perf: perf interface for uprobes
in [Kernel]
|
Prev: xstat: Add a pair of system calls to make extended file stats available [ver #4]
Next: [PATCH] Fixed division by zero bug in kernel/padata.c
From: Srikar Dronamraju on 2 Jul 2010 07:40 > Masami Hiramatsu wrote: > > Srikar Dronamraju wrote: > >> perf: perf interface for uprobes > >> > >> Changelog from v4: Merged to 2.6.35-rc3-tip. > >> > >> Changelog from v3: (addressed comments from Masami Hiramatsu) > >> * Every process id has a different group name. > >> * event name starts with function name. > >> * If vaddr is specified, event name has vaddr appended > >> along with function name, (this is to avoid subsequent probes > >> using same event name.) > >> * warning if -p and --list options are used together. > >> > >> Also dso can either be a short name or absolute path. > >> > >> Enhances perf probe to accept pid and user vaddr. > >> Provides very basic support for uprobes. > > > > It's unsure what will happen if user sets a probe by line number... > > I'd like to suggest you if need_dwarf && uid, it should show an error message. > > But other parts are OK for me. > > I've checked that perf probe was stopped by SEGV in convert_name_to_addr() with > below options. > > # perf probe -u 2403 hoge.c:100 It should have been perf probe -p 2403 hode.c:100 However I have taken your comment and fixed that part of the code. So if it uprobe based probes and need_dwarf is set, it errors out. > > because pp->function == NULL. > > Thank you, -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Srikar Dronamraju on 9 Jul 2010 01:00 perf: perf interface for uprobes Changelog: Fixed a compilation issue reported by Christoph Hellwig. Changelog from v6: Changelog from v6: Fixed a bug reported by Masami. i.e Throw an error message and exit if perf probe is for a dwarf based probes. Changelog from v4: Merged to 2.6.35-rc3-tip. Changelog from v3: (addressed comments from Masami Hiramatsu) * Every process id has a different group name. * event name starts with function name. * If vaddr is specified, event name has vaddr appended along with function name, (this is to avoid subsequent probes using same event name.) * warning if -p and --list options are used together. Also dso can either be a short name or absolute path. Enhances perf probe to accept pid and user vaddr. Provides very basic support for uprobes. TODO: Update perf-probes.txt. Global tracing. Signed-off-by: Srikar Dronamraju <srikar(a)linux.vnet.ibm.com> --- Here is a terminal snapshot of placing, using and removing a probe on a process with pid 3591 (corresponding to zsh) [ Probing a function in the executable using function name ] ------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree(a)zsh Added new event: probe_3591:zfree (on 0x446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:zfree (on 3591:0x0000000000446420) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/zfree 3591:0x0000000000446420 [root(a)ABCD]# perf record -f -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.039 MB perf.data (~1716 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree Remove event: probe_3591:zfree [root(a)ABCD]# perf report # Samples: 447 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a function + offset ] ------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree(a)zsh+5 Added new event: probe_3591:zfree (on 0x446425) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:zfree (on 3591:0x0000000000446425) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/zfree 3591:0x0000000000446425 [root(a)ABCD]# perf record -f -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.036 MB perf.data (~1590 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree Remove event: probe_3591:zfree [root(a)ABCD]# perf report # Samples: 18 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a library function using function name ] -------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 write(a)libc-2.5.so Added new event: probe_3591:write (on 0x36010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:write -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:write (on 3591:0x00000036010c6060) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/write 3591:0x00000036010c6060 [root(a)ABCD]# perf record -f -e probe_3591:write -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1738 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:write Remove event: probe_3591:write [root(a)ABCD]# perf report # Samples: 11 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a library function using function name and absolute path ] --------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 write@/lib64/libc-2.5.so Added new event: probe_3591:write (on 0x36010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:write -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:write (on 3591:0x00000036010c6060) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/write 3591:0x00000036010c6060 [root(a)ABCD]# perf record -f -e probe_3591:write -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1738 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:write Remove event: probe_3591:write [root(a)ABCD]# perf report # Samples: 11 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing using vaddr 0x0000000000446420 (corresponding to zfree)] ------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 0x0000000000446420 Added new event: probe_3591:zfree_446420 (on 0x0000000000446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree_446420 -a sleep 1 [root(a)ABCD]# perf record -e probe_3591:zfree_446420 -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.041 MB perf.data (~1797 samples) ] [root(a)ABCD]# perf report # # Samples: 628 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# perf report --sort comm,dso # Samples: 628 # # Overhead Command Shared Object # ........ ............... ............. # 100.00% zsh zsh [root(a)ABCD]# perf probe --list probe_3591:zfree_446420 (on 3591:0x0000000000446420) [root(a)ABCD]# perf list | grep probe probe_3591:zfree_446420 [Tracepoint event] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree_446420 Remove event: probe_3591:zfree_446420 [root(a)ABCD]# Another example for a shared library: write stub in libc. (corresponds to 0x00000036010c6060) on a vaddr [ Probing a libc vaddr 0x00000036010c6060 (corresponding to write) ] [root(a)ABCD]# perf probe -p 3591 0x00000036010c6060 dded new event: probe_3591:__GI___libc_write_36010c6060 (on 0x00000036010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:__GI___libc_write_36010c6060 -a sleep 1 [root(a)ABCD]# perf record -f -e probe_3591:__GI___libc_write_36010c6060 -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1748 samples) ] [root(a)ABCD]# perf report # Samples: 24 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# [ Probing using a function without specifying a dso (corresponding to zfree)] ------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree Added new event: probe_3591:zfree (on 0x0000000000446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf record -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.041 MB perf.data (~1797 samples) ] [root(a)ABCD]# perf report # # Samples: 628 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# tools/perf/builtin-probe.c | 13 + tools/perf/builtin-top.c | 20 -- tools/perf/util/event.c | 20 ++ tools/perf/util/event.h | 1 tools/perf/util/probe-event.c | 520 +++++++++++++++++++++++++++++----------- tools/perf/util/probe-event.h | 37 +-- tools/perf/util/probe-finder.c | 34 +-- tools/perf/util/probe-finder.h | 10 - 8 files changed, 445 insertions(+), 210 deletions(-) diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c index 5455186..cb915a5 100644 --- a/tools/perf/builtin-probe.c +++ b/tools/perf/builtin-probe.c @@ -55,6 +55,7 @@ static struct { struct strlist *dellist; struct line_range line_range; int max_probe_points; + pid_t upid; } params; @@ -73,6 +74,7 @@ static int parse_probe_event(const char *str) /* Parse a perf-probe command into event */ ret = parse_perf_probe_command(str, pev); pr_debug("%d arguments\n", pev->nargs); + pev->upid = params.upid; return ret; } @@ -188,6 +190,8 @@ static const struct option options[] = { OPT__DRY_RUN(&probe_event_dry_run), OPT_INTEGER('\0', "max-probes", ¶ms.max_probe_points, "Set how many probe points can be found for a probe."), + OPT_INTEGER('p', "pid", ¶ms.upid, + "specify a pid for a uprobes based probe"), OPT_END() }; @@ -225,6 +229,10 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) pr_err(" Error: Don't use --list with --line.\n"); usage_with_options(probe_usage, options); } + if (params.upid) { + pr_warning(" Error: Don't use --list with --pid.\n"); + usage_with_options(probe_usage, options); + } ret = show_perf_probe_events(); if (ret < 0) pr_err(" Error: Failed to show event list. (%d)\n", @@ -233,7 +241,7 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) } #ifdef DWARF_SUPPORT - if (params.show_lines) { + if (params.show_lines && !params.upid) { if (params.nevents != 0 || params.dellist) { pr_warning(" Error: Don't use --line with" " --add/--del.