. For example, to create a new "schedtest" synthetic event: + + ret = synth_event_create("schedtest", sched_fields, + ARRAY_SIZE(sched_fields), THIS_MODULE); + +The sched_fields param in this example points to an array of struct +synth_field_desc, each of which describes an event field by type and +name: + + static struct synth_field_desc sched_fields[] = { + { .type = "pid_t", .name = "next_pid_field" }, + { .type = "char[16]", .name = "next_comm_field" }, + { .type = "u64", .name = "ts_ns" }, + { .type = "u64", .name = "ts_ms" }, + { .type = "unsigned int", .name = "cpu" }, + { .type = "char[64]", .name = "my_string_field" }, + { .type = "int", .name = "my_int_field" }, + }; + +See synth_field_size() for available types. If field_name contains [n] +the field is considered to be an array. + +If the event is created from within a module, a pointer to the module +must be passed to synth_event_create(). This will ensure that the +trace buffer won't contain unreadable events when the module is +removed. + +At this point, the event object is ready to be used for generating new +events. + +In the second method, the event is created in several steps. This +allows events to be created dynamically and without the need to create +and populate an array of fields beforehand. + +To use this method, an empty or partially empty synthetic event should +first be created using synth_event_gen_cmd_start() or +synth_event_gen_cmd_array_start(). For synth_event_gen_cmd_start(), +the name of the event along with one or more pairs of args each pair +representing a 'type field_name;' field specification should be +supplied. For synth_event_gen_cmd_array_start(), the name of the +event along with an array of struct synth_field_desc should be +supplied. Before calling synth_event_gen_cmd_start() or +synth_event_gen_cmd_array_start(), the user should create and +initialize a dynevent_cmd object using synth_event_cmd_init(). + +For example, to create a new "schedtest" synthetic event with two +fields: + + struct dynevent_cmd cmd; + char *buf; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + + /* Before generating the command, initialize the cmd object */ + synth_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + ret = synth_event_gen_cmd_start(&cmd, "schedtest", THIS_MODULE, + "pid_t", "next_pid_field", + "u64", "ts_ns"); + +Alternatively, using an array of struct synth_field_desc fields +containing the same information: + + ret = synth_event_gen_cmd_array_start(&cmd, "schedtest", THIS_MODULE, + fields, n_fields); + +Once the synthetic event object has been created, it can then be +populated with more fields. Fields are added one by one using +synth_event_add_field(), supplying the dynevent_cmd object, a field +type, and a field name. For example, to add a new int field named +"intfield", the following call should be made: + + ret = synth_event_add_field(&cmd, "int", "intfield"); + +See synth_field_size() for available types. If field_name contains [n] +the field is considered to be an array. + +A group of fields can also be added all at once using an array of +synth_field_desc with add_synth_fields(). For example, this would add +just the first four sched_fields: + + ret = synth_event_add_fields(&cmd, sched_fields, 4); + +If you already have a string of the form 'type field_name', +synth_event_add_field_str() can be used to add it as-is; it will +also automatically append a ';' to the string. + +Once all the fields have been added, the event should be finalized and +registered by calling the synth_event_gen_cmd_end() function: + + ret = synth_event_gen_cmd_end(&cmd); + +At this point, the event object is ready to be used for tracing new +events. + +6.3.3 Tracing synthetic events from in-kernel code +-------------------------------------------------- + +To trace a synthetic event, there are several options. The first +option is to trace the event in one call, using synth_event_trace() +with a variable number of values, or synth_event_trace_array() with an +array of values to be set. A second option can be used to avoid the +need for a pre-formed array of values or list of arguments, via +synth_event_trace_start() and synth_event_trace_end() along with +synth_event_add_next_val() or synth_event_add_val() to add the values +piecewise. + +6.3.3.1 Tracing a synthetic event all at once +--------------------------------------------- + +To trace a synthetic event all at once, the synth_event_trace() or +synth_event_trace_array() functions can be used. + +The synth_event_trace() function is passed the trace_event_file +representing the synthetic event (which can be retrieved using +trace_get_event_file() using the synthetic event name, "synthetic" as +the system name, and the trace instance name (NULL if using the global +trace array)), along with an variable number of u64 args, one for each +synthetic event field, and the number of values being passed. + +So, to trace an event corresponding to the synthetic event definition +above, code like the following could be used: + + ret = synth_event_trace(create_synth_test, 7, /* number of values */ + 444, /* next_pid_field */ + (u64)"clackers", /* next_comm_field */ + 1000000, /* ts_ns */ + 1000, /* ts_ms */ + smp_processor_id(),/* cpu */ + (u64)"Thneed", /* my_string_field */ + 999); /* my_int_field */ + +All vals should be cast to u64, and string vals are just pointers to +strings, cast to u64. Strings will be copied into space reserved in +the event for the string, using these pointers. + +Alternatively, the synth_event_trace_array() function can be used to +accomplish the same thing. It is passed the trace_event_file +representing the synthetic event (which can be retrieved using +trace_get_event_file() using the synthetic event name, "synthetic" as +the system name, and the trace instance name (NULL if using the global +trace array)), along with an array of u64, one for each synthetic +event field. + +To trace an event corresponding to the synthetic event definition +above, code like the following could be used: + + u64 vals[7]; + + vals[0] = 777; /* next_pid_field */ + vals[1] = (u64)"tiddlywinks"; /* next_comm_field */ + vals[2] = 1000000; /* ts_ns */ + vals[3] = 1000; /* ts_ms */ + vals[4] = smp_processor_id(); /* cpu */ + vals[5] = (u64)"thneed"; /* my_string_field */ + vals[6] = 398; /* my_int_field */ + +The 'vals' array is just an array of u64, the number of which must +match the number of field in the synthetic event, and which must be in +the same order as the synthetic event fields. + +All vals should be cast to u64, and string vals are just pointers to +strings, cast to u64. Strings will be copied into space reserved in +the event for the string, using these pointers. + +In order to trace a synthetic event, a pointer to the trace event file +is needed. The trace_get_event_file() function can be used to get +it - it will find the file in the given trace instance (in this case +NULL since the top trace array is being used) while at the same time +preventing the instance containing it from going away: + + schedtest_event_file = trace_get_event_file(NULL, "synthetic", + "schedtest"); + +Before tracing the event, it should be enabled in some way, otherwise +the synthetic event won't actually show up in the trace buffer. + +To enable a synthetic event from the kernel, trace_array_set_clr_event() +can be used (which is not specific to synthetic events, so does need +the "synthetic" system name to be specified explicitly). + +To enable the event, pass 'true' to it: + + trace_array_set_clr_event(schedtest_event_file->tr, + "synthetic", "schedtest", true); + +To disable it pass false: + + trace_array_set_clr_event(schedtest_event_file->tr, + "synthetic", "schedtest", false); + +Finally, synth_event_trace_array() can be used to actually trace the +event, which should be visible in the trace buffer afterwards: + + ret = synth_event_trace_array(schedtest_event_file, vals, + ARRAY_SIZE(vals)); + +To remove the synthetic event, the event should be disabled, and the +trace instance should be 'put' back using trace_put_event_file(): + + trace_array_set_clr_event(schedtest_event_file->tr, + "synthetic", "schedtest", false); + trace_put_event_file(schedtest_event_file); + +If those have been successful, synth_event_delete() can be called to +remove the event: + + ret = synth_event_delete("schedtest"); + +6.3.3.1 Tracing a synthetic event piecewise +------------------------------------------- + +To trace a synthetic using the piecewise method described above, the +synth_event_trace_start() function is used to 'open' the synthetic +event trace: + + struct synth_trace_state trace_state; + + ret = synth_event_trace_start(schedtest_event_file, &trace_state); + +It's passed the trace_event_file representing the synthetic event +using the same methods as described above, along with a pointer to a +struct synth_trace_state object, which will be zeroed before use and +used to maintain state between this and following calls. + +Once the event has been opened, which means space for it has been +reserved in the trace buffer, the individual fields can be set. There +are two ways to do that, either one after another for each field in +the event, which requires no lookups, or by name, which does. The +tradeoff is flexibility in doing the assignments vs the cost of a +lookup per field. + +To assign the values one after the other without lookups, +synth_event_add_next_val() should be used. Each call is passed the +same synth_trace_state object used in the synth_event_trace_start(), +along with the value to set the next field in the event. After each +field is set, the 'cursor' points to the next field, which will be set +by the subsequent call, continuing until all the fields have been set +in order. The same sequence of calls as in the above examples using +this method would be (without error-handling code): + + /* next_pid_field */ + ret = synth_event_add_next_val(777, &trace_state); + + /* next_comm_field */ + ret = synth_event_add_next_val((u64)"slinky", &trace_state); + + /* ts_ns */ + ret = synth_event_add_next_val(1000000, &trace_state); + + /* ts_ms */ + ret = synth_event_add_next_val(1000, &trace_state); + + /* cpu */ + ret = synth_event_add_next_val(smp_processor_id(), &trace_state); + + /* my_string_field */ + ret = synth_event_add_next_val((u64)"thneed_2.01", &trace_state); + + /* my_int_field */ + ret = synth_event_add_next_val(395, &trace_state); + +To assign the values in any order, synth_event_add_val() should be +used. Each call is passed the same synth_trace_state object used in +the synth_event_trace_start(), along with the field name of the field +to set and the value to set it to. The same sequence of calls as in +the above examples using this method would be (without error-handling +code): + + ret = synth_event_add_val("next_pid_field", 777, &trace_state); + ret = synth_event_add_val("next_comm_field", (u64)"silly putty", + &trace_state); + ret = synth_event_add_val("ts_ns", 1000000, &trace_state); + ret = synth_event_add_val("ts_ms", 1000, &trace_state); + ret = synth_event_add_val("cpu", smp_processor_id(), &trace_state); + ret = synth_event_add_val("my_string_field", (u64)"thneed_9", + &trace_state); + ret = synth_event_add_val("my_int_field", 3999, &trace_state); + +Note that synth_event_add_next_val() and synth_event_add_val() are +incompatible if used within the same trace of an event - either one +can be used but not both at the same time. + +Finally, the event won't be actually traced until it's 'closed', +which is done using synth_event_trace_end(), which takes only the +struct synth_trace_state object used in the previous calls: + + ret = synth_event_trace_end(&trace_state); + +Note that synth_event_trace_end() must be called at the end regardless +of whether any of the add calls failed (say due to a bad field name +being passed in). + +6.3.4 Dyamically creating kprobe and kretprobe event definitions +---------------------------------------------------------------- + +To create a kprobe or kretprobe trace event from kernel code, the +kprobe_event_gen_cmd_start() or kretprobe_event_gen_cmd_start() +functions can be used. + +To create a kprobe event, an empty or partially empty kprobe event +should first be created using kprobe_event_gen_cmd_start(). The name +of the event and the probe location should be specfied along with one +or args each representing a probe field should be supplied to this +function. Before calling kprobe_event_gen_cmd_start(), the user +should create and initialize a dynevent_cmd object using +kprobe_event_cmd_init(). + +For example, to create a new "schedtest" kprobe event with two fields: + + struct dynevent_cmd cmd; + char *buf; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + + /* Before generating the command, initialize the cmd object */ + kprobe_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + /* + * Define the gen_kprobe_test event with the first 2 kprobe + * fields. + */ + ret = kprobe_event_gen_cmd_start(&cmd, "gen_kprobe_test", "do_sys_open", + "dfd=%ax", "filename=%dx"); + +Once the kprobe event object has been created, it can then be +populated with more fields. Fields can be added using +kprobe_event_add_fields(), supplying the dynevent_cmd object along +with a variable arg list of probe fields. For example, to add a +couple additional fields, the following call could be made: + + ret = kprobe_event_add_fields(&cmd, "flags=%cx", "mode=+4($stack)"); + +Once all the fields have been added, the event should be finalized and +registered by calling the kprobe_event_gen_cmd_end() or +kretprobe_event_gen_cmd_end() functions, depending on whether a kprobe +or kretprobe command was started: + + ret = kprobe_event_gen_cmd_end(&cmd); + +or + + ret = kretprobe_event_gen_cmd_end(&cmd); + +At this point, the event object is ready to be used for tracing new +events. + +Similarly, a kretprobe event can be created using +kretprobe_event_gen_cmd_start() with a probe name and location and +additional params such as $retval: + + ret = kretprobe_event_gen_cmd_start(&cmd, "gen_kretprobe_test", + "do_sys_open", "$retval"); + +Similar to the synthetic event case, code like the following can be +used to enable the newly created kprobe event: + + gen_kprobe_test = trace_get_event_file(NULL, "kprobes", "gen_kprobe_test"); + + ret = trace_array_set_clr_event(gen_kprobe_test->tr, + "kprobes", "gen_kprobe_test", true); + +Finally, also similar to synthetic events, the following code can be +used to give the kprobe event file back and delete the event: + + trace_put_event_file(gen_kprobe_test); + + ret = kprobe_event_delete("gen_kprobe_test"); + +6.3.4 The "dynevent_cmd" low-level API +-------------------------------------- + +Both the in-kernel synthetic event and kprobe interfaces are built on +top of a lower-level "dynevent_cmd" interface. This interface is +meant to provide the basis for higher-level interfaces such as the +synthetic and kprobe interfaces, which can be used as examples. + +The basic idea is simple and amounts to providing a general-purpose +layer that can be used to generate trace event commands. The +generated command strings can then be passed to the command-parsing +and event creation code that already exists in the trace event +subystem for creating the corresponding trace events. + +In a nutshell, the way it works is that the higher-level interface +code creates a struct dynevent_cmd object, then uses a couple +functions, dynevent_arg_add() and dynevent_arg_pair_add() to build up +a command string, which finally causes the command to be executed +using the dynevent_create() function. The details of the interface +are described below. + +The first step in building a new command string is to create and +initialize an instance of a dynevent_cmd. Here, for instance, we +create a dynevent_cmd on the stack and initialize it: + + struct dynevent_cmd cmd; + char *buf; + int ret; + + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + + dynevent_cmd_init(cmd, buf, maxlen, DYNEVENT_TYPE_FOO, + foo_event_run_command); + +The dynevent_cmd initialization needs to be given a user-specified +buffer and the length of the buffer (MAX_DYNEVENT_CMD_LEN can be used +for this purpose - at 2k it's generally too big to be comfortably put +on the stack, so is dynamically allocated), a dynevent type id, which +is meant to be used to check that further API calls are for the +correct command type, and a pointer to an event-specific run_command() +callback that will be called to actually execute the event-specific +command function. + +Once that's done, the command string can by built up by successive +calls to argument-adding functions. + +To add a single argument, define and initialize a struct dynevent_arg +or struct dynevent_arg_pair object. Here's an example of the simplest +possible arg addition, which is simply to append the given string as +a whitespace-separated argument to the command: + + struct dynevent_arg arg; + + dynevent_arg_init(&arg, NULL, 0); + + arg.str = name; + + ret = dynevent_arg_add(cmd, &arg); + +The arg object is first initialized using dynevent_arg_init() and in +this case the parameters are NULL or 0, which means there's no +optional sanity-checking function or separator appended to the end of +the arg. + +Here's another more complicated example using an 'arg pair', which is +used to create an argument that consists of a couple components added +together as a unit, for example, a 'type field_name;' arg or a simple +expression arg e.g. 'flags=%cx': + + struct dynevent_arg_pair arg_pair; + + dynevent_arg_pair_init(&arg_pair, dynevent_foo_check_arg_fn, 0, ';'); + + arg_pair.lhs = type; + arg_pair.rhs = name; + + ret = dynevent_arg_pair_add(cmd, &arg_pair); + +Again, the arg_pair is first initialized, in this case with a callback +function used to check the sanity of the args (for example, that +neither part of the pair is NULL), along with a character to be used +to add an operator between the pair (here none) and a separator to be +appended onto the end of the arg pair (here ';'). + +There's also a dynevent_str_add() function that can be used to simply +add a string as-is, with no spaces, delimeters, or arg check. + +Any number of dynevent_*_add() calls can be made to build up the string +(until its length surpasses cmd->maxlen). When all the arguments have +been added and the command string is complete, the only thing left to +do is run the command, which happens by simply calling +dynevent_create(): + + ret = dynevent_create(&cmd); + +At that point, if the return value is 0, the dynamic event has been +created and is ready to use. + +See the dynevent_cmd function definitions themselves for the details +of the API. diff --git a/Documentation/trace/index.rst b/Documentation/trace/index.rst index 04acd277c5f6..fa9e1c730f6a 100644 --- a/Documentation/trace/index.rst +++ b/Documentation/trace/index.rst @@ -19,6 +19,7 @@ Linux Tracing Technologies events-msr mmiotrace histogram + boottime-trace hwlat_detector intel_th stm diff --git a/Documentation/trace/kprobetrace.rst b/Documentation/trace/kprobetrace.rst index 55993055902c..705d73087099 100644 --- a/Documentation/trace/kprobetrace.rst +++ b/Documentation/trace/kprobetrace.rst @@ -97,6 +97,7 @@ which shows given pointer in "symbol+offset" style. For $comm, the default type is "string"; any other type is invalid. .. _user_mem_access: + User Memory Access ------------------ Kprobe events supports user-space memory access. For that purpose, you can use diff --git a/MAINTAINERS b/MAINTAINERS index 777f87ac2551..5c87b133dfa6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15934,6 +15934,15 @@ S: Supported F: Documentation/networking/device_drivers/stmicro/ F: drivers/net/ethernet/stmicro/stmmac/ +EXTRA BOOT CONFIG +M: Masami Hiramatsu <mhiramat@kernel.org> +S: Maintained +F: lib/bootconfig.c +F: fs/proc/bootconfig.c +F: include/linux/bootconfig.h +F: tools/bootconfig/* +F: Documentation/admin-guide/bootconfig.rst + SUN3/3X M: Sam Creasey <sammy@sammy.net> W: http://sammy.net/sun3/ diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index eda2633a393d..9210a95cb4e6 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -32,7 +32,7 @@ #define OP_BUFFER_FLAGS 0 -static struct ring_buffer *op_ring_buffer; +static struct trace_buffer *op_ring_buffer; DEFINE_PER_CPU(struct oprofile_cpu_buffer, op_cpu_buffer); static void wq_sync_buffer(struct work_struct *work); diff --git a/fs/proc/Makefile b/fs/proc/Makefile index ead487e80510..bd08616ed8ba 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -33,3 +33,4 @@ proc-$(CONFIG_PROC_KCORE) += kcore.o proc-$(CONFIG_PROC_VMCORE) += vmcore.o proc-$(CONFIG_PRINTK) += kmsg.o proc-$(CONFIG_PROC_PAGE_MONITOR) += page.o +proc-$(CONFIG_BOOT_CONFIG) += bootconfig.o diff --git a/fs/proc/bootconfig.c b/fs/proc/bootconfig.c new file mode 100644 index 000000000000..9955d75c0585 --- /dev/null +++ b/fs/proc/bootconfig.c @@ -0,0 +1,89 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * /proc/bootconfig - Extra boot configuration + */ +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/printk.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/bootconfig.h> +#include <linux/slab.h> + +static char *saved_boot_config; + +static int boot_config_proc_show(struct seq_file *m, void *v) +{ + if (saved_boot_config) + seq_puts(m, saved_boot_config); + return 0; +} + +/* Rest size of buffer */ +#define rest(dst, end) ((end) > (dst) ? (end) - (dst) : 0) + +/* Return the needed total length if @size is 0 */ +static int __init copy_xbc_key_value_list(char *dst, size_t size) +{ + struct xbc_node *leaf, *vnode; + const char *val; + char *key, *end = dst + size; + int ret = 0; + + key = kzalloc(XBC_KEYLEN_MAX, GFP_KERNEL); + + xbc_for_each_key_value(leaf, val) { + ret = xbc_node_compose_key(leaf, key, XBC_KEYLEN_MAX); + if (ret < 0) + break; + ret = snprintf(dst, rest(dst, end), "%s = ", key); + if (ret < 0) + break; + dst += ret; + vnode = xbc_node_get_child(leaf); + if (vnode && xbc_node_is_array(vnode)) { + xbc_array_for_each_value(vnode, val) { + ret = snprintf(dst, rest(dst, end), "\"%s\"%s", + val, vnode->next ? ", " : "\n"); + if (ret < 0) + goto out; + dst += ret; + } + } else { + ret = snprintf(dst, rest(dst, end), "\"%s\"\n", val); + if (ret < 0) + break; + dst += ret; + } + } +out: + kfree(key); + + return ret < 0 ? ret : dst - (end - size); +} + +static int __init proc_boot_config_init(void) +{ + int len; + + len = copy_xbc_key_value_list(NULL, 0); + if (len < 0) + return len; + + if (len > 0) { + saved_boot_config = kzalloc(len + 1, GFP_KERNEL); + if (!saved_boot_config) + return -ENOMEM; + + len = copy_xbc_key_value_list(saved_boot_config, len + 1); + if (len < 0) { + kfree(saved_boot_config); + return len; + } + } + + proc_create_single("bootconfig", 0, NULL, boot_config_proc_show); + + return 0; +} +fs_initcall(proc_boot_config_init); diff --git a/include/linux/bootconfig.h b/include/linux/bootconfig.h new file mode 100644 index 000000000000..7e18c939663e --- /dev/null +++ b/include/linux/bootconfig.h @@ -0,0 +1,224 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_XBC_H +#define _LINUX_XBC_H +/* + * Extra Boot Config + * Copyright (C) 2019 Linaro Ltd. + * Author: Masami Hiramatsu <mhiramat@kernel.org> + */ + +#include <linux/kernel.h> +#include <linux/types.h> + +/* XBC tree node */ +struct xbc_node { + u16 next; + u16 child; + u16 parent; + u16 data; +} __attribute__ ((__packed__)); + +#define XBC_KEY 0 +#define XBC_VALUE (1 << 15) +/* Maximum size of boot config is 32KB - 1 */ +#define XBC_DATA_MAX (XBC_VALUE - 1) + +#define XBC_NODE_MAX 1024 +#define XBC_KEYLEN_MAX 256 +#define XBC_DEPTH_MAX 16 + +/* Node tree access raw APIs */ +struct xbc_node * __init xbc_root_node(void); +int __init xbc_node_index(struct xbc_node *node); +struct xbc_node * __init xbc_node_get_parent(struct xbc_node *node); +struct xbc_node * __init xbc_node_get_child(struct xbc_node *node); +struct xbc_node * __init xbc_node_get_next(struct xbc_node *node); +const char * __init xbc_node_get_data(struct xbc_node *node); + +/** + * xbc_node_is_value() - Test the node is a value node + * @node: An XBC node. + * + * Test the @node is a value node and return true if a value node, false if not. + */ +static inline __init bool xbc_node_is_value(struct xbc_node *node) +{ + return node->data & XBC_VALUE; +} + +/** + * xbc_node_is_key() - Test the node is a key node + * @node: An XBC node. + * + * Test the @node is a key node and return true if a key node, false if not. + */ +static inline __init bool xbc_node_is_key(struct xbc_node *node) +{ + return !xbc_node_is_value(node); +} + +/** + * xbc_node_is_array() - Test the node is an arraied value node + * @node: An XBC node. + * + * Test the @node is an arraied value node. + */ +static inline __init bool xbc_node_is_array(struct xbc_node *node) +{ + return xbc_node_is_value(node) && node->next != 0; +} + +/** + * xbc_node_is_leaf() - Test the node is a leaf key node + * @node: An XBC node. + * + * Test the @node is a leaf key node which is a key node and has a value node + * or no child. Returns true if it is a leaf node, or false if not. + */ +static inline __init bool xbc_node_is_leaf(struct xbc_node *node) +{ + return xbc_node_is_key(node) && + (!node->child || xbc_node_is_value(xbc_node_get_child(node))); +} + +/* Tree-based key-value access APIs */ +struct xbc_node * __init xbc_node_find_child(struct xbc_node *parent, + const char *key); + +const char * __init xbc_node_find_value(struct xbc_node *parent, + const char *key, + struct xbc_node **vnode); + +struct xbc_node * __init xbc_node_find_next_leaf(struct xbc_node *root, + struct xbc_node *leaf); + +const char * __init xbc_node_find_next_key_value(struct xbc_node *root, + struct xbc_node **leaf); + +/** + * xbc_find_value() - Find a value which matches the key + * @key: Search key + * @vnode: A container pointer of XBC value node. + * + * Search a value whose key matches @key from whole of XBC tree and return + * the value if found. Found value node is stored in *@vnode. + * Note that this can return 0-length string and store NULL in *@vnode for + * key-only (non-value) entry. + */ +static inline const char * __init +xbc_find_value(const char *key, struct xbc_node **vnode) +{ + return xbc_node_find_value(NULL, key, vnode); +} + +/** + * xbc_find_node() - Find a node which matches the key + * @key: Search key + * + * Search a (key) node whose key matches @key from whole of XBC tree and + * return the node if found. If not found, returns NULL. + */ +static inline struct xbc_node * __init xbc_find_node(const char *key) +{ + return xbc_node_find_child(NULL, key); +} + +/** + * xbc_array_for_each_value() - Iterate value nodes on an array + * @anode: An XBC arraied value node + * @value: A value + * + * Iterate array value nodes and values starts from @anode. This is expected to + * be used with xbc_find_value() and xbc_node_find_value(), so that user can + * process each array entry node. + */ +#define xbc_array_for_each_value(anode, value) \ + for (value = xbc_node_get_data(anode); anode != NULL ; \ + anode = xbc_node_get_next(anode), \ + value = anode ? xbc_node_get_data(anode) : NULL) + +/** + * xbc_node_for_each_child() - Iterate child nodes + * @parent: An XBC node. + * @child: Iterated XBC node. + * + * Iterate child nodes of @parent. Each child nodes are stored to @child. + */ +#define xbc_node_for_each_child(parent, child) \ + for (child = xbc_node_get_child(parent); child != NULL ; \ + child = xbc_node_get_next(child)) + +/** + * xbc_node_for_each_array_value() - Iterate array entries of geven key + * @node: An XBC node. + * @key: A key string searched under @node + * @anode: Iterated XBC node of array entry. + * @value: Iterated value of array entry. + * + * Iterate array entries of given @key under @node. Each array entry node + * is stroed to @anode and @value. If the @node doesn't have @key node, + * it does nothing. + * Note that even if the found key node has only one value (not array) + * this executes block once. Hoever, if the found key node has no value + * (key-only node), this does nothing. So don't use this for testing the + * key-value pair existence. + */ +#define xbc_node_for_each_array_value(node, key, anode, value) \ + for (value = xbc_node_find_value(node, key, &anode); value != NULL; \ + anode = xbc_node_get_next(anode), \ + value = anode ? xbc_node_get_data(anode) : NULL) + +/** + * xbc_node_for_each_key_value() - Iterate key-value pairs under a node + * @node: An XBC node. + * @knode: Iterated key node + * @value: Iterated value string + * + * Iterate key-value pairs under @node. Each key node and value string are + * stored in @knode and @value respectively. + */ +#define xbc_node_for_each_key_value(node, knode, value) \ + for (knode = NULL, value = xbc_node_find_next_key_value(node, &knode);\ + knode != NULL; value = xbc_node_find_next_key_value(node, &knode)) + +/** + * xbc_for_each_key_value() - Iterate key-value pairs + * @knode: Iterated key node + * @value: Iterated value string + * + * Iterate key-value pairs in whole XBC tree. Each key node and value string + * are stored in @knode and @value respectively. + */ +#define xbc_for_each_key_value(knode, value) \ + xbc_node_for_each_key_value(NULL, knode, value) + +/* Compose partial key */ +int __init xbc_node_compose_key_after(struct xbc_node *root, + struct xbc_node *node, char *buf, size_t size); + +/** + * xbc_node_compose_key() - Compose full key string of the XBC node + * @node: An XBC node. + * @buf: A buffer to store the key. + * @size: The size of the @buf. + * + * Compose the full-length key of the @node into @buf. Returns the total + * length of the key stored in @buf. Or returns -EINVAL if @node is NULL, + * and -ERANGE if the key depth is deeper than max depth. + */ +static inline int __init xbc_node_compose_key(struct xbc_node *node, + char *buf, size_t size) +{ + return xbc_node_compose_key_after(NULL, node, buf, size); +} + +/* XBC node initializer */ +int __init xbc_init(char *buf); + +/* XBC cleanup data structures */ +void __init xbc_destroy_all(void); + +/* Debug dump functions */ +void __init xbc_debug_dump(void); + +#endif diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 6d4c22aee384..cf65763af0cb 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -582,7 +582,7 @@ struct swevent_hlist { #define PERF_ATTACH_ITRACE 0x10 struct perf_cgroup; -struct ring_buffer; +struct perf_buffer; struct pmu_event_list { raw_spinlock_t lock; @@ -694,7 +694,7 @@ struct perf_event { struct mutex mmap_mutex; atomic_t mmap_count; - struct ring_buffer *rb; + struct perf_buffer *rb; struct list_head rb_entry; unsigned long rcu_batches; int rcu_pending; @@ -854,7 +854,7 @@ struct perf_cpu_context { struct perf_output_handle { struct perf_event *event; - struct ring_buffer *rb; + struct perf_buffer *rb; unsigned long wakeup; unsigned long size; u64 aux_flags; diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 1a40277b512c..df0124eabece 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -6,7 +6,7 @@ #include <linux/seq_file.h> #include <linux/poll.h> -struct ring_buffer; +struct trace_buffer; struct ring_buffer_iter; /* @@ -77,13 +77,13 @@ u64 ring_buffer_event_time_stamp(struct ring_buffer_event *event); * else * ring_buffer_unlock_commit(buffer, event); */ -void ring_buffer_discard_commit(struct ring_buffer *buffer, +void ring_buffer_discard_commit(struct trace_buffer *buffer, struct ring_buffer_event *event); /* * size is in bytes for each per CPU buffer. */ -struct ring_buffer * +struct trace_buffer * __ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *key); /* @@ -97,38 +97,38 @@ __ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *k __ring_buffer_alloc((size), (flags), &__key); \ }) -int ring_buffer_wait(struct ring_buffer *buffer, int cpu, int full); -__poll_t ring_buffer_poll_wait(struct ring_buffer *buffer, int cpu, +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full); +__poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, struct file *filp, poll_table *poll_table); #define RING_BUFFER_ALL_CPUS -1 -void ring_buffer_free(struct ring_buffer *buffer); +void ring_buffer_free(struct trace_buffer *buffer); -int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size, int cpu); +int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, int cpu); -void ring_buffer_change_overwrite(struct ring_buffer *buffer, int val); +void ring_buffer_change_overwrite(struct trace_buffer *buffer, int val); -struct ring_buffer_event *ring_buffer_lock_reserve(struct ring_buffer *buffer, +struct ring_buffer_event *ring_buffer_lock_reserve(struct trace_buffer *buffer, unsigned long length); -int ring_buffer_unlock_commit(struct ring_buffer *buffer, +int ring_buffer_unlock_commit(struct trace_buffer *buffer, struct ring_buffer_event *event); -int ring_buffer_write(struct ring_buffer *buffer, +int ring_buffer_write(struct trace_buffer *buffer, unsigned long length, void *data); -void ring_buffer_nest_start(struct ring_buffer *buffer); -void ring_buffer_nest_end(struct ring_buffer *buffer); +void ring_buffer_nest_start(struct trace_buffer *buffer); +void ring_buffer_nest_end(struct trace_buffer *buffer); struct ring_buffer_event * -ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts, +ring_buffer_peek(struct trace_buffer *buffer, int cpu, u64 *ts, unsigned long *lost_events); struct ring_buffer_event * -ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, +ring_buffer_consume(struct trace_buffer *buffer, int cpu, u64 *ts, unsigned long *lost_events); struct ring_buffer_iter * -ring_buffer_read_prepare(struct ring_buffer *buffer, int cpu, gfp_t flags); +ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags); void ring_buffer_read_prepare_sync(void); void ring_buffer_read_start(struct ring_buffer_iter *iter); void ring_buffer_read_finish(struct ring_buffer_iter *iter); @@ -140,59 +140,59 @@ ring_buffer_read(struct ring_buffer_iter *iter, u64 *ts); void ring_buffer_iter_reset(struct ring_buffer_iter *iter); int ring_buffer_iter_empty(struct ring_buffer_iter *iter); -unsigned long ring_buffer_size(struct ring_buffer *buffer, int cpu); +unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu); -void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu); -void ring_buffer_reset(struct ring_buffer *buffer); +void ring_buffer_reset_cpu(struct trace_buffer *buffer, int cpu); +void ring_buffer_reset(struct trace_buffer *buffer); #ifdef CONFIG_RING_BUFFER_ALLOW_SWAP -int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, - struct ring_buffer *buffer_b, int cpu); +int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, + struct trace_buffer *buffer_b, int cpu); #else static inline int -ring_buffer_swap_cpu(struct ring_buffer *buffer_a, - struct ring_buffer *buffer_b, int cpu) +ring_buffer_swap_cpu(struct trace_buffer *buffer_a, + struct trace_buffer *buffer_b, int cpu) { return -ENODEV; } #endif -bool ring_buffer_empty(struct ring_buffer *buffer); -bool ring_buffer_empty_cpu(struct ring_buffer *buffer, int cpu); - -void ring_buffer_record_disable(struct ring_buffer *buffer); -void ring_buffer_record_enable(struct ring_buffer *buffer); -void ring_buffer_record_off(struct ring_buffer *buffer); -void ring_buffer_record_on(struct ring_buffer *buffer); -bool ring_buffer_record_is_on(struct ring_buffer *buffer); -bool ring_buffer_record_is_set_on(struct ring_buffer *buffer); -void ring_buffer_record_disable_cpu(struct ring_buffer *buffer, int cpu); -void ring_buffer_record_enable_cpu(struct ring_buffer *buffer, int cpu); - -u64 ring_buffer_oldest_event_ts(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_bytes_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_entries(struct ring_buffer *buffer); -unsigned long ring_buffer_overruns(struct ring_buffer *buffer); -unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_dropped_events_cpu(struct ring_buffer *buffer, int cpu); -unsigned long ring_buffer_read_events_cpu(struct ring_buffer *buffer, int cpu); - -u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu); -void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, +bool ring_buffer_empty(struct trace_buffer *buffer); +bool ring_buffer_empty_cpu(struct trace_buffer *buffer, int cpu); + +void ring_buffer_record_disable(struct trace_buffer *buffer); +void ring_buffer_record_enable(struct trace_buffer *buffer); +void ring_buffer_record_off(struct trace_buffer *buffer); +void ring_buffer_record_on(struct trace_buffer *buffer); +bool ring_buffer_record_is_on(struct trace_buffer *buffer); +bool ring_buffer_record_is_set_on(struct trace_buffer *buffer); +void ring_buffer_record_disable_cpu(struct trace_buffer *buffer, int cpu); +void ring_buffer_record_enable_cpu(struct trace_buffer *buffer, int cpu); + +u64 ring_buffer_oldest_event_ts(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_bytes_cpu(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_entries(struct trace_buffer *buffer); +unsigned long ring_buffer_overruns(struct trace_buffer *buffer); +unsigned long ring_buffer_entries_cpu(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_overrun_cpu(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_commit_overrun_cpu(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_dropped_events_cpu(struct trace_buffer *buffer, int cpu); +unsigned long ring_buffer_read_events_cpu(struct trace_buffer *buffer, int cpu); + +u64 ring_buffer_time_stamp(struct trace_buffer *buffer, int cpu); +void ring_buffer_normalize_time_stamp(struct trace_buffer *buffer, int cpu, u64 *ts); -void ring_buffer_set_clock(struct ring_buffer *buffer, +void ring_buffer_set_clock(struct trace_buffer *buffer, u64 (*clock)(void)); -void ring_buffer_set_time_stamp_abs(struct ring_buffer *buffer, bool abs); -bool ring_buffer_time_stamp_abs(struct ring_buffer *buffer); +void ring_buffer_set_time_stamp_abs(struct trace_buffer *buffer, bool abs); +bool ring_buffer_time_stamp_abs(struct trace_buffer *buffer); -size_t ring_buffer_nr_pages(struct ring_buffer *buffer, int cpu); -size_t ring_buffer_nr_dirty_pages(struct ring_buffer *buffer, int cpu); +size_t ring_buffer_nr_pages(struct trace_buffer *buffer, int cpu); +size_t ring_buffer_nr_dirty_pages(struct trace_buffer *buffer, int cpu); -void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu); -void ring_buffer_free_read_page(struct ring_buffer *buffer, int cpu, void *data); -int ring_buffer_read_page(struct ring_buffer *buffer, void **data_page, +void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu); +void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data); +int ring_buffer_read_page(struct trace_buffer *buffer, void **data_page, size_t len, int cpu, int full); struct trace_seq; diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index 13ea7f7d54ac..af2c85d3a1dd 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -11,7 +11,7 @@ #include <linux/tracepoint.h> struct trace_array; -struct trace_buffer; +struct array_buffer; struct tracer; struct dentry; struct bpf_prog; @@ -79,7 +79,7 @@ struct trace_entry { struct trace_iterator { struct trace_array *tr; struct tracer *trace; - struct trace_buffer *trace_buffer; + struct array_buffer *array_buffer; void *private; int cpu_file; struct mutex mutex; @@ -153,7 +153,7 @@ void tracing_generic_entry_update(struct trace_entry *entry, struct trace_event_file; struct ring_buffer_event * -trace_event_buffer_lock_reserve(struct ring_buffer **current_buffer, +trace_event_buffer_lock_reserve(struct trace_buffer **current_buffer, struct trace_event_file *trace_file, int type, unsigned long len, unsigned long flags, int pc); @@ -226,12 +226,13 @@ extern int trace_event_reg(struct trace_event_call *event, enum trace_reg type, void *data); struct trace_event_buffer { - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct ring_buffer_event *event; struct trace_event_file *trace_file; void *entry; unsigned long flags; int pc; + struct pt_regs *regs; }; void *trace_event_buffer_reserve(struct trace_event_buffer *fbuffer, @@ -364,6 +365,128 @@ enum { EVENT_FILE_FL_WAS_ENABLED_BIT, }; +extern struct trace_event_file *trace_get_event_file(const char *instance, + const char *system, + const