Re: [PATCH 3/5] MIPS: perf: Fix perf with MT counting other threads

[kbuild test robot <lkp@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 3092 bytes --]

Hi Matt,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on tip/perf/core]
[also build test ERROR on v4.16 next-20180403]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Matt-Redfearn/MIPS-perf-MT-fixes-and-improvements/20180404-011026
config: mips-gpr_defconfig (attached as .config)
compiler: mipsel-linux-gnu-gcc (Debian 7.2.0-11) 7.2.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # save the attached .config to linux build tree
        make.cross ARCH=mips 

All errors (new ones prefixed by >>):

   arch/mips/kernel/perf_event_mipsxx.c: In function 'mipsxx_pmu_enable_event':
>> arch/mips/kernel/perf_event_mipsxx.c:343:22: error: expected expression before ')' token
     } else if (range > V) {
                         ^
   arch/mips/kernel/perf_event_mipsxx.c: In function 'mipspmu_perf_event_encode':
>> arch/mips/kernel/perf_event_mipsxx.c:675:28: error: 'const struct mips_perf_event' has no member named 'range'
      return ((unsigned int)pev->range << 24) |
                               ^~

vim +343 arch/mips/kernel/perf_event_mipsxx.c

   325	
   326	static void mipsxx_pmu_enable_event(struct hw_perf_event *evt, int idx)
   327	{
   328		struct perf_event *event = container_of(evt, struct perf_event, hw);
   329		struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
   330		unsigned int range = evt->event_base >> 24;
   331	
   332		WARN_ON(idx < 0 || idx >= mipspmu.num_counters);
   333	
   334		cpuc->saved_ctrl[idx] = M_PERFCTL_EVENT(evt->event_base & 0xff) |
   335			(evt->config_base & M_PERFCTL_CONFIG_MASK) |
   336			/* Make sure interrupt enabled. */
   337			MIPS_PERFCTRL_IE;
   338	
   339		if (IS_ENABLED(CONFIG_CPU_BMIPS5000)) {
   340			/* enable the counter for the calling thread */
   341			cpuc->saved_ctrl[idx] |=
   342				(1 << (12 + vpe_id())) | BRCM_PERFCTRL_TC;
 > 343		} else if (range > V) {
   344			/* The counter is processor wide. Set it up to count all TCs. */
   345			pr_debug("Enabling perf counter for all TCs\n");
   346			cpuc->saved_ctrl[idx] |= M_TC_EN_ALL;
   347		} else {
   348			unsigned int cpu, ctrl;
   349	
   350			/*
   351			 * Set up the counter for a particular CPU when event->cpu is
   352			 * a valid CPU number. Otherwise set up the counter for the CPU
   353			 * scheduling this thread.
   354			 */
   355			cpu = (event->cpu >= 0) ? event->cpu : smp_processor_id();
   356	
   357			ctrl = M_PERFCTL_VPEID(cpu_vpe_id(&cpu_data[cpu]));
   358			ctrl |= M_TC_EN_VPE;
   359			cpuc->saved_ctrl[idx] |= ctrl;
   360			pr_debug("Enabling perf counter for CPU%d\n", cpu);
   361		}
   362		/*
   363		 * We do not actually let the counter run. Leave it until start().
   364		 */
   365	}
   366	

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 19550 bytes --]