\n"); @@ -248,7 +256,7 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) #endif if (params.dellist) { - ret = del_perf_probe_events(params.dellist); + ret = del_perf_probe_events(params.dellist, params.upid); strlist__delete(params.dellist); if (ret < 0) { pr_err(" Error: Failed to delete events. (%d)\n", ret); @@ -267,4 +275,3 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) } return 0; } - diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 1e8e92e..b513e40 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -1082,26 +1082,6 @@ static void event__process_sample(const event_t *self, } } -static int event__process(event_t *event, struct perf_session *session) -{ - switch (event->header.type) { - case PERF_RECORD_COMM: - event__process_comm(event, session); - break; - case PERF_RECORD_MMAP: - event__process_mmap(event, session); - break; - case PERF_RECORD_FORK: - case PERF_RECORD_EXIT: - event__process_task(event, session); - break; - default: - break; - } - - return 0; -} - struct mmap_data { int counter; void *base; diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index d7f21d7..d93e0bb 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -552,6 +552,26 @@ int event__process_task(event_t *self, struct perf_session *session) return 0; } +int event__process(event_t *event, struct perf_session *session) +{ + switch (event->header.type) { + case PERF_RECORD_COMM: + event__process_comm(event, session); + break; + case PERF_RECORD_MMAP: + event__process_mmap(event, session); + break; + case PERF_RECORD_FORK: + case PERF_RECORD_EXIT: + event__process_task(event, session); + break; + default: + break; + } + + return 0; +} + void thread__find_addr_map(struct thread *self, struct perf_session *session, u8 cpumode, enum map_type type, pid_t pid, u64 addr, diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 887ee63..8e790da 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -154,6 +154,7 @@ int event__process_comm(event_t *self, struct perf_session *session); int event__process_lost(event_t *self, struct perf_session *session); int event__process_mmap(event_t *self, struct perf_session *session); int event__process_task(event_t *self, struct perf_session *session); +int event__process(event_t *event, struct perf_session *session); struct addr_location; int event__preprocess_sample(const event_t *self, struct perf_session *session, diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 09cf546..ef7c2d5 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -1,5 +1,5 @@ /* - * probe-event.c : perf-probe definition to kprobe_events format converter + * probe-event.c : perf-probe definition to probe_events format converter * * Written by Masami Hiramatsu <mhiramat(a)redhat.com> * @@ -46,6 +46,7 @@ #include "trace-event.h" /* For __unused */ #include "probe-event.h" #include "probe-finder.h" +#include "session.h" #define MAX_CMDLEN 256 #define MAX_PROBE_ARGS 128 @@ -72,6 +73,7 @@ static int e_snprintf(char *str, size_t size, const char *format, ...) } static char *synthesize_perf_probe_point(struct perf_probe_point *pp); +static int convert_name_to_addr(struct perf_probe_event *pev); static struct machine machine; /* Initialize symbol maps and path of vmlinux */ @@ -109,6 +111,18 @@ out: return ret; } +static int convert_to_perf_probe_point(struct probe_trace_point *tp, + struct perf_probe_point *pp) +{ + pp->function = strdup(tp->symbol); + if (pp->function == NULL) + return -ENOMEM; + pp->offset = tp->offset; + pp->retprobe = tp->retprobe; + + return 0; +} + #ifdef DWARF_SUPPORT static int open_vmlinux(void) { @@ -120,8 +134,12 @@ static int open_vmlinux(void) return open(machine.vmlinux_maps[MAP__FUNCTION]->dso->long_name, O_RDONLY); } -/* Convert trace point to probe point with debuginfo */ -static int convert_to_perf_probe_point(struct kprobe_trace_point *tp, +/* + * Convert trace point to probe point with debuginfo + * Currently only handles kprobes. + */ + +static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp, struct perf_probe_point *pp) { struct symbol *sym; @@ -151,13 +169,21 @@ static int convert_to_perf_probe_point(struct kprobe_trace_point *tp, } /* Try to find perf_probe_event with debuginfo */ -static int try_to_find_kprobe_trace_events(struct perf_probe_event *pev, - struct kprobe_trace_event **tevs, +static int try_to_find_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs, int max_tevs) { bool need_dwarf = perf_probe_event_need_dwarf(pev); int fd, ntevs; + if (pev->upid) { + if (need_dwarf) { + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; + } + return convert_name_to_addr(pev); + } + fd = open_vmlinux(); if (fd < 0) { if (need_dwarf) { @@ -169,11 +195,11 @@ static int try_to_find_kprobe_trace_events(struct perf_probe_event *pev, } /* Searching trace events corresponding to probe event */ - ntevs = find_kprobe_trace_events(fd, pev, tevs, max_tevs); + ntevs = find_probe_trace_events(fd, pev, tevs, max_tevs); close(fd); if (ntevs > 0) { /* Succeeded to find trace events */ - pr_debug("find %d kprobe_trace_events.\n", ntevs); + pr_debug("find %d probe_trace_events.\n", ntevs); return ntevs; } @@ -308,26 +334,23 @@ end: #else /* !DWARF_SUPPORT */ -static int convert_to_perf_probe_point(struct kprobe_trace_point *tp, - struct perf_probe_point *pp) +static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp, + struct perf_probe_point *pp) { - pp->function = strdup(tp->symbol); - if (pp->function == NULL) - return -ENOMEM; - pp->offset = tp->offset; - pp->retprobe = tp->retprobe; - - return 0; + return convert_to_perf_probe_point(tp, pp); } -static int try_to_find_kprobe_trace_events(struct perf_probe_event *pev, - struct kprobe_trace_event **tevs __unused, +static int try_to_find_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs __unused, int max_tevs __unused) { if (perf_probe_event_need_dwarf(pev)) { pr_warning("Debuginfo-analysis is not supported.\n"); return -ENOSYS; } + if (pev->upid) + return convert_name_to_addr(pev); + return 0; } @@ -403,6 +426,115 @@ static bool check_event_name(const char *name) return true; } +/* + * uprobe_events only accepts address: + * Convert function and any offset to address + */ +static int convert_name_to_addr(struct perf_probe_event *pev) +{ + struct perf_probe_point *pp = &pev->point; + struct perf_session *session; + struct thread *thread; + struct symbol *sym; + struct map *map; + char *name; + int ret = -EINVAL; + unsigned long long vaddr; + + /* check if user has specifed a virtual address */ + vaddr = strtoul(pp->function, NULL, 0); + session = perf_session__new(NULL, O_WRONLY, false, false); + if (!session) { + pr_warning("Cannot initialize perf session.\n"); + return -ENOMEM; + } + + symbol_conf.try_vmlinux_path = false; + if (!vaddr) + symbol_conf.sort_by_name = true; + if (symbol__init() < 0) { + pr_warning("Cannot initialize symbols.\n"); + goto free_session; + } + + event__synthesize_thread(pev->upid, event__process, session); + thread = perf_session__findnew(session, pev->upid); + if (!thread) { + pr_warning("Cannot initialize perf session.\n"); + goto free_session; + } + + if (vaddr) { + if (pev->event) { + ret = 0; + goto free_session; + } + + pev->event = zalloc(sizeof(char *) * MAX_PROBE_ARGS); + if (!pev->event) { + ret = -ENOMEM; + pr_warning("Cannot allocate memory for event.\n"); + goto free_session; + } + + sym = map_groups__find_symbol(&thread->mg, MAP__FUNCTION, + vaddr, NULL, NULL); + if (!sym) + snprintf(pev->event, MAX_PROBE_ARGS, "p_%llx", vaddr); + else + snprintf(pev->event, MAX_PROBE_ARGS, "%s_%llx", + sym->name, vaddr); + ret = 0; + goto free_session; + } + + if (!pp->file) + /* Lets find the function in the executable. */ + name = thread->comm; + else + name = basename(make_absolute_path(pp->file)); + + if (!name) { + pr_debug("Please check DSO and retry\n"); + goto free_session; + } + + map = map_groups__find_by_name(&thread->mg, MAP__FUNCTION, name); + if (!map) { + pr_warning("Cannot find appropriate DSO.\n"); + goto free_session; + } + + sym = map__find_symbol_by_name(map, pp->function, NULL); + if (!sym) { + pr_warning("Cannot find appropriate DSO.\n"); + goto free_session; + } + + if (map->start > sym->start) + vaddr = map->start; + vaddr += sym->start + pp->offset + map->pgoff; + pp->offset = 0; + + if (!pev->event) + pev->event = pp->function; + else + free(pp->function); + pp->function = zalloc(sizeof(char *) * MAX_PROBE_ARGS); + if (!pp->function) { + ret = -ENOMEM; + pr_warning("Failed to allocate memory by zalloc.\n"); + goto free_session; + } + e_snprintf(pp->function, MAX_PROBE_ARGS, "0x%llx", vaddr); + ret = 0; + +free_session: + if (session) + perf_session__delete(session); + return ret; +} + /* Parse probepoint definition. */ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) { @@ -542,6 +674,11 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) return -EINVAL; } + if (pev->upid && !pp->function) { + semantic_error("No function specified for uprobes"); + return -EINVAL; + } + if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) { semantic_error("Offset/Line/Lazy pattern can't be used with " "return probe."); @@ -551,6 +688,14 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n", pp->function, pp->file, pp->line, pp->offset, pp->retprobe, pp->lazy_line); + + if (pev->upid && perf_probe_event_need_dwarf(pev)) { + semantic_error("no dwarf based probes for uprobes."); + return -EINVAL; + } + + if (pev->upid) + return convert_name_to_addr(pev); return 0; } @@ -702,7 +847,8 @@ bool perf_probe_event_need_dwarf(struct perf_probe_event *pev) { int i; - if (pev->point.file || pev->point.line || pev->point.lazy_line) + if ((pev->point.file && !pev->upid) || pev->point.line || + pev->point.lazy_line) return true; for (i = 0; i < pev->nargs; i++) @@ -712,16 +858,17 @@ bool perf_probe_event_need_dwarf(struct perf_probe_event *pev) return false; } -/* Parse kprobe_events event into struct probe_point */ -int parse_kprobe_trace_command(const char *cmd, struct kprobe_trace_event *tev) +/* Parse probe_events (uprobe_events) event into struct probe_point */ +static int parse_probe_trace_command(const char *cmd, + struct probe_trace_event *tev) { - struct kprobe_trace_point *tp = &tev->point; + struct probe_trace_point *tp = &tev->point; char pr; char *p; int ret, i, argc; char **argv; - pr_debug("Parsing kprobe_events: %s\n", cmd); + pr_debug("Parsing probe_events: %s\n", cmd); argv = argv_split(cmd, &argc); if (!argv) { pr_debug("Failed to split arguments.