From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S966888AbcAZRIY (ORCPT ); Tue, 26 Jan 2016 12:08:24 -0500 Received: from mail-oi0-f51.google.com ([209.85.218.51]:36659 "EHLO mail-oi0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934100AbcAZRIW (ORCPT ); Tue, 26 Jan 2016 12:08:22 -0500 MIME-Version: 1.0 In-Reply-To: <20160125212955.GG22501@kernel.org> References: <1452807977-8069-1-git-send-email-mathieu.poirier@linaro.org> <1452807977-8069-21-git-send-email-mathieu.poirier@linaro.org> <20160125211242.GF22501@kernel.org> <20160125212955.GG22501@kernel.org> Date: Tue, 26 Jan 2016 10:08:21 -0700 Message-ID: Subject: Re: [PATCH V8 20/23] perf tools: making function set_max_cpu_num() non static From: Mathieu Poirier To: Arnaldo Carvalho de Melo Cc: linux-doc@vger.kernel.org, Ingo Molnar , "linux-arm-kernel@lists.infradead.org" , "linux-kernel@vger.kernel.org" , Peter Zijlstra Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 25 January 2016 at 14:29, Arnaldo Carvalho de Melo wrote: > Em Mon, Jan 25, 2016 at 06:12:42PM -0300, Arnaldo Carvalho de Melo escreveu: >> Em Mon, Jan 25, 2016 at 01:46:22PM -0700, Mathieu Poirier escreveu: >> > On 14 January 2016 at 14:46, Mathieu Poirier wrote: >> > > For memory allocation purposes, code located in other places >> > > then util/cpumap.c may want to know how many CPUs the system has. >> > > >> > > This patch is making function set_max_cpu_num() available to >> > > other parts of the perf tool so that global variable >> > > 'max_cpu_num' gets the right value when referenced by cpu__max_cpu(). >> > > >> > > Cc: Peter Zijlstra >> > > Cc: Ingo Molnar >> > > Cc: Arnaldo Carvalho de Melo >> > > Signed-off-by: Mathieu Poirier >> > > --- >> > > tools/perf/util/cpumap.c | 2 +- >> > > tools/perf/util/cpumap.h | 1 + >> > > 2 files changed, 2 insertions(+), 1 deletion(-) >> > > >> > > diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c >> > > index 10af1e7524fb..ae179320c0c0 100644 >> > > --- a/tools/perf/util/cpumap.c >> > > +++ b/tools/perf/util/cpumap.c >> > > @@ -380,7 +380,7 @@ out: >> > > } >> > > >> > > /* Determine highest possible cpu in the system for sparse allocation */ >> > > -static void set_max_cpu_num(void) >> > > +void set_max_cpu_num(void) >> > > { >> > > const char *mnt; >> > > char path[PATH_MAX]; >> > > diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h >> > > index 85f7772457fa..45fa963345eb 100644 >> > > --- a/tools/perf/util/cpumap.h >> > > +++ b/tools/perf/util/cpumap.h >> > > @@ -14,6 +14,7 @@ struct cpu_map { >> > > int map[]; >> > > }; >> > > >> > > +void set_max_cpu_num(void); >> > > struct cpu_map *cpu_map__new(const char *cpu_list); >> > > struct cpu_map *cpu_map__empty_new(int nr); >> > > struct cpu_map *cpu_map__dummy_new(void); >> > > -- >> > > 2.1.4 >> > > >> > >> > Arnaldo, >> > >> > I can't queue this patch for 4.6 without at least a reviewed by from you. >> >> This one I remember, looks ugly, the name set_max_cpu_num() looks >> strange, when that was restricted (static) to that cpumap.c file, it >> wasn't a problem, exporting it for wider usage looks bad. >> >> You've been waiting for this for quite a while, it seems, lemme stop >> what I am doing to check this... > > So, please check the patch below, what you need then is just to use > cpu__max_cpu(). I like your approach - thanks for the review. I will spin V9 when I have received Adrian's comments. Mathieu > > - Arnaldo > > diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c > index fa935093a599..9bcf2bed3a6d 100644 > --- a/tools/perf/util/cpumap.c > +++ b/tools/perf/util/cpumap.c > @@ -8,6 +8,10 @@ > #include > #include "asm/bug.h" > > +static int max_cpu_num; > +static int max_node_num; > +static int *cpunode_map; > + > static struct cpu_map *cpu_map__default_new(void) > { > struct cpu_map *cpus; > @@ -486,6 +490,32 @@ out: > pr_err("Failed to read max nodes, using default of %d\n", max_node_num); > } > > +int cpu__max_node(void) > +{ > + if (unlikely(!max_node_num)) > + set_max_node_num(); > + > + return max_node_num; > +} > + > +int cpu__max_cpu(void) > +{ > + if (unlikely(!max_cpu_num)) > + set_max_cpu_num(); > + > + return max_cpu_num; > +} > + > +int cpu__get_node(int cpu) > +{ > + if (unlikely(cpunode_map == NULL)) { > + pr_debug("cpu_map not initialized\n"); > + return -1; > + } > + > + return cpunode_map[cpu]; > +} > + > static int init_cpunode_map(void) > { > int i; > diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h > index 71c41b9efabb..81a2562aaa2b 100644 > --- a/tools/perf/util/cpumap.h > +++ b/tools/perf/util/cpumap.h > @@ -57,37 +57,11 @@ static inline bool cpu_map__empty(const struct cpu_map *map) > return map ? map->map[0] == -1 : true; > } > > -int max_cpu_num; > -int max_node_num; > -int *cpunode_map; > - > int cpu__setup_cpunode_map(void); > > -static inline int cpu__max_node(void) > -{ > - if (unlikely(!max_node_num)) > - pr_debug("cpu_map not initialized\n"); > - > - return max_node_num; > -} > - > -static inline int cpu__max_cpu(void) > -{ > - if (unlikely(!max_cpu_num)) > - pr_debug("cpu_map not initialized\n"); > - > - return max_cpu_num; > -} > - > -static inline int cpu__get_node(int cpu) > -{ > - if (unlikely(cpunode_map == NULL)) { > - pr_debug("cpu_map not initialized\n"); > - return -1; > - } > - > - return cpunode_map[cpu]; > -} > +int cpu__max_node(void); > +int cpu__max_cpu(void); > +int cpu__get_node(int cpu); > > int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res, > int (*f)(struct cpu_map *map, int cpu, void *data),