char *event); +extern void trace_put_event_file(struct trace_event_file *file); + +#define MAX_DYNEVENT_CMD_LEN (2048) + +enum dynevent_type { + DYNEVENT_TYPE_SYNTH = 1, + DYNEVENT_TYPE_KPROBE, + DYNEVENT_TYPE_NONE, +}; + +struct dynevent_cmd; + +typedef int (*dynevent_create_fn_t)(struct dynevent_cmd *cmd); + +struct dynevent_cmd { + struct seq_buf seq; + const char *event_name; + unsigned int n_fields; + enum dynevent_type type; + dynevent_create_fn_t run_command; + void *private_data; +}; + +extern int dynevent_create(struct dynevent_cmd *cmd); + +extern int synth_event_delete(const char *name); + +extern void synth_event_cmd_init(struct dynevent_cmd *cmd, + char *buf, int maxlen); + +extern int __synth_event_gen_cmd_start(struct dynevent_cmd *cmd, + const char *name, + struct module *mod, ...); + +#define synth_event_gen_cmd_start(cmd, name, mod, ...) \ + __synth_event_gen_cmd_start(cmd, name, mod, ## __VA_ARGS__, NULL) + +struct synth_field_desc { + const char *type; + const char *name; +}; + +extern int synth_event_gen_cmd_array_start(struct dynevent_cmd *cmd, + const char *name, + struct module *mod, + struct synth_field_desc *fields, + unsigned int n_fields); +extern int synth_event_create(const char *name, + struct synth_field_desc *fields, + unsigned int n_fields, struct module *mod); + +extern int synth_event_add_field(struct dynevent_cmd *cmd, + const char *type, + const char *name); +extern int synth_event_add_field_str(struct dynevent_cmd *cmd, + const char *type_name); +extern int synth_event_add_fields(struct dynevent_cmd *cmd, + struct synth_field_desc *fields, + unsigned int n_fields); + +#define synth_event_gen_cmd_end(cmd) \ + dynevent_create(cmd) + +struct synth_event; + +struct synth_event_trace_state { + struct trace_event_buffer fbuffer; + struct synth_trace_event *entry; + struct trace_buffer *buffer; + struct synth_event *event; + unsigned int cur_field; + unsigned int n_u64; + bool enabled; + bool add_next; + bool add_name; +}; + +extern int synth_event_trace(struct trace_event_file *file, + unsigned int n_vals, ...); +extern int synth_event_trace_array(struct trace_event_file *file, u64 *vals, + unsigned int n_vals); +extern int synth_event_trace_start(struct trace_event_file *file, + struct synth_event_trace_state *trace_state); +extern int synth_event_add_next_val(u64 val, + struct synth_event_trace_state *trace_state); +extern int synth_event_add_val(const char *field_name, u64 val, + struct synth_event_trace_state *trace_state); +extern int synth_event_trace_end(struct synth_event_trace_state *trace_state); + +extern int kprobe_event_delete(const char *name); + +extern void kprobe_event_cmd_init(struct dynevent_cmd *cmd, + char *buf, int maxlen); + +#define kprobe_event_gen_cmd_start(cmd, name, loc, ...) \ + __kprobe_event_gen_cmd_start(cmd, false, name, loc, ## __VA_ARGS__, NULL) + +#define kretprobe_event_gen_cmd_start(cmd, name, loc, ...) \ + __kprobe_event_gen_cmd_start(cmd, true, name, loc, ## __VA_ARGS__, NULL) + +extern int __kprobe_event_gen_cmd_start(struct dynevent_cmd *cmd, + bool kretprobe, + const char *name, + const char *loc, ...); + +#define kprobe_event_add_fields(cmd, ...) \ + __kprobe_event_add_fields(cmd, ## __VA_ARGS__, NULL) + +#define kprobe_event_add_field(cmd, field) \ + __kprobe_event_add_fields(cmd, field, NULL) + +extern int __kprobe_event_add_fields(struct dynevent_cmd *cmd, ...); + +#define kprobe_event_gen_cmd_end(cmd) \ + dynevent_create(cmd) + +#define kretprobe_event_gen_cmd_end(cmd) \ + dynevent_create(cmd) + /* * Event file flags: * ENABLED - The event is enabled diff --git a/include/trace/trace_events.h b/include/trace/trace_events.h index 96d77e5e0664..502c7be50b8d 100644 --- a/include/trace/trace_events.h +++ b/include/trace/trace_events.h @@ -2,7 +2,8 @@ /* * Stage 1 of the trace events. * - * Override the macros in <trace/trace_events.h> to include the following: + * Override the macros in the event tracepoint header <trace/events/XXX.h> + * to include the following: * * struct trace_event_raw_<call> { * struct trace_entry ent; @@ -223,7 +224,8 @@ TRACE_MAKE_SYSTEM_STR(); /* * Stage 3 of the trace events. * - * Override the macros in <trace/trace_events.h> to include the following: + * Override the macros in the event tracepoint header <trace/events/XXX.h> + * to include the following: * * enum print_line_t * trace_raw_output_<call>(struct trace_iterator *iter, int flags) @@ -533,7 +535,8 @@ static inline notrace int trace_event_get_offsets_##call( \ /* * Stage 4 of the trace events. * - * Override the macros in <trace/trace_events.h> to include the following: + * Override the macros in the event tracepoint header <trace/events/XXX.h> + * to include the following: * * For those macros defined with TRACE_EVENT: * @@ -548,7 +551,7 @@ static inline notrace int trace_event_get_offsets_##call( \ * enum event_trigger_type __tt = ETT_NONE; * struct ring_buffer_event *event; * struct trace_event_raw_<call> *entry; <-- defined in stage 1 - * struct ring_buffer *buffer; + * struct trace_buffer *buffer; * unsigned long irq_flags; * int __data_size; * int pc; diff --git a/init/Kconfig b/init/Kconfig index 24b23d843df1..cfee56c151f1 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1224,6 +1224,20 @@ source "usr/Kconfig" endif +config BOOT_CONFIG + bool "Boot config support" + depends on BLK_DEV_INITRD + select LIBXBC + default y + help + Extra boot config allows system admin to pass a config file as + complemental extension of kernel cmdline when booting. + The boot config file must be attached at the end of initramfs + with checksum and size. + See <file:Documentation/admin-guide/bootconfig.rst> for details. + + If unsure, say Y. + choice prompt "Compiler optimization level" default CC_OPTIMIZE_FOR_PERFORMANCE diff --git a/init/main.c b/init/main.c index d8c7e86c2d28..cc0ee4873419 100644 --- a/init/main.c +++ b/init/main.c @@ -28,6 +28,7 @@ #include <linux/initrd.h> #include <linux/memblock.h> #include <linux/acpi.h> +#include <linux/bootconfig.h> #include <linux/console.h> #include <linux/nmi.h> #include <linux/percpu.h> @@ -136,8 +137,10 @@ char __initdata boot_command_line[COMMAND_LINE_SIZE]; char *saved_command_line; /* Command line for parameter parsing */ static char *static_command_line; -/* Command line for per-initcall parameter parsing */ -static char *initcall_command_line; +/* Untouched extra command line */ +static char *extra_command_line; +/* Extra init arguments */ +static char *extra_init_args; static char *execute_command; static char *ramdisk_execute_command; @@ -245,6 +248,156 @@ static int __init loglevel(char *str) early_param("loglevel", loglevel); +#ifdef CONFIG_BOOT_CONFIG + +char xbc_namebuf[XBC_KEYLEN_MAX] __initdata; + +#define rest(dst, end) ((end) > (dst) ? (end) - (dst) : 0) + +static int __init xbc_snprint_cmdline(char *buf, size_t size, + struct xbc_node *root) +{ + struct xbc_node *knode, *vnode; + char *end = buf + size; + char c = '\"'; + const char *val; + int ret; + + xbc_node_for_each_key_value(root, knode, val) { + ret = xbc_node_compose_key_after(root, knode, + xbc_namebuf, XBC_KEYLEN_MAX); + if (ret < 0) + return ret; + + vnode = xbc_node_get_child(knode); + ret = snprintf(buf, rest(buf, end), "%s%c", xbc_namebuf, + vnode ? '=' : ' '); + if (ret < 0) + return ret; + buf += ret; + if (!vnode) + continue; + + c = '\"'; + xbc_array_for_each_value(vnode, val) { + ret = snprintf(buf, rest(buf, end), "%c%s", c, val); + if (ret < 0) + return ret; + buf += ret; + c = ','; + } + if (rest(buf, end) > 2) + strcpy(buf, "\" "); + buf += 2; + } + + return buf - (end - size); +} +#undef rest + +/* Make an extra command line under given key word */ +static char * __init xbc_make_cmdline(const char *key) +{ + struct xbc_node *root; + char *new_cmdline; + int ret, len = 0; + + root = xbc_find_node(key); + if (!root) + return NULL; + + /* Count required buffer size */ + len = xbc_snprint_cmdline(NULL, 0, root); + if (len <= 0) + return NULL; + + new_cmdline = memblock_alloc(len + 1, SMP_CACHE_BYTES); + if (!new_cmdline) { + pr_err("Failed to allocate memory for extra kernel cmdline.\n"); + return NULL; + } + + ret = xbc_snprint_cmdline(new_cmdline, len + 1, root); + if (ret < 0 || ret > len) { + pr_err("Failed to print extra kernel cmdline.\n"); + return NULL; + } + + return new_cmdline; +} + +u32 boot_config_checksum(unsigned char *p, u32 size) +{ + u32 ret = 0; + + while (size--) + ret += *p++; + + return ret; +} + +static void __init setup_boot_config(const char *cmdline) +{ + u32 size, csum; + char *data, *copy; + const char *p; + u32 *hdr; + int ret; + + p = strstr(cmdline, "bootconfig"); + if (!p || (p != cmdline && !isspace(*(p-1))) || + (p[10] && !isspace(p[10]))) + return; + + if (!initrd_end) + goto not_found; + + hdr = (u32 *)(initrd_end - 8); + size = hdr[0]; + csum = hdr[1]; + + if (size >= XBC_DATA_MAX) { + pr_err("bootconfig size %d greater than max size %d\n", + size, XBC_DATA_MAX); + return; + } + + data = ((void *)hdr) - size; + if ((unsigned long)data < initrd_start) + goto not_found; + + if (boot_config_checksum((unsigned char *)data, size) != csum) { + pr_err("bootconfig checksum failed\n"); + return; + } + + copy = memblock_alloc(size + 1, SMP_CACHE_BYTES); + if (!copy) { + pr_err("Failed to allocate memory for bootconfig\n"); + return; + } + + memcpy(copy, data, size); + copy[size] = '\0'; + + ret = xbc_init(copy); + if (ret < 0) + pr_err("Failed to parse bootconfig\n"); + else { + pr_info("Load bootconfig: %d bytes %d nodes\n", size, ret); + /* keys starting with "kernel." are passed via cmdline */ + extra_command_line = xbc_make_cmdline("kernel"); + /* Also, "init." keys are init arguments */ + extra_init_args = xbc_make_cmdline("init"); + } + return; +not_found: + pr_err("'bootconfig' found on command line, but no bootconfig found\n"); +} +#else +#define setup_boot_config(cmdline) do { } while (0) +#endif + /* Change NUL term back to "=", to make "param" the whole string. */ static void __init repair_env_string(char *param, char *val) { @@ -373,22 +526,50 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { } */ static void __init setup_command_line(char *command_line) { - size_t len = strlen(boot_command_line) + 1; + size_t len, xlen = 0, ilen = 0; - saved_command_line = memblock_alloc(len, SMP_CACHE_BYTES); - if (!saved_command_line) - panic("%s: Failed to allocate %zu bytes\n", __func__, len); + if (extra_command_line) + xlen = strlen(extra_command_line); + if (extra_init_args) + ilen = strlen(extra_init_args) + 4; /* for " -- " */ - initcall_command_line = memblock_alloc(len, SMP_CACHE_BYTES); - if (!initcall_command_line) - panic("%s: Failed to allocate %zu bytes\n", __func__, len); + len = xlen + strlen(boot_command_line) + 1; + + saved_command_line = memblock_alloc(len + ilen, SMP_CACHE_BYTES); + if (!saved_command_line) + panic("%s: Failed to allocate %zu bytes\n", __func__, len + ilen); static_command_line = memblock_alloc(len, SMP_CACHE_BYTES); if (!static_command_line) panic("%s: Failed to allocate %zu bytes\n", __func__, len); - strcpy(saved_command_line, boot_command_line); - strcpy(static_command_line, command_line); + if (xlen) { + /* + * We have to put extra_command_line before boot command + * lines because there could be dashes (separator of init + * command line) in the command lines. + */ + strcpy(saved_command_line, extra_command_line); + strcpy(static_command_line, extra_command_line); + } + strcpy(saved_command_line + xlen, boot_command_line); + strcpy(static_command_line + xlen, command_line); + + if (ilen) { + /* + * Append supplemental init boot args to saved_command_line + * so that user can check what command line options passed + * to init. + */ + len = strlen(saved_command_line); + if (!strstr(boot_command_line, " -- ")) { + strcpy(saved_command_line + len, " -- "); + len += 4; + } else + saved_command_line[len++] = ' '; + + strcpy(saved_command_line + len, extra_init_args); + } } /* @@ -595,6 +776,7 @@ asmlinkage __visible void __init start_kernel(void) pr_notice("%s", linux_banner); early_security_init(); setup_arch(&command_line); + setup_boot_config(command_line); setup_command_line(command_line); setup_nr_cpu_ids(); setup_per_cpu_areas(); @@ -604,7 +786,7 @@ asmlinkage __visible void __init start_kernel(void) build_all_zonelists(NULL); page_alloc_init(); - pr_notice("Kernel command line: %s\n", boot_command_line); + pr_notice("Kernel command line: %s\n", saved_command_line); /* parameters may set static keys */ jump_label_init(); parse_early_param(); @@ -615,6 +797,9 @@ asmlinkage __visible void __init start_kernel(void) if (!IS_ERR_OR_NULL(after_dashes)) parse_args("Setting init args", after_dashes, NULL, 0, -1, -1, NULL, set_init_arg); + if (extra_init_args) + parse_args("Setting extra init args", extra_init_args, + NULL, 0, -1, -1, NULL, set_init_arg); /* * These use large bootmem allocations and must precede @@ -996,13 +1181,12 @@ static int __init ignore_unknown_bootoption(char *param, char *val, return 0; } -static void __init do_initcall_level(int level) +static void __init do_initcall_level(int level, char *command_line) { initcall_entry_t *fn; - strcpy(initcall_command_line, saved_command_line); parse_args(initcall_level_names[level], - initcall_command_line, __start___param, + command_line, __start___param, __stop___param - __start___param, level, level, NULL, ignore_unknown_bootoption); @@ -1015,9 +1199,20 @@ static void __init do_initcall_level(int level) static void __init do_initcalls(void) { int level; + size_t len = strlen(saved_command_line) + 1; + char *command_line; + + command_line = kzalloc(len, GFP_KERNEL); + if (!command_line) + panic("%s: Failed to allocate %zu bytes\n", __func__, len); + + for (level = 0; level < ARRAY_SIZE(initcall_levels) - 1; level++) { + /* Parser modifies command_line, restore it each time */ + strcpy(command_line, saved_command_line); + do_initcall_level(level, command_line); + } - for (level = 0; level < ARRAY_SIZE(initcall_levels) - 1; level++) - do_initcall_level(level); + kfree(command_line); } /* diff --git a/kernel/events/core.c b/kernel/events/core.c index dc9c643bce94..17f9a4a909eb 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4373,7 +4373,7 @@ static void free_event_rcu(struct rcu_head *head) } static void ring_buffer_attach(struct perf_event *event, - struct ring_buffer *rb); + struct perf_buffer *rb); static void detach_sb_event(struct perf_event *event) { @@ -5054,7 +5054,7 @@ perf_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) static __poll_t perf_poll(struct file *file, poll_table *wait) { struct perf_event *event = file->private_data; - struct ring_buffer *rb; + struct perf_buffer *rb; __poll_t events = EPOLLHUP; poll_wait(file, &event->waitq, wait); @@ -5296,7 +5296,7 @@ static long _perf_ioctl(struct perf_event *event, unsigned int cmd, unsigned lon return perf_event_set_bpf_prog(event, arg); case PERF_EVENT_IOC_PAUSE_OUTPUT: { - struct ring_buffer *rb; + struct perf_buffer *rb; rcu_read_lock(); rb = rcu_dereference(event->rb); @@ -5432,7 +5432,7 @@ static void calc_timer_values(struct perf_event *event, static void perf_event_init_userpage(struct perf_event *event) { struct perf_event_mmap_page *userpg; - struct ring_buffer *rb; + struct perf_buffer *rb; rcu_read_lock(); rb = rcu_dereference(event->rb); @@ -5464,7 +5464,7 @@ void __weak arch_perf_update_userpage( void perf_event_update_userpage(struct perf_event *event) { struct perf_event_mmap_page *userpg; - struct ring_buffer *rb; + struct perf_buffer *rb; u64 enabled, running, now; rcu_read_lock(); @@ -5515,7 +5515,7 @@ EXPORT_SYMBOL_GPL(perf_event_update_userpage); static vm_fault_t perf_mmap_fault(struct vm_fault *vmf) { struct perf_event *event = vmf->vma->vm_file->private_data; - struct ring_buffer *rb; + struct perf_buffer *rb; vm_fault_t ret = VM_FAULT_SIGBUS; if (vmf->flags & FAULT_FLAG_MKWRITE) { @@ -5548,9 +5548,9 @@ unlock: } static void ring_buffer_attach(struct perf_event *event, - struct ring_buffer *rb) + struct perf_buffer *rb) { - struct ring_buffer *old_rb = NULL; + struct perf_buffer *old_rb = NULL; unsigned long flags; if (event->rb) { @@ -5608,7 +5608,7 @@ static void ring_buffer_attach(struct perf_event *event, static void ring_buffer_wakeup(struct perf_event *event) { - struct ring_buffer *rb; + struct perf_buffer *rb; rcu_read_lock(); rb = rcu_dereference(event->rb); @@ -5619,9 +5619,9 @@ static void ring_buffer_wakeup(struct perf_event *event) rcu_read_unlock(); } -struct ring_buffer *ring_buffer_get(struct perf_event *event) +struct perf_buffer *ring_buffer_get(struct perf_event *event) { - struct ring_buffer *rb; + struct perf_buffer *rb; rcu_read_lock(); rb = rcu_dereference(event->rb); @@ -5634,7 +5634,7 @@ struct ring_buffer *ring_buffer_get(struct perf_event *event) return rb; } -void ring_buffer_put(struct ring_buffer *rb) +void ring_buffer_put(struct perf_buffer *rb) { if (!refcount_dec_and_test(&rb->refcount)) return; @@ -5672,7 +5672,7 @@ static void perf_mmap_close(struct vm_area_struct *vma) { struct perf_event *event = vma->vm_file->private_data; - struct ring_buffer *rb = ring_buffer_get(event); + struct perf_buffer *rb = ring_buffer_get(event); struct user_struct *mmap_user = rb->mmap_user; int mmap_locked = rb->mmap_locked; unsigned long size = perf_data_size(rb); @@ -5790,8 +5790,8 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) struct perf_event *event = file->private_data; unsigned long user_locked, user_lock_limit; struct user_struct *user = current_user(); + struct perf_buffer *rb = NULL; unsigned long locked, lock_limit; - struct ring_buffer *rb = NULL; unsigned long vma_size; unsigned long nr_pages; long user_extra = 0, extra = 0; @@ -6266,7 +6266,7 @@ static unsigned long perf_prepare_sample_aux(struct perf_event *event, size_t size) { struct perf_event *sampler = event->aux_event; - struct ring_buffer *rb; + struct perf_buffer *rb; data->aux_size = 0; @@ -6299,7 +6299,7 @@ out: return data->aux_size; } -long perf_pmu_snapshot_aux(struct ring_buffer *rb, +long perf_pmu_snapshot_aux(struct perf_buffer *rb, struct perf_event *event, struct perf_output_handle *handle, unsigned long size) @@ -6338,8 +6338,8 @@ static void perf_aux_sample_output(struct perf_event *event, struct perf_sample_data *data) { struct perf_event *sampler = event->aux_event; + struct perf_buffer *rb; unsigned long pad; - struct ring_buffer *rb; long size; if (WARN_ON_ONCE(!sampler || !data->aux_size)) @@ -6707,7 +6707,7 @@ void perf_output_sample(struct perf_output_handle *handle, int wakeup_events = event->attr.wakeup_events; if (wakeup_events) { - struct ring_buffer *rb = handle->rb; + struct perf_buffer *rb = handle->rb; int events = local_inc_return(&rb->events); if (events >= wakeup_events) { @@ -7150,7 +7150,7 @@ void perf_event_exec(void) } struct remote_output { - struct ring_buffer *rb; + struct perf_buffer *rb; int err; }; @@ -7158,7 +7158,7 @@ static void __perf_event_output_stop(struct perf_event *event, void *data) { struct perf_event *parent = event->parent; struct remote_output *ro = data; - struct ring_buffer *rb = ro->rb; + struct perf_buffer *rb = ro->rb; struct stop_event_data sd = { .event = event, }; @@ -10998,7 +10998,7 @@ err_size: static int perf_event_set_output(struct perf_event *event, struct perf_event *output_event) { - struct ring_buffer *rb = NULL; + struct perf_buffer *rb = NULL; int ret = -EINVAL; if (!output_event) diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 747d67f130cb..f16f66b6b655 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -10,7 +10,7 @@ #define RING_BUFFER_WRITABLE 0x01 -struct ring_buffer { +struct perf_buffer { refcount_t refcount; struct rcu_head rcu_head; #ifdef CONFIG_PERF_USE_VMALLOC @@ -58,17 +58,17 @@ struct ring_buffer { void *data_pages[0]; }; -extern void rb_free(struct ring_buffer *rb); +extern void rb_free(struct perf_buffer *rb); static inline void rb_free_rcu(struct rcu_head *rcu_head) { - struct ring_buffer *rb; + struct perf_buffer *rb; - rb = container_of(rcu_head, struct ring_buffer, rcu_head); + rb = container_of(rcu_head, struct perf_buffer, rcu_head); rb_free(rb); } -static inline void rb_toggle_paused(struct ring_buffer *rb, bool pause) +static inline void rb_toggle_paused(struct perf_buffer *rb, bool pause) { if (!pause && rb->nr_pages) rb->paused = 0; @@ -76,16 +76,16 @@ static inline void rb_toggle_paused(struct ring_buffer *rb, bool pause) rb->paused = 1; } -extern struct ring_buffer * +extern struct perf_buffer * rb_alloc(int nr_pages, long watermark, int cpu, int flags); extern void perf_event_wakeup(struct perf_event *event); -extern int rb_alloc_aux(struct ring_buffer *rb, struct perf_event *event, +extern int rb_alloc_aux(struct perf_buffer *rb, struct perf_event *event, pgoff_t pgoff, int nr_pages, long watermark, int flags); -extern void rb_free_aux(struct ring_buffer *rb); -extern struct ring_buffer *ring_buffer_get(struct perf_event *event); -extern void ring_buffer_put(struct ring_buffer *rb); +extern void rb_free_aux(struct perf_buffer *rb); +extern struct perf_buffer *ring_buffer_get(struct perf_event *event); +extern void ring_buffer_put(struct perf_buffer *rb); -static inline bool rb_has_aux(struct ring_buffer *rb) +static inline bool rb_has_aux(struct perf_buffer *rb) { return !!rb->aux_nr_pages; } @@ -94,7 +94,7 @@ void perf_event_aux_event(struct perf_event *event, unsigned long head, unsigned long size, u64 flags); extern struct page * -perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff); +perf_mmap_to_page(struct perf_buffer *rb, unsigned long pgoff); #ifdef CONFIG_PERF_USE_VMALLOC /* @@ -103,25 +103,25 @@ perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff); * Required for architectures that have d-cache aliasing issues. */ -static inline int page_order(struct ring_buffer *rb) +static inline int page_order(struct perf_buffer *rb) { return rb->page_order; } #else -static inline int page_order(struct ring_buffer *rb) +static inline int page_order(struct perf_buffer *rb) { return 0; } #endif -static inline unsigned long perf_data_size(struct ring_buffer *rb) +static inline unsigned long perf_data_size(struct perf_buffer *rb) { return rb->nr_pages << (PAGE_SHIFT + page_order(rb)); } -static inline unsigned long perf_aux_size(struct ring_buffer *rb) +static inline unsigned long perf_aux_size(struct perf_buffer *rb) { return rb->aux_nr_pages << PAGE_SHIFT; } @@ -141,7 +141,7 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) buf += written; \ handle->size -= written; \ if (!handle->size) { \ - struct ring_buffer *rb = handle->rb; \ + struct perf_buffer *rb = handle->rb; \ \ handle->page++; \ handle->page &= rb->nr_pages - 1; \ diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 7ffd5c763f93..192b8abc6330 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -35,7 +35,7 @@ static void perf_output_wakeup(struct perf_output_handle *handle) */ static void perf_output_get_handle(struct perf_output_handle *handle) { - struct ring_buffer *rb = handle->rb; + struct perf_buffer *rb = handle->rb; preempt_disable(); @@ -49,7 +49,7 @@ static void perf_output_get_handle(struct perf_output_handle *handle) static void perf_output_put_handle(struct perf_output_handle *handle) { - struct ring_buffer *rb = handle->rb; + struct perf_buffer *rb = handle->rb; unsigned long head; unsigned int nest; @@ -150,7 +150,7 @@ __perf_output_begin(struct perf_output_handle *handle, struct perf_event *event, unsigned int size, bool backward) { - struct ring_buffer *rb; + struct perf_buffer *rb; unsigned long tail, offset, head; int have_lost, page_shift; struct { @@ -301,7 +301,7 @@ void perf_output_end(struct perf_output_handle *handle) } static void -ring_buffer_init(struct ring_buffer *rb, long watermark, int flags) +ring_buffer_init(struct perf_buffer *rb, long watermark, int flags) { long max_size = perf_data_size(rb); @@ -361,7 +361,7 @@ void *perf_aux_output_begin(struct perf_output_handle *handle, { struct perf_event *output_event = event; unsigned long aux_head, aux_tail; - struct ring_buffer *rb; + struct perf_buffer *rb; unsigned int nest; if (output_event->parent) @@ -449,7 +449,7 @@ err: } EXPORT_SYMBOL_GPL(perf_aux_output_begin); -static __always_inline bool rb_need_aux_wakeup(struct ring_buffer *rb) +static __always_inline bool rb_need_aux_wakeup(struct perf_buffer *rb) { if (rb->aux_overwrite) return false; @@ -475,7 +475,7 @@ static __always_inline bool rb_need_aux_wakeup(struct ring_buffer *rb) void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size) { bool wakeup = !!(handle->aux_flags & PERF_AUX_FLAG_TRUNCATED); - struct ring_buffer *rb = handle->rb; + struct perf_buffer *rb = handle->rb; unsigned long aux_head; /* in overwrite mode, driver provides aux_head via handle */ @@ -532,7 +532,7 @@ EXPORT_SYMBOL_GPL(perf_aux_output_end); */ int perf_aux_output_skip(struct perf_output_handle *handle, unsigned long size) { - struct ring_buffer *rb = handle->rb; + struct perf_buffer *rb = handle->rb; if (size > handle->size) return -ENOSPC; @@ -569,8 +569,8 @@ long perf_output_copy_aux(struct perf_output_handle *aux_handle, struct perf_output_handle *handle, unsigned long from, unsigned long to) { + struct perf_buffer *rb = aux_handle->rb; unsigned long tocopy, remainder, len = 0; - struct ring_buffer *rb = aux_handle->rb; void *addr; from &= (rb->aux_nr_pages << PAGE_SHIFT) - 1; @@ -626,7 +626,7 @@ static struct page *rb_alloc_aux_page(int node, int order) return page; } -static void rb_free_aux_page(struct ring_buffer *rb, int idx) +static void rb_free_aux_page(struct perf_buffer *rb, int idx) { struct page *page = virt_to_page(rb->aux_pages[idx]); @@ -635,7 +635,7 @@ static void rb_free_aux_page(struct ring_buffer *rb, int idx) __free_page(page); } -static void __rb_free_aux(struct ring_buffer *rb) +static void __rb_free_aux(struct perf_buffer *rb) { int pg; @@ -662,7 +662,7 @@ static void __rb_free_aux(struct ring_buffer *rb) } } -int rb_alloc_aux(struct ring_buffer *rb, struct perf_event *event, +int rb_alloc_aux(struct perf_buffer *rb, struct perf_event *event, pgoff_t pgoff, int nr_pages, long watermark, int flags) { bool overwrite = !(flags & RING_BUFFER_WRITABLE); @@ -753,7 +753,7 @@ out: return ret; } -void rb_free_aux(struct ring_buffer *rb) +void rb_free_aux(struct perf_buffer *rb) { if (refcount_dec_and_test(&rb->aux_refcount)) __rb_free_aux(rb); @@ -766,7 +766,7 @@ void rb_free_aux(struct ring_buffer *rb) */ static struct page * -__perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) +__perf_mmap_to_page(struct perf_buffer *rb, unsigned long pgoff) { if (pgoff > rb->nr_pages) return NULL; @@ -798,13 +798,13 @@ static void perf_mmap_free_page(void *addr) __free_page(page); } -struct ring_buffer *rb_alloc(int nr_pages, long watermark, int cpu, int flags) +struct perf_buffer *rb_alloc(int nr_pages, long watermark, int cpu, int flags) { - struct ring_buffer *rb; + struct perf_buffer *rb; unsigned long size; int i; - size = sizeof(struct ring_buffer); + size = sizeof(struct perf_buffer); size += nr_pages * sizeof(void *); if (order_base_2(size) >= PAGE_SHIFT+MAX_ORDER) @@ -843,7 +843,7 @@ fail: return NULL; } -void rb_free(struct ring_buffer *rb) +void rb_free(struct perf_buffer *rb) { int i; @@ -854,13 +854,13 @@ void rb_free(struct ring_buffer *rb) } #else -static int data_page_nr(struct ring_buffer *rb) +static int data_page_nr(struct perf_buffer *rb) { return rb->nr_pages << page_order(rb); } static struct page * -__perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) +__perf_mmap_to_page(struct perf_buffer *rb, unsigned long pgoff) { /* The '>' counts in the user page. */ if (pgoff > data_page_nr(rb)) @@ -878,11 +878,11 @@ static void perf_mmap_unmark_page(void *addr) static void rb_free_work(struct work_struct *work) { - struct ring_buffer *rb; + struct perf_buffer *rb; void *base; int i, nr; - rb = container_of(work, struct ring_buffer, work); + rb = container_of(work, struct perf_buffer, work); nr = data_page_nr(rb); base = rb->user_page; @@ -894,18 +894,18 @@ static void rb_free_work(struct work_struct *work) kfree(rb); } -void rb_free(struct ring_buffer *rb) +void rb_free(struct perf_buffer *rb) { schedule_work(&rb->work); } -struct ring_buffer *rb_alloc(int nr_pages, long watermark, int cpu, int flags) +struct perf_buffer *rb_alloc(int nr_pages, long watermark, int cpu, int flags) { - struct ring_buffer *rb; + struct perf_buffer *rb; unsigned long size; void *all_buf; - size = sizeof(struct ring_buffer); + size = sizeof(struct perf_buffer); size += sizeof(void *); rb = kzalloc(size, GFP_KERNEL); @@ -939,7 +939,7 @@ fail: #endif struct page * -perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff) +perf_mmap_to_page(struct perf_buffer *rb, unsigned long pgoff) { if (rb->aux_nr_pages) { /* above AUX space */ diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 25a0fcfa7a5d..91e885194dbc 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -141,6 +141,15 @@ menuconfig FTRACE if FTRACE +config BOOTTIME_TRACING + bool "Boot-time Tracing support" + depends on BOOT_CONFIG && TRACING + default y + help + Enable developer to setup ftrace subsystem via supplemental + kernel cmdline at boot time for debugging (tracing) driver + initialization and boot process. + config FUNCTION_TRACER bool "Kernel Function Tracer" depends on HAVE_FUNCTION_TRACER @@ -172,6 +181,77 @@ config FUNCTION_GRAPH_TRACER the return value. This is done by setting the current return address on the current task structure into a stack of calls. +config DYNAMIC_FTRACE + bool "enable/disable function tracing dynamically" + depends on FUNCTION_TRACER + depends on HAVE_DYNAMIC_FTRACE + default y + help + This option will modify all the calls to function tracing + dynamically (will patch them out of the binary image and + replace them with a No-Op instruction) on boot up. During + compile time, a table is made of all the locations that ftrace + can function trace, and this table is linked into the kernel + image. When this is enabled, functions can be individually + enabled, and the functions not enabled will not affect + performance of the system. + + See the files in /sys/kernel/debug/tracing: + available_filter_functions + set_ftrace_filter + set_ftrace_notrace + + This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but + otherwise has native performance as long as no tracing is active. + +config DYNAMIC_FTRACE_WITH_REGS + def_bool y + depends on DYNAMIC_FTRACE + depends on HAVE_DYNAMIC_FTRACE_WITH_REGS + +config DYNAMIC_FTRACE_WITH_DIRECT_CALLS + def_bool y + depends on DYNAMIC_FTRACE + depends on HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS + +config FUNCTION_PROFILER + bool "Kernel function profiler" + depends on FUNCTION_TRACER + default n + help + This option enables the kernel function profiler. A file is created + in debugfs called function_profile_enabled which defaults to zero. + When a 1 is echoed into this file profiling begins, and when a + zero is entered, profiling stops. A "functions" file is created in + the trace_stat directory; this file shows the list of functions that + have been hit and their counters. + + If in doubt, say N. + +config STACK_TRACER + bool "Trace max stack" + depends on HAVE_FUNCTION_TRACER + select FUNCTION_TRACER + select STACKTRACE + select KALLSYMS + help + This special tracer records the maximum stack footprint of the + kernel and displays it in /sys/kernel/debug/tracing/stack_trace. + + This tracer works by hooking into every function call that the + kernel executes, and keeping a maximum stack depth value and + stack-trace saved. If this is configured with DYNAMIC_FTRACE + then it will not have any overhead while the stack tracer + is disabled. + + To enable the stack tracer on bootup, pass in 'stacktrace' + on the kernel command line. + + The stack tracer can also be enabled or disabled via the + sysctl kernel.stack_tracer_enabled + + Say N if unsure. + config TRACE_PREEMPT_TOGGLE bool help @@ -282,6 +362,19 @@ config HWLAT_TRACER file. Every time a latency is greater than tracing_thresh, it will be recorded into the ring buffer. +config MMIOTRACE + bool "Memory mapped IO tracing" + depends on HAVE_MMIOTRACE_SUPPORT && PCI + select GENERIC_TRACER + help + Mmiotrace traces Memory Mapped I/O access and is meant for + debugging and reverse engineering. It is called from the ioremap + implementation and works via page faults. Tracing is disabled by + default and can be enabled at run-time. + + See Documentation/trace/mmiotrace.rst. + If you are not helping to develop drivers, say N. + config ENABLE_DEFAULT_TRACERS bool "Trace process context switches and events" depends on !GENERIC_TRACER @@ -410,30 +503,6 @@ config BRANCH_TRACER Say N if unsure. -config STACK_TRACER - bool "Trace max stack" - depends on HAVE_FUNCTION_TRACER - select FUNCTION_TRACER - select STACKTRACE - select KALLSYMS - help - This special tracer records the maximum stack footprint of the - kernel and displays it in /sys/kernel/debug/tracing/stack_trace. - - This tracer works by hooking into every function call that the - kernel executes, and keeping a maximum stack depth value and - stack-trace saved. If this is configured with DYNAMIC_FTRACE - then it will not have any overhead while the stack tracer - is disabled. - - To enable the stack tracer on bootup, pass in 'stacktrace' - on the kernel command line. - - The stack tracer can also be enabled or disabled via the - sysctl kernel.stack_tracer_enabled - - Say N if unsure. - config BLK_DEV_IO_TRACE bool "Support for tracing block IO actions" depends on SYSFS @@ -531,53 +600,6 @@ config DYNAMIC_EVENTS config PROBE_EVENTS def_bool n -config DYNAMIC_FTRACE - bool "enable/disable function tracing dynamically" - depends on FUNCTION_TRACER - depends on HAVE_DYNAMIC_FTRACE - default y - help - This option will modify all the calls to function tracing - dynamically (will patch them out of the binary image and - replace them with a No-Op instruction) on boot up. During - compile time, a table is made of all the locations that ftrace - can function trace, and this table is linked into the kernel - image. When this is enabled, functions can be individually - enabled, and the functions not enabled will not affect - performance of the system. - - See the files in /sys/kernel/debug/tracing: - available_filter_functions - set_ftrace_filter - set_ftrace_notrace - - This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but - otherwise has native performance as long as no tracing is active. - -config DYNAMIC_FTRACE_WITH_REGS - def_bool y - depends on DYNAMIC_FTRACE - depends on HAVE_DYNAMIC_FTRACE_WITH_REGS - -config DYNAMIC_FTRACE_WITH_DIRECT_CALLS - def_bool y - depends on DYNAMIC_FTRACE - depends on HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS - -config FUNCTION_PROFILER - bool "Kernel function profiler" - depends on FUNCTION_TRACER - default n - help - This option enables the kernel function profiler. A file is created - in debugfs called function_profile_enabled which defaults to zero. - When a 1 is echoed into this file profiling begins, and when a - zero is entered, profiling stops. A "functions" file is created in - the trace_stat directory; this file shows the list of functions that - have been hit and their counters. - - If in doubt, say N. - config BPF_KPROBE_OVERRIDE bool "Enable BPF programs to override a kprobed function" depends on BPF_EVENTS @@ -592,54 +614,6 @@ config FTRACE_MCOUNT_RECORD depends on DYNAMIC_FTRACE depends on HAVE_FTRACE_MCOUNT_RECORD -config FTRACE_SELFTEST - bool - -config FTRACE_STARTUP_TEST - bool "Perform a startup test on ftrace" - depends on GENERIC_TRACER - select FTRACE_SELFTEST - help - This option performs a series of startup tests on ftrace. On bootup - a series of tests are made to verify that the tracer is - functioning properly. It will do tests on all the configured - tracers of ftrace. - -config EVENT_TRACE_STARTUP_TEST - bool "Run selftest on trace events" - depends on FTRACE_STARTUP_TEST - default y - help - This option performs a test on all trace events in the system. - It basically just enables each event and runs some code that - will trigger events (not necessarily the event it enables) - This may take some time run as there are a lot of events. - -config EVENT_TRACE_TEST_SYSCALLS - bool "Run selftest on syscall events" - depends on EVENT_TRACE_STARTUP_TEST - help - This option will also enable testing every syscall event. - It only enables the event and disables it and runs various loads - with the event enabled. This adds a bit more time for kernel boot - up since it runs this on every system call defined. - - TBD - enable a way to actually call the syscalls as we test their - events - -config MMIOTRACE - bool "Memory mapped IO tracing" - depends on HAVE_MMIOTRACE_SUPPORT && PCI - select GENERIC_TRACER - help - Mmiotrace traces Memory Mapped I/O access and is meant for - debugging and reverse engineering. It is called from the ioremap - implementation and works via page faults. Tracing is disabled by - default and can be enabled at run-time. - - See Documentation/trace/mmiotrace.rst. - If you are not helping to develop drivers, say N. - config TRACING_MAP bool depends on ARCH_HAVE_NMI_SAFE_CMPXCHG @@ -680,16 +654,6 @@ config TRACE_EVENT_INJECT If unsure, say N. -config MMIOTRACE_TEST - tristate "Test module for mmiotrace" - depends on MMIOTRACE && m - help - This is a dumb module for testing mmiotrace. It is very dangerous - as it will write garbage to IO memory starting at a given address. - However, it should be safe to use on e.g. unused portion of VRAM. - - Say N, unless you absolutely know what you are doing. - config TRACEPOINT_BENCHMARK bool "Add tracepoint that benchmarks tracepoints" help @@ -736,6 +700,81 @@ config RING_BUFFER_BENCHMARK If unsure, say N. +config TRACE_EVAL_MAP_FILE + bool "Show eval mappings for trace events" + depends on TRACING + help + The "print fmt" of the trace events will show the enum/sizeof names + instead of their values. This can cause problems for user space tools + that use this string to parse the raw data as user space does not know + how to convert the string to its value. + + To fix this, there's a special macro in the kernel that can be used + to convert an enum/sizeof into its value. If this macro is used, then + the print fmt strings will be converted to their values. + + If something does not get converted properly, this option can be + used to show what enums/sizeof the kernel tried to convert. + + This option is for debugging the conversions. A file is created + in the tracing directory called "eval_map" that will show the + names matched with their values and what trace event system they + belong too. + + Normally, the mapping of the strings to values will be freed after + boot up or module load. With this option, they will not be freed, as + they are needed for the "eval_map" file. Enabling this option will + increase the memory footprint of the running kernel. + + If unsure, say N. + +config GCOV_PROFILE_FTRACE + bool "Enable GCOV profiling on ftrace subsystem" + depends on GCOV_KERNEL + help + Enable GCOV profiling on ftrace subsystem for checking + which functions/lines are tested. + + If unsure, say N. + + Note that on a kernel compiled with this config, ftrace will + run significantly slower. + +config FTRACE_SELFTEST + bool + +config FTRACE_STARTUP_TEST + bool "Perform a startup test on ftrace" + depends on GENERIC_TRACER + select FTRACE_SELFTEST + help + This option performs a series of startup tests on ftrace. On bootup + a series of tests are made to verify that the tracer is + functioning properly. It will do tests on all the configured + tracers of ftrace. + +config EVENT_TRACE_STARTUP_TEST + bool "Run selftest on trace events" + depends on FTRACE_STARTUP_TEST + default y + help + This option performs a test on all trace events in the system. + It basically just enables each event and runs some code that + will trigger events (not necessarily the event it enables) + This may take some time run as there are a lot of events. + +config EVENT_TRACE_TEST_SYSCALLS + bool "Run selftest on syscall events" + depends on EVENT_TRACE_STARTUP_TEST + help + This option will also enable testing