\n"); @@ -753,7 +900,7 @@ int parse_kprobe_trace_command(const char *cmd, struct kprobe_trace_event *tev) tp->offset = 0; tev->nargs = argc - 2; - tev->args = zalloc(sizeof(struct kprobe_trace_arg) * tev->nargs); + tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs); if (tev->args == NULL) { ret = -ENOMEM; goto out; @@ -899,13 +1046,13 @@ char *synthesize_perf_probe_command(struct perf_probe_event *pev) } #endif -static int __synthesize_kprobe_trace_arg_ref(struct kprobe_trace_arg_ref *ref, +static int __synthesize_probe_trace_arg_ref(struct probe_trace_arg_ref *ref, char **buf, size_t *buflen, int depth) { int ret; if (ref->next) { - depth = __synthesize_kprobe_trace_arg_ref(ref->next, buf, + depth = __synthesize_probe_trace_arg_ref(ref->next, buf, buflen, depth + 1); if (depth < 0) goto out; @@ -923,10 +1070,10 @@ out: } -static int synthesize_kprobe_trace_arg(struct kprobe_trace_arg *arg, +static int synthesize_probe_trace_arg(struct probe_trace_arg *arg, char *buf, size_t buflen) { - struct kprobe_trace_arg_ref *ref = arg->ref; + struct probe_trace_arg_ref *ref = arg->ref; int ret, depth = 0; char *tmp = buf; @@ -946,7 +1093,7 @@ static int synthesize_kprobe_trace_arg(struct kprobe_trace_arg *arg, /* Dereferencing arguments */ if (ref) { - depth = __synthesize_kprobe_trace_arg_ref(ref, &buf, + depth = __synthesize_probe_trace_arg_ref(ref, &buf, &buflen, 1); if (depth < 0) return depth; @@ -982,9 +1129,9 @@ static int synthesize_kprobe_trace_arg(struct kprobe_trace_arg *arg, return buf - tmp; } -char *synthesize_kprobe_trace_command(struct kprobe_trace_event *tev) +char *synthesize_probe_trace_command(struct probe_trace_event *tev) { - struct kprobe_trace_point *tp = &tev->point; + struct probe_trace_point *tp = &tev->point; char *buf; int i, len, ret; @@ -992,15 +1139,27 @@ char *synthesize_kprobe_trace_command(struct kprobe_trace_event *tev) if (buf == NULL) return NULL; - len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu", - tp->retprobe ? 'r' : 'p', - tev->group, tev->event, - tp->symbol, tp->offset); + if (tev->upid) + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %d:%s\n", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tev->upid, tp->symbol); + else if (tp->offset) + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->symbol, tp->offset); + else + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->symbol); + if (len <= 0) goto error; for (i = 0; i < tev->nargs; i++) { - ret = synthesize_kprobe_trace_arg(&tev->args[i], buf + len, + ret = synthesize_probe_trace_arg(&tev->args[i], buf + len, MAX_CMDLEN - len); if (ret <= 0) goto error; @@ -1013,8 +1172,8 @@ error: return NULL; } -int convert_to_perf_probe_event(struct kprobe_trace_event *tev, - struct perf_probe_event *pev) +static int convert_to_perf_probe_event(struct probe_trace_event *tev, + struct perf_probe_event *pev, bool is_kprobe) { char buf[64] = ""; int i, ret; @@ -1026,7 +1185,11 @@ int convert_to_perf_probe_event(struct kprobe_trace_event *tev, return -ENOMEM; /* Convert trace_point to probe_point */ - ret = convert_to_perf_probe_point(&tev->point, &pev->point); + if (is_kprobe) + ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point); + else + ret = convert_to_perf_probe_point(&tev->point, &pev->point); + if (ret < 0) return ret; @@ -1039,7 +1202,7 @@ int convert_to_perf_probe_event(struct kprobe_trace_event *tev, if (tev->args[i].name) pev->args[i].name = strdup(tev->args[i].name); else { - ret = synthesize_kprobe_trace_arg(&tev->args[i], + ret = synthesize_probe_trace_arg(&tev->args[i], buf, 64); pev->args[i].name = strdup(buf); } @@ -1090,9 +1253,9 @@ void clear_perf_probe_event(struct perf_probe_event *pev) memset(pev, 0, sizeof(*pev)); } -void clear_kprobe_trace_event(struct kprobe_trace_event *tev) +static void clear_probe_trace_event(struct probe_trace_event *tev) { - struct kprobe_trace_arg_ref *ref, *next; + struct probe_trace_arg_ref *ref, *next; int i; if (tev->event) @@ -1120,7 +1283,7 @@ void clear_kprobe_trace_event(struct kprobe_trace_event *tev) memset(tev, 0, sizeof(*tev)); } -static int open_kprobe_events(bool readwrite) +static int open_probe_events(bool readwrite, bool is_kprobe) { char buf[PATH_MAX]; const char *__debugfs; @@ -1131,8 +1294,13 @@ static int open_kprobe_events(bool readwrite) pr_warning("Debugfs is not mounted.\n"); return -ENOENT; } + if (is_kprobe) + ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", + __debugfs); + else + ret = e_snprintf(buf, PATH_MAX, "%stracing/uprobe_events", + __debugfs); - ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", __debugfs); if (ret >= 0) { pr_debug("Opening %s write=%d\n", buf, readwrite); if (readwrite && !probe_event_dry_run) @@ -1143,17 +1311,30 @@ static int open_kprobe_events(bool readwrite) if (ret < 0) { if (errno == ENOENT) - pr_warning("kprobe_events file does not exist - please" - " rebuild kernel with CONFIG_KPROBE_EVENT.\n"); + pr_warning("%s file does not exist - please" + " rebuild kernel with CONFIG_%s_EVENT.\n", + is_kprobe ? "kprobe_events" : "uprobe_events", + is_kprobe ? "KPROBE" : "UPROBE"); else - pr_warning("Failed to open kprobe_events file: %s\n", - strerror(errno)); + pr_warning("Failed to open %s file: %s\n", + is_kprobe ? "kprobe_events" : "uprobe_events", + strerror(errno)); } return ret; } -/* Get raw string list of current kprobe_events */ -static struct strlist *get_kprobe_trace_command_rawlist(int fd) +static int open_kprobe_events(bool readwrite) +{ + return open_probe_events(readwrite, 1); +} + +static int open_uprobe_events(bool readwrite) +{ + return open_probe_events(readwrite, 0); +} + +/* Get raw string list of current kprobe_events or uprobe_events */ +static struct strlist *get_probe_trace_command_rawlist(int fd) { int ret, idx; FILE *fp; @@ -1217,64 +1398,78 @@ static int show_perf_probe_event(struct perf_probe_event *pev) return ret; } -/* List up current perf-probe events */ -int show_perf_probe_events(void) +static int __show_perf_probe_events(int fd, bool is_kprobe) { - int fd, ret; - struct kprobe_trace_event tev; + int ret = 0; + struct probe_trace_event tev; struct perf_probe_event pev; struct strlist *rawlist; struct str_node *ent; - setup_pager(); - ret = init_vmlinux(); - if (ret < 0) - return ret; - memset(&tev, 0, sizeof(tev)); memset(&pev, 0, sizeof(pev)); - fd = open_kprobe_events(false); - if (fd < 0) - return fd; - - rawlist = get_kprobe_trace_command_rawlist(fd); - close(fd); + rawlist = get_probe_trace_command_rawlist(fd); if (!rawlist) return -ENOENT; strlist__for_each(ent, rawlist) { - ret = parse_kprobe_trace_command(ent->s, &tev); + ret = parse_probe_trace_command(ent->s, &tev); if (ret >= 0) { - ret = convert_to_perf_probe_event(&tev, &pev); + ret = convert_to_perf_probe_event(&tev, &pev, + is_kprobe); if (ret >= 0) ret = show_perf_probe_event(&pev); } clear_perf_probe_event(&pev); - clear_kprobe_trace_event(&tev); + clear_probe_trace_event(&tev); if (ret < 0) break; } strlist__delete(rawlist); + return ret; +} + +/* List up current perf-probe events */ +int show_perf_probe_events(void) +{ + int fd, ret; + + setup_pager(); + fd = open_kprobe_events(false); + if (fd < 0) + return fd; + + ret = init_vmlinux(); + if (ret < 0) + return ret; + + ret = __show_perf_probe_events(fd, true); + close(fd); + + fd = open_uprobe_events(false); + if (fd >= 0) { + ret = __show_perf_probe_events(fd, false); + close(fd); + } return ret; } /* Get current perf-probe event names */ -static struct strlist *get_kprobe_trace_event_names(int fd, bool include_group) +static struct strlist *get_probe_trace_event_names(int fd, bool include_group) { char buf[128]; struct strlist *sl, *rawlist; struct str_node *ent; - struct kprobe_trace_event tev; + struct probe_trace_event tev; int ret = 0; memset(&tev, 0, sizeof(tev)); - - rawlist = get_kprobe_trace_command_rawlist(fd); + rawlist = get_probe_trace_command_rawlist(fd); sl = strlist__new(true, NULL); strlist__for_each(ent, rawlist) { - ret = parse_kprobe_trace_command(ent->s, &tev); + ret = parse_probe_trace_command(ent->s, &tev); if (ret < 0) break; if (include_group) { @@ -1284,7 +1479,7 @@ static struct strlist *get_kprobe_trace_event_names(int fd, bool include_group) ret = strlist__add(sl, buf); } else ret = strlist__add(sl, tev.event); - clear_kprobe_trace_event(&tev); + clear_probe_trace_event(&tev); if (ret < 0) break; } @@ -1297,13 +1492,13 @@ static struct strlist *get_kprobe_trace_event_names(int fd, bool include_group) return sl; } -static int write_kprobe_trace_event(int fd, struct kprobe_trace_event *tev) +static int write_probe_trace_event(int fd, struct probe_trace_event *tev) { int ret = 0; - char *buf = synthesize_kprobe_trace_command(tev); + char *buf = synthesize_probe_trace_command(tev); if (!buf) { - pr_debug("Failed to synthesize kprobe trace event.\n"); + pr_debug("Failed to synthesize probe trace event.\n"); return -EINVAL; } @@ -1356,21 +1551,24 @@ static int get_new_event_name(char *buf, size_t len, const char *base, return ret; } -static int __add_kprobe_trace_events(struct perf_probe_event *pev, - struct kprobe_trace_event *tevs, +static int __add_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event *tevs, int ntevs, bool allow_suffix) { int i, fd, ret; - struct kprobe_trace_event *tev = NULL; + struct probe_trace_event *tev = NULL; char buf[64]; const char *event, *group; struct strlist *namelist; - fd = open_kprobe_events(true); + if (pev->upid) + fd = open_uprobe_events(true); + else + fd = open_kprobe_events(true); if (fd < 0) return fd; /* Get current event names */ - namelist = get_kprobe_trace_event_names(fd, false); + namelist = get_probe_trace_event_names(fd, false); if (!namelist) { pr_debug("Failed to get current event list.