perf kmem: Measure kmalloc/kfree CPU ping-pong call-sites

perf kmem: Measure kmalloc/kfree CPU ping-pong call-sites

Show statistics for allocations and frees on different cpus:

------------------------------------------------------------------------------------------------------
Callsite                           | Total_alloc/Per | Total_req/Per   | Hit   | Ping-pong | Frag
------------------------------------------------------------------------------------------------------
 perf_event_alloc.clone.0+0         |      7504/682   |      7128/648   |     11 |        0 |  5.011%
 alloc_buffer_head+16               |       288/57    |       280/56    |      5 |        0 |  2.778%
 radix_tree_preload+51              |       296/296   |       288/288   |      1 |        0 |  2.703%
 tracepoint_add_probe+32e           |       157/31    |       154/30    |      5 |        0 |  1.911%
 do_maps_open+0                     |       796/12    |       792/12    |     66 |        0 |  0.503%
 sock_alloc_send_pskb+16e           |     23780/495   |     23744/494   |     48 |       38 |  0.151%
 anon_vma_prepare+9a                |      3744/44    |      3740/44    |     85 |        0 |  0.107%
 d_alloc+21                         |     64948/164   |     64944/164   |    396 |        0 |  0.006%
 proc_alloc_inode+23                |    262292/676   |    262288/676   |    388 |        0 |  0.002%
 create_object+28                   |    459600/200   |    459600/200   |   2298 |       71 |  0.000%
 journal_start+67                   |     14440/40    |     14440/40    |    361 |        0 |  0.000%
 get_empty_filp+df                  |     53504/256   |     53504/256   |    209 |        0 |  0.000%
 getname+2a                         |    823296/4096  |    823296/4096  |    201 |        0 |  0.000%
 seq_read+2b0                       |    544768/4096  |    544768/4096  |    133 |        0 |  0.000%
 seq_open+6d                        |     17024/128   |     17024/128   |    133 |        0 |  0.000%
 mmap_region+2e6                    |     11704/88    |     11704/88    |    133 |        0 |  0.000%
 single_open+0                      |      1072/16    |      1072/16    |     67 |        0 |  0.000%
 __alloc_skb+2e                     |     12544/256   |     12544/256   |     49 |       38 |  0.000%
 __sigqueue_alloc+4a                |      1296/144   |      1296/144   |      9 |        8 |  0.000%
 tracepoint_add_probe+6f            |        80/16    |        80/16    |      5 |        0 |  0.000%
------------------------------------------------------------------------------------------------------
...

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: linux-mm@kvack.org <linux-mm@kvack.org>
LKML-Reference: <4B0B6E9F.6020309@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>