every syscall event. + It only enables the event and disables it and runs various loads + with the event enabled. This adds a bit more time for kernel boot + up since it runs this on every system call defined. + + TBD - enable a way to actually call the syscalls as we test their + events + config RING_BUFFER_STARTUP_TEST bool "Ring buffer startup self test" depends on RING_BUFFER @@ -759,8 +798,18 @@ config RING_BUFFER_STARTUP_TEST If unsure, say N +config MMIOTRACE_TEST + tristate "Test module for mmiotrace" + depends on MMIOTRACE && m + help + This is a dumb module for testing mmiotrace. It is very dangerous + as it will write garbage to IO memory starting at a given address. + However, it should be safe to use on e.g. unused portion of VRAM. + + Say N, unless you absolutely know what you are doing. + config PREEMPTIRQ_DELAY_TEST - tristate "Preempt / IRQ disable delay thread to test latency tracers" + tristate "Test module to create a preempt / IRQ disable delay thread to test latency tracers" depends on m help Select this option to build a test module that can help test latency @@ -774,45 +823,30 @@ config PREEMPTIRQ_DELAY_TEST If unsure, say N -config TRACE_EVAL_MAP_FILE - bool "Show eval mappings for trace events" - depends on TRACING - help - The "print fmt" of the trace events will show the enum/sizeof names - instead of their values. This can cause problems for user space tools - that use this string to parse the raw data as user space does not know - how to convert the string to its value. - - To fix this, there's a special macro in the kernel that can be used - to convert an enum/sizeof into its value. If this macro is used, then - the print fmt strings will be converted to their values. - - If something does not get converted properly, this option can be - used to show what enums/sizeof the kernel tried to convert. - - This option is for debugging the conversions. A file is created - in the tracing directory called "eval_map" that will show the - names matched with their values and what trace event system they - belong too. +config SYNTH_EVENT_GEN_TEST + tristate "Test module for in-kernel synthetic event generation" + depends on HIST_TRIGGERS + help + This option creates a test module to check the base + functionality of in-kernel synthetic event definition and + generation. - Normally, the mapping of the strings to values will be freed after - boot up or module load. With this option, they will not be freed, as - they are needed for the "eval_map" file. Enabling this option will - increase the memory footprint of the running kernel. + To test, insert the module, and then check the trace buffer + for the generated sample events. - If unsure, say N. + If unsure, say N. -config GCOV_PROFILE_FTRACE - bool "Enable GCOV profiling on ftrace subsystem" - depends on GCOV_KERNEL +config KPROBE_EVENT_GEN_TEST + tristate "Test module for in-kernel kprobe event generation" + depends on KPROBE_EVENTS help - Enable GCOV profiling on ftrace subsystem for checking - which functions/lines are tested. + This option creates a test module to check the base + functionality of in-kernel kprobe event definition. - If unsure, say N. + To test, insert the module, and then check the trace buffer + for the generated kprobe events. - Note that on a kernel compiled with this config, ftrace will - run significantly slower. + If unsure, say N. endif # FTRACE diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 0e63db62225f..f9dcd19165fa 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -44,6 +44,8 @@ obj-$(CONFIG_TRACING) += trace_stat.o obj-$(CONFIG_TRACING) += trace_printk.o obj-$(CONFIG_TRACING_MAP) += tracing_map.o obj-$(CONFIG_PREEMPTIRQ_DELAY_TEST) += preemptirq_delay_test.o +obj-$(CONFIG_SYNTH_EVENT_GEN_TEST) += synth_event_gen_test.o +obj-$(CONFIG_KPROBE_EVENT_GEN_TEST) += kprobe_event_gen_test.o obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o obj-$(CONFIG_PREEMPTIRQ_TRACEPOINTS) += trace_preemptirq.o @@ -83,6 +85,7 @@ endif obj-$(CONFIG_DYNAMIC_EVENTS) += trace_dynevent.o obj-$(CONFIG_PROBE_EVENTS) += trace_probe.o obj-$(CONFIG_UPROBE_EVENTS) += trace_uprobe.o +obj-$(CONFIG_BOOTTIME_TRACING) += trace_boot.o obj-$(CONFIG_TRACEPOINT_BENCHMARK) += trace_benchmark.o diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 475e29498bca..0735ae8545d8 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -68,14 +68,14 @@ static void trace_note(struct blk_trace *bt, pid_t pid, int action, { struct blk_io_trace *t; struct ring_buffer_event *event = NULL; - struct ring_buffer *buffer = NULL; + struct trace_buffer *buffer = NULL; int pc = 0; int cpu = smp_processor_id(); bool blk_tracer = blk_tracer_enabled; ssize_t cgid_len = cgid ? sizeof(cgid) : 0; if (blk_tracer) { - buffer = blk_tr->trace_buffer.buffer; + buffer = blk_tr->array_buffer.buffer; pc = preempt_count(); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, sizeof(*t) + len + cgid_len, @@ -215,7 +215,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, { struct task_struct *tsk = current; struct ring_buffer_event *event = NULL; - struct ring_buffer *buffer = NULL; + struct trace_buffer *buffer = NULL; struct blk_io_trace *t; unsigned long flags = 0; unsigned long *sequence; @@ -248,7 +248,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, if (blk_tracer) { tracing_record_cmdline(current); - buffer = blk_tr->trace_buffer.buffer; + buffer = blk_tr->array_buffer.buffer; pc = preempt_count(); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, sizeof(*t) + pdu_len + cgid_len, diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 9bf1f2cd515e..3f7ee102868a 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -62,8 +62,6 @@ }) /* hash bits for specific function selection */ -#define FTRACE_HASH_BITS 7 -#define FTRACE_FUNC_HASHSIZE (1 << FTRACE_HASH_BITS) #define FTRACE_HASH_DEFAULT_BITS 10 #define FTRACE_HASH_MAX_BITS 12 @@ -146,7 +144,7 @@ static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip, { struct trace_array *tr = op->private; - if (tr && this_cpu_read(tr->trace_buffer.data->ftrace_ignore_pid)) + if (tr && this_cpu_read(tr->array_buffer.data->ftrace_ignore_pid)) return; op->saved_func(ip, parent_ip, op, regs); @@ -1103,9 +1101,6 @@ struct ftrace_page { #define ENTRY_SIZE sizeof(struct dyn_ftrace) #define ENTRIES_PER_PAGE (PAGE_SIZE / ENTRY_SIZE) -/* estimate from running different kernels */ -#define NR_TO_INIT 10000 - static struct ftrace_page *ftrace_pages_start; static struct ftrace_page *ftrace_pages; @@ -5464,7 +5459,7 @@ static void __init set_ftrace_early_graph(char *buf, int enable) struct ftrace_hash *hash; hash = alloc_ftrace_hash(FTRACE_HASH_DEFAULT_BITS); - if (WARN_ON(!hash)) + if (MEM_FAIL(!hash, "Failed to allocate hash\n")) return; while (buf) { @@ -5596,8 +5591,8 @@ static const struct file_operations ftrace_notrace_fops = { static DEFINE_MUTEX(graph_lock); -struct ftrace_hash *ftrace_graph_hash = EMPTY_HASH; -struct ftrace_hash *ftrace_graph_notrace_hash = EMPTY_HASH; +struct ftrace_hash __rcu *ftrace_graph_hash = EMPTY_HASH; +struct ftrace_hash __rcu *ftrace_graph_notrace_hash = EMPTY_HASH; enum graph_filter_type { GRAPH_FILTER_NOTRACE = 0, @@ -5872,8 +5867,15 @@ ftrace_graph_release(struct inode *inode, struct file *file) mutex_unlock(&graph_lock); - /* Wait till all users are no longer using the old hash */ - synchronize_rcu(); + /* + * We need to do a hard force of sched synchronization. + * This is because we use preempt_disable() to do RCU, but + * the function tracers can be called where RCU is not watching + * (like before user_exit()). We can not rely on the RCU + * infrastructure to do the synchronization, thus we must do it + * ourselves. + */ + schedule_on_each_cpu(ftrace_sync); free_ftrace_hash(old_hash); } @@ -6596,7 +6598,7 @@ static void add_to_clear_hash_list(struct list_head *clear_list, func = kmalloc(sizeof(*func), GFP_KERNEL); if (!func) { - WARN_ONCE(1, "alloc failure, ftrace filter could be stale\n"); + MEM_FAIL(1, "alloc failure, ftrace filter could be stale\n"); return; } @@ -6922,7 +6924,7 @@ ftrace_filter_pid_sched_switch_probe(void *data, bool preempt, pid_list = rcu_dereference_sched(tr->function_pids); - this_cpu_write(tr->trace_buffer.data->ftrace_ignore_pid, + this_cpu_write(tr->array_buffer.data->ftrace_ignore_pid, trace_ignore_this_task(pid_list, next)); } @@ -6976,7 +6978,7 @@ static void clear_ftrace_pids(struct trace_array *tr) unregister_trace_sched_switch(ftrace_filter_pid_sched_switch_probe, tr); for_each_possible_cpu(cpu) - per_cpu_ptr(tr->trace_buffer.data, cpu)->ftrace_ignore_pid = false; + per_cpu_ptr(tr->array_buffer.data, cpu)->ftrace_ignore_pid = false; rcu_assign_pointer(tr->function_pids, NULL); @@ -7031,9 +7033,10 @@ static void *fpid_next(struct seq_file *m, void *v, loff_t *pos) struct trace_array *tr = m->private; struct trace_pid_list *pid_list = rcu_dereference_sched(tr->function_pids); - if (v == FTRACE_NO_PIDS) + if (v == FTRACE_NO_PIDS) { + (*pos)++; return NULL; - + } return trace_pid_next(pid_list, v, pos); } @@ -7100,7 +7103,7 @@ static void ignore_task_cpu(void *data) pid_list = rcu_dereference_protected(tr->function_pids, mutex_is_locked(&ftrace_lock)); - this_cpu_write(tr->trace_buffer.data->ftrace_ignore_pid, + this_cpu_write(tr->array_buffer.data->ftrace_ignore_pid, trace_ignore_this_task(pid_list, current)); } diff --git a/kernel/trace/kprobe_event_gen_test.c b/kernel/trace/kprobe_event_gen_test.c new file mode 100644 index 000000000000..18b0f1cbb947 --- /dev/null +++ b/kernel/trace/kprobe_event_gen_test.c @@ -0,0 +1,225 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Test module for in-kernel kprobe event creation and generation. + * + * Copyright (C) 2019 Tom Zanussi <zanussi@kernel.org> + */ + +#include <linux/module.h> +#include <linux/trace_events.h> + +/* + * This module is a simple test of basic functionality for in-kernel + * kprobe/kretprobe event creation. The first test uses + * kprobe_event_gen_cmd_start(), kprobe_event_add_fields() and + * kprobe_event_gen_cmd_end() to create a kprobe event, which is then + * enabled in order to generate trace output. The second creates a + * kretprobe event using kretprobe_event_gen_cmd_start() and + * kretprobe_event_gen_cmd_end(), and is also then enabled. + * + * To test, select CONFIG_KPROBE_EVENT_GEN_TEST and build the module. + * Then: + * + * # insmod kernel/trace/kprobe_event_gen_test.ko + * # cat /sys/kernel/debug/tracing/trace + * + * You should see many instances of the "gen_kprobe_test" and + * "gen_kretprobe_test" events in the trace buffer. + * + * To remove the events, remove the module: + * + * # rmmod kprobe_event_gen_test + * + */ + +static struct trace_event_file *gen_kprobe_test; +static struct trace_event_file *gen_kretprobe_test; + +/* + * Test to make sure we can create a kprobe event, then add more + * fields. + */ +static int __init test_gen_kprobe_cmd(void) +{ + struct dynevent_cmd cmd; + char *buf; + int ret; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + /* Before generating the command, initialize the cmd object */ + kprobe_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + /* + * Define the gen_kprobe_test event with the first 2 kprobe + * fields. + */ + ret = kprobe_event_gen_cmd_start(&cmd, "gen_kprobe_test", + "do_sys_open", + "dfd=%ax", "filename=%dx"); + if (ret) + goto free; + + /* Use kprobe_event_add_fields to add the rest of the fields */ + + ret = kprobe_event_add_fields(&cmd, "flags=%cx", "mode=+4($stack)"); + if (ret) + goto free; + + /* + * This actually creates the event. + */ + ret = kprobe_event_gen_cmd_end(&cmd); + if (ret) + goto free; + + /* + * Now get the gen_kprobe_test event file. We need to prevent + * the instance and event from disappearing from underneath + * us, which trace_get_event_file() does (though in this case + * we're using the top-level instance which never goes away). + */ + gen_kprobe_test = trace_get_event_file(NULL, "kprobes", + "gen_kprobe_test"); + if (IS_ERR(gen_kprobe_test)) { + ret = PTR_ERR(gen_kprobe_test); + goto delete; + } + + /* Enable the event or you won't see anything */ + ret = trace_array_set_clr_event(gen_kprobe_test->tr, + "kprobes", "gen_kprobe_test", true); + if (ret) { + trace_put_event_file(gen_kprobe_test); + goto delete; + } + out: + return ret; + delete: + /* We got an error after creating the event, delete it */ + ret = kprobe_event_delete("gen_kprobe_test"); + free: + kfree(buf); + + goto out; +} + +/* + * Test to make sure we can create a kretprobe event. + */ +static int __init test_gen_kretprobe_cmd(void) +{ + struct dynevent_cmd cmd; + char *buf; + int ret; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + /* Before generating the command, initialize the cmd object */ + kprobe_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + /* + * Define the kretprobe event. + */ + ret = kretprobe_event_gen_cmd_start(&cmd, "gen_kretprobe_test", + "do_sys_open", + "$retval"); + if (ret) + goto free; + + /* + * This actually creates the event. + */ + ret = kretprobe_event_gen_cmd_end(&cmd); + if (ret) + goto free; + + /* + * Now get the gen_kretprobe_test event file. We need to + * prevent the instance and event from disappearing from + * underneath us, which trace_get_event_file() does (though in + * this case we're using the top-level instance which never + * goes away). + */ + gen_kretprobe_test = trace_get_event_file(NULL, "kprobes", + "gen_kretprobe_test"); + if (IS_ERR(gen_kretprobe_test)) { + ret = PTR_ERR(gen_kretprobe_test); + goto delete; + } + + /* Enable the event or you won't see anything */ + ret = trace_array_set_clr_event(gen_kretprobe_test->tr, + "kprobes", "gen_kretprobe_test", true); + if (ret) { + trace_put_event_file(gen_kretprobe_test); + goto delete; + } + out: + return ret; + delete: + /* We got an error after creating the event, delete it */ + ret = kprobe_event_delete("gen_kretprobe_test"); + free: + kfree(buf); + + goto out; +} + +static int __init kprobe_event_gen_test_init(void) +{ + int ret; + + ret = test_gen_kprobe_cmd(); + if (ret) + return ret; + + ret = test_gen_kretprobe_cmd(); + if (ret) { + WARN_ON(trace_array_set_clr_event(gen_kretprobe_test->tr, + "kprobes", + "gen_kretprobe_test", false)); + trace_put_event_file(gen_kretprobe_test); + WARN_ON(kprobe_event_delete("gen_kretprobe_test")); + } + + return ret; +} + +static void __exit kprobe_event_gen_test_exit(void) +{ + /* Disable the event or you can't remove it */ + WARN_ON(trace_array_set_clr_event(gen_kprobe_test->tr, + "kprobes", + "gen_kprobe_test", false)); + + /* Now give the file and instance back */ + trace_put_event_file(gen_kprobe_test); + + /* Now unregister and free the event */ + WARN_ON(kprobe_event_delete("gen_kprobe_test")); + + /* Disable the event or you can't remove it */ + WARN_ON(trace_array_set_clr_event(gen_kprobe_test->tr, + "kprobes", + "gen_kretprobe_test", false)); + + /* Now give the file and instance back */ + trace_put_event_file(gen_kretprobe_test); + + /* Now unregister and free the event */ + WARN_ON(kprobe_event_delete("gen_kretprobe_test")); +} + +module_init(kprobe_event_gen_test_init) +module_exit(kprobe_event_gen_test_exit) + +MODULE_AUTHOR("Tom Zanussi"); +MODULE_DESCRIPTION("kprobe event generation test"); +MODULE_LICENSE("GPL v2"); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 3f655371eaf6..61f0e92ace99 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -300,8 +300,6 @@ u64 ring_buffer_event_time_stamp(struct ring_buffer_event *event) /* Missed count stored at end */ #define RB_MISSED_STORED (1 << 30) -#define RB_MISSED_FLAGS (RB_MISSED_EVENTS|RB_MISSED_STORED) - struct buffer_data_page { u64 time_stamp; /* page time stamp */ local_t commit; /* write committed index */ @@ -443,7 +441,7 @@ enum { struct ring_buffer_per_cpu { int cpu; atomic_t record_disabled; - struct ring_buffer *buffer; + struct trace_buffer *buffer; raw_spinlock_t reader_lock; /* serialize readers */ arch_spinlock_t lock; struct lock_class_key lock_key; @@ -482,7 +480,7 @@ struct ring_buffer_per_cpu { struct rb_irq_work irq_work; }; -struct ring_buffer { +struct trace_buffer { unsigned flags; int cpus; atomic_t record_disabled; @@ -518,7 +516,7 @@ struct ring_buffer_iter { * * Returns the number of pages used by a per_cpu buffer of the ring buffer. */ -size_t ring_buffer_nr_pages(struct ring_buffer *buffer, int cpu) +size_t ring_buffer_nr_pages(struct trace_buffer *buffer, int cpu) { return buffer->buffers[cpu]->nr_pages; } @@ -530,7 +528,7 @@ size_t ring_buffer_nr_pages(struct ring_buffer *buffer, int cpu) * * Returns the number of pages that have content in the ring buffer. */ -size_t ring_buffer_nr_dirty_pages(struct ring_buffer *buffer, int cpu) +size_t ring_buffer_nr_dirty_pages(struct trace_buffer *buffer, int cpu) { size_t read; size_t cnt; @@ -573,7 +571,7 @@ static void rb_wake_up_waiters(struct irq_work *work) * as data is added to any of the @buffer's cpu buffers. Otherwise * it will wait for data to be added to a specific cpu buffer. */ -int ring_buffer_wait(struct ring_buffer *buffer, int cpu, int full) +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full) { struct ring_buffer_per_cpu *uninitialized_var(cpu_buffer); DEFINE_WAIT(wait); @@ -684,7 +682,7 @@ int ring_buffer_wait(struct ring_buffer *buffer, int cpu, int full) * Returns EPOLLIN | EPOLLRDNORM if data exists in the buffers, * zero otherwise. */ -__poll_t ring_buffer_poll_wait(struct ring_buffer *buffer, int cpu, +__poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, struct file *filp, poll_table *poll_table) { struct ring_buffer_per_cpu *cpu_buffer; @@ -742,13 +740,13 @@ __poll_t ring_buffer_poll_wait(struct ring_buffer *buffer, int cpu, /* Up this if you want to test the TIME_EXTENTS and normalization */ #define DEBUG_SHIFT 0 -static inline u64 rb_time_stamp(struct ring_buffer *buffer) +static inline u64 rb_time_stamp(struct trace_buffer *buffer) { /* shift to debug/test normalization and TIME_EXTENTS */ return buffer->clock() << DEBUG_SHIFT; } -u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu) +u64 ring_buffer_time_stamp(struct trace_buffer *buffer, int cpu) { u64 time; @@ -760,7 +758,7 @@ u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu) } EXPORT_SYMBOL_GPL(ring_buffer_time_stamp); -void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer, +void ring_buffer_normalize_time_stamp(struct trace_buffer *buffer, int cpu, u64 *ts) { /* Just stupid testing the normalize function and deltas */ @@ -1283,7 +1281,7 @@ static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer, } static struct ring_buffer_per_cpu * -rb_allocate_cpu_buffer(struct ring_buffer *buffer, long nr_pages, int cpu) +rb_allocate_cpu_buffer(struct trace_buffer *buffer, long nr_pages, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; struct buffer_page *bpage; @@ -1368,16 +1366,17 @@ static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer) * __ring_buffer_alloc - allocate a new ring_buffer * @size: the size in bytes per cpu that is needed. * @flags: attributes to set for the ring buffer. + * @key: ring buffer reader_lock_key. * * Currently the only flag that is available is the RB_FL_OVERWRITE * flag. This flag means that the buffer will overwrite old data * when the buffer wraps. If this flag is not set, the buffer will * drop data when the tail hits the head. */ -struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, +struct trace_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *key) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; long nr_pages; int bsize; int cpu; @@ -1447,7 +1446,7 @@ EXPORT_SYMBOL_GPL(__ring_buffer_alloc); * @buffer: the buffer to free. */ void -ring_buffer_free(struct ring_buffer *buffer) +ring_buffer_free(struct trace_buffer *buffer) { int cpu; @@ -1463,18 +1462,18 @@ ring_buffer_free(struct ring_buffer *buffer) } EXPORT_SYMBOL_GPL(ring_buffer_free); -void ring_buffer_set_clock(struct ring_buffer *buffer, +void ring_buffer_set_clock(struct trace_buffer *buffer, u64 (*clock)(void)) { buffer->clock = clock; } -void ring_buffer_set_time_stamp_abs(struct ring_buffer *buffer, bool abs) +void ring_buffer_set_time_stamp_abs(struct trace_buffer *buffer, bool abs) { buffer->time_stamp_abs = abs; } -bool ring_buffer_time_stamp_abs(struct ring_buffer *buffer) +bool ring_buffer_time_stamp_abs(struct trace_buffer *buffer) { return buffer->time_stamp_abs; } @@ -1712,7 +1711,7 @@ static void update_pages_handler(struct work_struct *work) * * Returns 0 on success and < 0 on failure. */ -int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size, +int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, int cpu_id) { struct ring_buffer_per_cpu *cpu_buffer; @@ -1891,7 +1890,7 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size, } EXPORT_SYMBOL_GPL(ring_buffer_resize); -void ring_buffer_change_overwrite(struct ring_buffer *buffer, int val) +void ring_buffer_change_overwrite(struct trace_buffer *buffer, int val) { mutex_lock(&buffer->mutex); if (val) @@ -2206,7 +2205,7 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, { struct buffer_page *tail_page = info->tail_page; struct buffer_page *commit_page = cpu_buffer->commit_page; - struct ring_buffer *buffer = cpu_buffer->buffer; + struct trace_buffer *buffer = cpu_buffer->buffer; struct buffer_page *next_page; int ret; @@ -2330,11 +2329,11 @@ static inline bool rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer, /** * rb_update_event - update event type and data + * @cpu_buffer: The per cpu buffer of the @event * @event: the event to update - * @type: the type of event - * @length: the size of the event field in the ring buffer + * @info: The info to update the @event with (contains length and delta) * - * Update the type and data fields of the event. The length + * Update the type and data fields of the @event. The length * is the actual size that is written to the ring buffer, * and with this, we can determine what to place into the * data field. @@ -2609,7 +2608,7 @@ static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer, } static __always_inline void -rb_wakeups(struct ring_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer) +rb_wakeups(struct trace_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer) { size_t nr_pages; size_t dirty; @@ -2733,7 +2732,7 @@ trace_recursive_unlock(struct ring_buffer_per_cpu *cpu_buffer) * Call this function before calling another ring_buffer_lock_reserve() and * call ring_buffer_nest_end() after the nested ring_buffer_unlock_commit(). */ -void ring_buffer_nest_start(struct ring_buffer *buffer) +void ring_buffer_nest_start(struct trace_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; int cpu; @@ -2753,7 +2752,7 @@ void ring_buffer_nest_start(struct ring_buffer *buffer) * Must be called after ring_buffer_nest_start() and after the * ring_buffer_unlock_commit(). */ -void ring_buffer_nest_end(struct ring_buffer *buffer) +void ring_buffer_nest_end(struct trace_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; int cpu; @@ -2775,7 +2774,7 @@ void ring_buffer_nest_end(struct ring_buffer *buffer) * * Must be paired with ring_buffer_lock_reserve. */ -int ring_buffer_unlock_commit(struct ring_buffer *buffer, +int ring_buffer_unlock_commit(struct trace_buffer *buffer, struct ring_buffer_event *event) { struct ring_buffer_per_cpu *cpu_buffer; @@ -2868,7 +2867,7 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, } static __always_inline struct ring_buffer_event * -rb_reserve_next_event(struct ring_buffer *buffer, +rb_reserve_next_event(struct trace_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer, unsigned long length) { @@ -2961,7 +2960,7 @@ rb_reserve_next_event(struct ring_buffer *buffer, * If NULL is returned, then nothing has been allocated or locked. */ struct ring_buffer_event * -ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length) +ring_buffer_lock_reserve(struct trace_buffer *buffer, unsigned long length) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; @@ -3062,7 +3061,7 @@ rb_decrement_entry(struct ring_buffer_per_cpu *cpu_buffer, * If this function is called, do not call ring_buffer_unlock_commit on * the event. */ -void ring_buffer_discard_commit(struct ring_buffer *buffer, +void ring_buffer_discard_commit(struct trace_buffer *buffer, struct ring_buffer_event *event) { struct ring_buffer_per_cpu *cpu_buffer; @@ -3113,7 +3112,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_discard_commit); * Note, like ring_buffer_lock_reserve, the length is the length of the data * and not the length of the event which would hold the header. */ -int ring_buffer_write(struct ring_buffer *buffer, +int ring_buffer_write(struct trace_buffer *buffer, unsigned long length, void *data) { @@ -3193,7 +3192,7 @@ static bool rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer) * * The caller should call synchronize_rcu() after this. */ -void ring_buffer_record_disable(struct ring_buffer *buffer) +void ring_buffer_record_disable(struct trace_buffer *buffer) { atomic_inc(&buffer->record_disabled); } @@ -3206,7 +3205,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_disable); * Note, multiple disables will need the same number of enables * to truly enable the writing (much like preempt_disable). */ -void ring_buffer_record_enable(struct ring_buffer *buffer) +void ring_buffer_record_enable(struct trace_buffer *buffer) { atomic_dec(&buffer->record_disabled); } @@ -3223,7 +3222,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_enable); * it works like an on/off switch, where as the disable() version * must be paired with a enable(). */ -void ring_buffer_record_off(struct ring_buffer *buffer) +void ring_buffer_record_off(struct trace_buffer *buffer) { unsigned int rd; unsigned int new_rd; @@ -3246,7 +3245,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_off); * it works like an on/off switch, where as the enable() version * must be paired with a disable(). */ -void ring_buffer_record_on(struct ring_buffer *buffer) +void ring_buffer_record_on(struct trace_buffer *buffer) { unsigned int rd; unsigned int new_rd; @@ -3264,7 +3263,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_on); * * Returns true if the ring buffer is in a state that it accepts writes. */ -bool ring_buffer_record_is_on(struct ring_buffer *buffer) +bool ring_buffer_record_is_on(struct trace_buffer *buffer) { return !atomic_read(&buffer->record_disabled); } @@ -3280,7 +3279,7 @@ bool ring_buffer_record_is_on(struct ring_buffer *buffer) * ring_buffer_record_disable(), as that is a temporary disabling of * the ring buffer. */ -bool ring_buffer_record_is_set_on(struct ring_buffer *buffer) +bool ring_buffer_record_is_set_on(struct trace_buffer *buffer) { return !(atomic_read(&buffer->record_disabled) & RB_BUFFER_OFF); } @@ -3295,7 +3294,7 @@ bool ring_buffer_record_is_set_on(struct ring_buffer *buffer) * * The caller should call synchronize_rcu() after this. */ -void ring_buffer_record_disable_cpu(struct ring_buffer *buffer, int cpu) +void ring_buffer_record_disable_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; @@ -3315,7 +3314,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_record_disable_cpu); * Note, multiple disables will need the same number of enables * to truly enable the writing (much like preempt_disable). */ -void ring_buffer_record_enable_cpu(struct ring_buffer *buffer, int cpu) +void ring_buffer_record_enable_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; @@ -3345,7 +3344,7 @@ rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer) * @buffer: The ring buffer * @cpu: The per CPU buffer to read from. */ -u64 ring_buffer_oldest_event_ts(struct ring_buffer *buffer, int cpu) +u64 ring_buffer_oldest_event_ts(struct trace_buffer *buffer, int cpu) { unsigned long flags; struct ring_buffer_per_cpu *cpu_buffer; @@ -3378,7 +3377,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_oldest_event_ts); * @buffer: The ring buffer * @cpu: The per CPU buffer to read from. */ -unsigned long ring_buffer_bytes_cpu(struct ring_buffer *buffer, int cpu) +unsigned long ring_buffer_bytes_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long ret; @@ -3398,7 +3397,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_bytes_cpu); * @buffer: The ring buffer * @cpu: The per CPU buffer to get the entries from. */ -unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu) +unsigned long ring_buffer_entries_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; @@ -3417,7 +3416,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu); * @buffer: The ring buffer * @cpu: The per CPU buffer to get the number of overruns from */ -unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu) +unsigned long ring_buffer_overrun_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long ret; @@ -3440,7 +3439,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_overrun_cpu); * @cpu: The per CPU buffer to get the number of overruns from */ unsigned long -ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu) +ring_buffer_commit_overrun_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long ret; @@ -3462,7 +3461,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_commit_overrun_cpu); * @cpu: The per CPU buffer to get the number of overruns from */ unsigned long -ring_buffer_dropped_events_cpu(struct ring_buffer *buffer, int cpu) +ring_buffer_dropped_events_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long ret; @@ -3483,7 +3482,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_dropped_events_cpu); * @cpu: The per CPU buffer to get the number of events read */ unsigned long -ring_buffer_read_events_cpu(struct ring_buffer *buffer, int cpu) +ring_buffer_read_events_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; @@ -3502,7 +3501,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_read_events_cpu); * Returns the total number of entries in the ring buffer * (all CPU entries) */ -unsigned long ring_buffer_entries(struct ring_buffer *buffer) +unsigned long ring_buffer_entries(struct trace_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long entries = 0; @@ -3525,7 +3524,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_entries); * Returns the total number of overruns in the ring buffer * (all CPU entries) */ -unsigned long ring_buffer_overruns(struct ring_buffer *buffer) +unsigned long ring_buffer_overruns(struct trace_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long overruns = 0; @@ -3949,7 +3948,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_peek); static struct ring_buffer_event * rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; int nr_loops = 0; @@ -4077,7 +4076,7 @@ rb_reader_unlock(struct ring_buffer_per_cpu *cpu_buffer, bool locked) * not consume the data. */ struct ring_buffer_event * -ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts, +ring_buffer_peek(struct trace_buffer *buffer, int cpu, u64 *ts, unsigned long *lost_events) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; @@ -4141,7 +4140,7 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) * and eventually empty the ring buffer if the producer is slower. */ struct ring_buffer_event * -ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, +ring_buffer_consume(struct trace_buffer *buffer, int cpu, u64 *ts, unsigned long *lost_events) { struct ring_buffer_per_cpu *cpu_buffer; @@ -4201,7 +4200,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_consume); * This overall must be paired with ring_buffer_read_finish. */ struct ring_buffer_iter * -ring_buffer_read_prepare(struct ring_buffer *buffer, int cpu, gfp_t flags) +ring_buffer_read_prepare(struct trace_buffer *buffer, int cpu, gfp_t flags) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_iter *iter; @@ -4331,8 +4330,9 @@ EXPORT_SYMBOL_GPL(ring_buffer_read); /** * ring_buffer_size - return the size of the ring buffer (in bytes) * @buffer: The ring buffer. + * @cpu: The CPU to get ring buffer size from. */ -unsigned long ring_buffer_size(struct ring_buffer *buffer, int cpu) +unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu) { /* * Earlier, this method returned @@ -4398,7 +4398,7 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) * @buffer: The ring buffer to reset a per cpu buffer of * @cpu: The CPU buffer to be reset */ -void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) +void ring_buffer_reset_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; unsigned long flags; @@ -4435,7 +4435,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_reset_cpu); * ring_buffer_reset - reset a ring buffer * @buffer: The ring buffer to reset all cpu buffers */ -void ring_buffer_reset(struct ring_buffer *buffer) +void ring_buffer_reset(struct trace_buffer *buffer) { int cpu; @@ -4448,7 +4448,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_reset); * rind_buffer_empty - is the ring buffer empty? * @buffer: The ring buffer to test */ -bool ring_buffer_empty(struct ring_buffer *buffer) +bool ring_buffer_empty(struct trace_buffer *buffer) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long flags; @@ -4478,7 +4478,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_empty); * @buffer: The ring buffer * @cpu: The CPU buffer to test */ -bool ring_buffer_empty_cpu(struct ring_buffer *buffer, int cpu) +bool ring_buffer_empty_cpu(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; unsigned long flags; @@ -4504,14 +4504,15 @@ EXPORT_SYMBOL_GPL(ring_buffer_empty_cpu); * ring_buffer_swap_cpu - swap a CPU buffer between two ring buffers * @buffer_a: One buffer to swap with * @buffer_b: The other buffer to swap with + * @cpu: the CPU of the buffers to swap * * This function is useful for tracers that want to take a "snapshot" * of a CPU buffer and has another back up buffer lying around. * it is expected that the tracer handles the cpu buffer not being * used at the moment. */ -int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, - struct ring_buffer *buffer_b, int cpu) +int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, + struct trace_buffer *buffer_b, int cpu) { struct ring_buffer_per_cpu *cpu_buffer_a; struct ring_buffer_per_cpu *cpu_buffer_b; @@ -4590,7 +4591,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); * Returns: * The page allocated, or ERR_PTR */ -void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) +void *ring_buffer_alloc_read_page(struct trace_buffer *buffer, int cpu) { struct ring_buffer_per_cpu *cpu_buffer; struct buffer_data_page *bpage = NULL; @@ -4637,7 +4638,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); * * Free a page allocated from ring_buffer_alloc_read_page. */ -void ring_buffer_free_read_page(struct ring_buffer *buffer, int cpu, void *data) +void ring_buffer_free_read_page(struct trace_buffer *buffer, int cpu, void *data) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct buffer_data_page *bpage = data; @@ -4697,7 +4698,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * >=0 if data has been transferred, returns the offset of consumed data. * <0 if no data has been transferred. */ -int ring_buffer_read_page(struct ring_buffer *buffer, +int ring_buffer_read_page(struct trace_buffer *buffer, void **data_page, size_t len, int cpu, int full) { struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; @@ -4868,12 +4869,12 @@ EXPORT_SYMBOL_GPL(ring_buffer_read_page); */ int trace_rb_cpu_prepare(unsigned int cpu, struct hlist_node *node) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; long nr_pages_same; int cpu_i; unsigned long nr_pages; - buffer = container_of(node, struct ring_buffer, node); + buffer = container_of(node, struct trace_buffer, node); if (cpumask_test_cpu(cpu, buffer->cpumask)) return 0; @@ -4923,7 +4924,7 @@ int trace_rb_cpu_prepare(unsigned int cpu, struct hlist_node *node) static struct task_struct *rb_threads[NR_CPUS] __initdata; struct rb_test_data { - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long events; unsigned long bytes_written; unsigned long bytes_alloc; @@ -5065,7 +5066,7 @@ static __init int rb_hammer_test(void *arg) static __init int test_ringbuffer(void) { struct task_struct *rb_hammer; - struct ring_buffer *buffer; + struct trace_buffer *buffer; int cpu; int ret = 0; diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index 32149e46551c..8df0aa810950 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -29,7 +29,7 @@ static int reader_finish; static DECLARE_COMPLETION(read_start); static DECLARE_COMPLETION(read_done); -static struct ring_buffer *buffer; +static struct trace_buffer *buffer; static struct task_struct *producer; static struct task_struct *consumer; static unsigned long read; diff --git a/kernel/trace/synth_event_gen_test.c b/kernel/trace/synth_event_gen_test.c new file mode 100644 index 000000000000..4aefe003cb7c --- /dev/null +++ b/kernel/trace/synth_event_gen_test.c @@ -0,0 +1,523 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Test module for in-kernel sythetic event creation and generation. + * + * Copyright (C) 2019 Tom Zanussi <zanussi@kernel.org> + */ + +#include <linux/module.h> +#include <linux/trace_events.h> + +/* + * This module is a simple test of basic functionality for in-kernel + * synthetic event creation and generation, the first and second tests + * using synth_event_gen_cmd_start() and synth_event_add_field(), the + * third uses synth_event_create() to do it all at once with a static + * field array. + * + * Following that are a few examples using the created events to test + * various ways of tracing a synthetic event. + * + * To test, select CONFIG_SYNTH_EVENT_GEN_TEST and build the module. + * Then: + * + * # insmod kernel/trace/synth_event_gen_test.ko + * # cat /sys/kernel/debug/tracing/trace + * + * You should see several events in the trace buffer - + * "create_synth_test", "empty_synth_test", and several instances of + * "gen_synth_test". + * + * To remove the events, remove the module: + * + * # rmmod synth_event_gen_test + * + */ + +static struct trace_event_file *create_synth_test; +static struct trace_event_file *empty_synth_test; +static struct trace_event_file *gen_synth_test; + +/* + * Test to make sure we can create a synthetic event, then add more + * fields. + */ +static int __init test_gen_synth_cmd(void) +{ + struct dynevent_cmd cmd; + u64 vals[7]; + char *buf; + int ret; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + /* Before generating the command, initialize the cmd object */ + synth_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + /* + * Create the empty gen_synth_test synthetic event with the + * first 4 fields. + */ + ret = synth_event_gen_cmd_start(&cmd, "gen_synth_test", THIS_MODULE, + "pid_t", "next_pid_field", + "char[16]", "next_comm_field", + "u64", "ts_ns", + "u64", "ts_ms"); + if (ret) + goto free; + + /* Use synth_event_add_field to add the rest of the fields */ + + ret = synth_event_add_field(&cmd, "unsigned int", "cpu"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "char[64]", "my_string_field"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "int", "my_int_field"); + if (ret) + goto free; + + ret = synth_event_gen_cmd_end(&cmd); + if (ret) + goto free; + + /* + * Now get the gen_synth_test event file. We need to prevent + * the instance and event from disappearing from underneath + * us, which trace_get_event_file() does (though in this case + * we're using the top-level instance which never goes away). + */ + gen_synth_test = trace_get_event_file(NULL, "synthetic", + "gen_synth_test"); + if (IS_ERR(gen_synth_test)) { + ret = PTR_ERR(gen_synth_test); + goto delete; + } + + /* Enable the event or you won't see anything */ + ret = trace_array_set_clr_event(gen_synth_test->tr, + "synthetic", "gen_synth_test", true); + if (ret) { + trace_put_event_file(gen_synth_test); + goto delete; + } + + /* Create some bogus values just for testing */ + + vals[0] = 777; /* next_pid_field */ + vals[1] = (u64)"hula hoops"; /* next_comm_field */ + vals[2] = 1000000; /* ts_ns */ + vals[3] = 1000; /* ts_ms */ + vals[4] = smp_processor_id(); /* cpu */ + vals[5] = (u64)"thneed"; /* my_string_field */ + vals[6] = 598; /* my_int_field */ + + /* Now generate a gen_synth_test event */ + ret = synth_event_trace_array(gen_synth_test, vals, ARRAY_SIZE(vals)); + out: + return ret; + delete: + /* We