\n"); return -EIO; @@ -1380,17 +1578,28 @@ static int __add_kprobe_trace_events(struct perf_probe_event *pev, printf("Add new event%s\n", (ntevs > 1) ? "s:" : ":"); for (i = 0; i < ntevs; i++) { tev = &tevs[i]; - if (pev->event) - event = pev->event; - else - if (pev->point.function) - event = pev->point.function; - else - event = tev->point.symbol; + if (pev->group) group = pev->group; - else + else if (!pev->upid) group = PERFPROBE_GROUP; + else { + /* + * For uprobes based probes create a group + * probe_<pid>. + */ + snprintf(buf, 64, "%s_%d", PERFPROBE_GROUP, pev->upid); + group = buf; + } + + tev->group = strdup(group); + + if (pev->event) + event = pev->event; + else if (pev->point.function) + event = pev->point.function; + else + event = tev->point.symbol; /* Get an unused new event name */ ret = get_new_event_name(buf, 64, event, @@ -1398,14 +1607,13 @@ static int __add_kprobe_trace_events(struct perf_probe_event *pev, if (ret < 0) break; event = buf; - tev->event = strdup(event); - tev->group = strdup(group); + if (tev->event == NULL || tev->group == NULL) { ret = -ENOMEM; break; } - ret = write_kprobe_trace_event(fd, tev); + ret = write_probe_trace_event(fd, tev); if (ret < 0) break; /* Add added event name to namelist */ @@ -1442,21 +1650,21 @@ static int __add_kprobe_trace_events(struct perf_probe_event *pev, return ret; } -static int convert_to_kprobe_trace_events(struct perf_probe_event *pev, - struct kprobe_trace_event **tevs, +static int convert_to_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs, int max_tevs) { struct symbol *sym; int ret = 0, i; - struct kprobe_trace_event *tev; + struct probe_trace_event *tev; /* Convert perf_probe_event with debuginfo */ - ret = try_to_find_kprobe_trace_events(pev, tevs, max_tevs); + ret = try_to_find_probe_trace_events(pev, tevs, max_tevs); if (ret != 0) return ret; /* Allocate trace event buffer */ - tev = *tevs = zalloc(sizeof(struct kprobe_trace_event)); + tev = *tevs = zalloc(sizeof(struct probe_trace_event)); if (tev == NULL) return -ENOMEM; @@ -1469,7 +1677,7 @@ static int convert_to_kprobe_trace_events(struct perf_probe_event *pev, tev->point.offset = pev->point.offset; tev->nargs = pev->nargs; if (tev->nargs) { - tev->args = zalloc(sizeof(struct kprobe_trace_arg) + tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs); if (tev->args == NULL) { ret = -ENOMEM; @@ -1498,6 +1706,11 @@ static int convert_to_kprobe_trace_events(struct perf_probe_event *pev, } } + if (pev->upid) { + tev->upid = pev->upid; + return 1; + } + /* Currently just checking function name from symbol map */ sym = map__find_symbol_by_name(machine.vmlinux_maps[MAP__FUNCTION], tev->point.symbol, NULL); @@ -1510,7 +1723,7 @@ static int convert_to_kprobe_trace_events(struct perf_probe_event *pev, return 1; error: - clear_kprobe_trace_event(tev); + clear_probe_trace_event(tev); free(tev); *tevs = NULL; return ret; @@ -1518,30 +1731,32 @@ error: struct __event_package { struct perf_probe_event *pev; - struct kprobe_trace_event *tevs; + struct probe_trace_event *tevs; int ntevs; }; int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, bool force_add, int max_tevs) { - int i, j, ret; + int i, j, ret = 0; struct __event_package *pkgs; pkgs = zalloc(sizeof(struct __event_package) * npevs); if (pkgs == NULL) return -ENOMEM; - /* Init vmlinux path */ - ret = init_vmlinux(); - if (ret < 0) - return ret; + if (!pevs->upid) { + /* Init vmlinux path */ + ret = init_vmlinux(); + if (ret < 0) + return ret; + } /* Loop 1: convert all events */ for (i = 0; i < npevs; i++) { pkgs[i].pev = &pevs[i]; /* Convert with or without debuginfo */ - ret = convert_to_kprobe_trace_events(pkgs[i].pev, + ret = convert_to_probe_trace_events(pkgs[i].pev, &pkgs[i].tevs, max_tevs); if (ret < 0) goto end; @@ -1550,24 +1765,24 @@ int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, /* Loop 2: add all events */ for (i = 0; i < npevs && ret >= 0; i++) - ret = __add_kprobe_trace_events(pkgs[i].pev, pkgs[i].tevs, + ret = __add_probe_trace_events(pkgs[i].pev, pkgs[i].tevs, pkgs[i].ntevs, force_add); end: /* Loop 3: cleanup trace events */ for (i = 0; i < npevs; i++) for (j = 0; j < pkgs[i].ntevs; j++) - clear_kprobe_trace_event(&pkgs[i].tevs[j]); + clear_probe_trace_event(&pkgs[i].tevs[j]); return ret; } -static int __del_trace_kprobe_event(int fd, struct str_node *ent) +static int __del_trace_probe_event(int fd, struct str_node *ent) { char *p; char buf[128]; int ret; - /* Convert from perf-probe event to trace-kprobe event */ + /* Convert from perf-probe event to trace-probe event */ ret = e_snprintf(buf, 128, "-:%s", ent->s); if (ret < 0) goto error; @@ -1593,24 +1808,16 @@ error: return ret; } -static int del_trace_kprobe_event(int fd, const char *group, - const char *event, struct strlist *namelist) +static int del_trace_probe_event(int fd, const char *buf, + struct strlist *namelist) { - char buf[128]; struct str_node *ent, *n; - int found = 0, ret = 0; - - ret = e_snprintf(buf, 128, "%s:%s", group, event); - if (ret < 0) { - pr_err("Failed to copy event."); - return ret; - } + int ret = -1; if (strpbrk(buf, "*?")) { /* Glob-exp */ strlist__for_each_safe(ent, n, namelist) if (strglobmatch(ent->s, buf)) { - found++; - ret = __del_trace_kprobe_event(fd, ent); + ret = __del_trace_probe_event(fd, ent); if (ret < 0) break; strlist__remove(namelist, ent); @@ -1618,40 +1825,44 @@ static int del_trace_kprobe_event(int fd, const char *group, } else { ent = strlist__find(namelist, buf); if (ent) { - found++; - ret = __del_trace_kprobe_event(fd, ent); + ret = __del_trace_probe_event(fd, ent); if (ret >= 0) strlist__remove(namelist, ent); } } - if (found == 0 && ret >= 0) - pr_info("Info: Event \"%s\" does not exist.\n", buf); - return ret; } -int del_perf_probe_events(struct strlist *dellist) +int del_perf_probe_events(struct strlist *dellist, pid_t pid) { - int fd, ret = 0; + int ret = -1, ufd = -1, kfd = -1; + char buf[128]; const char *group, *event; char *p, *str; struct str_node *ent; - struct strlist *namelist; + struct strlist *namelist = NULL, *unamelist = NULL; - fd = open_kprobe_events(true); - if (fd < 0) - return fd; /* Get current event names */ - namelist = get_kprobe_trace_event_names(fd, true); - if (namelist == NULL) - return -EINVAL; + if (!pid) { + kfd = open_kprobe_events(true); + if (kfd < 0) + return kfd; + namelist = get_probe_trace_event_names(kfd, true); + } + + ufd = open_uprobe_events(true); + if (ufd >= 0) + unamelist = get_probe_trace_event_names(ufd, true); + + if (namelist == NULL && unamelist == NULL) + goto error; strlist__for_each(ent, dellist) { str = strdup(ent->s); if (str == NULL) { ret = -ENOMEM; - break; + goto error; } pr_debug("Parsing: %s\n", str); p = strchr(str, ':'); @@ -1663,15 +1874,36 @@ int del_perf_probe_events(struct strlist *dellist) group = "*"; event = str; } + + ret = e_snprintf(buf, 128, "%s:%s", group, event); + if (ret < 0) { + pr_err("Failed to copy event."); + free(str); + goto error; + } + pr_debug("Group: %s, Event: %s\n", group, event); - ret = del_trace_kprobe_event(fd, group, event, namelist); + if (!pid && namelist) + ret = del_trace_probe_event(kfd, buf, namelist); + if (unamelist && ret != 0) + ret = del_trace_probe_event(ufd, buf, unamelist); + free(str); - if (ret < 0) - break; + if (ret != 0) + pr_info("Info: Event \"%s\" does not exist.\n", buf); } - strlist__delete(namelist); - close(fd); +error: + if (kfd >= 0) { + if (namelist) + strlist__delete(namelist); + close(kfd); + } + + if (ufd >= 0) { + if (unamelist) + strlist__delete(unamelist); + close(ufd); + } return ret; } - diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h index bc06d3e..f69f98e 100644 --- a/tools/perf/util/probe-event.h +++ b/tools/perf/util/probe-event.h @@ -6,34 +6,35 @@ extern bool probe_event_dry_run; -/* kprobe-tracer tracing point */ -struct kprobe_trace_point { +/* kprobe-tracer and uprobe-tracer tracing point */ +struct probe_trace_point { char *symbol; /* Base symbol */ unsigned long offset; /* Offset from symbol */ bool retprobe; /* Return probe flag */ }; -/* kprobe-tracer tracing argument referencing offset */ -struct kprobe_trace_arg_ref { - struct kprobe_trace_arg_ref *next; /* Next reference */ +/* probe-tracer tracing argument referencing offset */ +struct probe_trace_arg_ref { + struct probe_trace_arg_ref *next; /* Next reference */ long offset; /* Offset value */ }; -/* kprobe-tracer tracing argument */ -struct kprobe_trace_arg { +/* kprobe-tracer and uprobe-tracer tracing argument */ +struct probe_trace_arg { char *name; /* Argument name */ char *value; /* Base value */ char *type; /* Type name */ - struct kprobe_trace_arg_ref *ref; /* Referencing offset */ + struct probe_trace_arg_ref *ref; /* Referencing offset */ }; -/* kprobe-tracer tracing event (point + arg) */ -struct kprobe_trace_event { +/* kprobe-tracer and uprobe-tracer tracing event (point + arg) */ +struct probe_trace_event { char *event; /* Event name */ char *group; /* Group name */ - struct kprobe_trace_point point; /* Trace point */ + struct probe_trace_point point; /* Trace point */ int nargs; /* Number of args */ - struct kprobe_trace_arg *args; /* Arguments */ + pid_t upid; /* uprobes only */ + struct probe_trace_arg *args; /* Arguments */ }; /* Perf probe probing point */ @@ -68,6 +69,7 @@ struct perf_probe_event { char *group; /* Group name */ struct perf_probe_point point; /* Probe point */ int nargs; /* Number of arguments */ + pid_t upid; struct perf_probe_arg *args; /* Arguments */ }; @@ -92,25 +94,18 @@ struct line_range { /* Command string to events */ extern int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev); -extern int parse_kprobe_trace_command(const char *cmd, - struct kprobe_trace_event *tev); /* Events to command string */ extern char *synthesize_perf_probe_command(struct perf_probe_event *pev); -extern char *synthesize_kprobe_trace_command(struct kprobe_trace_event *tev); +extern char *synthesize_probe_trace_command(struct probe_trace_event *tev); extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len); /* Check the perf_probe_event needs debuginfo */ extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev); -/* Convert from kprobe_trace_event to perf_probe_event */ -extern int convert_to_perf_probe_event(struct kprobe_trace_event *tev, - struct perf_probe_event *pev); - /* Release event contents */ extern void clear_perf_probe_event(struct perf_probe_event *pev); -extern void clear_kprobe_trace_event(struct kprobe_trace_event *tev); /* Command string to line-range */ extern int parse_line_range_desc(const char *cmd, struct line_range *lr); @@ -118,7 +113,7 @@ extern int parse_line_range_desc(const char *cmd, struct line_range *lr); extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, bool force_add, int max_probe_points); -extern int del_perf_probe_events(struct strlist *dellist); +extern int del_perf_probe_events(struct strlist *dellist, pid_t pid); extern int show_perf_probe_events(void); extern int show_line_range(struct line_range *lr); diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 3e64e1f..f3a024a 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -406,10 +406,10 @@ static Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name, * Probe finder related functions */ -static struct kprobe_trace_arg_ref *alloc_trace_arg_ref(long offs) +static struct probe_trace_arg_ref *alloc_trace_arg_ref(long offs) { - struct kprobe_trace_arg_ref *ref; - ref = zalloc(sizeof(struct kprobe_trace_arg_ref)); + struct probe_trace_arg_ref *ref; + ref = zalloc(sizeof(struct probe_trace_arg_ref)); if (ref != NULL) ref->offset = offs; return ref; @@ -425,7 +425,7 @@ static int convert_variable_location(Dwarf_Die *vr_die, struct probe_finder *pf) Dwarf_Word offs = 0; bool ref = false; const char *regs; - struct kprobe_trace_arg *tvar = pf->tvar; + struct probe_trace_arg *tvar = pf->tvar; int ret; /* TODO: handle more than 1 exprs */ @@ -499,10 +499,10 @@ static int convert_variable_location(Dwarf_Die *vr_die, struct probe_finder *pf) } static int convert_variable_type(Dwarf_Die *vr_die, - struct kprobe_trace_arg *tvar, + struct probe_trace_arg *tvar, const char *cast) { - struct kprobe_trace_arg_ref **ref_ptr = &tvar->ref; + struct probe_trace_arg_ref **ref_ptr = &tvar->ref; Dwarf_Die type; char buf[16]; int ret; @@ -540,7 +540,7 @@ static int convert_variable_type(Dwarf_Die *vr_die, while (*ref_ptr) ref_ptr = &(*ref_ptr)->next; /* Add new reference with offset +0 */ - *ref_ptr = zalloc(sizeof(struct kprobe_trace_arg_ref)); + *ref_ptr = zalloc(sizeof(struct probe_trace_arg_ref)); if (*ref_ptr == NULL) { pr_warning("Out of memory error\n"); return -ENOMEM; @@ -585,10 +585,10 @@ static int convert_variable_type(Dwarf_Die *vr_die, static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, struct perf_probe_arg_field *field, - struct kprobe_trace_arg_ref **ref_ptr, + struct probe_trace_arg_ref **ref_ptr, Dwarf_Die *die_mem) { - struct kprobe_trace_arg_ref *ref = *ref_ptr; + struct probe_trace_arg_ref *ref = *ref_ptr; Dwarf_Die type; Dwarf_Word offs; int ret, tag; @@ -614,7 +614,7 @@ static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, pr_debug2("Array real type: (%x)\n", (unsigned)dwarf_dieoffset(&type)); if (tag == DW_TAG_pointer_type) { - ref = zalloc(sizeof(struct kprobe_trace_arg_ref)); + ref = zalloc(sizeof(struct probe_trace_arg_ref)); if (ref == NULL) return -ENOMEM; if (*ref_ptr) @@ -645,7 +645,7 @@ static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, return -EINVAL; } - ref = zalloc(sizeof(struct kprobe_trace_arg_ref)); + ref = zalloc(sizeof(struct probe_trace_arg_ref)); if (ref == NULL) return -ENOMEM; if (*ref_ptr) @@ -778,7 +778,7 @@ static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf) /* Show a probe point to output buffer */ static int convert_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf) { - struct kprobe_trace_event *tev; + struct probe_trace_event *tev; Dwarf_Addr eaddr; Dwarf_Die die_mem; const char *name; @@ -843,7 +843,7 @@ static int convert_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf) /* Find each argument */ tev->nargs = pf->pev->nargs; - tev->args = zalloc(sizeof(struct kprobe_trace_arg) * tev->nargs); + tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs); if (tev->args == NULL) return -ENOMEM; for (i = 0; i < pf->pev->nargs; i++) { @@ -1100,9 +1100,9 @@ static int find_probe_point_by_func(struct probe_finder *pf) return _param.retval; } -/* Find kprobe_trace_events specified by perf_probe_event from debuginfo */ -int find_kprobe_trace_events(int fd, struct perf_probe_event *pev, - struct kprobe_trace_event **tevs, int max_tevs) +/* Find probe_trace_events specified by perf_probe_event from debuginfo */ +int find_probe_trace_events(int fd, struct perf_probe_event *pev, + struct probe_trace_event **tevs, int max_tevs) { struct probe_finder pf = {.pev = pev, .max_tevs = max_tevs}; struct perf_probe_point *pp = &pev->point; @@ -1112,7 +1112,7 @@ int find_kprobe_trace_events(int fd, struct perf_probe_event *pev, Dwarf *dbg; int ret = 0; - pf.tevs = zalloc(sizeof(struct kprobe_trace_event) * max_tevs); + pf.tevs = zalloc(sizeof(struct probe_trace_event) * max_tevs); if (pf.tevs == NULL) return -ENOMEM; *tevs = pf.tevs; diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h index e1f61dc..4507d51 100644 --- a/tools/perf/util/probe-finder.h +++ b/tools/perf/util/probe-finder.h @@ -16,9 +16,9 @@ static inline int is_c_varname(const char *name) } #ifdef DWARF_SUPPORT -/* Find kprobe_trace_events specified by perf_probe_event from debuginfo */ -extern int find_kprobe_trace_events(int fd, struct perf_probe_event *pev, - struct kprobe_trace_event **tevs, +/* Find probe_trace_events specified by perf_probe_event from debuginfo */ +extern int find_probe_trace_events(int fd, struct perf_probe_event *pev, + struct probe_trace_event **tevs, int max_tevs); /* Find a perf_probe_point from debuginfo */ @@ -33,7 +33,7 @@ extern int find_line_range(int fd, struct line_range *lr); struct probe_finder { struct perf_probe_event *pev; /* Target probe event */ - struct kprobe_trace_event *tevs; /* Result trace events */ + struct probe_trace_event *tevs; /* Result trace events */ int ntevs; /* Number of trace events */ int max_tevs; /* Max number of trace events */ @@ -50,7 +50,7 @@ struct probe_finder { #endif Dwarf_Op *fb_ops; /* Frame base attribute */ struct perf_probe_arg *pvar; /* Current target variable */ - struct kprobe_trace_arg *tvar; /* Current result variable */ + struct probe_trace_arg *tvar; /* Current result variable */ }; struct line_finder { -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Srikar Dronamraju on 9 Jul 2010 01:50 Hi Christoph, > > Here is a terminal snapshot of placing, using and removing a probe on a > > process with pid 3591 (corresponding to zsh) > > Btw, I think this interface is not the most the best one for typical > uses cases. To make perf probe support more useful we'll need at least > a way to create a probe in all intances of a given binary/DSO, and > a way to run a binary with pre-defined breakpoints. > Yes Peter had already asked for this and we have it in plan. These are listed in the todos I listed corresponding to file based probes. - Allowing probes per-executable/per dso. - Allowing probes across fork. I have some thoughts on how to achieve these. However I am not sure which items I should work first - file based probes or on - automatically trigger perf uprobes on high overhead functions that Ingo suggested. I am not too clear on later so I might as well start thinking on file based probes. -- Thanks and Regards Srikar -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Srikar Dronamraju on 12 Jul 2010 13:50 > Before some more comments: this is getting really nice! Kudos! > Thanks. > > + > > + /* check if user has specifed a virtual address */ > > + vaddr = strtoul(pp->function, NULL, 0); > > + session = perf_session__new(NULL, O_WRONLY, false, false); > > At first creating a session here looks too much, lets see below... Okay. > > > + if (!vaddr) > > + symbol_conf.sort_by_name = true; > > + if (symbol__init() < 0) { > > + pr_warning("Cannot initialize symbols.\n"); > > + goto free_session; > > + } > > Configuring the symbol lib on a library function is a no-go, this > function (symbol__init()) should be marked with the equivalent of > "module_init()" on tools that need the symbol library, i.e. be called > from the cmd_{top,report,probe,etc} level. Oh okay, I then might be doing the same thing in patch 13/13. I will correct there too. > > > + event__synthesize_thread(pev->upid, event__process, session); > > + thread = perf_session__findnew(session, pev->upid); > > + if (!thread) { > > + pr_warning("Cannot initialize perf session.\n"); > > + goto free_session; > > + } > > Got it, you want to read an existing thread, get it into the > perf_session threads rb_tree to then use what was parsed from /proc. > > I think you should change event__synthesize_thread somehow to achieve > taht same goal instead of going in such a roundabout way, unless you > need the session for some other need. Right, I need the session for the thread. > > Probably we could change it to create a thread instance that then would > be used to synthesize the MMAP and COMM events... but then for the > existing use case we would be creating such events just to trow those > objects away right after synthesizing the PERF_RECORD_{MMAP,COMM} > events... perhaps duplicate them after all :-\ > > If I don't come with something for this quickly we can go on using what > you coded and later refactor it to remove the fat. Okay. I am fine either way. > > + struct probe_trace_event *tev) > > { > > - struct kprobe_trace_point *tp = &tev->point; > > + struct probe_trace_point *tp = &tev->point; > > char pr; > > char *p; > > int ret, i, argc; > > char **argv; > > > > - pr_debug("Parsing kprobe_events: %s\n", cmd); > > + pr_debug("Parsing probe_events: %s\n", cmd); > > I suggest you put these s/kprobe/probe/g parts in a separate patch for > easing review :) > Okay . -- Thanks and Regards Srikar -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Srikar Dronamraju on 29 Jul 2010 10:20
