已安装用于麦克风的PAPI，正确发现组件，但PAPI_add_event失败

Question

我编译并安装了最新的PAPI 5.4.0.0 for MIC。我的配置选项是：

CC=icc CXX=icpc ./configure --with-mic --host=x86_64-k1om-linux --with-arch=k1om --with-ffsll --with-walltimer=cycle --with-tls=__thread --with-virtualtimer=clock_thread_cputime_id 

问题是PAPI_add_event总是失败，无论是在我自己的程序中还是在测试程序中。报告的错误是“无效参数”。知道为什么吗？我尝试重新配置和重新安装，但问题仍然存在。

[root@mic2 bin]# ./papi_cost
Cost of execution for PAPI start/stop, read and accum.
This test takes a while. Please be patient...
cost.c                                       FAILED
Line # 231
Error in PAPI_add_event: Invalid argument

这些是'api_component_avail‘显示的结果。对我来说这看起来很正常。它也与这个线程相匹配：https://software.intel.com/en-us/forums/topic/530598

[root@mic2 bin]# ./papi_component_avail
Available components and hardware information.
--------------------------------------------------------------------------------
PAPI Version             : 5.4.0.0
 Vendor string and code   : GenuineIntel (1)
 Model string and code    : 0b/01 (1)
 CPU Revision             : 3.000000
 CPUID Info               : Family: 11  Model: 1  Stepping: 3
 CPU Max Megahertz        : 1052
 CPU Min Megahertz        : 842
 Hdw Threads per core     : 4
 Cores per Socket         : 60
 Sockets                  : 1
 CPUs per Node            : 240
 Total CPUs               : 240
 Running in a VM          : no
 Number Hardware Counters : 2
 Max Multiplex Counters   : 128
--------------------------------------------------------------------------------

 Compiled-in components:
 Name:   perf_event              Linux perf_event CPU counters
 Name:   perf_event_uncore       Linux perf_event CPU uncore and northbridge
   \-> Disabled: No uncore PMUs or events found

 Active components:
 Name:   perf_event              Linux perf_event CPU counters
                                 Native: 140, Preset: 14, Counters: 2
                                 PMU's supported: perf, knc, perf_raw


--------------------------------------------------------------------------------
 component.c                             PASSED

Answer 1

我将PAPI的版本从最新的5.4.0版本降低到之前的5.3.2版本，问题就解决了。因此，我认为这是当前版本的一个bug。