got an error after creating the event, delete it */ + synth_event_delete("gen_synth_test"); + free: + kfree(buf); + + goto out; +} + +/* + * Test to make sure we can create an initially empty synthetic event, + * then add all the fields. + */ +static int __init test_empty_synth_event(void) +{ + struct dynevent_cmd cmd; + u64 vals[7]; + char *buf; + int ret; + + /* Create a buffer to hold the generated command */ + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + /* Before generating the command, initialize the cmd object */ + synth_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + /* + * Create the empty_synth_test synthetic event with no fields. + */ + ret = synth_event_gen_cmd_start(&cmd, "empty_synth_test", THIS_MODULE); + if (ret) + goto free; + + /* Use synth_event_add_field to add all of the fields */ + + ret = synth_event_add_field(&cmd, "pid_t", "next_pid_field"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "char[16]", "next_comm_field"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "u64", "ts_ns"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "u64", "ts_ms"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "unsigned int", "cpu"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "char[64]", "my_string_field"); + if (ret) + goto free; + + ret = synth_event_add_field(&cmd, "int", "my_int_field"); + if (ret) + goto free; + + /* All fields have been added, close and register the synth event */ + + ret = synth_event_gen_cmd_end(&cmd); + if (ret) + goto free; + + /* + * Now get the empty_synth_test event file. We need to + * prevent the instance and event from disappearing from + * underneath us, which trace_get_event_file() does (though in + * this case we're using the top-level instance which never + * goes away). + */ + empty_synth_test = trace_get_event_file(NULL, "synthetic", + "empty_synth_test"); + if (IS_ERR(empty_synth_test)) { + ret = PTR_ERR(empty_synth_test); + goto delete; + } + + /* Enable the event or you won't see anything */ + ret = trace_array_set_clr_event(empty_synth_test->tr, + "synthetic", "empty_synth_test", true); + if (ret) { + trace_put_event_file(empty_synth_test); + goto delete; + } + + /* Create some bogus values just for testing */ + + vals[0] = 777; /* next_pid_field */ + vals[1] = (u64)"tiddlywinks"; /* next_comm_field */ + vals[2] = 1000000; /* ts_ns */ + vals[3] = 1000; /* ts_ms */ + vals[4] = smp_processor_id(); /* cpu */ + vals[5] = (u64)"thneed_2.0"; /* my_string_field */ + vals[6] = 399; /* my_int_field */ + + /* Now trace an empty_synth_test event */ + ret = synth_event_trace_array(empty_synth_test, vals, ARRAY_SIZE(vals)); + out: + return ret; + delete: + /* We got an error after creating the event, delete it */ + synth_event_delete("empty_synth_test"); + free: + kfree(buf); + + goto out; +} + +static struct synth_field_desc create_synth_test_fields[] = { + { .type = "pid_t", .name = "next_pid_field" }, + { .type = "char[16]", .name = "next_comm_field" }, + { .type = "u64", .name = "ts_ns" }, + { .type = "u64", .name = "ts_ms" }, + { .type = "unsigned int", .name = "cpu" }, + { .type = "char[64]", .name = "my_string_field" }, + { .type = "int", .name = "my_int_field" }, +}; + +/* + * Test synthetic event creation all at once from array of field + * descriptors. + */ +static int __init test_create_synth_event(void) +{ + u64 vals[7]; + int ret; + + /* Create the create_synth_test event with the fields above */ + ret = synth_event_create("create_synth_test", + create_synth_test_fields, + ARRAY_SIZE(create_synth_test_fields), + THIS_MODULE); + if (ret) + goto out; + + /* + * Now get the create_synth_test event file. We need to + * prevent the instance and event from disappearing from + * underneath us, which trace_get_event_file() does (though in + * this case we're using the top-level instance which never + * goes away). + */ + create_synth_test = trace_get_event_file(NULL, "synthetic", + "create_synth_test"); + if (IS_ERR(create_synth_test)) { + ret = PTR_ERR(create_synth_test); + goto delete; + } + + /* Enable the event or you won't see anything */ + ret = trace_array_set_clr_event(create_synth_test->tr, + "synthetic", "create_synth_test", true); + if (ret) { + trace_put_event_file(create_synth_test); + goto delete; + } + + /* Create some bogus values just for testing */ + + vals[0] = 777; /* next_pid_field */ + vals[1] = (u64)"tiddlywinks"; /* next_comm_field */ + vals[2] = 1000000; /* ts_ns */ + vals[3] = 1000; /* ts_ms */ + vals[4] = smp_processor_id(); /* cpu */ + vals[5] = (u64)"thneed"; /* my_string_field */ + vals[6] = 398; /* my_int_field */ + + /* Now generate a create_synth_test event */ + ret = synth_event_trace_array(create_synth_test, vals, ARRAY_SIZE(vals)); + out: + return ret; + delete: + /* We got an error after creating the event, delete it */ + ret = synth_event_delete("create_synth_test"); + + goto out; +} + +/* + * Test tracing a synthetic event by reserving trace buffer space, + * then filling in fields one after another. + */ +static int __init test_add_next_synth_val(void) +{ + struct synth_event_trace_state trace_state; + int ret; + + /* Start by reserving space in the trace buffer */ + ret = synth_event_trace_start(gen_synth_test, &trace_state); + if (ret) + return ret; + + /* Write some bogus values into the trace buffer, one after another */ + + /* next_pid_field */ + ret = synth_event_add_next_val(777, &trace_state); + if (ret) + goto out; + + /* next_comm_field */ + ret = synth_event_add_next_val((u64)"slinky", &trace_state); + if (ret) + goto out; + + /* ts_ns */ + ret = synth_event_add_next_val(1000000, &trace_state); + if (ret) + goto out; + + /* ts_ms */ + ret = synth_event_add_next_val(1000, &trace_state); + if (ret) + goto out; + + /* cpu */ + ret = synth_event_add_next_val(smp_processor_id(), &trace_state); + if (ret) + goto out; + + /* my_string_field */ + ret = synth_event_add_next_val((u64)"thneed_2.01", &trace_state); + if (ret) + goto out; + + /* my_int_field */ + ret = synth_event_add_next_val(395, &trace_state); + out: + /* Finally, commit the event */ + ret = synth_event_trace_end(&trace_state); + + return ret; +} + +/* + * Test tracing a synthetic event by reserving trace buffer space, + * then filling in fields using field names, which can be done in any + * order. + */ +static int __init test_add_synth_val(void) +{ + struct synth_event_trace_state trace_state; + int ret; + + /* Start by reserving space in the trace buffer */ + ret = synth_event_trace_start(gen_synth_test, &trace_state); + if (ret) + return ret; + + /* Write some bogus values into the trace buffer, using field names */ + + ret = synth_event_add_val("ts_ns", 1000000, &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("ts_ms", 1000, &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("cpu", smp_processor_id(), &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("next_pid_field", 777, &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("next_comm_field", (u64)"silly putty", + &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("my_string_field", (u64)"thneed_9", + &trace_state); + if (ret) + goto out; + + ret = synth_event_add_val("my_int_field", 3999, &trace_state); + out: + /* Finally, commit the event */ + ret = synth_event_trace_end(&trace_state); + + return ret; +} + +/* + * Test tracing a synthetic event all at once from array of values. + */ +static int __init test_trace_synth_event(void) +{ + int ret; + + /* Trace some bogus values just for testing */ + ret = synth_event_trace(create_synth_test, 7, /* number of values */ + 444, /* next_pid_field */ + (u64)"clackers", /* next_comm_field */ + 1000000, /* ts_ns */ + 1000, /* ts_ms */ + smp_processor_id(), /* cpu */ + (u64)"Thneed", /* my_string_field */ + 999); /* my_int_field */ + return ret; +} + +static int __init synth_event_gen_test_init(void) +{ + int ret; + + ret = test_gen_synth_cmd(); + if (ret) + return ret; + + ret = test_empty_synth_event(); + if (ret) { + WARN_ON(trace_array_set_clr_event(gen_synth_test->tr, + "synthetic", + "gen_synth_test", false)); + trace_put_event_file(gen_synth_test); + WARN_ON(synth_event_delete("gen_synth_test")); + goto out; + } + + ret = test_create_synth_event(); + if (ret) { + WARN_ON(trace_array_set_clr_event(gen_synth_test->tr, + "synthetic", + "gen_synth_test", false)); + trace_put_event_file(gen_synth_test); + WARN_ON(synth_event_delete("gen_synth_test")); + + WARN_ON(trace_array_set_clr_event(empty_synth_test->tr, + "synthetic", + "empty_synth_test", false)); + trace_put_event_file(empty_synth_test); + WARN_ON(synth_event_delete("empty_synth_test")); + goto out; + } + + ret = test_add_next_synth_val(); + WARN_ON(ret); + + ret = test_add_synth_val(); + WARN_ON(ret); + + ret = test_trace_synth_event(); + WARN_ON(ret); + out: + return ret; +} + +static void __exit synth_event_gen_test_exit(void) +{ + /* Disable the event or you can't remove it */ + WARN_ON(trace_array_set_clr_event(gen_synth_test->tr, + "synthetic", + "gen_synth_test", false)); + + /* Now give the file and instance back */ + trace_put_event_file(gen_synth_test); + + /* Now unregister and free the synthetic event */ + WARN_ON(synth_event_delete("gen_synth_test")); + + /* Disable the event or you can't remove it */ + WARN_ON(trace_array_set_clr_event(empty_synth_test->tr, + "synthetic", + "empty_synth_test", false)); + + /* Now give the file and instance back */ + trace_put_event_file(empty_synth_test); + + /* Now unregister and free the synthetic event */ + WARN_ON(synth_event_delete("empty_synth_test")); + + /* Disable the event or you can't remove it */ + WARN_ON(trace_array_set_clr_event(create_synth_test->tr, + "synthetic", + "create_synth_test", false)); + + /* Now give the file and instance back */ + trace_put_event_file(create_synth_test); + + /* Now unregister and free the synthetic event */ + WARN_ON(synth_event_delete("create_synth_test")); +} + +module_init(synth_event_gen_test_init) +module_exit(synth_event_gen_test_exit) + +MODULE_AUTHOR("Tom Zanussi"); +MODULE_DESCRIPTION("synthetic event generation test"); +MODULE_LICENSE("GPL v2"); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 5e02d65505c1..c797a15a1fc7 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -162,8 +162,8 @@ union trace_eval_map_item { static union trace_eval_map_item *trace_eval_maps; #endif /* CONFIG_TRACE_EVAL_MAP_FILE */ -static int tracing_set_tracer(struct trace_array *tr, const char *buf); -static void ftrace_trace_userstack(struct ring_buffer *buffer, +int tracing_set_tracer(struct trace_array *tr, const char *buf); +static void ftrace_trace_userstack(struct trace_buffer *buffer, unsigned long flags, int pc); #define MAX_TRACER_SIZE 100 @@ -338,7 +338,7 @@ int tracing_check_open_get_tr(struct trace_array *tr) } int call_filter_check_discard(struct trace_event_call *call, void *rec, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event) { if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) && @@ -603,7 +603,7 @@ int trace_pid_write(struct trace_pid_list *filtered_pids, return read; } -static u64 buffer_ftrace_now(struct trace_buffer *buf, int cpu) +static u64 buffer_ftrace_now(struct array_buffer *buf, int cpu) { u64 ts; @@ -619,7 +619,7 @@ static u64 buffer_ftrace_now(struct trace_buffer *buf, int cpu) u64 ftrace_now(int cpu) { - return buffer_ftrace_now(&global_trace.trace_buffer, cpu); + return buffer_ftrace_now(&global_trace.array_buffer, cpu); } /** @@ -747,22 +747,22 @@ static inline void trace_access_lock_init(void) #endif #ifdef CONFIG_STACKTRACE -static void __ftrace_trace_stack(struct ring_buffer *buffer, +static void __ftrace_trace_stack(struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs); static inline void ftrace_trace_stack(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs); #else -static inline void __ftrace_trace_stack(struct ring_buffer *buffer, +static inline void __ftrace_trace_stack(struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs) { } static inline void ftrace_trace_stack(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs) { @@ -780,7 +780,7 @@ trace_event_setup(struct ring_buffer_event *event, } static __always_inline struct ring_buffer_event * -__trace_buffer_lock_reserve(struct ring_buffer *buffer, +__trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, unsigned long flags, int pc) @@ -796,8 +796,8 @@ __trace_buffer_lock_reserve(struct ring_buffer *buffer, void tracer_tracing_on(struct trace_array *tr) { - if (tr->trace_buffer.buffer) - ring_buffer_record_on(tr->trace_buffer.buffer); + if (tr->array_buffer.buffer) + ring_buffer_record_on(tr->array_buffer.buffer); /* * This flag is looked at when buffers haven't been allocated * yet, or by some tracers (like irqsoff), that just want to @@ -825,7 +825,7 @@ EXPORT_SYMBOL_GPL(tracing_on); static __always_inline void -__buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event) +__buffer_unlock_commit(struct trace_buffer *buffer, struct ring_buffer_event *event) { __this_cpu_write(trace_taskinfo_save, true); @@ -848,7 +848,7 @@ __buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *eve int __trace_puts(unsigned long ip, const char *str, int size) { struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct print_entry *entry; unsigned long irq_flags; int alloc; @@ -865,11 +865,14 @@ int __trace_puts(unsigned long ip, const char *str, int size) alloc = sizeof(*entry) + size + 2; /* possible \n added */ local_save_flags(irq_flags); - buffer = global_trace.trace_buffer.buffer; + buffer = global_trace.array_buffer.buffer; + ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, alloc, irq_flags, pc); - if (!event) - return 0; + if (!event) { + size = 0; + goto out; + } entry = ring_buffer_event_data(event); entry->ip = ip; @@ -885,7 +888,8 @@ int __trace_puts(unsigned long ip, const char *str, int size) __buffer_unlock_commit(buffer, event); ftrace_trace_stack(&global_trace, buffer, irq_flags, 4, pc, NULL); - + out: + ring_buffer_nest_end(buffer); return size; } EXPORT_SYMBOL_GPL(__trace_puts); @@ -898,10 +902,11 @@ EXPORT_SYMBOL_GPL(__trace_puts); int __trace_bputs(unsigned long ip, const char *str) { struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct bputs_entry *entry; unsigned long irq_flags; int size = sizeof(struct bputs_entry); + int ret = 0; int pc; if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) @@ -913,11 +918,13 @@ int __trace_bputs(unsigned long ip, const char *str) return 0; local_save_flags(irq_flags); - buffer = global_trace.trace_buffer.buffer; + buffer = global_trace.array_buffer.buffer; + + ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_BPUTS, size, irq_flags, pc); if (!event) - return 0; + goto out; entry = ring_buffer_event_data(event); entry->ip = ip; @@ -926,7 +933,10 @@ int __trace_bputs(unsigned long ip, const char *str) __buffer_unlock_commit(buffer, event); ftrace_trace_stack(&global_trace, buffer, irq_flags, 4, pc, NULL); - return 1; + ret = 1; + out: + ring_buffer_nest_end(buffer); + return ret; } EXPORT_SYMBOL_GPL(__trace_bputs); @@ -1036,9 +1046,9 @@ void *tracing_cond_snapshot_data(struct trace_array *tr) } EXPORT_SYMBOL_GPL(tracing_cond_snapshot_data); -static int resize_buffer_duplicate_size(struct trace_buffer *trace_buf, - struct trace_buffer *size_buf, int cpu_id); -static void set_buffer_entries(struct trace_buffer *buf, unsigned long val); +static int resize_buffer_duplicate_size(struct array_buffer *trace_buf, + struct array_buffer *size_buf, int cpu_id); +static void set_buffer_entries(struct array_buffer *buf, unsigned long val); int tracing_alloc_snapshot_instance(struct trace_array *tr) { @@ -1048,7 +1058,7 @@ int tracing_alloc_snapshot_instance(struct trace_array *tr) /* allocate spare buffer */ ret = resize_buffer_duplicate_size(&tr->max_buffer, - &tr->trace_buffer, RING_BUFFER_ALL_CPUS); + &tr->array_buffer, RING_BUFFER_ALL_CPUS); if (ret < 0) return ret; @@ -1251,8 +1261,8 @@ EXPORT_SYMBOL_GPL(tracing_snapshot_cond_disable); void tracer_tracing_off(struct trace_array *tr) { - if (tr->trace_buffer.buffer) - ring_buffer_record_off(tr->trace_buffer.buffer); + if (tr->array_buffer.buffer) + ring_buffer_record_off(tr->array_buffer.buffer); /* * This flag is looked at when buffers haven't been allocated * yet, or by some tracers (like irqsoff), that just want to @@ -1294,8 +1304,8 @@ void disable_trace_on_warning(void) */ bool tracer_tracing_is_on(struct trace_array *tr) { - if (tr->trace_buffer.buffer) - return ring_buffer_record_is_on(tr->trace_buffer.buffer); + if (tr->array_buffer.buffer) + return ring_buffer_record_is_on(tr->array_buffer.buffer); return !tr->buffer_disabled; } @@ -1590,8 +1600,8 @@ void latency_fsnotify(struct trace_array *tr) static void __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) { - struct trace_buffer *trace_buf = &tr->trace_buffer; - struct trace_buffer *max_buf = &tr->max_buffer; + struct array_buffer *trace_buf = &tr->array_buffer; + struct array_buffer *max_buf = &tr->max_buffer; struct trace_array_cpu *data = per_cpu_ptr(trace_buf->data, cpu); struct trace_array_cpu *max_data = per_cpu_ptr(max_buf->data, cpu); @@ -1649,8 +1659,8 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu, arch_spin_lock(&tr->max_lock); - /* Inherit the recordable setting from trace_buffer */ - if (ring_buffer_record_is_set_on(tr->trace_buffer.buffer)) + /* Inherit the recordable setting from array_buffer */ + if (ring_buffer_record_is_set_on(tr->array_buffer.buffer)) ring_buffer_record_on(tr->max_buffer.buffer); else ring_buffer_record_off(tr->max_buffer.buffer); @@ -1659,7 +1669,7 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu, if (tr->cond_snapshot && !tr->cond_snapshot->update(tr, cond_data)) goto out_unlock; #endif - swap(tr->trace_buffer.buffer, tr->max_buffer.buffer); + swap(tr->array_buffer.buffer, tr->max_buffer.buffer); __update_max_tr(tr, tsk, cpu); @@ -1692,7 +1702,7 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) arch_spin_lock(&tr->max_lock); - ret = ring_buffer_swap_cpu(tr->max_buffer.buffer, tr->trace_buffer.buffer, cpu); + ret = ring_buffer_swap_cpu(tr->max_buffer.buffer, tr->array_buffer.buffer, cpu); if (ret == -EBUSY) { /* @@ -1718,7 +1728,7 @@ static int wait_on_pipe(struct trace_iterator *iter, int full) if (trace_buffer_iter(iter, iter->cpu_file)) return 0; - return ring_buffer_wait(iter->trace_buffer->buffer, iter->cpu_file, + return ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full); } @@ -1769,7 +1779,7 @@ static int run_tracer_selftest(struct tracer *type) * internal tracing to verify that everything is in order. * If we fail, we do not register this tracer. */ - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); tr->current_trace = type; @@ -1795,7 +1805,7 @@ static int run_tracer_selftest(struct tracer *type) return -1; } /* Only reset on passing, to avoid touching corrupted buffers */ - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); #ifdef CONFIG_TRACER_MAX_TRACE if (type->use_max_tr) { @@ -1962,9 +1972,9 @@ int __init register_tracer(struct tracer *type) return ret; } -static void tracing_reset_cpu(struct trace_buffer *buf, int cpu) +static void tracing_reset_cpu(struct array_buffer *buf, int cpu) { - struct ring_buffer *buffer = buf->buffer; + struct trace_buffer *buffer = buf->buffer; if (!buffer) return; @@ -1978,9 +1988,9 @@ static void tracing_reset_cpu(struct trace_buffer *buf, int cpu) ring_buffer_record_enable(buffer); } -void tracing_reset_online_cpus(struct trace_buffer *buf) +void tracing_reset_online_cpus(struct array_buffer *buf) { - struct ring_buffer *buffer = buf->buffer; + struct trace_buffer *buffer = buf->buffer; int cpu; if (!buffer) @@ -2008,7 +2018,7 @@ void tracing_reset_all_online_cpus(void) if (!tr->clear_trace) continue; tr->clear_trace = false; - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); #ifdef CONFIG_TRACER_MAX_TRACE tracing_reset_online_cpus(&tr->max_buffer); #endif @@ -2098,7 +2108,7 @@ int is_tracing_stopped(void) */ void tracing_start(void) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long flags; if (tracing_disabled) @@ -2117,7 +2127,7 @@ void tracing_start(void) /* Prevent the buffers from switching */ arch_spin_lock(&global_trace.max_lock); - buffer = global_trace.trace_buffer.buffer; + buffer = global_trace.array_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); @@ -2135,7 +2145,7 @@ void tracing_start(void) static void tracing_start_tr(struct trace_array *tr) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long flags; if (tracing_disabled) @@ -2156,7 +2166,7 @@ static void tracing_start_tr(struct trace_array *tr) goto out; } - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); @@ -2172,7 +2182,7 @@ static void tracing_start_tr(struct trace_array *tr) */ void tracing_stop(void) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long flags; raw_spin_lock_irqsave(&global_trace.start_lock, flags); @@ -2182,7 +2192,7 @@ void tracing_stop(void) /* Prevent the buffers from switching */ arch_spin_lock(&global_trace.max_lock); - buffer = global_trace.trace_buffer.buffer; + buffer = global_trace.array_buffer.buffer; if (buffer) ring_buffer_record_disable(buffer); @@ -2200,7 +2210,7 @@ void tracing_stop(void) static void tracing_stop_tr(struct trace_array *tr) { - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long flags; /* If global, we need to also stop the max tracer */ @@ -2211,7 +2221,7 @@ static void tracing_stop_tr(struct trace_array *tr) if (tr->stop_count++) goto out; - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_disable(buffer); @@ -2442,7 +2452,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned short type, EXPORT_SYMBOL_GPL(tracing_generic_entry_update); struct ring_buffer_event * -trace_buffer_lock_reserve(struct ring_buffer *buffer, +trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, unsigned long flags, int pc) @@ -2561,10 +2571,10 @@ void trace_buffered_event_disable(void) preempt_enable(); } -static struct ring_buffer *temp_buffer; +static struct trace_buffer *temp_buffer; struct ring_buffer_event * -trace_event_buffer_lock_reserve(struct ring_buffer **current_rb, +trace_event_buffer_lock_reserve(struct trace_buffer **current_rb, struct trace_event_file *trace_file, int type, unsigned long len, unsigned long flags, int pc) @@ -2572,7 +2582,7 @@ trace_event_buffer_lock_reserve(struct ring_buffer **current_rb, struct ring_buffer_event *entry; int val; - *current_rb = trace_file->tr->trace_buffer.buffer; + *current_rb = trace_file->tr->array_buffer.buffer; if (!ring_buffer_time_stamp_abs(*current_rb) && (trace_file->flags & (EVENT_FILE_FL_SOFT_DISABLED | EVENT_FILE_FL_FILTERED)) && @@ -2610,6 +2620,7 @@ static DEFINE_MUTEX(tracepoint_printk_mutex); static void output_printk(struct trace_event_buffer *fbuffer) { struct trace_event_call *event_call; + struct trace_event_file *file; struct trace_event *event; unsigned long flags; struct trace_iterator *iter = tracepoint_print_iter; @@ -2623,6 +2634,12 @@ static void output_printk(struct trace_event_buffer *fbuffer) !event_call->event.funcs->trace) return; + file = fbuffer->trace_file; + if (test_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags) || + (unlikely(file->flags & EVENT_FILE_FL_FILTERED) && + !filter_match_preds(file->filter, fbuffer->entry))) + return; + event = &fbuffer->trace_file->event_call->event; spin_lock_irqsave(&tracepoint_iter_lock, flags); @@ -2673,9 +2690,9 @@ void trace_event_buffer_commit(struct trace_event_buffer *fbuffer) if (static_key_false(&tracepoint_printk_key.key)) output_printk(fbuffer); - event_trigger_unlock_commit(fbuffer->trace_file, fbuffer->buffer, + event_trigger_unlock_commit_regs(fbuffer->trace_file, fbuffer->buffer, fbuffer->event, fbuffer->entry, - fbuffer->flags, fbuffer->pc); + fbuffer->flags, fbuffer->pc, fbuffer->regs); } EXPORT_SYMBOL_GPL(trace_event_buffer_commit); @@ -2689,7 +2706,7 @@ EXPORT_SYMBOL_GPL(trace_event_buffer_commit); # define STACK_SKIP 3 void trace_buffer_unlock_commit_regs(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, unsigned long flags, int pc, struct pt_regs *regs) @@ -2710,7 +2727,7 @@ void trace_buffer_unlock_commit_regs(struct trace_array *tr, * Similar to trace_buffer_unlock_commit_regs() but do not dump stack. */ void -trace_buffer_unlock_commit_nostack(struct ring_buffer *buffer, +trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, struct ring_buffer_event *event) { __buffer_unlock_commit(buffer, event); @@ -2845,7 +2862,7 @@ trace_function(struct trace_array *tr, int pc) { struct trace_event_call *call = &event_function; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct ftrace_entry *entry; @@ -2883,7 +2900,7 @@ struct ftrace_stacks { static DEFINE_PER_CPU(struct ftrace_stacks, ftrace_stacks); static DEFINE_PER_CPU(int, ftrace_stack_reserve); -static void __ftrace_trace_stack(struct ring_buffer *buffer, +static void __ftrace_trace_stack(struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs) { @@ -2958,7 +2975,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer, } static inline void ftrace_trace_stack(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, unsigned long flags, int skip, int pc, struct pt_regs *regs) { @@ -2971,7 +2988,7 @@ static inline void ftrace_trace_stack(struct trace_array *tr, void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, int pc) { - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; if (rcu_is_watching()) { __ftrace_trace_stack(buffer, flags, skip, pc, NULL); @@ -3009,7 +3026,7 @@ void trace_dump_stack(int skip) /* Skip 1 to skip this function. */ skip++; #endif - __ftrace_trace_stack(global_trace.trace_buffer.buffer, + __ftrace_trace_stack(global_trace.array_buffer.buffer, flags, skip, preempt_count(), NULL); } EXPORT_SYMBOL_GPL(trace_dump_stack); @@ -3018,7 +3035,7 @@ EXPORT_SYMBOL_GPL(trace_dump_stack); static DEFINE_PER_CPU(int, user_stack_count); static void -ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) +ftrace_trace_userstack(struct trace_buffer *buffer, unsigned long flags, int pc) { struct trace_event_call *call = &event_user_stack; struct ring_buffer_event *event; @@ -3063,7 +3080,7 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc) preempt_enable(); } #else /* CONFIG_USER_STACKTRACE_SUPPORT */ -static void ftrace_trace_userstack(struct ring_buffer *buffer, +static void ftrace_trace_userstack(struct trace_buffer *buffer, unsigned long flags, int pc) { } @@ -3109,7 +3126,7 @@ static int alloc_percpu_trace_buffer(void) struct trace_buffer_struct *buffers; buffers = alloc_percpu(struct trace_buffer_struct); - if (WARN(!buffers, "Could not allocate percpu trace_printk buffer")) + if (MEM_FAIL(!buffers, "Could not allocate percpu trace_printk buffer")) return -ENOMEM; trace_percpu_buffer = buffers; @@ -3154,7 +3171,7 @@ void trace_printk_init_buffers(void) * directly here. If the global_trace.buffer is already * allocated here, then this was called by module code. */ - if (global_trace.trace_buffer.buffer) + if (global_trace.array_buffer.buffer) tracing_start_cmdline_record(); } EXPORT_SYMBOL_GPL(trace_printk_init_buffers); @@ -3188,7 +3205,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) { struct trace_event_call *call = &event_bprint; struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct trace_array *tr = &global_trace; struct bprint_entry *entry; unsigned long flags; @@ -3213,11 +3230,12 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) len = vbin_printf((u32 *)tbuffer, TRACE_BUF_SIZE/sizeof(int), fmt, args); if (len > TRACE_BUF_SIZE/sizeof(int) || len < 0) - goto out; + goto out_put; local_save_flags(flags); size = sizeof(*entry) + sizeof(u32) * len; - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; + ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_BPRINT, size, flags, pc); if (!event) @@ -3233,6 +3251,8 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) } out: + ring_buffer_nest_end(buffer); +out_put: put_trace_buf(); out_nobuffer: @@ -3245,7 +3265,7 @@ EXPORT_SYMBOL_GPL(trace_vbprintk); __printf(3, 0) static int -__trace_array_vprintk(struct ring_buffer *buffer, +__trace_array_vprintk(struct trace_buffer *buffer, unsigned long ip, const char *fmt, va_list args) { struct trace_event_call *call = &event_print; @@ -3275,6 +3295,7 @@ __trace_array_vprintk(struct ring_buffer *buffer, local_save_flags(flags); size = sizeof(*entry) + len + 1; + ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, flags, pc); if (!event) @@ -3289,6 +3310,7 @@ __trace_array_vprintk(struct ring_buffer *buffer, } out: + ring_buffer_nest_end(buffer); put_trace_buf(); out_nobuffer: @@ -3302,7 +3324,7 @@ __printf(3, 0) int trace_array_vprintk(struct trace_array *tr, unsigned long ip, const char *fmt, va_list args) { - return __trace_array_vprintk(tr->trace_buffer.buffer, ip, fmt, args); + return __trace_array_vprintk(tr->array_buffer.buffer, ip, fmt, args); } __printf(3, 0) @@ -3326,7 +3348,7 @@ int trace_array_printk(struct trace_array *tr, EXPORT_SYMBOL_GPL(trace_array_printk); __printf(3, 4) -int trace_array_printk_buf(struct ring_buffer *buffer, +int trace_array_printk_buf(struct trace_buffer *buffer, unsigned long ip, const char *fmt, ...) { int ret; @@ -3367,7 +3389,7 @@ peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts, if (buf_iter) event = ring_buffer_iter_peek(buf_iter, ts); else - event = ring_buffer_peek(iter->trace_buffer->buffer, cpu, ts, + event = ring_buffer_peek(iter->array_buffer->buffer, cpu, ts, lost_events); if (event) { @@ -3382,7 +3404,7 @@ static struct trace_entry * __find_next_entry(struct trace_iterator *iter, int *ent_cpu, unsigned long *missing_events, u64 *ent_ts) { - struct ring_buffer *buffer = iter->trace_buffer->buffer; + struct trace_buffer *buffer = iter->array_buffer->buffer; struct trace_entry *ent, *next = NULL; unsigned long lost_events = 0, next_lost = 0; int cpu_file = iter->cpu_file; @@ -3459,7 +3481,7 @@ void *trace_find_next_entry_inc(struct trace_iterator *iter) static void trace_consume(struct trace_iterator *iter) { - ring_buffer_consume(iter->trace_buffer->buffer, iter->cpu, &iter->ts, + ring_buffer_consume(iter->array_buffer->buffer, iter->cpu, &iter->ts, &iter->lost_events); } @@ -3497,7 +3519,7 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu) unsigned long entries = 0; u64 ts; - per_cpu_ptr(iter->trace_buffer->data, cpu)->skipped_entries = 0; + per_cpu_ptr(iter->array_buffer->data, cpu)->skipped_entries = 0; buf_iter = trace_buffer_iter(iter, cpu); if (!buf_iter) @@ -3511,13 +3533,13 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu) * by the timestamp being before the start of the buffer. */ while ((event = ring_buffer_iter_peek(buf_iter, &ts))) { - if (ts >= iter->trace_buffer->time_start) + if (ts >= iter->array_buffer->time_start) break; entries++; ring_buffer_read(buf_iter, NULL); } - per_cpu_ptr(iter->trace_buffer->data, cpu)->skipped_entries = entries; + per_cpu_ptr(iter->array_buffer->data, cpu)->skipped_entries = entries; } /* @@ -3602,7 +3624,7 @@ static void s_stop(struct seq_file *m, void *p) } static void -get_total_entries_cpu(struct trace_buffer *buf, unsigned long *total, +get_total_entries_cpu(struct array_buffer *buf, unsigned long *total, unsigned long *entries, int cpu) { unsigned long count; @@ -3624,7 +3646,7 @@ get_total_entries_cpu(struct trace_buffer *buf, unsigned long *total, } static void -get_total_entries(struct trace_buffer *buf, +get_total_entries(struct array_buffer *buf, unsigned long *total, unsigned long *entries) { unsigned long t, e; @@ -3647,7 +3669,7 @@ unsigned long trace_total_entries_cpu(struct trace_array *tr, int cpu) if (!tr) tr = &global_trace; - get_total_entries_cpu(&tr->trace_buffer, &total, &entries, cpu); + get_total_entries_cpu(&tr->array_buffer, &total, &entries, cpu); return entries; } @@ -3659,7 +3681,7 @@ unsigned long trace_total_entries(struct trace_array *tr) if (!tr) tr = &global_trace; - get_total_entries(&tr->trace_buffer, &total, &entries); + get_total_entries(&tr->array_buffer, &total, &entries); return entries; } @@ -3676,7 +3698,7 @@ static void print_lat_help_header(struct seq_file *m) "# \\ / ||||| \\ | / \n"); } -static void print_event_info(struct trace_buffer *buf, struct seq_file *m) +static void print_event_info(struct array_buffer *buf, struct seq_file *m) { unsigned long total; unsigned long entries; @@ -3687,7 +3709,7 @@ static void print_event_info(struct trace_buffer *buf, struct seq_file *m) seq_puts(m, "#\n"); } -static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m, +static void print_func_help_header(struct array_buffer *buf, struct seq_file *m, unsigned int flags) { bool tgid = flags & TRACE_ITER_RECORD_TGID; @@ -3698,7 +3720,7 @@ static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m, seq_printf(m, "# | | %s | | |\n", tgid ? " | " : ""); } -static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file *m, +static void print_func_help_header_irq(struct array_buffer *buf, struct seq_file *m, unsigned int flags) { bool tgid = flags & TRACE_ITER_RECORD_TGID; @@ -3720,7 +3742,7 @@ void print_trace_header(struct seq_file *m, struct trace_iterator *iter) { unsigned long sym_flags = (global_trace.trace_flags & TRACE_ITER_SYM_MASK); - struct trace_buffer *buf = iter->trace_buffer; + struct array_buffer *buf = iter->array_buffer; struct trace_array_cpu *data = per_cpu_ptr(buf->data, buf->cpu); struct tracer *type = iter->trace; unsigned long entries; @@ -3795,7 +3817,7 @@ static void test_cpu_buff_start(struct trace_iterator *iter) cpumask_test_cpu(iter->cpu, iter->started)) return; - if (per_cpu_ptr(iter->trace_buffer->data, iter->cpu)->skipped_entries) + if (per_cpu_ptr(iter->array_buffer->data, iter->cpu)->skipped_entries) return; if (cpumask_available(iter->started)) @@ -3929,7 +3951,7 @@ int trace_empty(struct trace_iterator *iter) if (!ring_buffer_iter_empty(buf_iter)) return 0; } else { - if (!ring_buffer_empty_cpu(iter->trace_buffer->buffer, cpu)) + if (!ring_buffer_empty_cpu(iter->array_buffer->buffer, cpu)) return 0; } return 1; @@ -3941,7 +3963,7 @@ int trace_empty(struct trace_iterator *iter) if (!ring_buffer_iter_empty(buf_iter)) return 0; } else { - if (!ring_buffer_empty_cpu(iter->trace_buffer->buffer, cpu)) + if (!ring_buffer_empty_cpu(iter->array_buffer->buffer, cpu)) return 0; } } @@ -4031,10 +4053,10 @@ void trace_default_header(struct seq_file *m) } else { if (!