Masami, Below patch should address the comments raised by you. -- Thanks and Regards Srikar --- From: Srikar Dronamraju <srikar(a)linux.vnet.ibm.com> Enhances perf probe to accept pid and user vaddr. Provides very basic support for uprobes. TODO: Update perf-probes.txt. Global tracing. Signed-off-by: Srikar Dronamraju <srikar(a)linux.vnet.ibm.com> --- Changelog from v9: Renaming common fields/functions to refer to probe instead of kprobe. This was suggested by Arnaldo. Changelog from v8: Fixed a build break reported by Christoph Hellwig. Changelog from v6: Changelog from v6: Fixed a bug reported by Masami. i.e Throw an error message and exit if perf probe is for a dwarf based probes. Changelog from v4: Merged to 2.6.35-rc3-tip. Changelog from v3: (addressed comments from Masami Hiramatsu) * Every process id has a different group name. * event name starts with function name. * If vaddr is specified, event name has vaddr appended along with function name, (this is to avoid subsequent probes using same event name.) * warning if -p and --list options are used together. Also dso can either be a short name or absolute path. Here is a terminal snapshot of placing, using and removing a probe on a process with pid 3591 (corresponding to zsh) [ Probing a function in the executable using function name ] ------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree(a)zsh Added new event: probe_3591:zfree (on 0x446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:zfree (on 3591:0x0000000000446420) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/zfree 3591:0x0000000000446420 [root(a)ABCD]# perf record -f -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.039 MB perf.data (~1716 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree Remove event: probe_3591:zfree [root(a)ABCD]# perf report # Samples: 447 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a function + offset ] ------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree(a)zsh+5 Added new event: probe_3591:zfree (on 0x446425) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:zfree (on 3591:0x0000000000446425) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/zfree 3591:0x0000000000446425 [root(a)ABCD]# perf record -f -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.036 MB perf.data (~1590 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree Remove event: probe_3591:zfree [root(a)ABCD]# perf report # Samples: 18 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a library function using function name ] -------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 write(a)libc-2.5.so Added new event: probe_3591:write (on 0x36010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:write -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:write (on 3591:0x00000036010c6060) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/write 3591:0x00000036010c6060 [root(a)ABCD]# perf record -f -e probe_3591:write -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1738 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:write Remove event: probe_3591:write [root(a)ABCD]# perf report # Samples: 11 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing a library function using function name and absolute path ] --------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 write@/lib64/libc-2.5.so Added new event: probe_3591:write (on 0x36010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:write -a sleep 1 [root(a)ABCD]# perf probe --list probe_3591:write (on 3591:0x00000036010c6060) [root(a)ABCD]# cat /sys/kernel/debug/tracing/uprobe_events p:probe_3591/write 3591:0x00000036010c6060 [root(a)ABCD]# perf record -f -e probe_3591:write -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1738 samples) ] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:write Remove event: probe_3591:write [root(a)ABCD]# perf report # Samples: 11 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [ Probing using vaddr 0x0000000000446420 (corresponding to zfree)] ------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 0x0000000000446420 Added new event: probe_3591:zfree_446420 (on 0x0000000000446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree_446420 -a sleep 1 [root(a)ABCD]# perf record -e probe_3591:zfree_446420 -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.041 MB perf.data (~1797 samples) ] [root(a)ABCD]# perf report # # Samples: 628 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# perf report --sort comm,dso # Samples: 628 # # Overhead Command Shared Object # ........ ............... ............. # 100.00% zsh zsh [root(a)ABCD]# perf probe --list probe_3591:zfree_446420 (on 3591:0x0000000000446420) [root(a)ABCD]# perf list | grep probe probe_3591:zfree_446420 [Tracepoint event] [root(a)ABCD]# perf probe -p 3591 --del probe_3591:zfree_446420 Remove event: probe_3591:zfree_446420 [root(a)ABCD]# Another example for a shared library: write stub in libc. (corresponds to 0x00000036010c6060) on a vaddr [ Probing a libc vaddr 0x00000036010c6060 (corresponding to write) ] [root(a)ABCD]# perf probe -p 3591 0x00000036010c6060 dded new event: probe_3591:__GI___libc_write_36010c6060 (on 0x00000036010c6060) You can now use it on all perf tools, such as: perf record -e probe_3591:__GI___libc_write_36010c6060 -a sleep 1 [root(a)ABCD]# perf record -f -e probe_3591:__GI___libc_write_36010c6060 -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.040 MB perf.data (~1748 samples) ] [root(a)ABCD]# perf report # Samples: 24 # # Overhead Command Shared Object Symbol # ........ ............... .................. ...... # 100.00% zsh libc-2.5.so [.] __GI___libc_write # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# [ Probing using a function without specifying a dso (corresponding to zfree)] ------------------------------------------------------------------- [root(a)ABCD]# perf probe -p 3591 zfree Added new event: probe_3591:zfree (on 0x0000000000446420) You can now use it on all perf tools, such as: perf record -e probe_3591:zfree -a sleep 1 [root(a)ABCD]# perf record -e probe_3591:zfree -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.041 MB perf.data (~1797 samples) ] [root(a)ABCD]# perf report # # Samples: 628 # # Overhead Command Shared Object Symbol # ........ ............... ............. ...... # 100.00% zsh zsh [.] zfree # # (For a higher level overview, try: perf report --sort comm,dso) # [root(a)ABCD]# tools/perf/builtin-probe.c | 12 + tools/perf/builtin-top.c | 20 -- tools/perf/util/event.c | 20 ++ tools/perf/util/event.h | 1 tools/perf/util/probe-event.c | 396 +++++++++++++++++++++++++++++++++-------- tools/perf/util/probe-event.h | 10 + 6 files changed, 353 insertions(+), 106 deletions(-) diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c index 199d5e1..cb915a5 100644 --- a/tools/perf/builtin-probe.c +++ b/tools/perf/builtin-probe.c @@ -55,6 +55,7 @@ static struct { struct strlist *dellist; struct line_range line_range; int max_probe_points; + pid_t upid; } params; @@ -73,6 +74,7 @@ static int parse_probe_event(const char *str) /* Parse a perf-probe command into event */ ret = parse_perf_probe_command(str, pev); pr_debug("%d arguments\n", pev->nargs); + pev->upid = params.upid; return ret; } @@ -188,6 +190,8 @@ static const struct option options[] = { OPT__DRY_RUN(&probe_event_dry_run), OPT_INTEGER('\0', "max-probes", ¶ms.max_probe_points, "Set how many probe points can be found for a probe."), + OPT_INTEGER('p', "pid", ¶ms.upid, + "specify a pid for a uprobes based probe"), OPT_END() }; @@ -225,6 +229,10 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) pr_err(" Error: Don't use --list with --line.\n"); usage_with_options(probe_usage, options); } + if (params.upid) { + pr_warning(" Error: Don't use --list with --pid.\n"); + usage_with_options(probe_usage, options); + } ret = show_perf_probe_events(); if (ret < 0) pr_err(" Error: Failed to show event list. (%d)\n", @@ -233,7 +241,7 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) } #ifdef DWARF_SUPPORT - if (params.show_lines) { + if (params.show_lines && !params.upid) { if (params.nevents != 0 || params.dellist) { pr_warning(" Error: Don't use --line with" " --add/--del.