Re: [PATCH tracing/kprobes v4] perf: Add perf probe subcommand for kprobe-event setup helper

[Masami Hiramatsu <mhiramat@redhat.com>]

Frédéric Weisbecker wrote:
> 2009/10/8 Masami Hiramatsu <mhiramat@redhat.com>:
>> Add perf probe subcommand for kprobe-event setup helper to perf command.
>> This allows user to define kprobe events by C expressions (C line numbers,
>> C function names, and C local variables).
>>
>> Usage
>> -----
>>  perf probe [<options>] -P 'PROBEDEF' [-P 'PROBEDEF' ...]
>>
>>    -k, --vmlinux <file>  vmlinux/module pathname
>>    -P, --probe <p|r:[GRP/]NAME FUNC[+OFFS][@SRC]|@SRC:LINE [ARG ...]>
>>                          probe point definition, where
>>                p:      kprobe probe
>>                r:      kretprobe probe
>>                GRP:    Group name (optional)
>>                NAME:   Event name
>>                FUNC:   Function name
>>                OFFS:   Offset from function entry (in byte)
>>                SRC:    Source code path
>>                LINE:   Line number
>>                ARG:    Probe argument (local variable name or
>>                        kprobe-tracer argument format is supported.)
>>
>> Changes in v4:
>>  - Add _GNU_SOURCE macro for strndup().
>>
>> Changes in v3:
>>  - Remove -r option because perf always be used for online kernel.
>>  - Check malloc/calloc results.
>>
>> Changes in v2:
>>  - Check synthesized string length.
>>  - Rename perf kprobe to perf probe.
>>  - Use spaces for separator and update usage comment.
>>  - Check error paths in parse_probepoint().
>>  - Check optimized-out variables.
>>
>> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
>> Cc: Frederic Weisbecker <fweisbec@gmail.com>
>> Cc: Ingo Molnar <mingo@elte.hu>
>> Cc: Thomas Gleixner <tglx@linutronix.de>
>> Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
>> Cc: Steven Rostedt <rostedt@goodmis.org>
>> Cc: Mike Galbraith <efault@gmx.de>
>> Cc: Paul Mackerras <paulus@samba.org>
>> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
>> Cc: Christoph Hellwig <hch@infradead.org>
>> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
>> Cc: Jim Keniston <jkenisto@us.ibm.com>
>> Cc: Frank Ch. Eigler <fche@redhat.com>
>> ---
>>
> [...]
>> +/* Default vmlinux search paths */
>> +#define NR_SEARCH_PATH 3
>> +const char *default_search_path[NR_SEARCH_PATH] = {
>> +"/lib/modules/%s/build/vmlinux",               /* Custom build kernel */
>> +"/usr/lib/debug/lib/modules/%s/vmlinux",       /* Red Hat debuginfo */
>> +"/boot/vmlinux-debug-%s",                      /* Ubuntu */
>> +};
> [...]
>> +static int open_default_vmlinux(void)
>> +{
>> +       struct utsname uts;
>> +       char fname[MAX_PATH_LEN];
>> +       int fd, ret, i;
>> +
>> +       ret = uname(&uts);
>> +       if (ret) {
>> +               debug("uname() failed.\n");
>> +               return -errno;
>> +       }
>> +       session.release = uts.release;
>> +       for (i = 0; i < NR_SEARCH_PATH; i++) {
>> +               ret = snprintf(fname, MAX_PATH_LEN,
>> +                              default_search_path[i], session.release);
>> +               if (ret >= MAX_PATH_LEN || ret < 0) {
>> +                       debug("Filename(%d,%s) is too long.\n", i, uts.release);
>> +                       errno = E2BIG;
>> +                       return -E2BIG;
>> +               }
>> +               debug("try to open %s\n", fname);
>> +               fd = open(fname, O_RDONLY);
>> +               if (fd >= 0)
>> +                       break;
>> +       }
>> +       return fd;
>> +}
> 
> 
> We have a kind of kernel path finder already inside perf. It might be
> encapsulated
> inside the load_kernel() helper, I don't remember exactly.
> 
> It would be better to make use of such centralized and already
> existing facility.

Agreed. AFAICS, load_kernel() is currently only for symbol resolver,
and we can make it for dwarf analyzer too.

> 
> The patchset looks good. I'll apply and push it soon.

Thanks!

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com