(trace_flags & TRACE_ITER_VERBOSE)) { if (trace_flags & TRACE_ITER_IRQ_INFO) - print_func_help_header_irq(iter->trace_buffer, + print_func_help_header_irq(iter->array_buffer, m, trace_flags); else - print_func_help_header(iter->trace_buffer, m, + print_func_help_header(iter->array_buffer, m, trace_flags); } } @@ -4192,21 +4214,21 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) #ifdef CONFIG_TRACER_MAX_TRACE /* Currently only the top directory has a snapshot */ if (tr->current_trace->print_max || snapshot) - iter->trace_buffer = &tr->max_buffer; + iter->array_buffer = &tr->max_buffer; else #endif - iter->trace_buffer = &tr->trace_buffer; + iter->array_buffer = &tr->array_buffer; iter->snapshot = snapshot; iter->pos = -1; iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); /* Notify the tracer early; before we stop tracing. */ - if (iter->trace && iter->trace->open) + if (iter->trace->open) iter->trace->open(iter); /* Annotate start of buffers if we had overruns */ - if (ring_buffer_overruns(iter->trace_buffer->buffer)) + if (ring_buffer_overruns(iter->array_buffer->buffer)) iter->iter_flags |= TRACE_FILE_ANNOTATE; /* Output in nanoseconds only if we are using a clock in nanoseconds. */ @@ -4220,7 +4242,7 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) if (iter->cpu_file == RING_BUFFER_ALL_CPUS) { for_each_tracing_cpu(cpu) { iter->buffer_iter[cpu] = - ring_buffer_read_prepare(iter->trace_buffer->buffer, + ring_buffer_read_prepare(iter->array_buffer->buffer, cpu, GFP_KERNEL); } ring_buffer_read_prepare_sync(); @@ -4231,7 +4253,7 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) } else { cpu = iter->cpu_file; iter->buffer_iter[cpu] = - ring_buffer_read_prepare(iter->trace_buffer->buffer, + ring_buffer_read_prepare(iter->array_buffer->buffer, cpu, GFP_KERNEL); ring_buffer_read_prepare_sync(); ring_buffer_read_start(iter->buffer_iter[cpu]); @@ -4357,7 +4379,7 @@ static int tracing_open(struct inode *inode, struct file *file) /* If this file was open for write, then erase contents */ if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { int cpu = tracing_get_cpu(inode); - struct trace_buffer *trace_buf = &tr->trace_buffer; + struct array_buffer *trace_buf = &tr->array_buffer; #ifdef CONFIG_TRACER_MAX_TRACE if (tr->current_trace->print_max) @@ -4554,20 +4576,13 @@ out_err: return count; } -static ssize_t -tracing_cpumask_write(struct file *filp, const char __user *ubuf, - size_t count, loff_t *ppos) +int tracing_set_cpumask(struct trace_array *tr, + cpumask_var_t tracing_cpumask_new) { - struct trace_array *tr = file_inode(filp)->i_private; - cpumask_var_t tracing_cpumask_new; - int err, cpu; - - if (!alloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL)) - return -ENOMEM; + int cpu; - err = cpumask_parse_user(ubuf, count, tracing_cpumask_new); - if (err) - goto err_unlock; + if (!tr) + return -EINVAL; local_irq_disable(); arch_spin_lock(&tr->max_lock); @@ -4578,24 +4593,47 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, */ if (cpumask_test_cpu(cpu, tr->tracing_cpumask) && !cpumask_test_cpu(cpu, tracing_cpumask_new)) { - atomic_inc(&per_cpu_ptr(tr->trace_buffer.data, cpu)->disabled); - ring_buffer_record_disable_cpu(tr->trace_buffer.buffer, cpu); + atomic_inc(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); + ring_buffer_record_disable_cpu(tr->array_buffer.buffer, cpu); } if (!cpumask_test_cpu(cpu, tr->tracing_cpumask) && cpumask_test_cpu(cpu, tracing_cpumask_new)) { - atomic_dec(&per_cpu_ptr(tr->trace_buffer.data, cpu)->disabled); - ring_buffer_record_enable_cpu(tr->trace_buffer.buffer, cpu); + atomic_dec(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); + ring_buffer_record_enable_cpu(tr->array_buffer.buffer, cpu); } } arch_spin_unlock(&tr->max_lock); local_irq_enable(); cpumask_copy(tr->tracing_cpumask, tracing_cpumask_new); + + return 0; +} + +static ssize_t +tracing_cpumask_write(struct file *filp, const char __user *ubuf, + size_t count, loff_t *ppos) +{ + struct trace_array *tr = file_inode(filp)->i_private; + cpumask_var_t tracing_cpumask_new; + int err; + + if (!alloc_cpumask_var(&tracing_cpumask_new, GFP_KERNEL)) + return -ENOMEM; + + err = cpumask_parse_user(ubuf, count, tracing_cpumask_new); + if (err) + goto err_free; + + err = tracing_set_cpumask(tr, tracing_cpumask_new); + if (err) + goto err_free; + free_cpumask_var(tracing_cpumask_new); return count; -err_unlock: +err_free: free_cpumask_var(tracing_cpumask_new); return err; @@ -4726,7 +4764,7 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) ftrace_pid_follow_fork(tr, enabled); if (mask == TRACE_ITER_OVERWRITE) { - ring_buffer_change_overwrite(tr->trace_buffer.buffer, enabled); + ring_buffer_change_overwrite(tr->array_buffer.buffer, enabled); #ifdef CONFIG_TRACER_MAX_TRACE ring_buffer_change_overwrite(tr->max_buffer.buffer, enabled); #endif @@ -4740,7 +4778,7 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) return 0; } -static int trace_set_options(struct trace_array *tr, char *option) +int trace_set_options(struct trace_array *tr, char *option) { char *cmp; int neg = 0; @@ -5361,14 +5399,12 @@ static void *eval_map_next(struct seq_file *m, void *v, loff_t *pos) * Paranoid! If ptr points to end, we don't want to increment past it. * This really should never happen. */ + (*pos)++; ptr = update_eval_map(ptr); if (WARN_ON_ONCE(!ptr)) return NULL; ptr++; - - (*pos)++; - ptr = update_eval_map(ptr); return ptr; @@ -5534,11 +5570,11 @@ tracing_set_trace_read(struct file *filp, char __user *ubuf, int tracer_init(struct tracer *t, struct trace_array *tr) { - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); return t->init(tr); } -static void set_buffer_entries(struct trace_buffer *buf, unsigned long val) +static void set_buffer_entries(struct array_buffer *buf, unsigned long val) { int cpu; @@ -5548,8 +5584,8 @@ static void set_buffer_entries(struct trace_buffer *buf, unsigned long val) #ifdef CONFIG_TRACER_MAX_TRACE /* resize @tr's buffer to the size of @size_tr's entries */ -static int resize_buffer_duplicate_size(struct trace_buffer *trace_buf, - struct trace_buffer *size_buf, int cpu_id) +static int resize_buffer_duplicate_size(struct array_buffer *trace_buf, + struct array_buffer *size_buf, int cpu_id) { int cpu, ret = 0; @@ -5587,10 +5623,10 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, ring_buffer_expanded = true; /* May be called before buffers are initialized */ - if (!tr->trace_buffer.buffer) + if (!tr->array_buffer.buffer) return 0; - ret = ring_buffer_resize(tr->trace_buffer.buffer, size, cpu); + ret = ring_buffer_resize(tr->array_buffer.buffer, size, cpu); if (ret < 0) return ret; @@ -5601,8 +5637,8 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, ret = ring_buffer_resize(tr->max_buffer.buffer, size, cpu); if (ret < 0) { - int r = resize_buffer_duplicate_size(&tr->trace_buffer, - &tr->trace_buffer, cpu); + int r = resize_buffer_duplicate_size(&tr->array_buffer, + &tr->array_buffer, cpu); if (r < 0) { /* * AARGH! We are left with different @@ -5633,15 +5669,15 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, #endif /* CONFIG_TRACER_MAX_TRACE */ if (cpu == RING_BUFFER_ALL_CPUS) - set_buffer_entries(&tr->trace_buffer, size); + set_buffer_entries(&tr->array_buffer, size); else - per_cpu_ptr(tr->trace_buffer.data, cpu)->entries = size; + per_cpu_ptr(tr->array_buffer.data, cpu)->entries = size; return ret; } -static ssize_t tracing_resize_ring_buffer(struct trace_array *tr, - unsigned long size, int cpu_id) +ssize_t tracing_resize_ring_buffer(struct trace_array *tr, + unsigned long size, int cpu_id) { int ret = size; @@ -5720,7 +5756,7 @@ static void add_tracer_options(struct trace_array *tr, struct tracer *t) create_trace_option_files(tr, t); } -static int tracing_set_tracer(struct trace_array *tr, const char *buf) +int tracing_set_tracer(struct trace_array *tr, const char *buf) { struct tracer *t; #ifdef CONFIG_TRACER_MAX_TRACE @@ -5979,7 +6015,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) iter->iter_flags |= TRACE_FILE_TIME_IN_NS; iter->tr = tr; - iter->trace_buffer = &tr->trace_buffer; + iter->array_buffer = &tr->array_buffer; iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); filp->private_data = iter; @@ -6039,7 +6075,7 @@ trace_poll(struct trace_iterator *iter, struct file *filp, poll_table *poll_tabl */ return EPOLLIN | EPOLLRDNORM; else - return ring_buffer_poll_wait(iter->trace_buffer->buffer, iter->cpu_file, + return ring_buffer_poll_wait(iter->array_buffer->buffer, iter->cpu_file, filp, poll_table); } @@ -6356,8 +6392,8 @@ tracing_entries_read(struct file *filp, char __user *ubuf, for_each_tracing_cpu(cpu) { /* fill in the size from first enabled cpu */ if (size == 0) - size = per_cpu_ptr(tr->trace_buffer.data, cpu)->entries; - if (size != per_cpu_ptr(tr->trace_buffer.data, cpu)->entries) { + size = per_cpu_ptr(tr->array_buffer.data, cpu)->entries; + if (size != per_cpu_ptr(tr->array_buffer.data, cpu)->entries) { buf_size_same = 0; break; } @@ -6373,7 +6409,7 @@ tracing_entries_read(struct file *filp, char __user *ubuf, } else r = sprintf(buf, "X\n"); } else - r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->trace_buffer.data, cpu)->entries >> 10); + r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->array_buffer.data, cpu)->entries >> 10); mutex_unlock(&trace_types_lock); @@ -6420,7 +6456,7 @@ tracing_total_entries_read(struct file *filp, char __user *ubuf, mutex_lock(&trace_types_lock); for_each_tracing_cpu(cpu) { - size += per_cpu_ptr(tr->trace_buffer.data, cpu)->entries >> 10; + size += per_cpu_ptr(tr->array_buffer.data, cpu)->entries >> 10; if (!ring_buffer_expanded) expanded_size += trace_buf_size >> 10; } @@ -6470,7 +6506,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, struct trace_array *tr = filp->private_data; struct ring_buffer_event *event; enum event_trigger_type tt = ETT_NONE; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct print_entry *entry; unsigned long irq_flags; ssize_t written; @@ -6499,7 +6535,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, if (cnt < FAULTED_SIZE) size += FAULTED_SIZE - cnt; - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, irq_flags, preempt_count()); if (unlikely(!event)) @@ -6550,7 +6586,7 @@ tracing_mark_raw_write(struct file *filp, const char __user *ubuf, { struct trace_array *tr = filp->private_data; struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct raw_data_entry *entry; unsigned long irq_flags; ssize_t written; @@ -6579,7 +6615,7 @@ tracing_mark_raw_write(struct file *filp, const char __user *ubuf, if (cnt < FAULT_SIZE_ID) size += FAULT_SIZE_ID - cnt; - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_RAW_DATA, size, irq_flags, preempt_count()); if (!event) @@ -6634,13 +6670,13 @@ int tracing_set_clock(struct trace_array *tr, const char *clockstr) tr->clock_id = i; - ring_buffer_set_clock(tr->trace_buffer.buffer, trace_clocks[i].func); + ring_buffer_set_clock(tr->array_buffer.buffer, trace_clocks[i].func); /* * New clock may not be consistent with the previous clock. * Reset the buffer so that it doesn't have incomparable timestamps. */ - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); #ifdef CONFIG_TRACER_MAX_TRACE if (tr->max_buffer.buffer) @@ -6703,7 +6739,7 @@ static int tracing_time_stamp_mode_show(struct seq_file *m, void *v) mutex_lock(&trace_types_lock); - if (ring_buffer_time_stamp_abs(tr->trace_buffer.buffer)) + if (ring_buffer_time_stamp_abs(tr->array_buffer.buffer)) seq_puts(m, "delta [absolute]\n"); else seq_puts(m, "[delta] absolute\n"); @@ -6748,7 +6784,7 @@ int tracing_set_time_stamp_abs(struct trace_array *tr, bool abs) goto out; } - ring_buffer_set_time_stamp_abs(tr->trace_buffer.buffer, abs); + ring_buffer_set_time_stamp_abs(tr->array_buffer.buffer, abs); #ifdef CONFIG_TRACER_MAX_TRACE if (tr->max_buffer.buffer) @@ -6797,7 +6833,7 @@ static int tracing_snapshot_open(struct inode *inode, struct file *file) ret = 0; iter->tr = tr; - iter->trace_buffer = &tr->max_buffer; + iter->array_buffer = &tr->max_buffer; iter->cpu_file = tracing_get_cpu(inode); m->private = iter; file->private_data = m; @@ -6860,7 +6896,7 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, #endif if (tr->allocated_snapshot) ret = resize_buffer_duplicate_size(&tr->max_buffer, - &tr->trace_buffer, iter->cpu_file); + &tr->array_buffer, iter->cpu_file); else ret = tracing_alloc_snapshot_instance(tr); if (ret < 0) @@ -6935,7 +6971,7 @@ static int snapshot_raw_open(struct inode *inode, struct file *filp) } info->iter.snapshot = true; - info->iter.trace_buffer = &info->iter.tr->max_buffer; + info->iter.array_buffer = &info->iter.tr->max_buffer; return ret; } @@ -7310,7 +7346,7 @@ static int tracing_buffers_open(struct inode *inode, struct file *filp) info->iter.tr = tr; info->iter.cpu_file = tracing_get_cpu(inode); info->iter.trace = tr->current_trace; - info->iter.trace_buffer = &tr->trace_buffer; + info->iter.array_buffer = &tr->array_buffer; info->spare = NULL; /* Force reading ring buffer for first read */ info->read = (unsigned int)-1; @@ -7355,7 +7391,7 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, #endif if (!info->spare) { - info->spare = ring_buffer_alloc_read_page(iter->trace_buffer->buffer, + info->spare = ring_buffer_alloc_read_page(iter->array_buffer->buffer, iter->cpu_file); if (IS_ERR(info->spare)) { ret = PTR_ERR(info->spare); @@ -7373,7 +7409,7 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, again: trace_access_lock(iter->cpu_file); - ret = ring_buffer_read_page(iter->trace_buffer->buffer, + ret = ring_buffer_read_page(iter->array_buffer->buffer, &info->spare, count, iter->cpu_file, 0); @@ -7423,7 +7459,7 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) __trace_array_put(iter->tr); if (info->spare) - ring_buffer_free_read_page(iter->trace_buffer->buffer, + ring_buffer_free_read_page(iter->array_buffer->buffer, info->spare_cpu, info->spare); kfree(info); @@ -7433,7 +7469,7 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) } struct buffer_ref { - struct ring_buffer *buffer; + struct trace_buffer *buffer; void *page; int cpu; refcount_t refcount; @@ -7528,7 +7564,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, again: trace_access_lock(iter->cpu_file); - entries = ring_buffer_entries_cpu(iter->trace_buffer->buffer, iter->cpu_file); + entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file); for (i = 0; i < spd.nr_pages_max && len && entries; i++, len -= PAGE_SIZE) { struct page *page; @@ -7541,7 +7577,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, } refcount_set(&ref->refcount, 1); - ref->buffer = iter->trace_buffer->buffer; + ref->buffer = iter->array_buffer->buffer; ref->page = ring_buffer_alloc_read_page(ref->buffer, iter->cpu_file); if (IS_ERR(ref->page)) { ret = PTR_ERR(ref->page); @@ -7569,7 +7605,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, spd.nr_pages++; *ppos += PAGE_SIZE; - entries = ring_buffer_entries_cpu(iter->trace_buffer->buffer, iter->cpu_file); + entries = ring_buffer_entries_cpu(iter->array_buffer->buffer, iter->cpu_file); } trace_access_unlock(iter->cpu_file); @@ -7613,7 +7649,7 @@ tracing_stats_read(struct file *filp, char __user *ubuf, { struct inode *inode = file_inode(filp); struct trace_array *tr = inode->i_private; - struct trace_buffer *trace_buf = &tr->trace_buffer; + struct array_buffer *trace_buf = &tr->array_buffer; int cpu = tracing_get_cpu(inode); struct trace_seq *s; unsigned long cnt; @@ -7894,7 +7930,7 @@ static struct dentry *tracing_dentry_percpu(struct trace_array *tr, int cpu) tr->percpu_dir = tracefs_create_dir("per_cpu", d_tracer); - WARN_ONCE(!tr->percpu_dir, + MEM_FAIL(!tr->percpu_dir, "Could not create tracefs directory 'per_cpu/%d'\n", cpu); return tr->percpu_dir; @@ -8215,7 +8251,7 @@ create_trace_option_files(struct trace_array *tr, struct tracer *tracer) for (cnt = 0; opts[cnt].name; cnt++) { create_trace_option_file(tr, &topts[cnt], flags, &opts[cnt]); - WARN_ONCE(topts[cnt].entry == NULL, + MEM_FAIL(topts[cnt].entry == NULL, "Failed to create trace option: %s", opts[cnt].name); } @@ -8272,7 +8308,7 @@ rb_simple_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { struct trace_array *tr = filp->private_data; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; unsigned long val; int ret; @@ -8362,7 +8398,7 @@ static void init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer); static int -allocate_trace_buffer(struct trace_array *tr, struct trace_buffer *buf, int size) +allocate_trace_buffer(struct trace_array *tr, struct array_buffer *buf, int size) { enum ring_buffer_flags rb_flags; @@ -8382,8 +8418,8 @@ allocate_trace_buffer(struct trace_array *tr, struct trace_buffer *buf, int size } /* Allocate the first page for all buffers */ - set_buffer_entries(&tr->trace_buffer, - ring_buffer_size(tr->trace_buffer.buffer, 0)); + set_buffer_entries(&tr->array_buffer, + ring_buffer_size(tr->array_buffer.buffer, 0)); return 0; } @@ -8392,18 +8428,18 @@ static int allocate_trace_buffers(struct trace_array *tr, int size) { int ret; - ret = allocate_trace_buffer(tr, &tr->trace_buffer, size); + ret = allocate_trace_buffer(tr, &tr->array_buffer, size); if (ret) return ret; #ifdef CONFIG_TRACER_MAX_TRACE ret = allocate_trace_buffer(tr, &tr->max_buffer, allocate_snapshot ? size : 1); - if (WARN_ON(ret)) { - ring_buffer_free(tr->trace_buffer.buffer); - tr->trace_buffer.buffer = NULL; - free_percpu(tr->trace_buffer.data); - tr->trace_buffer.data = NULL; + if (MEM_FAIL(ret, "Failed to allocate trace buffer\n")) { + ring_buffer_free(tr->array_buffer.buffer); + tr->array_buffer.buffer = NULL; + free_percpu(tr->array_buffer.data); + tr->array_buffer.data = NULL; return -ENOMEM; } tr->allocated_snapshot = allocate_snapshot; @@ -8417,7 +8453,7 @@ static int allocate_trace_buffers(struct trace_array *tr, int size) return 0; } -static void free_trace_buffer(struct trace_buffer *buf) +static void free_trace_buffer(struct array_buffer *buf) { if (buf->buffer) { ring_buffer_free(buf->buffer); @@ -8432,7 +8468,7 @@ static void free_trace_buffers(struct trace_array *tr) if (!tr) return; - free_trace_buffer(&tr->trace_buffer); + free_trace_buffer(&tr->array_buffer); #ifdef CONFIG_TRACER_MAX_TRACE free_trace_buffer(&tr->max_buffer); @@ -8463,6 +8499,34 @@ static void update_tracer_options(struct trace_array *tr) mutex_unlock(&trace_types_lock); } +/* Must have trace_types_lock held */ +struct trace_array *trace_array_find(const char *instance) +{ + struct trace_array *tr, *found = NULL; + + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (tr->name && strcmp(tr->name, instance) == 0) { + found = tr; + break; + } + } + + return found; +} + +struct trace_array *trace_array_find_get(const char *instance) +{ + struct trace_array *tr; + + mutex_lock(&trace_types_lock); + tr = trace_array_find(instance); + if (tr) + tr->ref++; + mutex_unlock(&trace_types_lock); + + return tr; +} + static struct trace_array *trace_array_create(const char *name) { struct trace_array *tr; @@ -8539,10 +8603,8 @@ static int instance_mkdir(const char *name) mutex_lock(&trace_types_lock); ret = -EEXIST; - list_for_each_entry(tr, &ftrace_trace_arrays, list) { - if (tr->name && strcmp(tr->name, name) == 0) - goto out_unlock; - } + if (trace_array_find(name)) + goto out_unlock; tr = trace_array_create(name); @@ -8564,6 +8626,10 @@ out_unlock: * NOTE: This function increments the reference counter associated with the * trace array returned. This makes sure it cannot be freed while in use. * Use trace_array_put() once the trace array is no longer needed. + * If the trace_array is to be freed, trace_array_destroy() needs to + * be called after the trace_array_put(), or simply let user space delete + * it from the tracefs instances directory. But until the + * trace_array_put() is called, user space can not delete it. * */ struct trace_array *trace_array_get_by_name(const char *name) @@ -8666,12 +8732,9 @@ static int instance_rmdir(const char *name) mutex_lock(&trace_types_lock); ret = -ENODEV; - list_for_each_entry(tr, &ftrace_trace_arrays, list) { - if (tr->name && strcmp(tr->name, name) == 0) { - ret = __remove_instance(tr); - break; - } - } + tr = trace_array_find(name); + if (tr) + ret = __remove_instance(tr); mutex_unlock(&trace_types_lock); mutex_unlock(&event_mutex); @@ -8684,7 +8747,7 @@ static __init void create_trace_instances(struct dentry *d_tracer) trace_instance_dir = tracefs_create_instance_dir("instances", d_tracer, instance_mkdir, instance_rmdir); - if (WARN_ON(!trace_instance_dir)) + if (MEM_FAIL(!trace_instance_dir, "Failed to create instances directory\n")) return; } @@ -8754,7 +8817,7 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) #endif if (ftrace_create_function_files(tr, d_tracer)) - WARN(1, "Could not allocate function filter files"); + MEM_FAIL(1, "Could not allocate function filter files"); #ifdef CONFIG_TRACER_SNAPSHOT trace_create_file("snapshot", 0644, d_tracer, @@ -9036,13 +9099,13 @@ void trace_init_global_iter(struct trace_iterator *iter) iter->tr = &global_trace; iter->trace = iter->tr->current_trace; iter->cpu_file = RING_BUFFER_ALL_CPUS; - iter->trace_buffer = &global_trace.trace_buffer; + iter->array_buffer = &global_trace.array_buffer; if (iter->trace && iter->trace->open) iter->trace->open(iter); /* Annotate start of buffers if we had overruns */ - if (ring_buffer_overruns(iter->trace_buffer->buffer)) + if (ring_buffer_overruns(iter->array_buffer->buffer)) iter->iter_flags |= TRACE_FILE_ANNOTATE; /* Output in nanoseconds only if we are using a clock in nanoseconds. */ @@ -9083,7 +9146,7 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) trace_init_global_iter(&iter); for_each_tracing_cpu(cpu) { - atomic_inc(&per_cpu_ptr(iter.trace_buffer->data, cpu)->disabled); + atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } old_userobj = tr->trace_flags & TRACE_ITER_SYM_USEROBJ; @@ -9151,7 +9214,7 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) tr->trace_flags |= old_userobj; for_each_tracing_cpu(cpu) { - atomic_dec(&per_cpu_ptr(iter.trace_buffer->data, cpu)->disabled); + atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } atomic_dec(&dump_running); printk_nmi_direct_exit(); @@ -9306,8 +9369,7 @@ __init static int tracer_alloc_buffers(void) /* TODO: make the number of buffers hot pluggable with CPUS */ if (allocate_trace_buffers(&global_trace, ring_buf_size) < 0) { - printk(KERN_ERR "tracer: failed to allocate ring buffer!\n"); - WARN_ON(1); + MEM_FAIL(1, "tracer: failed to allocate ring buffer!\n"); goto out_free_savedcmd; } @@ -9380,7 +9442,8 @@ void __init early_trace_init(void) if (tracepoint_printk) { tracepoint_print_iter = kmalloc(sizeof(*tracepoint_print_iter), GFP_KERNEL); - if (WARN_ON(!tracepoint_print_iter)) + if (MEM_FAIL(!tracepoint_print_iter, + "Failed to allocate trace iterator\n")) tracepoint_printk = 0; else static_key_enable(&tracepoint_printk_key.key); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index a98dce1b3334..99372dd7d168 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -93,6 +93,18 @@ enum trace_type { #include "trace_entries.h" +/* Use this for memory failure errors */ +#define MEM_FAIL(condition, fmt, ...) ({ \ + static bool __section(.data.once) __warned; \ + int __ret_warn_once = !!(condition); \ + \ + if (unlikely(__ret_warn_once && !__warned)) { \ + __warned = true; \ + pr_err("ERROR: " fmt, ##__VA_ARGS__); \ + } \ + unlikely(__ret_warn_once); \ +}) + /* * syscalls are special, and need special handling, this is why * they are not included in trace_entries.h @@ -175,9 +187,9 @@ struct trace_array_cpu { struct tracer; struct trace_option_dentry; -struct trace_buffer { +struct array_buffer { struct trace_array *tr; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct trace_array_cpu __percpu *data; u64 time_start; int cpu; @@ -248,7 +260,7 @@ struct cond_snapshot { struct trace_array { struct list_head list; char *name; - struct trace_buffer trace_buffer; + struct array_buffer array_buffer; #ifdef CONFIG_TRACER_MAX_TRACE /* * The max_buffer is used to snapshot the trace when a maximum @@ -256,12 +268,12 @@ struct trace_array { * Some tracers will use this to store a maximum trace while * it continues examining live traces. * - * The buffers for the max_buffer are set up the same as the trace_buffer + * The buffers for the max_buffer are set up the same as the array_buffer * When a snapshot is taken, the buffer of the max_buffer is swapped - * with the buffer of the trace_buffer and the buffers are reset for - * the trace_buffer so the tracing can continue. + * with the buffer of the array_buffer and the buffers are reset for + * the array_buffer so the tracing can continue. */ - struct trace_buffer max_buffer; + struct array_buffer max_buffer; bool allocated_snapshot; #endif #if defined(CONFIG_TRACER_MAX_TRACE) || defined(CONFIG_HWLAT_TRACER) @@ -345,6 +357,8 @@ extern struct mutex trace_types_lock; extern int trace_array_get(struct trace_array *tr); extern int tracing_check_open_get_tr(struct trace_array *tr); +extern struct trace_array *trace_array_find(const char *instance); +extern struct trace_array *trace_array_find_get(const char *instance); extern int tracing_set_time_stamp_abs(struct trace_array *tr, bool abs); extern int tracing_set_clock(struct trace_array *tr, const char *clockstr); @@ -684,7 +698,7 @@ trace_buffer_iter(struct trace_iterator *iter, int cpu) int tracer_init(struct tracer *t, struct trace_array *tr); int tracing_is_enabled(void); -void tracing_reset_online_cpus(struct trace_buffer *buf); +void tracing_reset_online_cpus(struct array_buffer *buf); void tracing_reset_current(int cpu); void tracing_reset_all_online_cpus(void); int tracing_open_generic(struct inode *inode, struct file *filp); @@ -704,7 +718,7 @@ struct dentry *tracing_init_dentry(void); struct ring_buffer_event; struct ring_buffer_event * -trace_buffer_lock_reserve(struct ring_buffer *buffer, +trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, unsigned long flags, @@ -716,7 +730,7 @@ struct trace_entry *tracing_get_trace_entry(struct trace_array *tr, struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, int *ent_cpu, u64 *ent_ts); -void trace_buffer_unlock_commit_nostack(struct ring_buffer *buffer, +void trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, struct ring_buffer_event *event); int trace_empty(struct trace_iterator *iter); @@ -872,7 +886,7 @@ trace_vprintk(unsigned long ip, const char *fmt, va_list args); extern int trace_array_vprintk(struct trace_array *tr, unsigned long ip, const char *fmt, va_list args); -int trace_array_printk_buf(struct ring_buffer *buffer, +int trace_array_printk_buf(struct trace_buffer *buffer, unsigned long ip, const char *fmt, ...); void trace_printk_seq(struct trace_seq *s); enum print_line_t print_trace_line(struct trace_iterator *iter); @@ -949,22 +963,31 @@ extern void __trace_graph_return(struct trace_array *tr, unsigned long flags, int pc); #ifdef CONFIG_DYNAMIC_FTRACE -extern struct ftrace_hash *ftrace_graph_hash; -extern struct ftrace_hash *ftrace_graph_notrace_hash; +extern struct ftrace_hash __rcu *ftrace_graph_hash; +extern struct ftrace_hash __rcu *ftrace_graph_notrace_hash; static inline int ftrace_graph_addr(struct ftrace_graph_ent *trace) { unsigned long addr = trace->func; int ret = 0; + struct ftrace_hash *hash; preempt_disable_notrace(); - if (ftrace_hash_empty(ftrace_graph_hash)) { + /* + * Have to open code "rcu_dereference_sched()" because the + * function graph tracer can be called when RCU is not + * "watching". + * Protected with schedule_on_each_cpu(ftrace_sync) + */ + hash = rcu_dereference_protected(ftrace_graph_hash, !preemptible()); + + if (ftrace_hash_empty(hash)) { ret = 1; goto out; } - if (ftrace_lookup_ip(ftrace_graph_hash, addr)) { + if (ftrace_lookup_ip(hash, addr)) { /* * This needs to be cleared on the return functions @@ -1000,10 +1023,20 @@ static inline void ftrace_graph_addr_finish(struct ftrace_graph_ret *trace) static inline int ftrace_graph_notrace_addr(unsigned long addr) { int ret = 0; + struct ftrace_hash *notrace_hash; preempt_disable_notrace(); - if (ftrace_lookup_ip(ftrace_graph_notrace_hash, addr)) + /* + * Have to open code "rcu_dereference_sched()" because the + * function graph tracer can be called when RCU is not + * "watching". + * Protected with schedule_on_each_cpu(ftrace_sync) + */ + notrace_hash = rcu_dereference_protected(ftrace_graph_notrace_hash, + !preemptible()); + + if (ftrace_lookup_ip(notrace_hash, addr)) ret = 1; preempt_enable_notrace(); @@ -1056,7 +1089,7 @@ struct ftrace_func_command { extern bool ftrace_filter_param __initdata; static inline int ftrace_trace_task(struct trace_array *tr) { - return !this_cpu_read(tr->trace_buffer.data->ftrace_ignore_pid); + return !this_cpu_read(tr->array_buffer.data->ftrace_ignore_pid); } extern int ftrace_is_dead(void); int ftrace_create_function_files(struct trace_array *tr, @@ -1144,6 +1177,11 @@ int unregister_ftrace_command(struct ftrace_func_command *cmd); void ftrace_create_filter_files(struct ftrace_ops *ops, struct dentry *parent); void ftrace_destroy_filter_files(struct ftrace_ops *ops); + +extern int ftrace_set_filter(struct ftrace_ops *ops, unsigned char *buf, + int len, int reset); +extern int ftrace_set_notrace(struct ftrace_ops *ops, unsigned char *buf, + int len, int reset); #else struct ftrace_func_command; @@ -1366,17 +1404,17 @@ struct trace_subsystem_dir { }; extern int call_filter_check_discard(struct trace_event_call *call, void *rec, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event); void trace_buffer_unlock_commit_regs(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, unsigned long flags, int pc, struct pt_regs *regs); static inline void trace_buffer_unlock_commit(struct trace_array *tr, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, unsigned long flags, int pc) { @@ -1389,7 +1427,7 @@ void trace_buffered_event_disable(void); void trace_buffered_event_enable(void); static inline void -__trace_event_discard_commit(struct ring_buffer *buffer, +__trace_event_discard_commit(struct trace_buffer *buffer, struct ring_buffer_event *event) { if (this_cpu_read(trace_buffered_event) == event) { @@ -1415,7 +1453,7 @@ __trace_event_discard_commit(struct ring_buffer *buffer, */ static inline bool __event_trigger_test_discard(struct trace_event_file *file, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, void *entry, enum event_trigger_type *tt) @@ -1450,7 +1488,7 @@ __event_trigger_test_discard(struct trace_event_file *file, */ static inline void event_trigger_unlock_commit(struct trace_event_file *file, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, void *entry, unsigned long irq_flags, int pc) { @@ -1481,7 +1519,7 @@ event_trigger_unlock_commit(struct trace_event_file *file, */ static inline void event_trigger_unlock_commit_regs(struct trace_event_file *file, - struct ring_buffer *buffer, + struct trace_buffer *buffer, struct ring_buffer_event *event, void *entry, unsigned long irq_flags, int pc, struct pt_regs *regs) @@ -1892,6 +1930,15 @@ void trace_printk_start_comm(void); int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set); int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled); +/* Used from boot time tracer */ +extern int trace_set_options(struct trace_array *tr, char *option); +extern int tracing_set_tracer(struct trace_array *tr, const char *buf); +extern ssize_t tracing_resize_ring_buffer(struct trace_array *tr, + unsigned long size, int cpu_id); +extern int tracing_set_cpumask(struct trace_array *tr, + cpumask_var_t tracing_cpumask_new); + + #define MAX_EVENT_NAME_LEN 64 extern int trace_run_command(const char *buf, int (*createfn)(int, char**)); @@ -1949,6 +1996,9 @@ static inline const char *get_syscall_name(int syscall) #ifdef CONFIG_EVENT_TRACING void trace_event_init(void); void trace_event_eval_update(struct trace_eval_map **map, int len); +/* Used from boot time tracer */ +extern int ftrace_set_clr_event(struct trace_array *tr, char *buf, int set); +extern int trigger_process_regex(struct trace_event_file *file, char *buff); #else static inline void __init trace_event_init(void) { } static inline void trace_event_eval_update(struct trace_eval_map **map, int len) { } diff --git a/kernel/trace/trace_boot.c b/kernel/trace/trace_boot.c new file mode 100644 index 000000000000..06d7feb5255f --- /dev/null +++ b/kernel/trace/trace_boot.c @@ -0,0 +1,334 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * trace_boot.c + * Tracing kernel boot-time + */ + +#define pr_fmt(fmt) "trace_boot: " fmt + +#include <linux/bootconfig.h> +#include <linux/cpumask.h> +#include <linux/ftrace.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/mutex.h> +#include <linux/string.h> +#include <linux/slab.h> +#include <linux/trace.h> +#include <linux/trace_events.h> + +#include "trace.h" + +#define MAX_BUF_LEN 256 + +static void __init +trace_boot_set_instance_options(struct trace_array *tr, struct xbc_node *node) +{ + struct xbc_node *anode; + const char *p; + char buf[MAX_BUF_LEN]; + unsigned long v = 0; + + /* Common ftrace options */ + xbc_node_for_each_array_value(node, "options", anode, p) { + if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) { + pr_err("String is too long: %s\n", p); + continue; + } + + if (trace_set_options(tr, buf) < 0) + pr_err("Failed to set option: %s\n", buf); + } + + p = xbc_node_find_value(node, "trace_clock", NULL); + if (p && *p != '\0') { + if (tracing_set_clock(tr, p) < 0) + pr_err("Failed to set trace clock: %s\n", p); + } + + p = xbc_node_find_value(node, "buffer_size", NULL); + if (p && *p != '\0') { + v = memparse(p, NULL); + if (v < PAGE_SIZE) + pr_err("Buffer size is too small: %s\n", p); + if (tracing_resize_ring_buffer(tr, v, RING_BUFFER_ALL_CPUS) < 0) + pr_err("Failed to resize trace buffer to %s\n", p); + } + + p = xbc_node_find_value(node, "cpumask", NULL); + if (p && *p != '\0') { + cpumask_var_t new_mask; + + if (alloc_cpumask_var(&new_mask, GFP_KERNEL)) { + if (cpumask_parse(p, new_mask) < 0 || + tracing_set_cpumask(tr, new_mask) < 0) + pr_err("Failed to set new CPU mask %s\n", p); + free_cpumask_var(new_mask); + } + } +} + +#ifdef CONFIG_EVENT_TRACING +static void __init +trace_boot_enable_events(struct trace_array *tr, struct xbc_node *node) +{ + struct xbc_node *anode; + char buf[MAX_BUF_LEN]; + const char *p; + + xbc_node_for_each_array_value(node, "events", anode, p) { + if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) { + pr_err("String is too long: %s\n", p); + continue; + } + + if (ftrace_set_clr_event(tr, buf, 1) < 0) + pr_err("Failed to enable event: %s\n", p); + } +} + +#ifdef CONFIG_KPROBE_EVENTS +static int __init +trace_boot_add_kprobe_event(struct xbc_node *node, const char *event) +{ + struct dynevent_cmd cmd; + struct xbc_node *anode; + char buf[MAX_BUF_LEN]; + const char *val; + int ret; + + kprobe_event_cmd_init(&cmd, buf, MAX_BUF_LEN); + + ret = kprobe_event_gen_cmd_start(&cmd, event, NULL); + if (ret) + return ret; + + xbc_node_for_each_array_value(node, "probes", anode, val) { + ret = kprobe_event_add_field(&cmd, val); + if (ret) + return ret; + } + + ret = kprobe_event_gen_cmd_end(&cmd); + if (ret) + pr_err("Failed to add probe: %s\n", buf); + + return ret; +} +#else +static inline int __init +trace_boot_add_kprobe_event(struct xbc_node *node, const char *event) +{ + pr_err("Kprobe event is not supported.\n"); + return -ENOTSUPP; +} +#endif + +#ifdef CONFIG_HIST_TRIGGERS +static int __init +trace_boot_add_synth_event(struct xbc_node *node, const char *event) +{ + struct dynevent_cmd cmd; + struct xbc_node *anode; + char buf[MAX_BUF_LEN]; + const char *p; + int ret; + + synth_event_cmd_init(&cmd, buf, MAX_BUF_LEN); + + ret = synth_event_gen_cmd_start(&cmd, event, NULL); + if (ret) + return ret; + + xbc_node_for_each_array_value(node, "fields", anode, p) { + ret = synth_event_add_field_str(&cmd, p); + if (ret) + return ret; + } + + ret = synth_event_gen_cmd_end(&cmd); + if (ret < 0) + pr_err("Failed to add synthetic event: %s\n", buf); + + return ret; +} +#else +static inline int __init +trace_boot_add_synth_event(struct xbc_node *node, const char *event) +{ + pr_err("Synthetic event is not supported.\n"); + return -ENOTSUPP; +} +#endif + +static void __init +trace_boot_init_one_event(struct trace_array *tr, struct xbc_node *gnode, + struct xbc_node *enode) +{ + struct trace_event_file *file; + struct xbc_node *anode; + char buf[MAX_BUF_LEN]; + const char *p, *group, *event; + + group = xbc_node_get_data(gnode); + event = xbc_node_get_data(enode); + + if (!strcmp(group, "kprobes")) + if (trace_boot_add_kprobe_event(enode, event) < 0) + return; + if (!strcmp(group, "synthetic")) + if (trace_boot_add_synth_event(enode, event) < 0) + return; + + mutex_lock(&event_mutex); + file = find_event_file(tr, group, event); + if (!file) { + pr_err("Failed to find event: %s:%s\n", group, event); + goto out; + } + + p = xbc_node_find_value(enode, "filter", NULL); + if (p && *p != '\0') { + if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) + pr_err("filter string is too long: %s\n", p); + else if (apply_event_filter(file, buf) < 0) + pr_err("Failed to apply filter: %s\n", buf); + } + + xbc_node_for_each_array_value(enode, "actions", anode, p) { + if (strlcpy(buf, p, ARRAY_SIZE(buf)) >= ARRAY_SIZE(buf)) + pr_err("action string is too long: %s\n", p); + else if (trigger_process_regex(file, buf) < 0) + pr_err("Failed to apply an action: %s\n", buf); + } + + if (xbc_node_find_value(enode, "enable", NULL)) { + if (trace_event_enable_disable(file, 1, 0) < 0) + pr_err("Failed to enable event node: %s:%s\n", + group, event); + } +out: + mutex_unlock(&event_mutex); +} + +static void __init +trace_boot_init_events(struct trace_array *tr, struct xbc_node *node) +{ + struct xbc_node *gnode, *enode; + + node = xbc_node_find_child(node, "event"); + if (!node) + return; + /* per-event key starts with "event.GROUP.EVENT" */ + xbc_node_for_each_child(node, gnode) + xbc_node_for_each_child(gnode, enode) + trace_boot_init_one_event(tr, gnode, enode); +} +#else +#define trace_boot_enable_events(tr, node) do {} while (0) +#define trace_boot_init_events(tr, node) do {} while (0) +#endif + +#ifdef CONFIG_DYNAMIC_FTRACE +static void __init +trace_boot_set_ftrace_filter(struct trace_array *tr, struct xbc_node *node) +{ + struct xbc_node *anode; + const char *p; + char *q; + + xbc_node_for_each_array_value(node, "ftrace.filters", anode, p) { + q = kstrdup(p, GFP_KERNEL); + if (!q) + return; + if (ftrace_set_filter(tr->ops, q, strlen(q), 0) < 0) + pr_err("Failed to add %s to ftrace filter\n", p); + else + ftrace_filter_param = true; + kfree(q); + } + xbc_node_for_each_array_value(node, "ftrace.notraces", anode, p) { + q = kstrdup(p, GFP_KERNEL); + if (!q) + return; + if (ftrace_set_notrace(tr->ops, q, strlen(q), 0) < 0) + pr_err("Failed to add %s to ftrace filter\n", p); + else + ftrace_filter_param = true; + kfree(q); + } +} +#else +#define trace_boot_set_ftrace_filter(tr, node) do {} while (0) +#endif + +static void __init +trace_boot_enable_tracer(struct trace_array *tr, struct xbc_node *node) +{ + const char *p; + + trace_boot_set_ftrace_filter(tr, node); + + p = xbc_node_find_value(node, "tracer", NULL); + if (p && *p != '\0') { + if (tracing_set_tracer(tr, p) < 0) + pr_err("Failed to set given tracer: %s\n", p); + } +} + +static void __init +trace_boot_init_one_instance(struct trace_array *tr, struct xbc_node *node) +{ + trace_boot_set_instance_options(tr, node); + trace_boot_init_events(tr, node); + trace_boot_enable_events(tr, node); + trace_boot_enable_tracer(tr, node); +} + +static void __init +trace_boot_init_instances(struct xbc_node *node) +{ + struct xbc_node *inode; + struct trace_array *tr; + const char *p; + + node = xbc_node_find_child(node, "instance"); + if (!node) + return; + + xbc_node_for_each_child(node, inode) { + p = xbc_node_get_data(inode); + if (!p || *p == '\0') + continue; + + tr = trace_array_get_by_name(p); + if (!tr) { + pr_err("Failed to get trace instance %s\n", p); + continue; + } + trace_boot_init_one_instance(tr, inode); + trace_array_put(tr); + } +} + +static int __init trace_boot_init(void) +{ + struct xbc_node *trace_node; + struct trace_array *tr; + + trace_node = xbc_find_node("ftrace"); + if (!trace_node) + return 0; + + tr = top_trace_array(); + if (!tr) + return 0; + + /* Global trace array is also one instance */ + trace_boot_init_one_instance(tr, trace_node); + trace_boot_init_instances(trace_node); + + return 0; +} + +fs_initcall(trace_boot_init); diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index 88e158d27965..eff099123aa2 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -32,10 +32,10 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) { struct trace_event_call *call = &event_branch; struct trace_array *tr = branch_tracer; + struct trace_buffer *buffer; struct trace_array_cpu *data; struct ring_buffer_event *event; struct trace_branch *entry; - struct ring_buffer *buffer; unsigned long flags; int pc; const char *p; @@ -55,12 +55,12 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) raw_local_irq_save(flags); current->trace_recursion |= TRACE_BRANCH_BIT; - data = this_cpu_ptr(tr->trace_buffer.data); + data = this_cpu_ptr(tr->array_buffer.data); if (atomic_read(&data->disabled)) goto out; pc = preempt_count(); - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_BRANCH, sizeof(*entry), flags, pc); if (!event) diff --git a/kernel/trace/trace_dynevent.c b/kernel/trace/trace_dynevent.c index 89779eb84a07..9f2e8520b748 100644 --- a/kernel/trace/trace_dynevent.c +++ b/kernel/trace/trace_dynevent.c @@ -223,3 +223,215 @@ static __init int init_dynamic_event(void) return 0; } fs_initcall(init_dynamic_event); + +/** + * dynevent_arg_add - Add an arg to a dynevent_cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event cmd + * @arg: The argument to append to the current cmd + * @check_arg: An (optional) pointer to a function checking arg sanity + * + * Append an argument to a dynevent_cmd. The argument string will be + * appended to the current cmd string, followed by a separator, if + * applicable. Before the argument is added, the @check_arg function, + * if present, will be used to check the sanity of the current arg + * string. + * + * The cmd string and separator should be set using the + * dynevent_arg_init() before any arguments are added using this + * function. + * + * Return: 0 if successful, error otherwise. + */ +int dynevent_arg_add(struct dynevent_cmd *cmd, + struct dynevent_arg *arg, + dynevent_check_arg_fn_t check_arg) +{ + int ret = 0; + + if (check_arg) { + ret = check_arg(arg); + if (ret) + return ret; + } + + ret = seq_buf_printf(&cmd->seq, " %s%c", arg->str, arg->separator); + if (ret) { + pr_err("String is too long: %s%c\n", arg->str, arg->separator); + return -E2BIG; + } + + return ret; +} + +/** + * dynevent_arg_pair_add - Add an arg pair to a dynevent_cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event cmd + * @arg_pair: The argument pair to append to the current cmd + * @check_arg: An (optional) pointer to a function checking arg sanity + * + * Append an argument pair to a dynevent_cmd. An argument pair + * consists of a left-hand-side argument and a right-hand-side + * argument separated by an operator, which can be whitespace, all + * followed by a separator, if applicable. This can be used to add + * arguments of the form 'type variable_name;' or 'x+y'. + * + * The lhs argument string will be appended to the current cmd string, + * followed by an operator, if applicable, followd by the rhs string, + * followed finally by a separator, if applicable. Before the + * argument is added, the @check_arg function, if present, will be + * used to check the sanity of the current arg strings. + * + * The cmd strings, operator, and separator should be set using the + * dynevent_arg_pair_init() before any arguments are added using this + * function. + * + * Return: 0 if successful, error otherwise. + */ +int dynevent_arg_pair_add(struct dynevent_cmd *cmd, + struct dynevent_arg_pair *arg_pair, + dynevent_check_arg_fn_t check_arg) +{ + int ret = 0; + + if (check_arg) { + ret = check_arg(arg_pair); + if (ret) + return ret; + } + + ret = seq_buf_printf(&cmd->seq, " %s%c%s%c", arg_pair->lhs, + arg_pair->operator, arg_pair->rhs, + arg_pair->separator); + if (ret) { + pr_err("field string is too long: %s%c%s%c\n", arg_pair->lhs, + arg_pair->operator, arg_pair->rhs, + arg_pair->separator); + return -E2BIG; + } + + return ret; +} + +/** + * dynevent_str_add - Add a string to a dynevent_cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event cmd + * @str: The string to append to the current cmd + * + * Append a string to a dynevent_cmd. The string will be appended to + * the current cmd string as-is, with nothing prepended or appended. + * + * Return: 0 if successful, error otherwise. + */ +int dynevent_str_add(struct dynevent_cmd *cmd, const char *str) +{ + int ret = 0; + + ret = seq_buf_puts(&cmd->seq, str); + if (ret) { + pr_err("String is too long: %s\n", str); + return -E2BIG; + } + + return ret; +} + +/** + * dynevent_cmd_init - Initialize a dynevent_cmd object + * @cmd: A pointer to the dynevent_cmd struct representing the cmd + * @buf: A pointer to the buffer to generate the command into + * @maxlen: The length of the buffer the command will be generated into + * @type: The type of the cmd, checked against further operations + * @run_command: The type-specific function that will actually run the command + * + * Initialize a dynevent_cmd. A dynevent_cmd is used to build up and + * run dynamic event creation commands, such as commands for creating + * synthetic and kprobe events. Before calling any of the functions + * used to build the command, a dynevent_cmd object should be + * instantiated and initialized using this function. + * + * The initialization sets things up by saving a pointer to the + * user-supplied buffer and its length via the @buf and @maxlen + * params, and by saving the cmd-specific @type and @run_command + * params which are used to check subsequent dynevent_cmd operations + * and actually