\n"); @@ -248,7 +256,7 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used) #endif if (params.dellist) { - ret = del_perf_probe_events(params.dellist); + ret = del_perf_probe_events(params.dellist, params.upid); strlist__delete(params.dellist); if (ret < 0) { pr_err(" Error: Failed to delete events. (%d)\n", ret); diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 1e8e92e..b513e40 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -1082,26 +1082,6 @@ static void event__process_sample(const event_t *self, } } -static int event__process(event_t *event, struct perf_session *session) -{ - switch (event->header.type) { - case PERF_RECORD_COMM: - event__process_comm(event, session); - break; - case PERF_RECORD_MMAP: - event__process_mmap(event, session); - break; - case PERF_RECORD_FORK: - case PERF_RECORD_EXIT: - event__process_task(event, session); - break; - default: - break; - } - - return 0; -} - struct mmap_data { int counter; void *base; diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index d7f21d7..d93e0bb 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -552,6 +552,26 @@ int event__process_task(event_t *self, struct perf_session *session) return 0; } +int event__process(event_t *event, struct perf_session *session) +{ + switch (event->header.type) { + case PERF_RECORD_COMM: + event__process_comm(event, session); + break; + case PERF_RECORD_MMAP: + event__process_mmap(event, session); + break; + case PERF_RECORD_FORK: + case PERF_RECORD_EXIT: + event__process_task(event, session); + break; + default: + break; + } + + return 0; +} + void thread__find_addr_map(struct thread *self, struct perf_session *session, u8 cpumode, enum map_type type, pid_t pid, u64 addr, diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 887ee63..8e790da 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -154,6 +154,7 @@ int event__process_comm(event_t *self, struct perf_session *session); int event__process_lost(event_t *self, struct perf_session *session); int event__process_mmap(event_t *self, struct perf_session *session); int event__process_task(event_t *self, struct perf_session *session); +int event__process(event_t *event, struct perf_session *session); struct addr_location; int event__preprocess_sample(const event_t *self, struct perf_session *session, diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 2e665cb..3759216 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -46,6 +46,7 @@ #include "trace-event.h" /* For __unused */ #include "probe-event.h" #include "probe-finder.h" +#include "session.h" #define MAX_CMDLEN 256 #define MAX_PROBE_ARGS 128 @@ -72,6 +73,7 @@ static int e_snprintf(char *str, size_t size, const char *format, ...) } static char *synthesize_perf_probe_point(struct perf_probe_point *pp); +static int convert_name_to_addr(struct perf_probe_event *pev); static struct machine machine; /* Initialize symbol maps and path of vmlinux */ @@ -109,6 +111,32 @@ out: return ret; } +static int init_perf_uprobes(void) +{ + int ret = 0; + + symbol_conf.try_vmlinux_path = false; + symbol_conf.sort_by_name = true; + ret = symbol__init(); + if (ret < 0) + pr_debug("Failed to init symbol map.\n"); + + return ret; +} + + +static int convert_to_perf_probe_point(struct probe_trace_point *tp, + struct perf_probe_point *pp) +{ + pp->function = strdup(tp->symbol); + if (pp->function == NULL) + return -ENOMEM; + pp->offset = tp->offset; + pp->retprobe = tp->retprobe; + + return 0; +} + #ifdef DWARF_SUPPORT static int open_vmlinux(void) { @@ -161,6 +189,15 @@ static int try_to_find_probe_trace_events(struct perf_probe_event *pev, bool need_dwarf = perf_probe_event_need_dwarf(pev); int fd, ntevs; + if (pev->upid) { + if (need_dwarf) { + pr_warning("Debuginfo-analysis is not yet supported" + " with -p/--pid option.\n"); + return -ENOSYS; + } + return convert_name_to_addr(pev); + } + fd = open_vmlinux(); if (fd < 0) { if (need_dwarf) { @@ -383,13 +420,7 @@ end: static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp, struct perf_probe_point *pp) { - pp->function = strdup(tp->symbol); - if (pp->function == NULL) - return -ENOMEM; - pp->offset = tp->offset; - pp->retprobe = tp->retprobe; - - return 0; + return convert_to_perf_probe_point(tp, pp); } static int try_to_find_probe_trace_events(struct perf_probe_event *pev, @@ -400,6 +431,9 @@ static int try_to_find_probe_trace_events(struct perf_probe_event *pev, pr_warning("Debuginfo-analysis is not supported.\n"); return -ENOSYS; } + if (pev->upid) + return convert_name_to_addr(pev); + return 0; } @@ -475,6 +509,109 @@ static bool check_event_name(const char *name) return true; } +/* + * uprobe_events only accepts address: + * Convert function and any offset to address + */ +static int convert_name_to_addr(struct perf_probe_event *pev) +{ + struct perf_probe_point *pp = &pev->point; + struct perf_session *session; + struct thread *thread; + struct symbol *sym; + struct map *map; + char *name = NULL, *tmpname = NULL; + int ret = -EINVAL; + unsigned long long vaddr; + + /* check if user has specifed a virtual address */ + vaddr = strtoul(pp->function, NULL, 0); + session = perf_session__new(NULL, O_WRONLY, false, false); + if (!session) { + pr_warning("Cannot initialize perf session.\n"); + return -ENOMEM; + } + + event__synthesize_thread(pev->upid, event__process, session); + thread = perf_session__findnew(session, pev->upid); + if (!thread) { + pr_warning("Cannot initialize perf session.\n"); + goto free_session; + } + + if (vaddr) { + if (pev->event) { + ret = 0; + goto free_session; + } + + pev->event = zalloc(sizeof(char *) * MAX_PROBE_ARGS); + if (!pev->event) { + ret = -ENOMEM; + pr_warning("Cannot allocate memory for event.\n"); + goto free_session; + } + + sym = map_groups__find_symbol(&thread->mg, MAP__FUNCTION, + vaddr, NULL, NULL); + if (!sym) + snprintf(pev->event, MAX_PROBE_ARGS, "p_%llx", vaddr); + else + snprintf(pev->event, MAX_PROBE_ARGS, "%s_%llx", + sym->name, vaddr); + ret = 0; + goto free_session; + } + + if (!pp->file) + /* Lets find the function in the executable. */ + name = thread->comm; + else { + tmpname = realpath(pp->file, NULL); + if (tmpname) + name = basename(tmpname); + else + name = pp->file; + } + + map = map_groups__find_by_name(&thread->mg, MAP__FUNCTION, name); + if (pp->file && tmpname) + free(tmpname); + if (!map) { + pr_warning("Cannot find appropriate DSO.\n"); + goto free_session; + } + + sym = map__find_symbol_by_name(map, pp->function, NULL); + if (!sym) { + pr_warning("Cannot find appropriate DSO.\n"); + goto free_session; + } + + if (map->start > sym->start) + vaddr = map->start; + vaddr += sym->start + pp->offset + map->pgoff; + pp->offset = 0; + + if (!pev->event) + pev->event = pp->function; + else + free(pp->function); + pp->function = zalloc(sizeof(char *) * MAX_PROBE_ARGS); + if (!pp->function) { + ret = -ENOMEM; + pr_warning("Failed to allocate memory by zalloc.\n"); + goto free_session; + } + e_snprintf(pp->function, MAX_PROBE_ARGS, "0x%llx", vaddr); + ret = 0; + +free_session: + if (session) + perf_session__delete(session); + return ret; +} + /* Parse probepoint definition. */ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) { @@ -614,6 +751,11 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) return -EINVAL; } + if (pev->upid && !pp->function) { + semantic_error("No function specified for uprobes"); + return -EINVAL; + } + if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) { semantic_error("Offset/Line/Lazy pattern can't be used with " "return probe."); @@ -623,6 +765,11 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n", pp->function, pp->file, pp->line, pp->offset, pp->retprobe, pp->lazy_line); + + if (pev->upid && perf_probe_event_need_dwarf(pev)) { + semantic_error("no dwarf based probes for uprobes."); + return -EINVAL; + } return 0; } @@ -774,7 +921,8 @@ bool perf_probe_event_need_dwarf(struct perf_probe_event *pev) { int i; - if (pev->point.file || pev->point.line || pev->point.lazy_line) + if ((pev->point.file && !pev->upid) || pev->point.line || + pev->point.lazy_line) return true; for (i = 0; i < pev->nargs; i++) @@ -1065,10 +1213,22 @@ char *synthesize_probe_trace_command(struct probe_trace_event *tev) if (buf == NULL) return NULL; - len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu", - tp->retprobe ? 'r' : 'p', - tev->group, tev->event, - tp->symbol, tp->offset); + if (tev->upid) + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %d:%s", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tev->upid, tp->symbol); + else if (tp->offset) + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s+%lu", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->symbol, tp->offset); + else + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->symbol); + if (len <= 0) goto error; @@ -1087,7 +1247,7 @@ error: } static int convert_to_perf_probe_event(struct probe_trace_event *tev, - struct perf_probe_event *pev) + struct perf_probe_event *pev, bool is_kprobe) { char buf[64] = ""; int i, ret; @@ -1099,7 +1259,11 @@ static int convert_to_perf_probe_event(struct probe_trace_event *tev, return -ENOMEM; /* Convert trace_point to probe_point */ - ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point); + if (is_kprobe) + ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point); + else + ret = convert_to_perf_probe_point(&tev->point, &pev->point); + if (ret < 0) return ret; @@ -1193,7 +1357,7 @@ static void clear_probe_trace_event(struct probe_trace_event *tev) memset(tev, 0, sizeof(*tev)); } -static int open_kprobe_events(bool readwrite) +static int open_probe_events(bool readwrite, bool is_kprobe) { char buf[PATH_MAX]; const char *__debugfs; @@ -1204,8 +1368,13 @@ static int open_kprobe_events(bool readwrite) pr_warning("Debugfs is not mounted.