Hi Wang,
On 2015/06/27 16:34, Wangnan (F) wrote:
>
>
> On 2015/6/27 15:29, Masami Hiramatsu wrote:
>> On 2015/06/25 19:37, Wang Nan wrote:
>>> Before this patch, add_perf_probe_events() init symbol maps only for
>>> uprobe if the first pev passed to it is a uprobe event. However, with
>>> the incoming BPF uprobe support, now it will be possible to pass an
>>> array with combined kprobe and uprobe events to add_perf_probe_events().
>> This description is not correct. Actually, add_perf_probe_events already
>> supports mix of uprobes and kprobes. However, from the command line syntax
>> constrains the first elements of the probe_event arrays must be kprobes.
>> So, if the array starts with uprobes, no kprobes should be there.
>>
>>> This patch check all pevs instead of the first one, and init kernel
>>> symbol if any events is not uprobe.
>> Anyway, I prefer to call init_symbol_maps() with "false" :)
>
> I also prefer "false", so I try to find whether all events are uprobe
> instead
> of init_symbol_maps(true).
>
> So for this patch only commit message needs to be corrected, the code is
> no problem, right?
>
No, I meant you just need to change one line, as below.
- ret = init_symbol_maps(pevs->uprobes);
+ ret = init_symbol_maps(false);
That's enough, I don't like to introduce a bool flag and loop.
Thanks,
> Thank you.
>
>> Thank you,
>>
>>> Signed-off-by: Wang Nan <wangn...@huawei.com>
>>> ---
>>> tools/perf/util/probe-event.c | 15 ++++++++++++++-
>>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
>>> index b386d2f..a2b3026 100644
>>> --- a/tools/perf/util/probe-event.c
>>> +++ b/tools/perf/util/probe-event.c
>>> @@ -2802,8 +2802,21 @@ int cleanup_perf_probe_event(struct perf_probe_event
>>> *pev)
>>> int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, bool
>>> cleanup)
>>> {
>>> int i, ret;
>>> + bool user_only = true;
>>>
>>> - ret = init_symbol_maps(pevs->uprobes);
>>> + /* If any pev is kprobe, init kernel symbols. */
>>> + for (i = 0; i < npevs; i++) {
>>> + if (!pevs[i].uprobes) {
>>> + user_only = false;
>>> + break;
>>> + }
>>> + }
>>> +
>>> + /*
>>> + * Compiler can drop user_only:
>>> + * ret = init_symbol_maps(i >= npevs);
>>> + */
>>> + ret = init_symbol_maps(user_only);
>>> if (ret < 0)
>>> return ret;
>>>
>>>
>>
>
>
>
--
Masami HIRAMATSU
Linux Technology Research Center, System Productivity Research Dept.
Center for Technology Innovation - Systems Engineering
Hitachi, Ltd., Research & Development Group
E-mail: masami.hiramatsu...@hitachi.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