run the command when complete. + */ +void dynevent_cmd_init(struct dynevent_cmd *cmd, char *buf, int maxlen, + enum dynevent_type type, + dynevent_create_fn_t run_command) +{ + memset(cmd, '\0', sizeof(*cmd)); + + seq_buf_init(&cmd->seq, buf, maxlen); + cmd->type = type; + cmd->run_command = run_command; +} + +/** + * dynevent_arg_init - Initialize a dynevent_arg object + * @arg: A pointer to the dynevent_arg struct representing the arg + * @separator: An (optional) separator, appended after adding the arg + * + * Initialize a dynevent_arg object. A dynevent_arg represents an + * object used to append single arguments to the current command + * string. After the arg string is successfully appended to the + * command string, the optional @separator is appended. If no + * separator was specified when initializing the arg, a space will be + * appended. + */ +void dynevent_arg_init(struct dynevent_arg *arg, + char separator) +{ + memset(arg, '\0', sizeof(*arg)); + + if (!separator) + separator = ' '; + arg->separator = separator; +} + +/** + * dynevent_arg_pair_init - Initialize a dynevent_arg_pair object + * @arg_pair: A pointer to the dynevent_arg_pair struct representing the arg + * @operator: An (optional) operator, appended after adding the first arg + * @separator: An (optional) separator, appended after adding the second arg + * + * Initialize a dynevent_arg_pair object. A dynevent_arg_pair + * represents an object used to append argument pairs such as 'type + * variable_name;' or 'x+y' to the current command string. An + * argument pair consists of a left-hand-side argument and a + * right-hand-side argument separated by an operator, which can be + * whitespace, all followed by a separator, if applicable. After the + * first arg string is successfully appended to the command string, + * the optional @operator is appended, followed by the second arg and + * and optional @separator. If no separator was specified when + * initializing the arg, a space will be appended. + */ +void dynevent_arg_pair_init(struct dynevent_arg_pair *arg_pair, + char operator, char separator) +{ + memset(arg_pair, '\0', sizeof(*arg_pair)); + + if (!operator) + operator = ' '; + arg_pair->operator = operator; + + if (!separator) + separator = ' '; + arg_pair->separator = separator; +} + +/** + * dynevent_create - Create the dynamic event contained in dynevent_cmd + * @cmd: The dynevent_cmd object containing the dynamic event creation command + * + * Once a dynevent_cmd object has been successfully built up via the + * dynevent_cmd_init(), dynevent_arg_add() and dynevent_arg_pair_add() + * functions, this function runs the final command to actually create + * the event. + * + * Return: 0 if the event was successfully created, error otherwise. + */ +int dynevent_create(struct dynevent_cmd *cmd) +{ + return cmd->run_command(cmd); +} +EXPORT_SYMBOL_GPL(dynevent_create); diff --git a/kernel/trace/trace_dynevent.h b/kernel/trace/trace_dynevent.h index 46898138d2df..d6857a254ede 100644 --- a/kernel/trace/trace_dynevent.h +++ b/kernel/trace/trace_dynevent.h @@ -117,4 +117,36 @@ int dyn_event_release(int argc, char **argv, struct dyn_event_operations *type); #define for_each_dyn_event_safe(pos, n) \ list_for_each_entry_safe(pos, n, &dyn_event_list, list) +extern void dynevent_cmd_init(struct dynevent_cmd *cmd, char *buf, int maxlen, + enum dynevent_type type, + dynevent_create_fn_t run_command); + +typedef int (*dynevent_check_arg_fn_t)(void *data); + +struct dynevent_arg { + const char *str; + char separator; /* e.g. ';', ',', or nothing */ +}; + +extern void dynevent_arg_init(struct dynevent_arg *arg, + char separator); +extern int dynevent_arg_add(struct dynevent_cmd *cmd, + struct dynevent_arg *arg, + dynevent_check_arg_fn_t check_arg); + +struct dynevent_arg_pair { + const char *lhs; + const char *rhs; + char operator; /* e.g. '=' or nothing */ + char separator; /* e.g. ';', ',', or nothing */ +}; + +extern void dynevent_arg_pair_init(struct dynevent_arg_pair *arg_pair, + char operator, char separator); + +extern int dynevent_arg_pair_add(struct dynevent_cmd *cmd, + struct dynevent_arg_pair *arg_pair, + dynevent_check_arg_fn_t check_arg); +extern int dynevent_str_add(struct dynevent_cmd *cmd, const char *str); + #endif diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index 3e9d81608284..f22746f3c132 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -164,7 +164,7 @@ FTRACE_ENTRY(kernel_stack, stack_entry, F_STRUCT( __field( int, size ) - __dynamic_array(unsigned long, caller ) + __array( unsigned long, caller, FTRACE_STACK_ENTRIES ) ), F_printk("\t=> %ps\n\t=> %ps\n\t=> %ps\n" diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index fcd3975a88d1..f38234ecea18 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -238,7 +238,7 @@ bool trace_event_ignore_this_pid(struct trace_event_file *trace_file) if (!pid_list) return false; - data = this_cpu_ptr(tr->trace_buffer.data); + data = this_cpu_ptr(tr->array_buffer.data); return data->ignore_pid; } @@ -273,6 +273,7 @@ void *trace_event_buffer_reserve(struct trace_event_buffer *fbuffer, if (!fbuffer->event) return NULL; + fbuffer->regs = NULL; fbuffer->entry = ring_buffer_event_data(fbuffer->event); return fbuffer->entry; } @@ -547,7 +548,7 @@ event_filter_pid_sched_switch_probe_pre(void *data, bool preempt, pid_list = rcu_dereference_sched(tr->filtered_pids); - this_cpu_write(tr->trace_buffer.data->ignore_pid, + this_cpu_write(tr->array_buffer.data->ignore_pid, trace_ignore_this_task(pid_list, prev) && trace_ignore_this_task(pid_list, next)); } @@ -561,7 +562,7 @@ event_filter_pid_sched_switch_probe_post(void *data, bool preempt, pid_list = rcu_dereference_sched(tr->filtered_pids); - this_cpu_write(tr->trace_buffer.data->ignore_pid, + this_cpu_write(tr->array_buffer.data->ignore_pid, trace_ignore_this_task(pid_list, next)); } @@ -572,12 +573,12 @@ event_filter_pid_sched_wakeup_probe_pre(void *data, struct task_struct *task) struct trace_pid_list *pid_list; /* Nothing to do if we are already tracing */ - if (!this_cpu_read(tr->trace_buffer.data->ignore_pid)) + if (!this_cpu_read(tr->array_buffer.data->ignore_pid)) return; pid_list = rcu_dereference_sched(tr->filtered_pids); - this_cpu_write(tr->trace_buffer.data->ignore_pid, + this_cpu_write(tr->array_buffer.data->ignore_pid, trace_ignore_this_task(pid_list, task)); } @@ -588,13 +589,13 @@ event_filter_pid_sched_wakeup_probe_post(void *data, struct task_struct *task) struct trace_pid_list *pid_list; /* Nothing to do if we are not tracing */ - if (this_cpu_read(tr->trace_buffer.data->ignore_pid)) + if (this_cpu_read(tr->array_buffer.data->ignore_pid)) return; pid_list = rcu_dereference_sched(tr->filtered_pids); /* Set tracing if current is enabled */ - this_cpu_write(tr->trace_buffer.data->ignore_pid, + this_cpu_write(tr->array_buffer.data->ignore_pid, trace_ignore_this_task(pid_list, current)); } @@ -626,7 +627,7 @@ static void __ftrace_clear_event_pids(struct trace_array *tr) } for_each_possible_cpu(cpu) - per_cpu_ptr(tr->trace_buffer.data, cpu)->ignore_pid = false; + per_cpu_ptr(tr->array_buffer.data, cpu)->ignore_pid = false; rcu_assign_pointer(tr->filtered_pids, NULL); @@ -1595,7 +1596,7 @@ static void ignore_task_cpu(void *data) pid_list = rcu_dereference_protected(tr->filtered_pids, mutex_is_locked(&event_mutex)); - this_cpu_write(tr->trace_buffer.data->ignore_pid, + this_cpu_write(tr->array_buffer.data->ignore_pid, trace_ignore_this_task(pid_list, current)); } @@ -2553,6 +2554,91 @@ find_event_file(struct trace_array *tr, const char *system, const char *event) return file; } +/** + * trace_get_event_file - Find and return a trace event file + * @instance: The name of the trace instance containing the event + * @system: The name of the system containing the event + * @event: The name of the event + * + * Return a trace event file given the trace instance name, trace + * system, and trace event name. If the instance name is NULL, it + * refers to the top-level trace array. + * + * This function will look it up and return it if found, after calling + * trace_array_get() to prevent the instance from going away, and + * increment the event's module refcount to prevent it from being + * removed. + * + * To release the file, call trace_put_event_file(), which will call + * trace_array_put() and decrement the event's module refcount. + * + * Return: The trace event on success, ERR_PTR otherwise. + */ +struct trace_event_file *trace_get_event_file(const char *instance, + const char *system, + const char *event) +{ + struct trace_array *tr = top_trace_array(); + struct trace_event_file *file = NULL; + int ret = -EINVAL; + + if (instance) { + tr = trace_array_find_get(instance); + if (!tr) + return ERR_PTR(-ENOENT); + } else { + ret = trace_array_get(tr); + if (ret) + return ERR_PTR(ret); + } + + mutex_lock(&event_mutex); + + file = find_event_file(tr, system, event); + if (!file) { + trace_array_put(tr); + ret = -EINVAL; + goto out; + } + + /* Don't let event modules unload while in use */ + ret = try_module_get(file->event_call->mod); + if (!ret) { + trace_array_put(tr); + ret = -EBUSY; + goto out; + } + + ret = 0; + out: + mutex_unlock(&event_mutex); + + if (ret) + file = ERR_PTR(ret); + + return file; +} +EXPORT_SYMBOL_GPL(trace_get_event_file); + +/** + * trace_put_event_file - Release a file from trace_get_event_file() + * @file: The trace event file + * + * If a file was retrieved using trace_get_event_file(), this should + * be called when it's no longer needed. It will cancel the previous + * trace_array_get() called by that function, and decrement the + * event's module refcount. + */ +void trace_put_event_file(struct trace_event_file *file) +{ + mutex_lock(&event_mutex); + module_put(file->event_call->mod); + mutex_unlock(&event_mutex); + + trace_array_put(file->tr); +} +EXPORT_SYMBOL_GPL(trace_put_event_file); + #ifdef CONFIG_DYNAMIC_FTRACE /* Avoid typos */ @@ -3409,8 +3495,8 @@ static void __init function_test_events_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct pt_regs *pt_regs) { + struct trace_buffer *buffer; struct ring_buffer_event *event; - struct ring_buffer *buffer; struct ftrace_entry *entry; unsigned long flags; long disabled; diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index f2896d13001b..e7ce7cdac62f 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -66,7 +66,12 @@ C(INVALID_SUBSYS_EVENT, "Invalid subsystem or event name"), \ C(INVALID_REF_KEY, "Using variable references in keys not supported"), \ C(VAR_NOT_FOUND, "Couldn't find variable"), \ - C(FIELD_NOT_FOUND, "Couldn't find field"), + C(FIELD_NOT_FOUND, "Couldn't find field"), \ + C(EMPTY_ASSIGNMENT, "Empty assignment"), \ + C(INVALID_SORT_MODIFIER,"Invalid sort modifier"), \ + C(EMPTY_SORT_FIELD, "Empty sort field"), \ + C(TOO_MANY_SORT_FIELDS, "Too many sort fields (Max = 2)"), \ + C(INVALID_SORT_FIELD, "Sort field must be a key or a val"), #undef C #define C(a, b) HIST_ERR_##a @@ -375,7 +380,7 @@ struct hist_trigger_data { unsigned int n_save_var_str; }; -static int synth_event_create(int argc, const char **argv); +static int create_synth_event(int argc, const char **argv); static int synth_event_show(struct seq_file *m, struct dyn_event *ev); static int synth_event_release(struct dyn_event *ev); static bool synth_event_is_busy(struct dyn_event *ev); @@ -383,7 +388,7 @@ static bool synth_event_match(const char *system, const char *event, int argc, const char **argv, struct dyn_event *ev); static struct dyn_event_operations synth_event_ops = { - .create = synth_event_create, + .create = create_synth_event, .show = synth_event_show, .is_busy = synth_event_is_busy, .free = synth_event_release, @@ -394,6 +399,7 @@ struct synth_field { char *type; char *name; size_t size; + unsigned int offset; bool is_signed; bool is_string; }; @@ -408,6 +414,7 @@ struct synth_event { struct trace_event_class class; struct trace_event_call call; struct tracepoint *tp; + struct module *mod; }; static bool is_synth_event(struct dyn_event *ev) @@ -470,11 +477,12 @@ struct action_data { * When a histogram trigger is hit, the values of any * references to variables, including variables being passed * as parameters to synthetic events, are collected into a - * var_ref_vals array. This var_ref_idx is the index of the - * first param in the array to be passed to the synthetic - * event invocation. + * var_ref_vals array. This var_ref_idx array is an array of + * indices into the var_ref_vals array, one for each synthetic + * event param, and is passed to the synthetic event + * invocation. */ - unsigned int var_ref_idx; + unsigned int var_ref_idx[TRACING_MAP_VARS_MAX]; struct synth_event *synth_event; bool use_trace_keyword; char *synth_event_name; @@ -608,7 +616,8 @@ static void last_cmd_set(struct trace_event_file *file, char *str) if (!str) return; - strncpy(last_cmd, str, MAX_FILTER_STR_VAL - 1); + strcpy(last_cmd, "hist:"); + strncat(last_cmd, str, MAX_FILTER_STR_VAL - 1 - sizeof("hist:")); if (file) { call = file->event_call; @@ -662,6 +671,8 @@ static int synth_event_define_fields(struct trace_event_call *call) if (ret) break; + event->fields[i]->offset = n_u64; + if (event->fields[i]->is_string) { offset += STR_VAR_LEN_MAX; n_u64 += STR_VAR_LEN_MAX / sizeof(u64); @@ -834,7 +845,7 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, fmt = synth_field_fmt(se->fields[i]->type); /* parameter types */ - if (tr->trace_flags & TRACE_ITER_VERBOSE) + if (tr && tr->trace_flags & TRACE_ITER_VERBOSE) trace_seq_printf(s, "%s ", fmt); snprintf(print_fmt, sizeof(print_fmt), "%%s=%s%%s", fmt); @@ -875,14 +886,14 @@ static struct trace_event_functions synth_event_funcs = { static notrace void trace_event_raw_event_synth(void *__data, u64 *var_ref_vals, - unsigned int var_ref_idx) + unsigned int *var_ref_idx) { struct trace_event_file *trace_file = __data; struct synth_trace_event *entry; struct trace_event_buffer fbuffer; - struct ring_buffer *buffer; + struct trace_buffer *buffer; struct synth_event *event; - unsigned int i, n_u64; + unsigned int i, n_u64, val_idx; int fields_size = 0; event = trace_file->event_call->data; @@ -896,7 +907,7 @@ static notrace void trace_event_raw_event_synth(void *__data, * Avoid ring buffer recursion detection, as this event * is being performed within another event. */ - buffer = trace_file->tr->trace_buffer.buffer; + buffer = trace_file->tr->array_buffer.buffer; ring_buffer_nest_start(buffer); entry = trace_event_buffer_reserve(&fbuffer, trace_file, @@ -905,15 +916,16 @@ static notrace void trace_event_raw_event_synth(void *__data, goto out; for (i = 0, n_u64 = 0; i < event->n_fields; i++) { + val_idx = var_ref_idx[i]; if (event->fields[i]->is_string) { - char *str_val = (char *)(long)var_ref_vals[var_ref_idx + i]; + char *str_val = (char *)(long)var_ref_vals[val_idx]; char *str_field = (char *)&entry->fields[n_u64]; strscpy(str_field, str_val, STR_VAR_LEN_MAX); n_u64 += STR_VAR_LEN_MAX / sizeof(u64); } else { struct synth_field *field = event->fields[i]; - u64 val = var_ref_vals[var_ref_idx + i]; + u64 val = var_ref_vals[val_idx]; switch (field->size) { case 1: @@ -1113,10 +1125,10 @@ static struct tracepoint *alloc_synth_tracepoint(char *name) } typedef void (*synth_probe_func_t) (void *__data, u64 *var_ref_vals, - unsigned int var_ref_idx); + unsigned int *var_ref_idx); static inline void trace_synth(struct synth_event *event, u64 *var_ref_vals, - unsigned int var_ref_idx) + unsigned int *var_ref_idx) { struct tracepoint *tp = event->tp; @@ -1293,6 +1305,273 @@ struct hist_var_data { struct hist_trigger_data *hist_data; }; +static int synth_event_check_arg_fn(void *data) +{ + struct dynevent_arg_pair *arg_pair = data; + int size; + + size = synth_field_size((char *)arg_pair->lhs); + + return size ? 0 : -EINVAL; +} + +/** + * synth_event_add_field - Add a new field to a synthetic event cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @type: The type of the new field to add + * @name: The name of the new field to add + * + * Add a new field to a synthetic event cmd object. Field ordering is in + * the same order the fields are added. + * + * See synth_field_size() for available types. If field_name contains + * [n] the field is considered to be an array. + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_add_field(struct dynevent_cmd *cmd, const char *type, + const char *name) +{ + struct dynevent_arg_pair arg_pair; + int ret; + + if (cmd->type != DYNEVENT_TYPE_SYNTH) + return -EINVAL; + + if (!type || !name) + return -EINVAL; + + dynevent_arg_pair_init(&arg_pair, 0, ';'); + + arg_pair.lhs = type; + arg_pair.rhs = name; + + ret = dynevent_arg_pair_add(cmd, &arg_pair, synth_event_check_arg_fn); + if (ret) + return ret; + + if (++cmd->n_fields > SYNTH_FIELDS_MAX) + ret = -EINVAL; + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_add_field); + +/** + * synth_event_add_field_str - Add a new field to a synthetic event cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @type_name: The type and name of the new field to add, as a single string + * + * Add a new field to a synthetic event cmd object, as a single + * string. The @type_name string is expected to be of the form 'type + * name', which will be appended by ';'. No sanity checking is done - + * what's passed in is assumed to already be well-formed. Field + * ordering is in the same order the fields are added. + * + * See synth_field_size() for available types. If field_name contains + * [n] the field is considered to be an array. + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_add_field_str(struct dynevent_cmd *cmd, const char *type_name) +{ + struct dynevent_arg arg; + int ret; + + if (cmd->type != DYNEVENT_TYPE_SYNTH) + return -EINVAL; + + if (!type_name) + return -EINVAL; + + dynevent_arg_init(&arg, ';'); + + arg.str = type_name; + + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + return ret; + + if (++cmd->n_fields > SYNTH_FIELDS_MAX) + ret = -EINVAL; + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_add_field_str); + +/** + * synth_event_add_fields - Add multiple fields to a synthetic event cmd + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @fields: An array of type/name field descriptions + * @n_fields: The number of field descriptions contained in the fields array + * + * Add a new set of fields to a synthetic event cmd object. The event + * fields that will be defined for the event should be passed in as an + * array of struct synth_field_desc, and the number of elements in the + * array passed in as n_fields. Field ordering will retain the + * ordering given in the fields array. + * + * See synth_field_size() for available types. If field_name contains + * [n] the field is considered to be an array. + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_add_fields(struct dynevent_cmd *cmd, + struct synth_field_desc *fields, + unsigned int n_fields) +{ + unsigned int i; + int ret = 0; + + for (i = 0; i < n_fields; i++) { + if (fields[i].type == NULL || fields[i].name == NULL) { + ret = -EINVAL; + break; + } + + ret = synth_event_add_field(cmd, fields[i].type, fields[i].name); + if (ret) + break; + } + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_add_fields); + +/** + * __synth_event_gen_cmd_start - Start a synthetic event command from arg list + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @name: The name of the synthetic event + * @mod: The module creating the event, NULL if not created from a module + * @args: Variable number of arg (pairs), one pair for each field + * + * NOTE: Users normally won't want to call this function directly, but + * rather use the synth_event_gen_cmd_start() wrapper, which + * automatically adds a NULL to the end of the arg list. If this + * function is used directly, make sure the last arg in the variable + * arg list is NULL. + * + * Generate a synthetic event command to be executed by + * synth_event_gen_cmd_end(). This function can be used to generate + * the complete command or only the first part of it; in the latter + * case, synth_event_add_field(), synth_event_add_field_str(), or + * synth_event_add_fields() can be used to add more fields following + * this. + * + * There should be an even number variable args, each pair consisting + * of a type followed by a field name. + * + * See synth_field_size() for available types. If field_name contains + * [n] the field is considered to be an array. + * + * Return: 0 if successful, error otherwise. + */ +int __synth_event_gen_cmd_start(struct dynevent_cmd *cmd, const char *name, + struct module *mod, ...) +{ + struct dynevent_arg arg; + va_list args; + int ret; + + cmd->event_name = name; + cmd->private_data = mod; + + if (cmd->type != DYNEVENT_TYPE_SYNTH) + return -EINVAL; + + dynevent_arg_init(&arg, 0); + arg.str = name; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + return ret; + + va_start(args, mod); + for (;;) { + const char *type, *name; + + type = va_arg(args, const char *); + if (!type) + break; + name = va_arg(args, const char *); + if (!name) + break; + + if (++cmd->n_fields > SYNTH_FIELDS_MAX) { + ret = -EINVAL; + break; + } + + ret = synth_event_add_field(cmd, type, name); + if (ret) + break; + } + va_end(args); + + return ret; +} +EXPORT_SYMBOL_GPL(__synth_event_gen_cmd_start); + +/** + * synth_event_gen_cmd_array_start - Start synthetic event command from an array + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @name: The name of the synthetic event + * @fields: An array of type/name field descriptions + * @n_fields: The number of field descriptions contained in the fields array + * + * Generate a synthetic event command to be executed by + * synth_event_gen_cmd_end(). This function can be used to generate + * the complete command or only the first part of it; in the latter + * case, synth_event_add_field(), synth_event_add_field_str(), or + * synth_event_add_fields() can be used to add more fields following + * this. + * + * The event fields that will be defined for the event should be + * passed in as an array of struct synth_field_desc, and the number of + * elements in the array passed in as n_fields. Field ordering will + * retain the ordering given in the fields array. + * + * See synth_field_size() for available types. If field_name contains + * [n] the field is considered to be an array. + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_gen_cmd_array_start(struct dynevent_cmd *cmd, const char *name, + struct module *mod, + struct synth_field_desc *fields, + unsigned int n_fields) +{ + struct dynevent_arg arg; + unsigned int i; + int ret = 0; + + cmd->event_name = name; + cmd->private_data = mod; + + if (cmd->type != DYNEVENT_TYPE_SYNTH) + return -EINVAL; + + if (n_fields > SYNTH_FIELDS_MAX) + return -EINVAL; + + dynevent_arg_init(&arg, 0); + arg.str = name; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + return ret; + + for (i = 0; i < n_fields; i++) { + if (fields[i].type == NULL || fields[i].name == NULL) + return -EINVAL; + + ret = synth_event_add_field(cmd, fields[i].type, fields[i].name); + if (ret) + break; + } + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_gen_cmd_array_start); + static int __create_synth_event(int argc, const char *name, const char **argv) { struct synth_field *field, *fields[SYNTH_FIELDS_MAX]; @@ -1361,29 +1640,123 @@ static int __create_synth_event(int argc, const char *name, const char **argv) goto out; } +/** + * synth_event_create - Create a new synthetic event + * @name: The name of the new sythetic event + * @fields: An array of type/name field descriptions + * @n_fields: The number of field descriptions contained in the fields array + * @mod: The module creating the event, NULL if not created from a module + * + * Create a new synthetic event with the given name under the + * trace/events/synthetic/ directory. The event fields that will be + * defined for the event should be passed in as an array of struct + * synth_field_desc, and the number elements in the array passed in as + * n_fields. Field ordering will retain the ordering given in the + * fields array. + * + * If the new synthetic event is being created from a module, the mod + * param must be non-NULL. This will ensure that the trace buffer + * won't contain unreadable events. + * + * The new synth event should be deleted using synth_event_delete() + * function. The new synthetic event can be generated from modules or + * other kernel code using trace_synth_event() and related functions. + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_create(const char *name, struct synth_field_desc *fields, + unsigned int n_fields, struct module *mod) +{ + struct dynevent_cmd cmd; + char *buf; + int ret; + + buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + synth_event_cmd_init(&cmd, buf, MAX_DYNEVENT_CMD_LEN); + + ret = synth_event_gen_cmd_array_start(&cmd, name, mod, + fields, n_fields); + if (ret) + goto out; + + ret = synth_event_gen_cmd_end(&cmd); + out: + kfree(buf); + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_create); + +static int destroy_synth_event(struct synth_event *se) +{ + int ret; + + if (se->ref) + ret = -EBUSY; + else { + ret = unregister_synth_event(se); + if (!ret) { + dyn_event_remove(&se->devent); + free_synth_event(se); + } + } + + return ret; +} + +/** + * synth_event_delete - Delete a synthetic event + * @event_name: The name of the new sythetic event + * + * Delete a synthetic event that was created with synth_event_create(). + * + * Return: 0 if successful, error otherwise. + */ +int synth_event_delete(const char *event_name) +{ + struct synth_event *se = NULL; + struct module *mod = NULL; + int ret = -ENOENT; + + mutex_lock(&event_mutex); + se = find_synth_event(event_name); + if (se) { + mod = se->mod; + ret = destroy_synth_event(se); + } + mutex_unlock(&event_mutex); + + if (mod) { + mutex_lock(&trace_types_lock); + /* + * It is safest to reset the ring buffer if the module + * being unloaded registered any events that were + * used. The only worry is if a new module gets + * loaded, and takes on the same id as the events of + * this module. When printing out the buffer, traced + * events left over from this module may be passed to + * the new module events and unexpected results may + * occur. + */ + tracing_reset_all_online_cpus(); + mutex_unlock(&trace_types_lock); + } + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_delete); + static int create_or_delete_synth_event(int argc, char **argv) { const char *name = argv[0]; - struct synth_event *event = NULL; int ret; /* trace_run_command() ensures argc != 0 */ if (name[0] == '!') { - mutex_lock(&event_mutex); - event = find_synth_event(name + 1); - if (event) { - if (event->ref) - ret = -EBUSY; - else { - ret = unregister_synth_event(event); - if (!ret) { - dyn_event_remove(&event->devent); - free_synth_event(event); - } - } - } else - ret = -ENOENT; - mutex_unlock(&event_mutex); + ret = synth_event_delete(name + 1); return ret; } @@ -1391,7 +1764,474 @@ static int create_or_delete_synth_event(int argc, char **argv) return ret == -ECANCELED ? -EINVAL : ret; } -static int synth_event_create(int argc, const char **argv) +static int synth_event_run_command(struct dynevent_cmd *cmd) +{ + struct synth_event *se; + int ret; + + ret = trace_run_command(cmd->seq.buffer, create_or_delete_synth_event); + if (ret) + return ret; + + se = find_synth_event(cmd->event_name); + if (WARN_ON(!se)) + return -ENOENT; + + se->mod = cmd->private_data; + + return ret; +} + +/** + * synth_event_cmd_init - Initialize a synthetic event command object + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @buf: A pointer to the buffer used to build the command + * @maxlen: The length of the buffer passed in @buf + * + * Initialize a synthetic event command object. Use this before + * calling any of the other dyenvent_cmd functions. + */ +void synth_event_cmd_init(struct dynevent_cmd *cmd, char *buf, int maxlen) +{ + dynevent_cmd_init(cmd, buf, maxlen, DYNEVENT_TYPE_SYNTH, + synth_event_run_command); +} +EXPORT_SYMBOL_GPL(synth_event_cmd_init); + +/** + * synth_event_trace - Trace a synthetic event + * @file: The trace_event_file representing the synthetic event + * @n_vals: The number of values in vals + * @args: Variable number of args containing the event values + * + * Trace a synthetic event using the values passed in the variable + * argument list. + * + * The argument list should be a list 'n_vals' u64 values. The number + * of vals must match the number of field in the synthetic event, and + * must be in the same order as the synthetic event fields. + * + * All vals should be cast to u64, and string vals are just pointers + * to strings, cast to u64. Strings will be copied into space + * reserved in the event for the string, using these pointers. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_trace(struct trace_event_file *file, unsigned int n_vals, ...) +{ + struct trace_event_buffer fbuffer; + struct synth_trace_event *entry; + struct trace_buffer *buffer; + struct synth_event *event; + unsigned int i, n_u64; + int fields_size = 0; + va_list args; + int ret = 0; + + /* + * Normal event generation doesn't get called at all unless + * the ENABLED bit is set (which attaches the probe thus + * allowing this code to be called, etc). Because this is + * called directly by the user, we don't have that but we + * still need to honor not logging when disabled. + */ + if (!(file->flags & EVENT_FILE_FL_ENABLED)) + return 0; + + event = file->event_call->data; + + if (n_vals != event->n_fields) + return -EINVAL; + + if (trace_trigger_soft_disabled(file)) + return -EINVAL; + + fields_size = event->n_u64 * sizeof(u64); + + /* + * Avoid ring buffer recursion detection, as this event + * is being performed within another event. + */ + buffer = file->tr->array_buffer.buffer; + ring_buffer_nest_start(buffer); + + entry = trace_event_buffer_reserve(&fbuffer, file, + sizeof(*entry) + fields_size); + if (!entry) { + ret = -EINVAL; + goto out; + } + + va_start(args, n_vals); + for (i = 0, n_u64 = 0; i < event->n_fields; i++) { + u64 val; + + val = va_arg(args, u64); + + if (event->fields[i]->is_string) { + char *str_val = (char *)(long)val; + char *str_field = (char *)&entry->fields[n_u64]; + + strscpy(str_field, str_val, STR_VAR_LEN_MAX); + n_u64 += STR_VAR_LEN_MAX / sizeof(u64); + } else { + entry->fields[n_u64] = val; + n_u64++; + } + } + va_end(args); + + trace_event_buffer_commit(&fbuffer); +out: + ring_buffer_nest_end(buffer); + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_trace); + +/** + * synth_event_trace_array - Trace a synthetic event from an array + * @file: The trace_event_file representing the synthetic event + * @vals: Array of values + * @n_vals: The number of values in vals + * + * Trace a synthetic event using the values passed in as 'vals'. + * + * The 'vals' array is just an array of 'n_vals' u64. The number of + * vals must match the number of field in the synthetic event, and + * must be in the same order as the synthetic event fields. + * + * All vals should be cast to u64, and string vals are just pointers + * to strings, cast to u64. Strings will be copied into space + * reserved in the event for the string, using these pointers. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_trace_array(struct trace_event_file *file, u64 *vals, + unsigned int n_vals) +{ + struct trace_event_buffer fbuffer; + struct synth_trace_event *entry; + struct trace_buffer *buffer; + struct synth_event *event; + unsigned int i, n_u64; + int fields_size = 0; + int ret = 0; + + /* + * Normal event generation doesn't get called at all unless + * the ENABLED bit is set (which attaches the probe thus + * allowing this code to be called, etc). Because this is + * called directly by the user, we don't have that but we + * still need to honor not logging when disabled. + */ + if (!(file->flags & EVENT_FILE_FL_ENABLED)) + return 0; + + event = file->event_call->data; + + if (n_vals != event->n_fields) + return -EINVAL; + + if (trace_trigger_soft_disabled(file)) + return -EINVAL; + + fields_size = event->n_u64 * sizeof(u64); + + /* + * Avoid ring buffer recursion detection, as this event + * is being performed within another event. + */ + buffer = file->tr->array_buffer.buffer; + ring_buffer_nest_start(buffer); + + entry = trace_event_buffer_reserve(&fbuffer, file, + sizeof(*entry) + fields_size); + if (!entry) { + ret = -EINVAL; + goto out; + } + + for (i = 0, n_u64 = 0; i < event->n_fields; i++) { + if (event->fields[i]->is_string) { + char *str_val = (char *)(long)vals[i]; + char *str_field = (char *)&entry->fields[n_u64]; + + strscpy(str_field, str_val, STR_VAR_LEN_MAX); + n_u64 += STR_VAR_LEN_MAX / sizeof(u64); + } else { + entry->fields[n_u64] = vals[i]; + n_u64++; + } + } + + trace_event_buffer_commit(&fbuffer); +out: + ring_buffer_nest_end(buffer); + + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_trace_array); + +/** + * synth_event_trace_start - Start piecewise synthetic event trace + * @file: The trace_event_file representing the synthetic event + * @trace_state: A pointer to object tracking the piecewise trace state + * + * Start the trace of a synthetic event field-by-field rather than all + * at once. + * + * This function 'opens' an event trace, which means space is reserved + * for the event in the trace buffer, after which the event's + * individual field values can be set through either + * synth_event_add_next_val() or synth_event_add_val(). + * + * A pointer to a trace_state object is passed in, which will keep + * track of the current event trace state until the event trace is + * closed (and the event finally traced) using + * synth_event_trace_end(). + * + * Note that synth_event_trace_end() must be called after all values + * have been added for each event trace, regardless of whether adding + * all field values succeeded or not. + * + * Note also that for a given event trace, all fields must be added + * using either synth_event_add_next_val() or synth_event_add_val() + * but not both together or interleaved. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_trace_start(struct trace_event_file *file, + struct synth_event_trace_state *trace_state) +{ + struct synth_trace_event *entry; + int fields_size = 0; + int ret = 0; + + if (!trace_state) { + ret = -EINVAL; + goto out; + } + + memset(trace_state, '\0', sizeof(*trace_state)); + + /* + * Normal event tracing doesn't get called at all unless the + * ENABLED bit is set (which attaches the probe thus allowing + * this code to be called, etc). Because this is called + * directly by the user, we don't have that but we still need + * to honor not logging when disabled. For the the iterated + * trace case, we save the enabed state upon start and just + * ignore the following data calls. + */ + if (!(file->flags & EVENT_FILE_FL_ENABLED)) { + trace_state->enabled = false; + goto out; + } + + trace_state->enabled = true; + + trace_state->event = file->event_call->data; + + if (trace_trigger_soft_disabled(file)) { + ret = -EINVAL; + goto out; + } + + fields_size = trace_state->event->n_u64 * sizeof(u64); + + /* + * Avoid ring buffer recursion detection, as this event + * is being performed within another event. + */ + trace_state->buffer = file->tr->array_buffer.buffer; + ring_buffer_nest_start(trace_state->buffer); + + entry = trace_event_buffer_reserve(&trace_state->fbuffer, file, + sizeof(*entry) + fields_size); + if (!entry) { + ret = -EINVAL; + goto out; + } + + trace_state->entry = entry; +out: + return ret; +} +EXPORT_SYMBOL_GPL(synth_event_trace_start); + +static int __synth_event_add_val(const char *field_name, u64 val, + struct synth_event_trace_state *trace_state) +{ + struct synth_field *field = NULL; + struct synth_trace_event *entry; + struct synth_event *event; + int i, ret = 0; + + if (!trace_state) { + ret = -EINVAL; + goto out; + } + + /* can't mix add_next_synth_val() with add_synth_val() */ + if (field_name) { + if (trace_state->add_next) { + ret = -EINVAL; + goto out; + } + trace_state->add_name = true; + } else { + if (trace_state->add_name) { + ret = -EINVAL; + goto out; + } + trace_state->add_next = true; + } + + if (!trace_state->enabled) + goto out; + + event = trace_state->event; + if (trace_state->add_name) { + for (i = 0; i < event->n_fields; i++) { + field = event->fields[i]; + if (strcmp(field->name, field_name) == 0) + break; + } + if (!field) { + ret = -EINVAL; + goto out; + } + } else { + if (trace_state->cur_field >= event->n_fields) { + ret = -EINVAL; + goto out; + } + field = event->fields[trace_state->cur_field++]; + } + + entry = trace_state->entry; + if (field->is_string) { + char *str_val = (char *)(long)val; + char *str_field; + + if (!str_val) { + ret = -EINVAL; + goto out; + } + + str_field = (char *)&entry->fields[field->offset]; + strscpy(str_field, str_val, STR_VAR_LEN_MAX); + } else + entry->fields[field->offset] = val; + out: + return ret; +} + +/** + * synth_event_add_next_val - Add the next field's value to an open synth trace + * @val: The value to set the next field to + * @trace_state: A pointer to object tracking the piecewise trace state + * + * Set the value of the next field in an event that's been opened by + * synth_event_trace_start(). + * + * The val param should be the value cast to u64. If the value points + * to a string, the val param should be a char * cast to u64. + * + * This function assumes all the fields in an event are to be set one + * after another - successive calls to this function are made, one for + * each field, in the order of the fields in the event, until all + * fields have been set. If you'd rather set each field individually + * without regard to ordering, synth_event_add_val() can be used + * instead. + * + * Note however that synth_event_add_next_val() and + * synth_event_add_val() can't be intermixed for a given event trace - + * one or the other but not both can be used at the same time. + * + * Note also that synth_event_trace_end() must be called after all + * values have been added for each event trace, regardless of whether + * adding all field values succeeded or not. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_add_next_val(u64 val, + struct synth_event_trace_state *trace_state) +{ + return __synth_event_add_val(NULL, val, trace_state); +} +EXPORT_SYMBOL_GPL(synth_event_add_next_val); + +/** + * synth_event_add_val - Add a named field's value to an open synth trace + * @field_name: The name of the synthetic event field value to set + * @val: The value to set the next field to + * @trace_state: A pointer to object tracking the piecewise trace state + * + * Set the value of the named field in an event that's been opened by + * synth_event_trace_start(). + * + * The val param should be the value cast to u64. If the value points + * to a string, the val param should be a char * cast to u64. + * + * This function looks up the field name, and if found, sets the field + * to the specified value. This lookup makes this function more + * expensive than synth_event_add_next_val(), so use that or the + * none-piecewise synth_event_trace() instead if efficiency is more + * important. + * + * Note however that synth_event_add_next_val() and + * synth_event_add_val() can't be intermixed for a given event trace - + * one or the other but not both can be used at the same time. + * + * Note also that synth_event_trace_end() must be called after all + * values have been added for each event trace, regardless of whether + * adding all field values succeeded or not. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_add_val(const char *field_name, u64 val, + struct synth_event_trace_state *trace_state) +{ + return __synth_event_add_val(field_name, val, trace_state); +} +EXPORT_SYMBOL_GPL(synth_event_add_val); + +/** + * synth_event_trace_end - End piecewise synthetic event trace + * @trace_state: A pointer to object tracking the piecewise trace state + * + * End the trace of a synthetic event opened by + * synth_event_trace__start(). + * + * This function 'closes' an event trace, which basically means that + * it commits the reserved event and cleans up other loose ends. + * + * A pointer to a trace_state object is passed in, which will keep + * track of the current event trace state opened with + * synth_event_trace_start(). + * + * Note that this function must be called after all values have been + * added for each event trace, regardless of whether adding all field + * values succeeded or not. + * + * Return: 0 on success, err otherwise. + */ +int synth_event_trace_end(struct synth_event_trace_state *trace_state) +{ + if (!trace_state) + return -EINVAL; + + trace_event_buffer_commit(&trace_state->fbuffer); + + ring_buffer_nest_end(trace_state->buffer); + + return 0; +} +EXPORT_SYMBOL_GPL(synth_event_trace_end); + +static int create_synth_event(int argc, const char **argv) { const char *name = argv[0]; int len; @@ -2041,12 +2881,6 @@ static int parse_map_size(char *str) unsigned long size, map_bits; int ret; - strsep(&str, "="); - if (!str) { - ret = -EINVAL; - goto out; - } - ret = kstrtoul(str, 0, &size); if (ret) goto out; @@ -2106,25 +2940,25 @@ static int parse_action(char *str, struct hist_trigger_attrs *attrs) static int parse_assignment(struct trace_array *tr, char *str, struct hist_trigger_attrs *attrs) { - int ret = 0; + int len, ret = 0; - if ((str_has_prefix(str, "key=")) || - (str_has_prefix(str, "keys="))) { - attrs->keys_str = kstrdup(str, GFP_KERNEL); + if ((len = str_has_prefix(str, "key=")) || + (len = str_has_prefix(str, "keys="))) { + attrs->keys_str = kstrdup(str + len, GFP_KERNEL); if (!attrs->keys_str) { ret = -ENOMEM; goto out; } - } else if ((str_has_prefix(str, "val=")) || - (str_has_prefix(str, "vals=")) || - (str_has_prefix(str, "values="))) { - attrs->vals_str = kstrdup(str, GFP_KERNEL); + } else if ((len = str_has_prefix(str, "val=")) || + (len = str_has_prefix(str, "vals=")) || + (len = str_has_prefix(str, "values="))) { + attrs->vals_str = kstrdup(str + len, GFP_KERNEL); if (!attrs->vals_str) { ret = -ENOMEM; goto out; } - } else if (str_has_prefix(str, "sort=")) { - attrs->sort_key_str = kstrdup(str, GFP_KERNEL); + } else if ((len = str_has_prefix(str, "sort="))) { + attrs->sort_key_str = kstrdup(str + len, GFP_KERNEL); if (!attrs->sort_key_str) { ret = -ENOMEM; goto out; @@ -2135,12 +2969,8 @@ static int parse_assignment(struct trace_array *tr, ret = -ENOMEM; goto out; } - } else if (str_has_prefix(str, "clock=")) { - strsep(&str, "="); - if (!str) { - ret = -EINVAL; - goto out; - } + } else if ((len = str_has_prefix(str, "clock="))) { + str += len; str = strstrip(str); attrs->clock = kstrdup(str, GFP_KERNEL); @@ -2148,8 +2978,8 @@ static int parse_assignment(struct trace_array *tr, ret = -ENOMEM; goto out; } - } else if (str_has_prefix(str, "size=")) { - int map_bits = parse_map_size(str); + } else if ((len = str_has_prefix(str, "size="))) { + int map_bits = parse_map_size(str + len); if (map_bits < 0) { ret = map_bits; @@ -2189,8 +3019,15 @@ parse_hist_trigger_attrs(struct trace_array *tr, char *trigger_str) while (trigger_str) { char *str = strsep(&trigger_str, ":"); + char *rhs; - if (strchr(str, '=')) { + rhs = strchr(str, '='); + if (rhs) { + if (!strlen(++rhs)) { + ret = -EINVAL; + hist_err(tr, HIST_ERR_EMPTY_ASSIGNMENT, errpos(str)); + goto free; + } ret = parse_assignment(tr, str, attrs); if (ret) goto free; @@ -2661,6 +3498,22 @@ static int init_var_ref(struct hist_field *ref_field, goto out; } +static int find_var_ref_idx(struct hist_trigger_data *hist_data, + struct hist_field *var_field) +{ + struct hist_field *ref_field; + int i; + + for (i = 0; i < hist_data->n_var_refs; i++) { + ref_field = hist_data->var_refs[i]; + if (ref_field->var.idx == var_field->var.idx && + ref_field->var.hist_data == var_field->hist_data) + return i; + } + + return -ENOENT; +} + /** * create_var_ref - Create a variable reference and attach it to trigger * @hist_data: The trigger that will be referencing the variable @@ -4146,8 +4999,11 @@ static int check_synth_field(struct synth_event *event, field = event->fields[field_pos]; - if (strcmp(field->type, hist_field->type) != 0) - return -EINVAL; + if (strcmp(field->type, hist_field->type) != 0) { + if (field->size != hist_field->size || + field->is_signed != hist_field->is_signed) + return -EINVAL; + } return 0; } @@ -4234,11 +5090,11 @@ static int trace_action_create(struct hist_trigger_data *hist_data, struct trace_array *tr = hist_data->event_file->tr; char *event_name, *param, *system = NULL; struct hist_field *hist_field, *var_ref; - unsigned int i, var_ref_idx; + unsigned int i; unsigned int field_pos = 0; struct synth_event *event; char *synth_event_name; - int ret = 0; + int var_ref_idx, ret = 0; lockdep_assert_held(&event_mutex); @@ -4255,8 +5111,6 @@ static int trace_action_create(struct hist_trigger_data *hist_data, event->ref++; - var_ref_idx = hist_data->n_var_refs; - for (i = 0; i < data->n_params; i++) { char *p; @@ -4305,6 +5159,14 @@ static int trace_action_create(struct hist_trigger_data *hist_data, goto err; } + var_ref_idx = find_var_ref_idx(hist_data, var_ref); + if (WARN_ON(var_ref_idx < 0)) { + ret = var_ref_idx; + goto err; + } + + data->var_ref_idx[i] = var_ref_idx; + field_pos++; kfree(p); continue; @@ -4323,7 +5185,6 @@ static int trace_action_create(struct hist_trigger_data *hist_data, } data->synth_event = event; - data->var_ref_idx = var_ref_idx; out: return ret; err: @@ -4542,10 +5403,6 @@ static int create_val_fields(struct hist_trigger_data *hist_data, if (!fields_str) goto out; - strsep(&fields_str, "="); - if (!fields_str) - goto out; - for (i = 0, j = 1; i < TRACING_MAP_VALS_MAX && j < TRACING_MAP_VALS_MAX; i++) { field_str = strsep(&fields_str, ","); @@ -4640,10 +5497,6 @@ static int create_key_fields(struct hist_trigger_data *hist_data, if (!fields_str) goto out; - strsep(&fields_str, "="); - if (!fields_str) - goto out; - for (i = n_vals; i < n_vals + TRACING_MAP_KEYS_MAX; i++) { field_str = strsep(&fields_str, ","); if (!field_str) @@ -4775,7 +5628,7 @@ static int create_hist_fields(struct hist_trigger_data *hist_data, return ret; } -static int is_descending(const char *str) +static int is_descending(struct trace_array *tr, const char *str) { if (!str) return 0; @@ -4786,11 +5639,14 @@ static int is_descending(const char *str) if (strcmp(str, "ascending") == 0) return 0; + hist_err(tr, HIST_ERR_INVALID_SORT_MODIFIER, errpos((char *)str)); + return -EINVAL; } static int create_sort_keys(struct hist_trigger_data *hist_data) { + struct trace_array *tr = hist_data->event_file->tr; char *fields_str = hist_data->attrs->sort_key_str; struct tracing_map_sort_key *sort_key; int descending, ret = 0; @@ -4801,12 +5657,6 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) if (!fields_str) goto out; - strsep(&fields_str, "="); - if (!fields_str) { - ret = -EINVAL; - goto out; - } - for (i = 0; i < TRACING_MAP_SORT_KEYS_MAX; i++) { struct hist_field *hist_field; char *field_str, *field_name; @@ -4815,25 +5665,30 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) sort_key = &hist_data->sort_keys[i]; field_str = strsep(&fields_str, ","); - if (!field_str) { - if (i == 0) - ret = -EINVAL; + if (!field_str) + break; + + if (!*field_str) { + ret = -EINVAL; + hist_err(tr, HIST_ERR_EMPTY_SORT_FIELD, errpos("sort=")); break; } if ((i == TRACING_MAP_SORT_KEYS_MAX - 1) && fields_str) { + hist_err(tr, HIST_ERR_TOO_MANY_SORT_FIELDS, errpos("sort=")); ret = -EINVAL; break; } field_name = strsep(&field_str, "."); - if (!field_name) { + if (!field_name || !*field_name) { ret = -EINVAL; + hist_err(tr, HIST_ERR_EMPTY_SORT_FIELD, errpos("sort=")); break; } if (strcmp(field_name, "hitcount") == 0) { - descending = is_descending(field_str); + descending = is_descending(tr, field_str); if (descending < 0) { ret = descending; break; @@ -4855,7 +5710,7 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) if (strcmp(field_name, test_name) == 0) { sort_key->field_idx = idx; - descending = is_descending(field_str); + descending = is_descending(tr, field_str); if (descending < 0) { ret = descending; goto out; @@ -4866,6 +5721,7 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) } if (j == hist_data->n_fields) { ret = -EINVAL; + hist_err(tr, HIST_ERR_INVALID_SORT_FIELD, errpos(field_name)); break; } } diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index 40106fff06a4..dd34a1b46a86 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -116,9 +116,10 @@ static void *trigger_next(struct seq_file *m, void *t, loff_t *pos) { struct trace_event_file *event_file = event_file_data(m->private); - if (t == SHOW_AVAILABLE_TRIGGERS) + if (t == SHOW_AVAILABLE_TRIGGERS) { + (*pos)++; return NULL; - + } return seq_list_next(t, &event_file->triggers, pos); } @@ -213,7 +214,7 @@ static int event_trigger_regex_open(struct inode *inode, struct file *file) return ret; } -static int trigger_process_regex(struct trace_event_file *file, char *buff) +int trigger_process_regex(struct trace_event_file *file, char *buff) { char *command, *next = buff; struct event_command *p; diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index b611cd36e22d..8a4c8d5c2c98 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -101,7 +101,7 @@ static int function_trace_init(struct trace_array *tr) ftrace_init_array_ops(tr, func); - tr->trace_buffer.cpu = get_cpu(); + tr->array_buffer.cpu = get_cpu(); put_cpu(); tracing_start_cmdline_record(); @@ -118,7 +118,7 @@ static void function_trace_reset(struct trace_array *tr) static void function_trace_start(struct trace_array *tr) { - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); } static void @@ -143,7 +143,7 @@ function_trace_call(unsigned long ip, unsigned long parent_ip, goto out; cpu = smp_processor_id(); - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); if (!atomic_read(&data->disabled)) { local_save_flags(flags); trace_function(tr, ip, parent_ip, flags, pc); @@ -192,7 +192,7 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, */ local_irq_save(flags); cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 78af97163147..7d71546ba00a 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -101,7 +101,7 @@ int __trace_graph_entry(struct trace_array *tr, { struct trace_event_call *call = &event_funcgraph_entry; struct ring_buffer_event *event; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ftrace_graph_ent_entry *entry; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, @@ -171,7 +171,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) local_irq_save(flags); cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { pc = preempt_count(); @@ -221,7 +221,7 @@ void __trace_graph_return(struct trace_array *tr, { struct trace_event_call *call = &event_funcgraph_exit; struct ring_buffer_event *event; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ftrace_graph_ret_entry *entry; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RET, @@ -252,7 +252,7 @@ void trace_graph_return(struct ftrace_graph_ret *trace) local_irq_save(flags); cpu = raw_smp_processor_id(); - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { pc = preempt_count(); @@ -444,9 +444,9 @@ get_return_for_leaf(struct trace_iterator *iter, * We need to consume the current entry to see * the next one. */ - ring_buffer_consume(iter->trace_buffer->buffer, iter->cpu, + ring_buffer_consume(iter->array_buffer->buffer, iter->cpu, NULL, NULL); - event = ring_buffer_peek(iter->trace_buffer->buffer, iter->cpu, + event = ring_buffer_peek(iter->array_buffer->buffer, iter->cpu, NULL, NULL); } @@ -503,7 +503,7 @@ print_graph_rel_time(struct trace_iterator *iter, struct trace_seq *s) { unsigned long long usecs; - usecs = iter->ts - iter->trace_buffer->time_start; + usecs = iter->ts - iter->array_buffer->time_start; do_div(usecs, NSEC_PER_USEC); trace_seq_printf(s, "%9llu us | ", usecs); diff --git a/kernel/trace/trace_hwlat.c b/kernel/trace/trace_hwlat.c index 402d022d0229..a48808c43249 100644 --- a/kernel/trace/trace_hwlat.c +++ b/kernel/trace/trace_hwlat.c @@ -104,7 +104,7 @@ static void trace_hwlat_sample(struct hwlat_sample *sample) { struct trace_array *tr = hwlat_trace; struct trace_event_call *call = &event_hwlat; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct hwlat_entry *entry; unsigned long flags; diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index a745b0cee5d3..10bbb0f381d5 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -122,7 +122,7 @@ static int func_prolog_dec(struct trace_array *tr, if (!irqs_disabled_flags(*flags) && !preempt_count()) return 0; - *data = per_cpu_ptr(tr->trace_buffer.data, cpu); + *data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&(*data)->disabled); if (likely(disabled == 1)) @@ -167,7 +167,7 @@ static int irqsoff_display_graph(struct trace_array *tr, int set) per_cpu(tracing_cpu, cpu) = 0; tr->max_latency = 0; - tracing_reset_online_cpus(&irqsoff_trace->trace_buffer); + tracing_reset_online_cpus(&irqsoff_trace->array_buffer); return start_irqsoff_tracer(irqsoff_trace, set); } @@ -382,7 +382,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) if (per_cpu(tracing_cpu, cpu)) return; - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); if (unlikely(!data) || atomic_read(&data->disabled)) return; @@ -420,7 +420,7 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) if (!tracer_enabled || !tracing_is_enabled()) return; - data = per_cpu_ptr(tr->trace_buffer.data, cpu); + data = per_cpu_ptr(tr->array_buffer.data, cpu); if (unlikely(!data) || !data->critical_start || atomic_read(&data->disabled)) diff --git a/kernel/trace/trace_kdb.c b/kernel/trace/trace_kdb.c index cca65044c14c..9da76104f7a2 100644 --- a/kernel/trace/trace_kdb.c +++ b/kernel/trace/trace_kdb.c @@ -43,7 +43,7 @@ static void ftrace_dump_buf(int skip_entries, long cpu_file) if (cpu_file == RING_BUFFER_ALL_CPUS) { for_each_tracing_cpu(cpu) { iter.buffer_iter[cpu] = - ring_buffer_read_prepare(iter.trace_buffer->buffer, + ring_buffer_read_prepare(iter.array_buffer->buffer, cpu, GFP_ATOMIC); ring_buffer_read_start(iter.buffer_iter[cpu]); tracing_iter_reset(&iter, cpu); @@ -51,7 +51,7 @@ static void ftrace_dump_buf(int skip_entries, long cpu_file) } else { iter.cpu_file = cpu_file; iter.buffer_iter[cpu_file] = - ring_buffer_read_prepare(iter.trace_buffer->buffer, + ring_buffer_read_prepare(iter.array_buffer->buffer, cpu_file, GFP_ATOMIC); ring_buffer_read_start(iter.buffer_iter[cpu_file]); tracing_iter_reset(&iter, cpu_file); @@ -124,7 +124,7 @@ static int kdb_ftdump(int argc, const char **argv) iter.buffer_iter = buffer_iter; for_each_tracing_cpu(cpu) { - atomic_inc(&per_cpu_ptr(iter.trace_buffer->data, cpu)->disabled); + atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } /* A negative skip_entries means skip all but the last entries */ @@ -139,7 +139,7 @@ static int kdb_ftdump(int argc, const char **argv) ftrace_dump_buf(skip_entries, cpu_file); for_each_tracing_cpu(cpu) { - atomic_dec(&per_cpu_ptr(iter.trace_buffer->data, cpu)->disabled); + atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } kdb_trap_printk--; diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index aa515d578c5b..d8264ebb9581 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -22,7 +22,6 @@ #define KPROBE_EVENT_SYSTEM "kprobes" #define KRETPROBE_MAXACTIVE_MAX 4096 -#define MAX_KPROBE_CMDLINE_SIZE 1024 /* Kprobe early definition from command line */ static char kprobe_boot_events_buf[COMMAND_LINE_SIZE] __initdata; @@ -902,6 +901,167 @@ static int create_or_delete_trace_kprobe(int argc, char **argv) return ret == -ECANCELED ? -EINVAL : ret; } +static int trace_kprobe_run_command(struct dynevent_cmd *cmd) +{ + return trace_run_command(cmd->seq.buffer, create_or_delete_trace_kprobe); +} + +/** + * kprobe_event_cmd_init - Initialize a kprobe event command object + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @buf: A pointer to the buffer used to build the command + * @maxlen: The length of the buffer passed in @buf + * + * Initialize a synthetic event command object. Use this before + * calling any of the other kprobe_event functions. + */ +void kprobe_event_cmd_init(struct dynevent_cmd *cmd, char *buf, int maxlen) +{ + dynevent_cmd_init(cmd, buf, maxlen, DYNEVENT_TYPE_KPROBE, + trace_kprobe_run_command); +} +EXPORT_SYMBOL_GPL(kprobe_event_cmd_init); + +/** + * __kprobe_event_gen_cmd_start - Generate a kprobe event command from arg list + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @name: The name of the kprobe event + * @loc: The location of the kprobe event + * @kretprobe: Is this a return probe? + * @args: Variable number of arg (pairs), one pair for each field + * + * NOTE: Users normally won't want to call this function directly, but + * rather use the kprobe_event_gen_cmd_start() wrapper, which automatically + * adds a NULL to the end of the arg list. If this function is used + * directly, make sure the last arg in the variable arg list is NULL. + * + * Generate a kprobe event command to be executed by + * kprobe_event_gen_cmd_end(). This function can be used to generate the + * complete command or only the first part of it; in the latter case, + * kprobe_event_add_fields() can be used to add more fields following this. + * + * Return: 0 if successful, error otherwise. + */ +int __kprobe_event_gen_cmd_start(struct dynevent_cmd *cmd, bool kretprobe, + const char *name, const char *loc, ...) +{ + char buf[MAX_EVENT_NAME_LEN]; + struct dynevent_arg arg; + va_list args; + int ret; + + if (cmd->type != DYNEVENT_TYPE_KPROBE) + return -EINVAL; + + if (kretprobe) + snprintf(buf, MAX_EVENT_NAME_LEN, "r:kprobes/%s", name); + else + snprintf(buf, MAX_EVENT_NAME_LEN, "p:kprobes/%s", name); + + ret = dynevent_str_add(cmd, buf); + if (ret) + return ret; + + dynevent_arg_init(&arg, 0); + arg.str = loc; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + return ret; + + va_start(args, loc); + for (;;) { + const char *field; + + field = va_arg(args, const char *); + if (!field) + break; + + if (++cmd->n_fields > MAX_TRACE_ARGS) { + ret = -EINVAL; + break; + } + + arg.str = field; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + break; + } + va_end(args); + + return ret; +} +EXPORT_SYMBOL_GPL(__kprobe_event_gen_cmd_start); + +/** + * __kprobe_event_add_fields - Add probe fields to a kprobe command from arg list + * @cmd: A pointer to the dynevent_cmd struct representing the new event + * @args: Variable number of arg (pairs), one pair for each field + * + * NOTE: Users normally won't want to call this function directly, but + * rather use the kprobe_event_add_fields() wrapper, which + * automatically adds a NULL to the end of the arg list. If this + * function is used directly, make sure the last arg in the variable + * arg list is NULL. + * + * Add probe fields to an existing kprobe command using a variable + * list of args. Fields are added in the same order they're listed. + * + * Return: 0 if successful, error otherwise. + */ +int __kprobe_event_add_fields(struct dynevent_cmd *cmd, ...) +{ + struct dynevent_arg arg; + va_list args; + int ret; + + if (cmd->type != DYNEVENT_TYPE_KPROBE) + return -EINVAL; + + dynevent_arg_init(&arg, 0); + + va_start(args, cmd); + for (;;) { + const char *field; + + field = va_arg(args, const char *); + if (!field) + break; + + if (++cmd->n_fields > MAX_TRACE_ARGS) { + ret = -EINVAL; + break; + } + + arg.str = field; + ret = dynevent_arg_add(cmd, &arg, NULL); + if (ret) + break; + } + va_end(args); + + return ret; +} +EXPORT_SYMBOL_GPL(__kprobe_event_add_fields); + +/** + * kprobe_event_delete - Delete a kprobe event + * @name: The name of the kprobe event to delete + * + * Delete a kprobe event with the give @name from kernel code rather + * than directly from the command line. + * + * Return: 0 if successful, error otherwise. + */ +int kprobe_event_delete(const char *name) +{ + char buf[MAX_EVENT_NAME_LEN]; + + snprintf(buf, MAX_EVENT_NAME_LEN, "-:%s", name); + + return trace_run_command(buf, create_or_delete_trace_kprobe); +} +EXPORT_SYMBOL_GPL(kprobe_event_delete); + static int trace_kprobe_release(struct dyn_event *ev) { struct trace_kprobe *tk = to_trace_kprobe(ev); @@ -1175,35 +1335,35 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, struct trace_event_file *trace_file) { struct kprobe_trace_entry_head *entry; - struct ring_buffer_event *event; - struct ring_buffer *buffer; - int size, dsize, pc; - unsigned long irq_flags; struct trace_event_call *call = trace_probe_event_call(&tk->tp); + struct trace_event_buffer fbuffer; + int dsize; WARN_ON(call != trace_file->event_call); if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(irq_flags); - pc = preempt_count(); + local_save_flags(fbuffer.flags); + fbuffer.pc = preempt_count(); + fbuffer.trace_file = trace_file; dsize = __get_data_size(&tk->tp, regs); - size = sizeof(*entry) + tk->tp.size + dsize; - event = trace_event_buffer_lock_reserve(&buffer, trace_file, - call->event.type, - size, irq_flags, pc); - if (!event) + fbuffer.event = + trace_event_buffer_lock_reserve(&fbuffer.buffer, trace_file, + call->event.type, + sizeof(*entry) + tk->tp.size + dsize, + fbuffer.flags, fbuffer.pc); + if (!fbuffer.event) return; - entry = ring_buffer_event_data(event); + fbuffer.regs = regs; + entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); entry->ip = (unsigned long)tk->rp.kp.addr; store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); - event_trigger_unlock_commit_regs(trace_file, buffer, event, - entry, irq_flags, pc, regs); + trace_event_buffer_commit(&fbuffer); } static void @@ -1223,36 +1383,35 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct trace_event_file *trace_file) { struct kretprobe_trace_entry_head *entry; - struct ring_buffer_event *event; - struct ring_buffer *buffer; - int size, pc, dsize; - unsigned long irq_flags; + struct trace_event_buffer fbuffer; struct trace_event_call *call = trace_probe_event_call(&tk->tp); + int dsize; WARN_ON(call != trace_file->event_call); if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(irq_flags); - pc = preempt_count(); + local_save_flags(fbuffer.flags); + fbuffer.pc = preempt_count(); + fbuffer.trace_file = trace_file; dsize = __get_data_size(&tk->tp, regs); - size = sizeof(*entry) + tk->tp.size + dsize; - - event = trace_event_buffer_lock_reserve(&buffer, trace_file, - call->event.type, - size, irq_flags, pc); - if (!event) + fbuffer.event = + trace_event_buffer_lock_reserve(&fbuffer.buffer, trace_file, + call->event.type, + sizeof(*entry) + tk->tp.size + dsize, + fbuffer.flags, fbuffer.pc); + if (!fbuffer.event) return; - entry = ring_buffer_event_data(event); + fbuffer.regs = regs; + entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); entry->func = (unsigned long)tk->rp.kp.addr; entry->ret_ip = (unsigned long)ri->ret_addr; store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); - event_trigger_unlock_commit_regs(trace_file, buffer, event, - entry, irq_flags, pc, regs); + trace_event_buffer_commit(&fbuffer); } static void @@ -1698,11 +1857,12 @@ static __init void setup_boot_kprobe_events(void) enable_boot_kprobe_events(); } -/* Make a tracefs interface for controlling probe points */ -static __init int init_kprobe_trace(void) +/* + * Register dynevent at subsys_initcall. This allows kernel to setup kprobe + * events in fs_initcall without tracefs. + */ +static __init int init_kprobe_trace_early(void) { - struct dentry *d_tracer; - struct dentry *entry; int ret; ret = dyn_event_register(&trace_kprobe_ops); @@ -1712,6 +1872,16 @@ static __init int init_kprobe_trace(void) if (register_module_notifier(&trace_kprobe_module_nb)) return -EINVAL; + return 0; +} +subsys_initcall(init_kprobe_trace_early); + +/* Make a tracefs interface for controlling probe points */ +static __init int init_kprobe_trace(void) +{ + struct dentry *d_tracer; + struct dentry *entry; + d_tracer = tracing_init_dentry(); if (IS_ERR(d_tracer)) return 0; diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index b0388016b687..84582bf1ed5f 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -32,7 +32,7 @@ static void mmio_reset_data(struct trace_array *tr) overrun_detected = false; prev_overruns = 0; - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); } static int mmio_trace_init(struct trace_array *tr) @@ -122,7 +122,7 @@ static void mmio_close(struct trace_iterator *iter) static unsigned long count_overruns(struct trace_iterator *iter) { unsigned long cnt = atomic_xchg(&dropped_count, 0); - unsigned long over = ring_buffer_overruns(iter->trace_buffer->buffer); + unsigned long over = ring_buffer_overruns(iter->array_buffer->buffer); if (over > prev_overruns) cnt += over - prev_overruns; @@ -297,7 +297,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, struct mmiotrace_rw *rw) { struct trace_event_call *call = &event_mmiotrace_rw; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_rw *entry; int pc = preempt_count(); @@ -318,7 +318,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, void mmio_trace_rw(struct mmiotrace_rw *rw) { struct trace_array *tr = mmio_trace_array; - struct trace_array_cpu *data = per_cpu_ptr(tr->trace_buffer.data, smp_processor_id()); + struct trace_array_cpu *data = per_cpu_ptr(tr->array_buffer.data, smp_processor_id()); __trace_mmiotrace_rw(tr, data, rw); } @@ -327,7 +327,7 @@ static void __trace_mmiotrace_map(struct trace_array *tr, struct mmiotrace_map *map) { struct trace_event_call *call = &event_mmiotrace_map; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_map *entry; int pc = preempt_count(); @@ -351,7 +351,7 @@ void mmio_trace_mapping(struct mmiotrace_map *map) struct trace_array_cpu *data; preempt_disable(); - data = per_cpu_ptr(tr->trace_buffer.data, smp_processor_id()); + data = per_cpu_ptr(tr->array_buffer.data, smp_processor_id()); __trace_mmiotrace_map(tr, data, map); preempt_enable(); } diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index d9b4b7c22db4..b4909082f6a4 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -538,7 +538,7 @@ lat_print_timestamp(struct trace_iterator *iter, u64 next_ts) struct trace_array *tr = iter->tr; unsigned long verbose = tr->trace_flags & TRACE_ITER_VERBOSE; unsigned long in_ns = iter->iter_flags & TRACE_FILE_TIME_IN_NS; - unsigned long long abs_ts = iter->ts - iter->trace_buffer->time_start; + unsigned long long abs_ts = iter->ts - iter->array_buffer->time_start; unsigned long long rel_ts = next_ts - iter->ts; struct trace_seq *s = &iter->seq; diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index e288168661e1..e304196d7c28 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -89,8 +89,10 @@ static void tracing_sched_unregister(void) static void tracing_start_sched_switch(int ops) { - bool sched_register = (!sched_cmdline_ref && !sched_tgid_ref); + bool sched_register; + mutex_lock(&sched_register_mutex); + sched_register = (!sched_cmdline_ref && !sched_tgid_ref); switch (ops) { case RECORD_CMDLINE: diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 617e297f46dc..97b10bb31a1f 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -82,7 +82,7 @@ func_prolog_preempt_disable(struct trace_array *tr, if (cpu != wakeup_current_cpu) goto out_enable; - *data = per_cpu_ptr(tr->trace_buffer.data, cpu); + *data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&(*data)->disabled); if (unlikely(disabled != 1)) goto out; @@ -378,7 +378,7 @@ tracing_sched_switch_trace(struct trace_array *tr, unsigned long flags, int pc) { struct trace_event_call *call = &event_context_switch; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct ctx_switch_entry *entry; @@ -408,7 +408,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, struct trace_event_call *call = &event_wakeup; struct ring_buffer_event *event; struct ctx_switch_entry *entry; - struct ring_buffer *buffer = tr->trace_buffer.buffer; + struct trace_buffer *buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_WAKE, sizeof(*entry), flags, pc); @@ -459,7 +459,7 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, /* disable local data, not wakeup_cpu data */ cpu = raw_smp_processor_id(); - disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->trace_buffer.data, cpu)->disabled); + disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); if (likely(disabled != 1)) goto out; @@ -471,7 +471,7 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, goto out_unlock; /* The task we are waiting for is waking up */ - data = per_cpu_ptr(wakeup_trace->trace_buffer.data, wakeup_cpu); + data = per_cpu_ptr(wakeup_trace->array_buffer.data, wakeup_cpu); __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); @@ -494,7 +494,7 @@ out_unlock: arch_spin_unlock(&wakeup_lock); local_irq_restore(flags); out: - atomic_dec(&per_cpu_ptr(wakeup_trace->trace_buffer.data, cpu)->disabled); + atomic_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); } static void __wakeup_reset(struct trace_array *tr) @@ -513,7 +513,7 @@ static void wakeup_reset(struct trace_array *tr) { unsigned long flags; - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); local_irq_save(flags); arch_spin_lock(&wakeup_lock); @@ -551,7 +551,7 @@ probe_wakeup(void *ignore, struct task_struct *p) return; pc = preempt_count(); - disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->trace_buffer.data, cpu)->disabled); + disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); if (unlikely(disabled != 1)) goto out; @@ -583,7 +583,7 @@ probe_wakeup(void *ignore, struct task_struct *p) local_save_flags(flags); - data = per_cpu_ptr(wakeup_trace->trace_buffer.data, wakeup_cpu); + data = per_cpu_ptr(wakeup_trace->array_buffer.data, wakeup_cpu); data->preempt_timestamp = ftrace_now(cpu); tracing_sched_wakeup_trace(wakeup_trace, p, current, flags, pc); __trace_stack(wakeup_trace, flags, 0, pc); @@ -598,7 +598,7 @@ probe_wakeup(void *ignore, struct task_struct *p) out_locked: arch_spin_unlock(&wakeup_lock); out: - atomic_dec(&per_cpu_ptr(wakeup_trace->trace_buffer.data, cpu)->disabled); + atomic_dec(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); } static void start_wakeup_tracer(struct trace_array *tr) diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 69ee8ef12cee..b5e3496cf803 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -23,7 +23,7 @@ static inline int trace_valid_entry(struct trace_entry *entry) return 0; } -static int trace_test_buffer_cpu(struct trace_buffer *buf, int cpu) +static int trace_test_buffer_cpu(struct array_buffer *buf, int cpu) { struct ring_buffer_event *event; struct trace_entry *entry; @@ -60,7 +60,7 @@ static int trace_test_buffer_cpu(struct trace_buffer *buf, int cpu) * Test the trace buffer to see if all the elements * are still sane. */ -static int __maybe_unused trace_test_buffer(struct trace_buffer *buf, unsigned long *count) +static int __maybe_unused trace_test_buffer(struct array_buffer *buf, unsigned long *count) { unsigned long flags, cnt = 0; int cpu, ret = 0; @@ -362,7 +362,7 @@ static int trace_selftest_startup_dynamic_tracing(struct tracer *trace, msleep(100); /* we should have nothing in the buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); + ret = trace_test_buffer(&tr->array_buffer, &count); if (ret) goto out; @@ -383,7 +383,7 @@ static int trace_selftest_startup_dynamic_tracing(struct tracer *trace, ftrace_enabled = 0; /* check the trace buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); + ret = trace_test_buffer(&tr->array_buffer, &count); ftrace_enabled = 1; tracing_start(); @@ -682,7 +682,7 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) ftrace_enabled = 0; /* check the trace buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); + ret = trace_test_buffer(&tr->array_buffer, &count); ftrace_enabled = 1; trace->reset(tr); @@ -768,7 +768,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, * Simulate the init() callback but we attach a watchdog callback * to detect and recover from possible hangs */ - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(&tr->array_buffer); set_graph_array(tr); ret = register_ftrace_graph(&fgraph_ops); if (ret) { @@ -790,7 +790,7 @@ trace_selftest_startup_function_graph(struct tracer *trace, tracing_stop(); /* check the trace buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); + ret = trace_test_buffer(&tr->array_buffer, &count); /* Need to also simulate the tr->reset to remove this fgraph_ops */ tracing_stop_cmdline_record(); @@ -848,7 +848,7 @@ trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ - ret = trace_test_buffer(&tr->trace_buffer, NULL); + ret = trace_test_buffer(&tr->array_buffer, NULL); if (!ret) ret = trace_test_buffer(&tr->max_buffer, &count); trace->reset(tr); @@ -910,7 +910,7 @@ trace_selftest_startup_preemptoff(struct tracer *trace, struct trace_array *tr) /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ - ret = trace_test_buffer(&tr->trace_buffer, NULL); + ret = trace_test_buffer(&tr->array_buffer, NULL); if (!ret) ret = trace_test_buffer(&tr->max_buffer, &count); trace->reset(tr); @@ -976,7 +976,7 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ - ret = trace_test_buffer(&tr->trace_buffer, NULL); + ret = trace_test_buffer(&tr->array_buffer, NULL); if (ret) goto out; @@ -1006,7 +1006,7 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ - ret = trace_test_buffer(&tr->trace_buffer, NULL); + ret = trace_test_buffer(&tr->array_buffer, NULL); if (ret) goto out; @@ -1136,7 +1136,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) /* stop the tracing. */ tracing_stop(); /* check both trace buffers */ - ret = trace_test_buffer(&tr->trace_buffer, NULL); + ret = trace_test_buffer(&tr->array_buffer, NULL); if (!ret) ret = trace_test_buffer(&tr->max_buffer, &count); @@ -1177,7 +1177,7 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) /* stop the tracing. */ tracing_stop(); /* check the trace buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); + ret = trace_test_buffer(&tr->array_buffer, &count); trace->reset(tr); tracing_start(); diff --git a/kernel/trace/trace_seq.c b/kernel/trace/trace_seq.c index 87de6edafd14..1d84fcc78e3e 100644 --- a/kernel/trace/trace_seq.c +++ b/kernel/trace/trace_seq.c @@ -30,9 +30,6 @@ /* How much buffer is left on the trace_seq? */ #define TRACE_SEQ_BUF_LEFT(s) seq_buf_buffer_left(&(s)->seq) -/* How much buffer is written? */ -#define TRACE_SEQ_BUF_USED(s) seq_buf_used(&(s)->seq) - /* * trace_seq should work with being initialized with 0s. */ diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c index 874f1274cf99..d1fa19773cc8 100644 --- a/kernel/trace/trace_stat.c +++ b/kernel/trace/trace_stat.c @@ -280,18 +280,22 @@ static int tracing_stat_init(void) d_tracing = tracing_init_dentry(); if (IS_ERR(d_tracing)) - return 0; + return -ENODEV; stat_dir = tracefs_create_dir("trace_stat", d_tracing); - if (!stat_dir) + if (!stat_dir) { pr_warn("Could not create tracefs 'trace_stat' entry\n"); + return -ENOMEM; + } return 0; } static int init_stat_file(struct stat_session *session) { - if (!stat_dir && tracing_stat_init()) - return -ENODEV; + int ret; + + if (!stat_dir && (ret = tracing_stat_init())) + return ret; session->file = tracefs_create_file(session->ts->name, 0644, stat_dir, @@ -304,7 +308,7 @@ static int init_stat_file(struct stat_session *session) int register_stat_tracer(struct tracer_stat *trace) { struct stat_session *session, *node; - int ret; + int ret = -EINVAL; if (!trace) return -EINVAL; @@ -315,17 +319,15 @@ int register_stat_tracer(struct tracer_stat *trace) /* Already registered? */ mutex_lock(&all_stat_sessions_mutex); list_for_each_entry(node, &all_stat_sessions, session_list) { - if (node->ts == trace) { - mutex_unlock(&all_stat_sessions_mutex); - return -EINVAL; - } + if (node->ts == trace) + goto out; } - mutex_unlock(&all_stat_sessions_mutex); + ret = -ENOMEM; /* Init the session */ session = kzalloc(sizeof(*session), GFP_KERNEL); if (!session) - return -ENOMEM; + goto out; session->ts = trace; INIT_LIST_HEAD(&session->session_list); @@ -334,15 +336,16 @@ int register_stat_tracer(struct tracer_stat *trace) ret = init_stat_file(session); if (ret) { destroy_session(session); - return ret; + goto out; } + ret = 0; /* Register */ - mutex_lock(&all_stat_sessions_mutex); list_add_tail(&session->session_list, &all_stat_sessions); + out: mutex_unlock(&all_stat_sessions_mutex); - return 0; + return ret; } void unregister_stat_tracer(struct tracer_stat *trace) diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 2978c29d87d4..d85a2f0f316b 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -297,7 +297,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) struct syscall_trace_enter *entry; struct syscall_metadata *sys_data; struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long irq_flags; unsigned long args[6]; int pc; @@ -325,7 +325,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) local_save_flags(irq_flags); pc = preempt_count(); - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, sys_data->enter_event->event.type, size, irq_flags, pc); if (!event) @@ -347,7 +347,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) struct syscall_trace_exit *entry; struct syscall_metadata *sys_data; struct ring_buffer_event *event; - struct ring_buffer *buffer; + struct trace_buffer *buffer; unsigned long irq_flags; int pc; int syscall_nr; @@ -371,7 +371,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) local_save_flags(irq_flags); pc = preempt_count(); - buffer = tr->trace_buffer.buffer; + buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, sys_data->exit_event->event.type, sizeof(*entry), irq_flags, pc); diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 7885ebd23d0c..18d16f3ef980 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -931,8 +931,8 @@ static void __uprobe_trace_func(struct trace_uprobe *tu, struct trace_event_file *trace_file) { struct uprobe_trace_entry_head *entry; + struct trace_buffer *buffer; struct ring_buffer_event *event; - struct ring_buffer *buffer; void *data; int size, esize; struct trace_event_call *call = trace_probe_event_call(&tu->tp); diff --git a/lib/Kconfig b/lib/Kconfig index bc7e56370129..0cf875fd627c 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -573,6 +573,9 @@ config DIMLIB config LIBFDT bool +config LIBXBC + bool + config OID_REGISTRY tristate help diff --git a/lib/Makefile b/lib/Makefile index 23ca78d43d24..8bb91176c2a1 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -230,6 +230,8 @@ $(foreach file, $(libfdt_files), \ $(eval CFLAGS_$(file) = -I $(srctree)/scripts/dtc/libfdt)) lib-$(CONFIG_LIBFDT) += $(libfdt_files) +lib-$(CONFIG_LIBXBC) += bootconfig.o + obj-$(CONFIG_RBTREE_TEST) += rbtree_test.o obj-$(CONFIG_INTERVAL_TREE_TEST) += interval_tree_test.o diff --git a/lib/bootconfig.c b/lib/bootconfig.c new file mode 100644 index 000000000000..afb2e767e6fe --- /dev/null +++ b/lib/bootconfig.c @@ -0,0 +1,814 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Extra Boot Config + * Masami Hiramatsu <mhiramat@kernel.org> + */ + +#define pr_fmt(fmt) "bootconfig: " fmt + +#include <linux/bug.h> +#include <linux/ctype.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/printk.h> +#include <linux/bootconfig.h> +#include <linux/string.h> + +/* + * Extra Boot Config (XBC) is given as tree-structured ascii text of + * key-value pairs on memory. + * xbc_parse() parses the text to build a simple tree. Each tree node is + * simply a key word or a value. A key node may have a next key node or/and + * a child node (both key and value). A value node may have a next value + * node (for array). + */ + +static struct xbc_node xbc_nodes[XBC_NODE_MAX] __initdata; +static int xbc_node_num __initdata; +static char *xbc_data __initdata; +static size_t xbc_data_size __initdata; +static struct xbc_node *last_parent __initdata; + +static int __init xbc_parse_error(const char *msg, const char *p) +{ + int pos = p - xbc_data; + + pr_err("Parse error at pos %d: %s\n", pos, msg); + return -EINVAL; +} + +/** + * xbc_root_node() - Get the root node of extended boot config + * + * Return the address of root node of extended boot config. If the + * extended boot config is not initiized, return NULL. + */ +struct xbc_node * __init xbc_root_node(void) +{ + if (unlikely(!xbc_data)) + return NULL; + + return xbc_nodes; +} + +/** + * xbc_node_index() - Get the index of XBC node + * @node: A target node of getting index. + * + * Return the index number of @node in XBC node list. + */ +int __init xbc_node_index(struct xbc_node *node) +{ + return node - &xbc_nodes[0]; +} + +/** + * xbc_node_get_parent() - Get the parent XBC node + * @node: An XBC node. + * + * Return the parent node of @node. If the node is top node of the tree, + * return NULL. + */ +struct xbc_node * __init xbc_node_get_parent(struct xbc_node *node) +{ + return node->parent == XBC_NODE_MAX ? NULL : &xbc_nodes[node->parent]; +} + +/** + * xbc_node_get_child() - Get the child XBC node + * @node: An XBC node. + * + * Return the first child node of @node. If the node has no child, return + * NULL. + */ +struct xbc_node * __init xbc_node_get_child(struct xbc_node *node) +{ + return node->child ? &xbc_nodes[node->child] : NULL; +} + +/** + * xbc_node_get_next() - Get the next sibling XBC node + * @node: An XBC node. + * + * Return the NEXT sibling node of @node. If the node has no next sibling, + * return NULL. Note that even if this returns NULL, it doesn't mean @node + * has no siblings. (You also has to check whether the parent's child node + * is @node or not.) + */ +struct xbc_node * __init xbc_node_get_next(struct xbc_node *node) +{ + return node->next ? &xbc_nodes[node->next] : NULL; +} + +/** + * xbc_node_get_data() - Get the data of XBC node + * @node: An XBC node. + * + * Return the data (which is always a null terminated string) of @node. + * If the node has invalid data, warn and return NULL. + */ +const char * __init xbc_node_get_data(struct xbc_node *node) +{ + int offset = node->data & ~XBC_VALUE; + + if (WARN_ON(offset >= xbc_data_size)) + return NULL; + + return xbc_data + offset; +} + +static bool __init +xbc_node_match_prefix(struct xbc_node *node, const char **prefix) +{ + const char *p = xbc_node_get_data(node); + int len = strlen(p); + + if (strncmp(*prefix, p, len)) + return false; + + p = *prefix + len; + if (*p == '.') + p++; + else if (*p != '\0') + return false; + *prefix = p; + + return true; +} + +/** + * xbc_node_find_child() - Find a child node which matches given key + * @parent: An XBC node. + * @key: A key string. + * + * Search a node under @parent which matches @key. The @key can contain + * several words jointed with '.'. If @parent is NULL, this searches the + * node from whole tree. Return NULL if no node is matched. + */ +struct xbc_node * __init +xbc_node_find_child(struct xbc_node *parent, const char *key) +{ + struct xbc_node *node; + + if (parent) + node = xbc_node_get_child(parent); + else + node = xbc_root_node(); + + while (node && xbc_node_is_key(node)) { + if (!xbc_node_match_prefix(node, &key)) + node = xbc_node_get_next(node); + else if (*key != '\0') + node = xbc_node_get_child(node); + else + break; + } + + return node; +} + +/** + * xbc_node_find_value() - Find a value node which matches given key + * @parent: An XBC node. + * @key: A key string. + * @vnode: A container pointer of found XBC node. + * + * Search a value node under @parent whose (parent) key node matches @key, + * store it in *@vnode, and returns the value string. + * The @key can contain several words jointed with '.'. If @parent is NULL, + * this searches the node from whole tree. Return the value string if a + * matched key found, return NULL if no node is matched. + * Note that this returns 0-length string and stores NULL in *@vnode if the + * key has no value. And also it will return the value of the first entry if + * the value is an array. + */ +const char * __init +xbc_node_find_value(struct xbc_node *parent, const char *key, + struct xbc_node **vnode) +{ + struct xbc_node *node = xbc_node_find_child(parent, key); + + if (!node || !xbc_node_is_key(node)) + return NULL; + + node = xbc_node_get_child(node); + if (node && !xbc_node_is_value(node)) + return NULL; + + if (vnode) + *vnode = node; + + return node ? xbc_node_get_data(node) : ""; +} + +/** + * xbc_node_compose_key_after() - Compose partial key string of the XBC node + * @root: Root XBC node + * @node: Target XBC node. + * @buf: A buffer to store the key. + * @size: The size of the @buf. + * + * Compose the partial key of the @node into @buf, which is starting right + * after @root (@root is not included.) If @root is NULL, this returns full + * key words of @node. + * Returns the total length of the key stored in @buf. Returns -EINVAL + * if @node is NULL or @root is not the ancestor of @node or @root is @node, + * or returns -ERANGE if the key depth is deeper than max depth. + * This is expected to be used with xbc_find_node() to list up all (child) + * keys under given key. + */ +int __init xbc_node_compose_key_after(struct xbc_node *root, + struct xbc_node *node, + char *buf, size_t size) +{ + u16 keys[XBC_DEPTH_MAX]; + int depth = 0, ret = 0, total = 0; + + if (!node || node == root) + return -EINVAL; + + if (xbc_node_is_value(node)) + node = xbc_node_get_parent(node); + + while (node && node != root) { + keys[depth++] = xbc_node_index(node); + if (depth == XBC_DEPTH_MAX) + return -ERANGE; + node = xbc_node_get_parent(node); + } + if (!node && root) + return -EINVAL; + + while (--depth >= 0) { + node = xbc_nodes + keys[depth]; + ret = snprintf(buf, size, "%s%s", xbc_node_get_data(node), + depth ? "." : ""); + if (ret < 0) + return ret; + if (ret > size) { + size = 0; + } else { + size -= ret; + buf += ret; + } + total += ret; + } + + return total; +} + +/** + * xbc_node_find_next_leaf() - Find the next leaf node under given node + * @root: An XBC root node + * @node: An XBC node which starts from. + * + * Search the next leaf node (which means the terminal key node) of @node + * under @root node (including @root node itself). + * Return the next node or NULL if next leaf node is not found. + */ +struct xbc_node * __init xbc_node_find_next_leaf(struct xbc_node *root, + struct xbc_node *node) +{ + if (unlikely(!xbc_data)) + return NULL; + + if (!node) { /* First try */ + node = root; + if (!node) + node = xbc_nodes; + } else { + if (node == root) /* @root was a leaf, no child node. */ + return NULL; + + while (!node->next) { + node = xbc_node_get_parent(node); + if (node == root) + return NULL; + /* User passed a node which is not uder parent */ + if (WARN_ON(!node)) + return NULL; + } + node = xbc_node_get_next(node); + } + + while (node && !xbc_node_is_leaf(node)) + node = xbc_node_get_child(node); + + return node; +} + +/** + * xbc_node_find_next_key_value() - Find the next key-value pair nodes + * @root: An XBC root node + * @leaf: A container pointer of XBC node which starts from. + * + * Search the next leaf node (which means the terminal key node) of *@leaf + * under @root node. Returns the value and update *@leaf if next leaf node + * is found, or NULL if no next leaf node is found. + * Note that this returns 0-length string if the key has no value, or + * the value of the first entry if the value is an array. + */ +const char * __init xbc_node_find_next_key_value(struct xbc_node *root, + struct xbc_node **leaf) +{ + /* tip must be passed */ + if (WARN_ON(!leaf)) + return NULL; + + *leaf = xbc_node_find_next_leaf(root, *leaf); + if (!*leaf) + return NULL; + if ((*leaf)->child) + return xbc_node_get_data(xbc_node_get_child(*leaf)); + else + return ""; /* No value key */ +} + +/* XBC parse and tree build */ + +static struct xbc_node * __init xbc_add_node(char *data, u32 flag) +{ + struct xbc_node *node; + unsigned long offset; + + if (xbc_node_num == XBC_NODE_MAX) + return NULL; + + node = &xbc_nodes[xbc_node_num++]; + offset = data - xbc_data; + node->data = (u16)offset; + if (WARN_ON(offset >= XBC_DATA_MAX)) + return NULL; + node->data |= flag; + node->child = 0; + node->next = 0; + + return node; +} + +static inline __init struct xbc_node *xbc_last_sibling(struct xbc_node *node) +{ + while (node->next) + node = xbc_node_get_next(node); + + return node; +} + +static struct xbc_node * __init xbc_add_sibling(char *data, u32 flag) +{ + struct xbc_node *sib, *node = xbc_add_node(data, flag); + + if (node) { + if (!last_parent) { + node->parent = XBC_NODE_MAX; + sib = xbc_last_sibling(xbc_nodes); + sib->next = xbc_node_index(node); + } else { + node->parent = xbc_node_index(last_parent); + if (!last_parent->child) { + last_parent->child = xbc_node_index(node); + } else { + sib = xbc_node_get_child(last_parent); + sib = xbc_last_sibling(sib); + sib->next = xbc_node_index(node); + } + } + } else + xbc_parse_error("Too many nodes", data); + + return node; +} + +static inline __init struct xbc_node *xbc_add_child(char *data, u32 flag) +{ + struct xbc_node *node = xbc_add_sibling(data, flag); + + if (node) + last_parent = node; + + return node; +} + +static inline __init bool xbc_valid_keyword(char *key) +{ + if (key[0] == '\0') + return false; + + while (isalnum(*key) || *key == '-' || *key == '_') + key++; + + return *key == '\0'; +} + +static char *skip_comment(char *p) +{ + char *ret; + + ret = strchr(p, '\n'); + if (!ret) + ret = p + strlen(p); + else + ret++; + + return ret; +} + +static char *skip_spaces_until_newline(char *p) +{ + while (isspace(*p) && *p != '\n') + p++; + return p; +} + +static int __init __xbc_open_brace(void) +{ + /* Mark the last key as open brace */ + last_parent->next = XBC_NODE_MAX; + + return 0; +} + +static int __init __xbc_close_brace(char *p) +{ + struct xbc_node *node; + + if (!last_parent || last_parent->next != XBC_NODE_MAX) + return xbc_parse_error("Unexpected closing brace", p); + + node = last_parent; + node->next = 0; + do { + node = xbc_node_get_parent(node); + } while (node && node->next != XBC_NODE_MAX); + last_parent = node; + + return 0; +} + +/* + * Return delimiter or error, no node added. As same as lib/cmdline.c, + * you can use " around spaces, but can't escape " for value. + */ +static int __init __xbc_parse_value(char **__v, char **__n) +{ + char *p, *v = *__v; + int c, quotes = 0; + + v = skip_spaces(v); + while (*v == '#') { + v = skip_comment(v); + v = skip_spaces(v); + } + if (*v == '"' || *v == '\'') { + quotes = *v; + v++; + } + p = v - 1; + while ((c = *++p)) { + if (!isprint(c) && !isspace(c)) + return xbc_parse_error("Non printable value", p); + if (quotes) { + if (c != quotes) + continue; + quotes = 0; + *p++ = '\0'; + p = skip_spaces_until_newline(p); + c = *p; + if (c && !strchr(",;\n#}", c)) + return xbc_parse_error("No value delimiter", p); + if (*p) + p++; + break; + } + if (strchr(",;\n#}", c)) { + v = strim(v); + *p++ = '\0'; + break; + } + } + if (quotes) + return xbc_parse_error("No closing quotes", p); + if (c == '#') { + p = skip_comment(p); + c = '\n'; /* A comment must be treated as a newline */ + } + *__n = p; + *__v = v; + + return c; +} + +static int __init xbc_parse_array(char **__v) +{ + struct xbc_node *node; + char *next; + int c = 0; + + do { + c = __xbc_parse_value(__v, &next); + if (c < 0) + return c; + + node = xbc_add_sibling(*__v, XBC_VALUE); + if (!node) + return -ENOMEM; + *__v = next; + } while (c == ','); + node->next = 0; + + return c; +} + +static inline __init +struct xbc_node *find_match_node(struct xbc_node *node, char *k) +{ + while (node) { + if (!strcmp(xbc_node_get_data(node), k)) + break; + node = xbc_node_get_next(node); + } + return node; +} + +static int __init __xbc_add_key(char *k) +{ + struct xbc_node *node; + + if (!xbc_valid_keyword(k)) + return xbc_parse_error("Invalid keyword", k); + + if (unlikely(xbc_node_num == 0)) + goto add_node; + + if (!last_parent) /* the first level */ + node = find_match_node(xbc_nodes, k); + else + node = find_match_node(xbc_node_get_child(last_parent), k); + + if (node) + last_parent = node; + else { +add_node: + node = xbc_add_child(k, XBC_KEY); + if (!node) + return -ENOMEM; + } + return 0; +} + +static int __init __xbc_parse_keys(char *k) +{ + char *p; + int ret; + + k = strim(k); + while ((p = strchr(k, '.'))) { + *p++ = '\0'; + ret = __xbc_add_key(k); + if (ret) + return ret; + k = p; + } + + return __xbc_add_key(k); +} + +static int __init xbc_parse_kv(char **k, char *v) +{ + struct xbc_node *prev_parent = last_parent; + struct xbc_node *node; + char *next; + int c, ret; + + ret = __xbc_parse_keys(*k); + if (ret) + return ret; + + c = __xbc_parse_value(&v, &next); + if (c < 0) + return c; + + node = xbc_add_sibling(v, XBC_VALUE); + if (!node) + return -ENOMEM; + + if (c == ',') { /* Array */ + c = xbc_parse_array(&next); + if (c < 0) + return c; + } + + last_parent = prev_parent; + + if (c == '}') { + ret = __xbc_close_brace(next - 1); + if (ret < 0) + return ret; + } + + *k = next; + + return 0; +} + +static int __init xbc_parse_key(char **k, char *n) +{ + struct xbc_node *prev_parent = last_parent; + int ret; + + *k = strim(*k); + if (**k != '\0') { + ret = __xbc_parse_keys(*k); + if (ret) + return ret; + last_parent = prev_parent; + } + *k = n; + + return 0; +} + +static int __init xbc_open_brace(char **k, char *n) +{ + int ret; + + ret = __xbc_parse_keys(*k); + if (ret) + return ret; + *k = n; + + return __xbc_open_brace(); +} + +static int __init xbc_close_brace(char **k, char *n) +{ + int ret; + + ret = xbc_parse_key(k, n); + if (ret) + return ret; + /* k is updated in xbc_parse_key() */ + + return __xbc_close_brace(n - 1); +} + +static int __init xbc_verify_tree(void) +{ + int i, depth, len, wlen; + struct xbc_node *n, *m; + + /* Empty tree */ + if (xbc_node_num == 0) { + xbc_parse_error("Empty config", xbc_data); + return -ENOENT; + } + + for (i = 0; i < xbc_node_num; i++) { + if (xbc_nodes[i].next > xbc_node_num) { + return xbc_parse_error("No closing brace", + xbc_node_get_data(xbc_nodes + i)); + } + } + + /* Key tree limitation check */ + n = &xbc_nodes[0]; + depth = 1; + len = 0; + + while (n) { + wlen = strlen(xbc_node_get_data(n)) + 1; + len += wlen; + if (len > XBC_KEYLEN_MAX) + return xbc_parse_error("Too long key length", + xbc_node_get_data(n)); + + m = xbc_node_get_child(n); + if (m && xbc_node_is_key(m)) { + n = m; + depth++; + if (depth > XBC_DEPTH_MAX) + return xbc_parse_error("Too many key words", + xbc_node_get_data(n)); + continue; + } + len -= wlen; + m = xbc_node_get_next(n); + while (!m) { + n = xbc_node_get_parent(n); + if (!n) + break; + len -= strlen(xbc_node_get_data(n)) + 1; + depth--; + m = xbc_node_get_next(n); + } + n = m; + } + + return 0; +} + +/** + * xbc_destroy_all() - Clean up all parsed bootconfig + * + * This clears all data structures of parsed bootconfig on memory. + * If you need to reuse xbc_init() with new boot config, you can + * use this. + */ +void __init xbc_destroy_all(void) +{ + xbc_data = NULL; + xbc_data_size = 0; + xbc_node_num = 0; + memset(xbc_nodes, 0, sizeof(xbc_nodes)); +} + +/** + * xbc_init() - Parse given XBC file and build XBC internal tree + * @buf: boot config text + * + * This parses the boot config text in @buf. @buf must be a + * null terminated string and smaller than XBC_DATA_MAX. + * Return the number of stored nodes (>0) if succeeded, or -errno + * if there is any error. + */ +int __init xbc_init(char *buf) +{ + char *p, *q; + int ret, c; + + if (xbc_data) { + pr_err("Error: bootconfig is already initialized.\n"); + return -EBUSY; + } + + ret = strlen(buf); + if (ret > XBC_DATA_MAX - 1 || ret == 0) { + pr_err("Error: Config data is %s.\n", + ret ? "too big" : "empty"); + return -ERANGE; + } + + xbc_data = buf; + xbc_data_size = ret + 1; + last_parent = NULL; + + p = buf; + do { + q = strpbrk(p, "{}=;\n#"); + if (!q) { + p = skip_spaces(p); + if (*p != '\0') + ret = xbc_parse_error("No delimiter", p); + break; + } + + c = *q; + *q++ = '\0'; + switch (c) { + case '=': + ret = xbc_parse_kv(&p, q); + break; + case '{': + ret = xbc_open_brace(&p, q); + break; + case '#': + q = skip_comment(q); + /* fall through */ + case ';': + case '\n': + ret = xbc_parse_key(&p, q); + break; + case '}': + ret = xbc_close_brace(&p, q); + break; + } + } while (!ret); + + if (!ret) + ret = xbc_verify_tree(); + + if (ret < 0) + xbc_destroy_all(); + else + ret = xbc_node_num; + + return ret; +} + +/** + * xbc_debug_dump() - Dump current XBC node list + * + * Dump the current XBC node list on printk buffer for debug. + */ +void __init xbc_debug_dump(void) +{ + int i; + + for (i = 0; i < xbc_node_num; i++) { + pr_debug("[%d] %s (%s) .next=%d, .child=%d .parent=%d\n", i, + xbc_node_get_data(xbc_nodes + i), + xbc_node_is_value(xbc_nodes + i) ? "value" : "key", + xbc_nodes[i].next, xbc_nodes[i].child, + xbc_nodes[i].parent); + } +} diff --git a/tools/Makefile b/tools/Makefile index 7e42f7b8bfa7..bd778812e915 100644 --- a/tools/Makefile +++ b/tools/Makefile @@ -28,6 +28,7 @@ help: @echo ' pci - PCI tools' @echo ' perf - Linux performance measurement and analysis tool' @echo ' selftests - various kernel selftests' + @echo ' bootconfig - boot config tool' @echo ' spi - spi tools' @echo ' tmon - thermal monitoring and tuning tool' @echo ' turbostat - Intel CPU idle stats and freq reporting tool' @@ -63,7 +64,7 @@ acpi: FORCE cpupower: FORCE $(call descend,power/$@) -cgroup firewire hv guest spi usb virtio vm bpf iio gpio objtool leds wmi pci firmware debugging: FORCE +cgroup firewire hv guest bootconfig spi usb virtio vm bpf iio gpio objtool leds wmi pci firmware debugging: FORCE $(call descend,$@) liblockdep: FORCE @@ -96,7 +97,7 @@ kvm_stat: FORCE $(call descend,kvm/$@) all: acpi cgroup cpupower gpio hv firewire liblockdep \ - perf selftests spi turbostat usb \ + perf selftests bootconfig spi turbostat usb \ virtio vm bpf x86_energy_perf_policy \ tmon freefall iio objtool kvm_stat wmi \ pci debugging @@ -107,7 +108,7 @@ acpi_install: cpupower_install: $(call descend,power/$(@:_install=),install) -cgroup_install firewire_install gpio_install hv_install iio_install perf_install spi_install usb_install virtio_install vm_install bpf_install objtool_install wmi_install pci_install debugging_install: +cgroup_install firewire_install gpio_install hv_install iio_install perf_install bootconfig_install spi_install usb_install virtio_install vm_install bpf_install objtool_install wmi_install pci_install debugging_install: $(call descend,$(@:_install=),install) liblockdep_install: @@ -141,7 +142,7 @@ acpi_clean: cpupower_clean: $(call descend,power/cpupower,clean) -cgroup_clean hv_clean firewire_clean spi_clean usb_clean virtio_clean vm_clean wmi_clean bpf_clean iio_clean gpio_clean objtool_clean leds_clean pci_clean firmware_clean debugging_clean: +cgroup_clean hv_clean firewire_clean bootconfig_clean spi_clean usb_clean virtio_clean vm_clean wmi_clean bpf_clean iio_clean gpio_clean objtool_clean leds_clean pci_clean firmware_clean debugging_clean: $(call descend,$(@:_clean=),clean) liblockdep_clean: @@ -176,7 +177,7 @@ build_clean: $(call descend,build,clean) clean: acpi_clean cgroup_clean cpupower_clean hv_clean firewire_clean \ - perf_clean selftests_clean turbostat_clean spi_clean usb_clean virtio_clean \ + perf_clean selftests_clean turbostat_clean bootconfig_clean spi_clean usb_clean virtio_clean \ vm_clean bpf_clean iio_clean x86_energy_perf_policy_clean tmon_clean \ freefall_clean build_clean libbpf_clean libsubcmd_clean liblockdep_clean \ gpio_clean objtool_clean leds_clean wmi_clean pci_clean firmware_clean debugging_clean \ diff --git a/tools/bootconfig/.gitignore b/tools/bootconfig/.gitignore new file mode 100644 index 000000000000..e7644dfaa4a7 --- /dev/null +++ b/tools/bootconfig/.gitignore @@ -0,0 +1 @@ +bootconfig diff --git a/tools/bootconfig/Makefile b/tools/bootconfig/Makefile new file mode 100644 index 000000000000..a6146ac64458 --- /dev/null +++ b/tools/bootconfig/Makefile @@ -0,0 +1,23 @@ +# SPDX-License-Identifier: GPL-2.0 +# Makefile for bootconfig command + +bindir ?= /usr/bin + +HEADER = include/linux/bootconfig.h +CFLAGS = -Wall -g -I./include + +PROGS = bootconfig + +all: $(PROGS) + +bootconfig: ../../lib/bootconfig.c main.c $(HEADER) + $(CC) $(filter %.c,$^) $(CFLAGS) -o $@ + +install: $(PROGS) + install bootconfig $(DESTDIR)$(bindir) + +test: bootconfig + ./test-bootconfig.sh + +clean: + $(RM) -f *.o bootconfig diff --git a/tools/bootconfig/include/linux/bootconfig.h b/tools/bootconfig/include/linux/bootconfig.h new file mode 100644 index 000000000000..078cbd2ba651 --- /dev/null +++ b/tools/bootconfig/include/linux/bootconfig.h @@ -0,0 +1,7 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _BOOTCONFIG_LINUX_BOOTCONFIG_H +#define _BOOTCONFIG_LINUX_BOOTCONFIG_H + +#include "../../../../include/linux/bootconfig.h" + +#endif diff --git a/tools/bootconfig/include/linux/bug.h b/tools/bootconfig/include/linux/bug.h new file mode 100644 index 000000000000..7b65a389c0dd --- /dev/null +++ b/tools/bootconfig/include/linux/bug.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_BUG_H +#define _SKC_LINUX_BUG_H + +#include <stdio.h> +#include <stdlib.h> + +#define WARN_ON(cond) \ + ((cond) ? printf("Internal warning(%s:%d, %s): %s\n", \ + __FILE__, __LINE__, __func__, #cond) : 0) + +#endif diff --git a/tools/bootconfig/include/linux/ctype.h b/tools/bootconfig/include/linux/ctype.h new file mode 100644 index 000000000000..c56ecc136448 --- /dev/null +++ b/tools/bootconfig/include/linux/ctype.h @@ -0,0 +1,7 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_CTYPE_H +#define _SKC_LINUX_CTYPE_H + +#include <ctype.h> + +#endif diff --git a/tools/bootconfig/include/linux/errno.h b/tools/bootconfig/include/linux/errno.h new file mode 100644 index 000000000000..5d9f91ec2fda --- /dev/null +++ b/tools/bootconfig/include/linux/errno.h @@ -0,0 +1,7 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_ERRNO_H +#define _SKC_LINUX_ERRNO_H + +#include <asm/errno.h> + +#endif diff --git a/tools/bootconfig/include/linux/kernel.h b/tools/bootconfig/include/linux/kernel.h new file mode 100644 index 000000000000..2d93320aa374 --- /dev/null +++ b/tools/bootconfig/include/linux/kernel.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_KERNEL_H +#define _SKC_LINUX_KERNEL_H + +#include <stdlib.h> +#include <stdbool.h> + +#include <linux/printk.h> + +typedef unsigned short u16; +typedef unsigned int u32; + +#define unlikely(cond) (cond) + +#define __init +#define __initdata + +#endif diff --git a/tools/bootconfig/include/linux/printk.h b/tools/bootconfig/include/linux/printk.h new file mode 100644 index 000000000000..017bcd6912a5 --- /dev/null +++ b/tools/bootconfig/include/linux/printk.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_PRINTK_H +#define _SKC_LINUX_PRINTK_H + +#include <stdio.h> + +/* controllable printf */ +extern int pr_output; +#define printk(fmt, ...) \ + (pr_output ? printf(fmt, __VA_ARGS__) : 0) + +#define pr_err printk +#define pr_warn printk +#define pr_info printk +#define pr_debug printk + +#endif diff --git a/tools/bootconfig/include/linux/string.h b/tools/bootconfig/include/linux/string.h new file mode 100644 index 000000000000..8267af75153a --- /dev/null +++ b/tools/bootconfig/include/linux/string.h @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _SKC_LINUX_STRING_H +#define _SKC_LINUX_STRING_H + +#include <string.h> + +/* Copied from lib/string.c */ +static inline char *skip_spaces(const char *str) +{ + while (isspace(*str)) + ++str; + return (char *)str; +} + +static inline char *strim(char *s) +{ + size_t size; + char *end; + + size = strlen(s); + if (!size) + return s; + + end = s + size - 1; + while (end >= s && isspace(*end)) + end--; + *(end + 1) = '\0'; + + return skip_spaces(s); +} + +#endif diff --git a/tools/bootconfig/main.c b/tools/bootconfig/main.c new file mode 100644 index 000000000000..47f488458328 --- /dev/null +++ b/tools/bootconfig/main.c @@ -0,0 +1,354 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Boot config tool for initrd image + */ +#include <stdio.h> +#include <stdlib.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> + +#include <linux/kernel.h> +#include <linux/bootconfig.h> + +int pr_output = 1; + +static int xbc_show_array(struct xbc_node *node) +{ + const char *val; + int i = 0; + + xbc_array_for_each_value(node, val) { + printf("\"%s\"%s", val, node->next ? ", " : ";\n"); + i++; + } + return i; +} + +static void xbc_show_compact_tree(void) +{ + struct xbc_node *node, *cnode; + int depth = 0, i; + + node = xbc_root_node(); + while (node && xbc_node_is_key(node)) { + for (i = 0; i < depth; i++) + printf("\t"); + cnode = xbc_node_get_child(node); + while (cnode && xbc_node_is_key(cnode) && !cnode->next) { + printf("%s.", xbc_node_get_data(node)); + node = cnode; + cnode = xbc_node_get_child(node); + } + if (cnode && xbc_node_is_key(cnode)) { + printf("%s {\n", xbc_node_get_data(node)); + depth++; + node = cnode; + continue; + } else if (cnode && xbc_node_is_value(cnode)) { + printf("%s = ", xbc_node_get_data(node)); + if (cnode->next) + xbc_show_array(cnode); + else + printf("\"%s\";\n", xbc_node_get_data(cnode)); + } else { + printf("%s;\n", xbc_node_get_data(node)); + } + + if (node->next) { + node = xbc_node_get_next(node); + continue; + } + while (!node->next) { + node = xbc_node_get_parent(node); + if (!node) + return; + if (!xbc_node_get_child(node)->next) + continue; + depth--; + for (i = 0; i < depth; i++) + printf("\t"); + printf("}\n"); + } + node = xbc_node_get_next(node); + } +} + +/* Simple real checksum */ +int checksum(unsigned char *buf, int len) +{ + int i, sum = 0; + + for (i = 0; i < len; i++) + sum += buf[i]; + + return sum; +} + +#define PAGE_SIZE 4096 + +int load_xbc_fd(int fd, char **buf, int size) +{ + int ret; + + *buf = malloc(size + 1); + if (!*buf) + return -ENOMEM; + + ret = read(fd, *buf, size); + if (ret < 0) + return -errno; + (*buf)[size] = '\0'; + + return ret; +} + +/* Return the read size or -errno */ +int load_xbc_file(const char *path, char **buf) +{ + struct stat stat; + int fd, ret; + + fd = open(path, O_RDONLY); + if (fd < 0) + return -errno; + ret = fstat(fd, &stat); + if (ret < 0) + return -errno; + + ret = load_xbc_fd(fd, buf, stat.st_size); + + close(fd); + + return ret; +} + +int load_xbc_from_initrd(int fd, char **buf) +{ + struct stat stat; + int ret; + u32 size = 0, csum = 0, rcsum; + + ret = fstat(fd, &stat); + if (ret < 0) + return -errno; + + if (stat.st_size < 8) + return 0; + + if (lseek(fd, -8, SEEK_END) < 0) { + printf("Failed to lseek: %d\n", -errno); + return -errno; + } + + if (read(fd, &size, sizeof(u32)) < 0) + return -errno; + + if (read(fd, &csum, sizeof(u32)) < 0) + return -errno; + + /* Wrong size, maybe no boot config here */ + if (stat.st_size < size + 8) + return 0; + + if (lseek(fd, stat.st_size - 8 - size, SEEK_SET) < 0) { + printf("Failed to lseek: %d\n", -errno); + return -errno; + } + + ret = load_xbc_fd(fd, buf, size); + if (ret < 0) + return ret; + + /* Wrong Checksum, maybe no boot config here */ + rcsum = checksum((unsigned char *)*buf, size); + if (csum != rcsum) { + printf("checksum error: %d != %d\n", csum, rcsum); + return 0; + } + + ret = xbc_init(*buf); + /* Wrong data, maybe no boot config here */ + if (ret < 0) + return 0; + + return size; +} + +int show_xbc(const char *path) +{ + int ret, fd; + char *buf = NULL; + + fd = open(path, O_RDONLY); + if (fd < 0) { + printf("Failed to open initrd %s: %d\n", path, fd); + return -errno; + } + + ret = load_xbc_from_initrd(fd, &buf); + if (ret < 0) + printf("Failed to load a boot config from initrd: %d\n", ret); + else + xbc_show_compact_tree(); + + close(fd); + free(buf); + + return ret; +} + +int delete_xbc(const char *path) +{ + struct stat stat; + int ret = 0, fd, size; + char *buf = NULL; + + fd = open(path, O_RDWR); + if (fd < 0) { + printf("Failed to open initrd %s: %d\n", path, fd); + return -errno; + } + + /* + * Suppress error messages in xbc_init() because it can be just a + * data which concidentally matches the size and checksum footer. + */ + pr_output = 0; + size = load_xbc_from_initrd(fd, &buf); + pr_output = 1; + if (size < 0) { + ret = size; + printf("Failed to load a boot config from initrd: %d\n", ret); + } else if (size > 0) { + ret = fstat(fd, &stat); + if (!ret) + ret = ftruncate(fd, stat.st_size - size - 8); + if (ret) + ret = -errno; + } /* Ignore if there is no boot config in initrd */ + + close(fd); + free(buf); + + return ret; +} + +int apply_xbc(const char *path, const char *xbc_path) +{ + u32 size, csum; + char *buf, *data; + int ret, fd; + + ret = load_xbc_file(xbc_path, &buf); + if (ret < 0) { + printf("Failed to load %s : %d\n", xbc_path, ret); + return ret; + } + size = strlen(buf) + 1; + csum = checksum((unsigned char *)buf, size); + + /* Prepare xbc_path data */ + data = malloc(size + 8); + if (!data) + return -ENOMEM; + strcpy(data, buf); + *(u32 *)(data + size) = size; + *(u32 *)(data + size + 4) = csum; + + /* Check the data format */ + ret = xbc_init(buf); + if (ret < 0) { + printf("Failed to parse %s: %d\n", xbc_path, ret); + free(data); + free(buf); + return ret; + } + printf("Apply %s to %s\n", xbc_path, path); + printf("\tNumber of nodes: %d\n", ret); + printf("\tSize: %u bytes\n", (unsigned int)size); + printf("\tChecksum: %d\n", (unsigned int)csum); + + /* TODO: Check the options by schema */ + xbc_destroy_all(); + free(buf); + + /* Remove old boot config if exists */ + ret = delete_xbc(path); + if (ret < 0) { + printf("Failed to delete previous boot config: %d\n", ret); + return ret; + } + + /* Apply new one */ + fd = open(path, O_RDWR | O_APPEND); + if (fd < 0) { + printf("Failed to open %s: %d\n", path, fd); + return fd; + } + /* TODO: Ensure the @path is initramfs/initrd image */ + ret = write(fd, data, size + 8); + if (ret < 0) { + printf("Failed to apply a boot config: %d\n", ret); + return ret; + } + close(fd); + free(data); + + return 0; +} + +int usage(void) +{ + printf("Usage: bootconfig [OPTIONS] <INITRD>\n" + " Apply, delete or show boot config to initrd.\n" + " Options:\n" + " -a <config>: Apply boot config to initrd\n" + " -d : Delete boot config file from initrd\n\n" + " If no option is given, show current applied boot config.\n"); + return -1; +} + +int main(int argc, char **argv) +{ + char *path = NULL; + char *apply = NULL; + bool delete = false; + int opt; + + while ((opt = getopt(argc, argv, "hda:")) != -1) { + switch (opt) { + case 'd': + delete = true; + break; + case 'a': + apply = optarg; + break; + case 'h': + default: + return usage(); + } + } + + if (apply && delete) { + printf("Error: You can not specify both -a and -d at once.\n"); + return usage(); + } + + if (optind >= argc) { + printf("Error: No initrd is specified.\n"); + return usage(); + } + + path = argv[optind]; + + if (apply) + return apply_xbc(path, apply); + else if (delete) + return delete_xbc(path); + + return show_xbc(path); +} diff --git a/tools/bootconfig/samples/bad-array-space-comment.bconf b/tools/bootconfig/samples/bad-array-space-comment.bconf new file mode 100644 index 000000000000..fda19e47d0db --- /dev/null +++ b/tools/bootconfig/samples/bad-array-space-comment.bconf @@ -0,0 +1,5 @@ +key = # comment + "value1", # comment1 + "value2" # comment2 +, + "value3" diff --git a/tools/bootconfig/samples/bad-array.bconf b/tools/bootconfig/samples/bad-array.bconf new file mode 100644 index 000000000000..0174af019d7f --- /dev/null +++ b/tools/bootconfig/samples/bad-array.bconf @@ -0,0 +1,2 @@ +# Array must be comma separated. +key = "value1" "value2" diff --git a/tools/bootconfig/samples/bad-dotword.bconf b/tools/bootconfig/samples/bad-dotword.bconf new file mode 100644 index 000000000000..ba5557b2bdd3 --- /dev/null +++ b/tools/bootconfig/samples/bad-dotword.bconf @@ -0,0 +1,4 @@ +# do not start keyword with . +key { + .word = 1 +} diff --git a/tools/bootconfig/samples/bad-empty.bconf b/tools/bootconfig/samples/bad-empty.bconf new file mode 100644 index 000000000000..2ba3f6cc6a47 --- /dev/null +++ b/tools/bootconfig/samples/bad-empty.bconf @@ -0,0 +1 @@ +# Wrong boot config: comment only diff --git a/tools/bootconfig/samples/bad-keyerror.bconf b/tools/bootconfig/samples/bad-keyerror.bconf new file mode 100644 index 000000000000..b6e247a099d0 --- /dev/null +++ b/tools/bootconfig/samples/bad-keyerror.bconf @@ -0,0 +1,2 @@ +# key word can not contain "," +key,word diff --git a/tools/bootconfig/samples/bad-longkey.bconf b/tools/bootconfig/samples/bad-longkey.bconf new file mode 100644 index 000000000000..eb97369f91a8 --- /dev/null +++ b/tools/bootconfig/samples/bad-longkey.bconf @@ -0,0 +1 @@ +key_word_is_too_long01234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345 diff --git a/tools/bootconfig/samples/bad-manywords.bconf b/tools/bootconfig/samples/bad-manywords.bconf new file mode 100644 index 000000000000..8db81967c48a --- /dev/null +++ b/tools/bootconfig/samples/bad-manywords.bconf @@ -0,0 +1 @@ +key1.is2.too3.long4.5.6.7.8.9.10.11.12.13.14.15.16.17 diff --git a/tools/bootconfig/samples/bad-no-keyword.bconf b/tools/bootconfig/samples/bad-no-keyword.bconf new file mode 100644 index 000000000000..eff26808566c --- /dev/null +++ b/tools/bootconfig/samples/bad-no-keyword.bconf @@ -0,0 +1,2 @@ +# No keyword +{} diff --git a/tools/bootconfig/samples/bad-nonprintable.bconf b/tools/bootconfig/samples/bad-nonprintable.bconf new file mode 100644 index 000000000000..3bb1a2864e52 --- /dev/null +++ b/tools/bootconfig/samples/bad-nonprintable.bconf @@ -0,0 +1,2 @@ +# Non printable +key = "" diff --git a/tools/bootconfig/samples/bad-spaceword.bconf b/tools/bootconfig/samples/bad-spaceword.bconf new file mode 100644 index 000000000000..90c703d32a9a --- /dev/null +++ b/tools/bootconfig/samples/bad-spaceword.bconf @@ -0,0 +1,2 @@ +# No space between words +key . word diff --git a/tools/bootconfig/samples/bad-tree.bconf b/tools/bootconfig/samples/bad-tree.bconf new file mode 100644 index 000000000000..5a6038edcd55 --- /dev/null +++ b/tools/bootconfig/samples/bad-tree.bconf @@ -0,0 +1,5 @@ +# brace is not closing +tree { + node { + value = 1 +} diff --git a/tools/bootconfig/samples/bad-value.bconf b/tools/bootconfig/samples/bad-value.bconf new file mode 100644 index 000000000000..a1217fed86cc --- /dev/null +++ b/tools/bootconfig/samples/bad-value.bconf @@ -0,0 +1,3 @@ +# Quotes error +value = "data + diff --git a/tools/bootconfig/samples/escaped.bconf b/tools/bootconfig/samples/escaped.bconf new file mode 100644 index 000000000000..9f72043b3216 --- /dev/null +++ b/tools/bootconfig/samples/escaped.bconf @@ -0,0 +1,3 @@ +key1 = "A\B\C" +key2 = '\'\'' +key3 = "\\" diff --git a/tools/bootconfig/samples/good-array-space-comment.bconf b/tools/bootconfig/samples/good-array-space-comment.bconf new file mode 100644 index 000000000000..45b938dc0695 --- /dev/null +++ b/tools/bootconfig/samples/good-array-space-comment.bconf @@ -0,0 +1,4 @@ +key = # comment + "value1", # comment1 + "value2" , # comment2 + "value3" diff --git a/tools/bootconfig/samples/good-comment-after-value.bconf b/tools/bootconfig/samples/good-comment-after-value.bconf new file mode 100644 index 000000000000..0d92a853df72 --- /dev/null +++ b/tools/bootconfig/samples/good-comment-after-value.bconf @@ -0,0 +1 @@ +key = "value" # comment diff --git a/tools/bootconfig/samples/good-printables.bconf b/tools/bootconfig/samples/good-printables.bconf new file mode 100644 index 000000000000..ebb985a66ed8 --- /dev/null +++ b/tools/bootconfig/samples/good-printables.bconf @@ -0,0 +1,2 @@ +key = " + !#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~" diff --git a/tools/bootconfig/samples/good-simple.bconf b/tools/bootconfig/samples/good-simple.bconf new file mode 100644 index 000000000000..37dd6d21c176 --- /dev/null +++ b/tools/bootconfig/samples/good-simple.bconf @@ -0,0 +1,11 @@ +# A good simple bootconfig + +key.word1 = 1 +key.word2=2 +key.word3 = 3; + +key { +word4 = 4 } + +key { word5 = 5; word6 = 6 } + diff --git a/tools/bootconfig/samples/good-single.bconf b/tools/bootconfig/samples/good-single.bconf new file mode 100644 index 000000000000..98e55ad8b711 --- /dev/null +++ b/tools/bootconfig/samples/good-single.bconf @@ -0,0 +1,4 @@ +# single key style +key = 1 +key2 = 2 +key3 = "alpha", "beta" diff --git a/tools/bootconfig/samples/good-space-after-value.bconf b/tools/bootconfig/samples/good-space-after-value.bconf new file mode 100644 index 000000000000..56c15cbc5741 --- /dev/null +++ b/tools/bootconfig/samples/good-space-after-value.bconf @@ -0,0 +1 @@ +key = "value" diff --git a/tools/bootconfig/samples/good-tree.bconf b/tools/bootconfig/samples/good-tree.bconf new file mode 100644 index 000000000000..f2ddefc8b52a --- /dev/null +++ b/tools/bootconfig/samples/good-tree.bconf @@ -0,0 +1,12 @@ +key { + word { + tree { + value = "0"} + } + word2 { + tree { + value = 1,2 } + } +} +other.tree { + value = 2; value2 = 3;} diff --git a/tools/bootconfig/test-bootconfig.sh b/tools/bootconfig/test-bootconfig.sh new file mode 100755 index 000000000000..87725e8723f8 --- /dev/null +++ b/tools/bootconfig/test-bootconfig.sh @@ -0,0 +1,105 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0-only + +echo "Boot config test script" + +BOOTCONF=./bootconfig +INITRD=`mktemp initrd-XXXX` +TEMPCONF=`mktemp temp-XXXX.bconf` +NG=0 + +cleanup() { + rm -f $INITRD $TEMPCONF + exit $NG +} + +trap cleanup EXIT TERM + +NO=1 + +xpass() { # pass test command + echo "test case $NO ($3)... " + if ! ($@ && echo "\t\t[OK]"); then + echo "\t\t[NG]"; NG=$((NG + 1)) + fi + NO=$((NO + 1)) +} + +xfail() { # fail test command + echo "test case $NO ($3)... " + if ! (! $@ && echo "\t\t[OK]"); then + echo "\t\t[NG]"; NG=$((NG + 1)) + fi + NO=$((NO + 1)) +} + +echo "Basic command test" +xpass $BOOTCONF $INITRD + +echo "Delete command should success without bootconfig" +xpass $BOOTCONF -d $INITRD + +dd if=/dev/zero of=$INITRD bs=4096 count=1 +echo "key = value;" > $TEMPCONF +bconf_size=$(stat -c %s $TEMPCONF) +initrd_size=$(stat -c %s $INITRD) + +echo "Apply command test" +xpass $BOOTCONF -a $TEMPCONF $INITRD +new_size=$(stat -c %s $INITRD) + +echo "File size check" +xpass test $new_size -eq $(expr $bconf_size + $initrd_size + 9) + +echo "Apply command repeat test" +xpass $BOOTCONF -a $TEMPCONF $INITRD + +echo "File size check" +xpass test $new_size -eq $(stat -c %s $INITRD) + +echo "Delete command check" +xpass $BOOTCONF -d $INITRD + +echo "File size check" +new_size=$(stat -c %s $INITRD) +xpass test $new_size -eq $initrd_size + +echo "Max node number check" + +echo -n > $TEMPCONF +for i in `seq 1 1024` ; do + echo "node$i" >> $TEMPCONF +done +xpass $BOOTCONF -a $TEMPCONF $INITRD + +echo "badnode" >> $TEMPCONF +xfail $BOOTCONF -a $TEMPCONF $INITRD + +echo "Max filesize check" + +# Max size is 32767 (including terminal byte) +echo -n "data = \"" > $TEMPCONF +dd if=/dev/urandom bs=768 count=32 | base64 -w0 >> $TEMPCONF +echo "\"" >> $TEMPCONF +xfail $BOOTCONF -a $TEMPCONF $INITRD + +truncate -s 32764 $TEMPCONF +echo "\"" >> $TEMPCONF # add 2 bytes + terminal ('\"\n\0') +xpass $BOOTCONF -a $TEMPCONF $INITRD + +echo "=== expected failure cases ===" +for i in samples/bad-* ; do + xfail $BOOTCONF -a $i $INITRD +done + +echo "=== expected success cases ===" +for i in samples/good-* ; do + xpass $BOOTCONF -a $i $INITRD +done + +echo +if [ $NG -eq 0 ]; then + echo "All tests passed" +else + echo "$NG tests failed" +fi diff --git a/tools/testing/selftests/ftrace/test.d/trigger/trigger-hist-syntax-errors.tc b/tools/testing/selftests/ftrace/test.d/trigger/trigger-hist-syntax-errors.tc new file mode 100644 index 000000000000..d44087a2f3d1 --- /dev/null +++ b/tools/testing/selftests/ftrace/test.d/trigger/trigger-hist-syntax-errors.tc @@ -0,0 +1,32 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 +# description: event trigger - test histogram parser errors + +if [ ! -f set_event -o ! -d events/kmem ]; then + echo "event tracing is not supported" + exit_unsupported +fi + +if [ ! -f events/kmem/kmalloc/trigger ]; then + echo "event trigger is not supported" + exit_unsupported +fi + +if [ ! -f events/kmem/kmalloc/hist ]; then + echo "hist trigger is not supported" + exit_unsupported +fi + +[ -f error_log ] || exit_unsupported + +check_error() { # command-with-error-pos-by-^ + ftrace_errlog_check 'hist:kmem:kmalloc' "$1" 'events/kmem/kmalloc/trigger' +} + +check_error 'hist:keys=common_pid:vals=bytes_req:sort=common_pid,^junk' # INVALID_SORT_FIELD +check_error 'hist:keys=common_pid:vals=bytes_req:^sort=' # EMPTY_ASSIGNMENT +check_error 'hist:keys=common_pid:vals=bytes_req:^sort=common_pid,' # EMPTY_SORT_FIELD +check_error 'hist:keys=common_pid:vals=bytes_req:sort=common_pid.^junk' # INVALID_SORT_MODIFIER +check_error 'hist:keys=common_pid:vals=bytes_req,bytes_alloc:^sort=common_pid,bytes_req,bytes_alloc' # TOO_MANY_SORT_FIELDS + +exit 0 |