linux-kernel - Re: [PATCH 6/7 V2] perf: Add support to dynamically get cacheline size

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20140530162042.GI191965@redhat.com>
Date:	Fri, 30 May 2014 12:20:42 -0400
From:	Don Zickus <dzickus@...hat.com>
To:	Arnaldo Carvalho de Melo <acme@...nel.org>
Cc:	jolsa@...hat.com, Peter Zijlstra <peterz@...radead.org>,
	LKML <linux-kernel@...r.kernel.org>, namhyung@...il.com,
	eranian@...gle.com, Andi Kleen <andi@...stfloor.org>
Subject: Re: [PATCH 6/7 V2] perf: Add support to dynamically get cacheline
 size

On Fri, May 30, 2014 at 12:28:30PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Fri, May 30, 2014 at 10:50:25AM -0400, Don Zickus escreveu:
> > Different arches may have different cacheline sizes.  Look it up and set
> > a global variable for reference.
> 
> [acme@zoo linux]$ strings `which getconf` | grep LINESIZE
> LEVEL1_ICACHE_LINESIZE
> LEVEL1_DCACHE_LINESIZE
> LEVEL2_CACHE_LINESIZE
> LEVEL3_CACHE_LINESIZE
> LEVEL4_CACHE_LINESIZE
> [acme@zoo linux]$
> 
> man sysconf
> 
> That is what I use in pahole, for instance:
> 
> [acme@zoo pahole]$ grep sysconf *.c
> dwarves_fprintf.c:		long sys_cacheline_size =
> sysconf(_SC_LEVEL1_DCACHE_LINESIZE);
> [acme@zoo pahole]$

Well that does look easier.  I'll hack that up to see if it gives me the
same answer.

Thanks!

Cheers,
Don

> 
> - Arnaldo
>  
> > Signed-off-by: Don Zickus <dzickus@...hat.com>
> > 
> > ---
> > V3: remove unneeded cpumap.h (Namhyung Kim)
> > V2: change to be global and setup in perf.c
> >     use filename__read_int for setup
> > ---
> >  tools/perf/perf.c        |  5 +++++
> >  tools/perf/util/cpumap.c | 27 +++++++++++++++++++++++++++
> >  tools/perf/util/cpumap.h |  3 +++
> >  tools/perf/util/util.c   |  1 +
> >  tools/perf/util/util.h   |  1 +
> >  5 files changed, 37 insertions(+)
> > 
> > diff --git a/tools/perf/perf.c b/tools/perf/perf.c
> > index 431798a..dabf08b 100644
> > --- a/tools/perf/perf.c
> > +++ b/tools/perf/perf.c
> > @@ -13,6 +13,7 @@
> >  #include "util/quote.h"
> >  #include "util/run-command.h"
> >  #include "util/parse-events.h"
> > +#include "util/cpumap.h"
> >  #include <api/fs/debugfs.h>
> >  #include <pthread.h>
> >  
> > @@ -459,6 +460,10 @@ int main(int argc, const char **argv)
> >  	/* The page_size is placed in util object. */
> >  	page_size = sysconf(_SC_PAGE_SIZE);
> >  
> > +	/* The cacheline_size is placed in util objet */
> > +	if (cpu__setup_cacheline_size() < 0) 
> > +		goto out;
> > +
> >  	cmd = perf_extract_argv0_path(argv[0]);
> >  	if (!cmd)
> >  		cmd = "perf-help";
> > diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
> > index c4e55b7..1029982 100644
> > --- a/tools/perf/util/cpumap.c
> > +++ b/tools/perf/util/cpumap.c
> > @@ -477,3 +477,30 @@ int cpu__setup_cpunode_map(void)
> >  	closedir(dir1);
> >  	return 0;
> >  }
> > +
> > +int cpu__setup_cacheline_size(void)
> > +{
> > +	const char *mnt;
> > +	char path[PATH_MAX];
> > +	int n, size;
> > +	
> > +
> > +	mnt = sysfs__mountpoint();
> > +	if (!mnt)
> > +		return -1;
> > +
> > +	n = snprintf(path, PATH_MAX, "%s/devices/system/cpu/cpu0/cache/index0/coherency_line_size", mnt);
> > +	if (n == PATH_MAX) {
> > +		pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
> > +		return -1;
> > +	}
> > +
> > +	if (filename__read_int(path, &size)) {
> > +		pr_err("Can not read cacheline size\n");
> > +		return -1;
> > +	}
> > +
> > +	cacheline_size = size;
> > +
> > +	return 0;
> > +}
> > diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h
> > index 61a6548..507d7fd 100644
> > --- a/tools/perf/util/cpumap.h
> > +++ b/tools/perf/util/cpumap.h
> > @@ -5,6 +5,7 @@
> >  #include <stdbool.h>
> >  
> >  #include "perf.h"
> > +#include "util/util.h"
> >  #include "util/debug.h"
> >  
> >  struct cpu_map {
> > @@ -81,4 +82,6 @@ static inline int cpu__get_node(int cpu)
> >  	return cpunode_map[cpu];
> >  }
> >  
> > +int cpu__setup_cacheline_size(void);
> > +
> >  #endif /* __PERF_CPUMAP_H */
> > diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
> > index 7fff6be..95aefa7 100644
> > --- a/tools/perf/util/util.c
> > +++ b/tools/perf/util/util.c
> > @@ -17,6 +17,7 @@
> >   * XXX We need to find a better place for these things...
> >   */
> >  unsigned int page_size;
> > +int cacheline_size;
> >  
> >  bool test_attr__enabled;
> >  
> > diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
> > index b03da44..6686436 100644
> > --- a/tools/perf/util/util.h
> > +++ b/tools/perf/util/util.h
> > @@ -304,6 +304,7 @@ char *rtrim(char *s);
> >  void dump_stack(void);
> >  
> >  extern unsigned int page_size;
> > +extern int cacheline_size;
> >  
> >  void get_term_dimensions(struct winsize *ws);
> >  
> > -- 
> > 1.7.11.7
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/