\n"); return -ENOENT; } + if (is_kprobe) + ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", + __debugfs); + else + ret = e_snprintf(buf, PATH_MAX, "%stracing/uprobe_events", + __debugfs); - ret = e_snprintf(buf, PATH_MAX, "%stracing/kprobe_events", __debugfs); if (ret >= 0) { pr_debug("Opening %s write=%d\n", buf, readwrite); if (readwrite && !probe_event_dry_run) @@ -1216,16 +1385,29 @@ static int open_kprobe_events(bool readwrite) if (ret < 0) { if (errno == ENOENT) - pr_warning("kprobe_events file does not exist - please" - " rebuild kernel with CONFIG_KPROBE_EVENT.\n"); + pr_warning("%s file does not exist - please" + " rebuild kernel with CONFIG_%s_EVENT.\n", + is_kprobe ? "kprobe_events" : "uprobe_events", + is_kprobe ? "KPROBE" : "UPROBE"); else - pr_warning("Failed to open kprobe_events file: %s\n", - strerror(errno)); + pr_warning("Failed to open %s file: %s\n", + is_kprobe ? "kprobe_events" : "uprobe_events", + strerror(errno)); } return ret; } -/* Get raw string list of current kprobe_events */ +static int open_kprobe_events(bool readwrite) +{ + return open_probe_events(readwrite, 1); +} + +static int open_uprobe_events(bool readwrite) +{ + return open_probe_events(readwrite, 0); +} + +/* Get raw string list of current kprobe_events or uprobe_events */ static struct strlist *get_probe_trace_command_rawlist(int fd) { int ret, idx; @@ -1290,36 +1472,26 @@ static int show_perf_probe_event(struct perf_probe_event *pev) return ret; } -/* List up current perf-probe events */ -int show_perf_probe_events(void) +static int __show_perf_probe_events(int fd, bool is_kprobe) { - int fd, ret; + int ret = 0; struct probe_trace_event tev; struct perf_probe_event pev; struct strlist *rawlist; struct str_node *ent; - setup_pager(); - ret = init_vmlinux(); - if (ret < 0) - return ret; - memset(&tev, 0, sizeof(tev)); memset(&pev, 0, sizeof(pev)); - fd = open_kprobe_events(false); - if (fd < 0) - return fd; - rawlist = get_probe_trace_command_rawlist(fd); - close(fd); if (!rawlist) return -ENOENT; strlist__for_each(ent, rawlist) { ret = parse_probe_trace_command(ent->s, &tev); if (ret >= 0) { - ret = convert_to_perf_probe_event(&tev, &pev); + ret = convert_to_perf_probe_event(&tev, &pev, + is_kprobe); if (ret >= 0) ret = show_perf_probe_event(&pev); } @@ -1329,6 +1501,31 @@ int show_perf_probe_events(void) break; } strlist__delete(rawlist); + return ret; +} + +/* List up current perf-probe events */ +int show_perf_probe_events(void) +{ + int fd, ret; + + setup_pager(); + fd = open_kprobe_events(false); + if (fd < 0) + return fd; + + ret = init_vmlinux(); + if (ret < 0) + return ret; + + ret = __show_perf_probe_events(fd, true); + close(fd); + + fd = open_uprobe_events(false); + if (fd >= 0) { + ret = __show_perf_probe_events(fd, false); + close(fd); + } return ret; } @@ -1438,7 +1635,10 @@ static int __add_probe_trace_events(struct perf_probe_event *pev, const char *event, *group; struct strlist *namelist; - fd = open_kprobe_events(true); + if (pev->upid) + fd = open_uprobe_events(true); + else + fd = open_kprobe_events(true); if (fd < 0) return fd; /* Get current event names */ @@ -1452,17 +1652,28 @@ static int __add_probe_trace_events(struct perf_probe_event *pev, printf("Add new event%s\n", (ntevs > 1) ? "s:" : ":"); for (i = 0; i < ntevs; i++) { tev = &tevs[i]; - if (pev->event) - event = pev->event; - else - if (pev->point.function) - event = pev->point.function; - else - event = tev->point.symbol; + if (pev->group) group = pev->group; - else + else if (!pev->upid) group = PERFPROBE_GROUP; + else { + /* + * For uprobes based probes create a group + * probe_<pid>. + */ + snprintf(buf, 64, "%s_%d", PERFPROBE_GROUP, pev->upid); + group = buf; + } + + tev->group = strdup(group); + + if (pev->event) + event = pev->event; + else if (pev->point.function) + event = pev->point.function; + else + event = tev->point.symbol; /* Get an unused new event name */ ret = get_new_event_name(buf, 64, event, @@ -1470,9 +1681,8 @@ static int __add_probe_trace_events(struct perf_probe_event *pev, if (ret < 0) break; event = buf; - tev->event = strdup(event); - tev->group = strdup(group); + if (tev->event == NULL || tev->group == NULL) { ret = -ENOMEM; break; @@ -1570,6 +1780,11 @@ static int convert_to_probe_trace_events(struct perf_probe_event *pev, } } + if (pev->upid) { + tev->upid = pev->upid; + return 1; + } + /* Currently just checking function name from symbol map */ sym = map__find_symbol_by_name(machine.vmlinux_maps[MAP__FUNCTION], tev->point.symbol, NULL); @@ -1597,15 +1812,19 @@ struct __event_package { int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, bool force_add, int max_tevs) { - int i, j, ret; + int i, j, ret = 0; struct __event_package *pkgs; pkgs = zalloc(sizeof(struct __event_package) * npevs); if (pkgs == NULL) return -ENOMEM; - /* Init vmlinux path */ - ret = init_vmlinux(); + if (!pevs->upid) + /* Init vmlinux path */ + ret = init_vmlinux(); + else + ret = init_perf_uprobes(); + if (ret < 0) return ret; @@ -1665,23 +1884,15 @@ error: return ret; } -static int del_trace_probe_event(int fd, const char *group, - const char *event, struct strlist *namelist) +static int del_trace_probe_event(int fd, const char *buf, + struct strlist *namelist) { - char buf[128]; struct str_node *ent, *n; - int found = 0, ret = 0; - - ret = e_snprintf(buf, 128, "%s:%s", group, event); - if (ret < 0) { - pr_err("Failed to copy event."); - return ret; - } + int ret = -1; if (strpbrk(buf, "*?")) { /* Glob-exp */ strlist__for_each_safe(ent, n, namelist) if (strglobmatch(ent->s, buf)) { - found++; ret = __del_trace_probe_event(fd, ent); if (ret < 0) break; @@ -1690,40 +1901,44 @@ static int del_trace_probe_event(int fd, const char *group, } else { ent = strlist__find(namelist, buf); if (ent) { - found++; ret = __del_trace_probe_event(fd, ent); if (ret >= 0) strlist__remove(namelist, ent); } } - if (found == 0 && ret >= 0) - pr_info("Info: Event \"%s\" does not exist.\n", buf); - return ret; } -int del_perf_probe_events(struct strlist *dellist) +int del_perf_probe_events(struct strlist *dellist, pid_t pid) { - int fd, ret = 0; + int ret = -1, ufd = -1, kfd = -1; + char buf[128]; const char *group, *event; char *p, *str; struct str_node *ent; - struct strlist *namelist; + struct strlist *namelist = NULL, *unamelist = NULL; - fd = open_kprobe_events(true); - if (fd < 0) - return fd; /* Get current event names */ - namelist = get_probe_trace_event_names(fd, true); - if (namelist == NULL) - return -EINVAL; + if (!pid) { + kfd = open_kprobe_events(true); + if (kfd < 0) + return kfd; + namelist = get_probe_trace_event_names(kfd, true); + } + + ufd = open_uprobe_events(true); + if (ufd >= 0) + unamelist = get_probe_trace_event_names(ufd, true); + + if (namelist == NULL && unamelist == NULL) + goto error; strlist__for_each(ent, dellist) { str = strdup(ent->s); if (str == NULL) { ret = -ENOMEM; - break; + goto error; } pr_debug("Parsing: %s\n", str); p = strchr(str, ':'); @@ -1735,15 +1950,36 @@ int del_perf_probe_events(struct strlist *dellist) group = "*"; event = str; } + + ret = e_snprintf(buf, 128, "%s:%s", group, event); + if (ret < 0) { + pr_err("Failed to copy event."); + free(str); + goto error; + } + pr_debug("Group: %s, Event: %s\n", group, event); - ret = del_trace_probe_event(fd, group, event, namelist); + if (!pid && namelist) + ret = del_trace_probe_event(kfd, buf, namelist); + if (unamelist && ret != 0) + ret = del_trace_probe_event(ufd, buf, unamelist); + free(str); - if (ret < 0) - break; + if (ret != 0) + pr_info("Info: Event \"%s\" does not exist.\n", buf); } - strlist__delete(namelist); - close(fd); +error: + if (kfd >= 0) { + if (namelist) + strlist__delete(namelist); + close(kfd); + } + + if (ufd >= 0) { + if (unamelist) + strlist__delete(unamelist); + close(ufd); + } return ret; } - diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h index 5af3924..27f052c 100644 --- a/tools/perf/util/probe-event.h +++ b/tools/perf/util/probe-event.h @@ -6,7 +6,7 @@ extern bool probe_event_dry_run; -/* kprobe-tracer tracing point */ +/* kprobe-tracer and uprobe-tracer tracing point */ struct probe_trace_point { char *symbol; /* Base symbol */ unsigned long offset; /* Offset from symbol */ @@ -19,7 +19,7 @@ struct probe_trace_arg_ref { long offset; /* Offset value */ }; -/* kprobe-tracer tracing argument */ +/* kprobe-tracer and uprobe-tracer tracing argument */ struct probe_trace_arg { char *name; /* Argument name */ char *value; /* Base value */ @@ -27,12 +27,13 @@ struct probe_trace_arg { struct probe_trace_arg_ref *ref; /* Referencing offset */ }; -/* kprobe-tracer tracing event (point + arg) */ +/* kprobe-tracer and uprobe-tracer tracing event (point + arg) */ struct probe_trace_event { char *event; /* Event name */ char *group; /* Group name */ struct probe_trace_point point; /* Trace point */ int nargs; /* Number of args */ + pid_t upid; /* uprobes only */ struct probe_trace_arg *args; /* Arguments */ }; @@ -68,6 +69,7 @@ struct perf_probe_event { char *group; /* Group name */ struct perf_probe_point point; /* Probe point */ int nargs; /* Number of arguments */ + pid_t upid; struct perf_probe_arg *args; /* Arguments */ }; @@ -112,7 +114,7 @@ extern int parse_line_range_desc(const char *cmd, struct line_range *lr); extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, bool force_add, int max_probe_points); -extern int del_perf_probe_events(struct strlist *dellist); +extern int del_perf_probe_events(struct strlist *dellist, pid_t pid); extern int show_perf_probe_events(void); extern int show_line_range(struct line_range *lr); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |