From: Songshan Gong <gongss@linux.vnet.ibm.com>
To: acme@kernel.org, jolsa@kernel.org
Cc: dsahern@gmail.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] [RFC V1]s390/perf: fix 'start' address of module's map
Date: Fri, 8 Jul 2016 10:17:30 +0800 [thread overview]
Message-ID: <5c24562a-17ee-d30e-0308-5ec141ff52f1@linux.vnet.ibm.com> (raw)
In-Reply-To: <1467856176-8712-1-git-send-email-gongss@linux.vnet.ibm.com>
在 7/7/2016 9:49 AM, Song Shan Gong 写道:
> At preset, when creating module's map, perf gets 'start' address by parsing
> 'proc/modules', but it's module base address, isn't the start address of
> '.text' section. In most archs, it's OK. But for s390, it places 'GOT' and
> 'PLT' relocations before '.text' section. So there exists an offset between
> module base address and '.text' section, which will incur wrong symbol
> resolution for modules.
>
> Fix this bug by getting 'start' address of module's map from parsing
> '/sys/module/[module name]/sections/.text', not from '/proc/modules'.
>
> Signed-off-by: Song Shan Gong <gongss@linux.vnet.ibm.com>
> ---
> tools/perf/arch/s390/util/Build | 2 ++
> tools/perf/arch/s390/util/sym-handling.c | 49 ++++++++++++++++++++++++++++++++
> tools/perf/util/machine.c | 6 ++++
> tools/perf/util/machine.h | 2 ++
> 4 files changed, 59 insertions(+)
> create mode 100644 tools/perf/arch/s390/util/sym-handling.c
>
> diff --git a/tools/perf/arch/s390/util/Build b/tools/perf/arch/s390/util/Build
> index 8a61372..5e322ed 100644
> --- a/tools/perf/arch/s390/util/Build
> +++ b/tools/perf/arch/s390/util/Build
> @@ -2,3 +2,5 @@ libperf-y += header.o
> libperf-y += kvm-stat.o
>
> libperf-$(CONFIG_DWARF) += dwarf-regs.o
> +
> +libperf-y += sym-handling.o
> diff --git a/tools/perf/arch/s390/util/sym-handling.c b/tools/perf/arch/s390/util/sym-handling.c
> new file mode 100644
> index 0000000..efe2a50
> --- /dev/null
> +++ b/tools/perf/arch/s390/util/sym-handling.c
> @@ -0,0 +1,49 @@
> +#include <stdio.h>
> +#include <stdlib.h>
> +#include <unistd.h>
> +#include "symbol.h"
> +#include "map.h"
> +#include "util.h"
> +#include "machine.h"
> +
> +int arch__fix_module_baseaddr(struct machine *machine,
> + u64 *start, const char *name)
> +{
> + char path[PATH_MAX];
> + char *module_name = strdup(name);
> + int len = strlen(module_name);
> + FILE *file;
> + int err = 0;
> + u64 text_start;
> + char *line = NULL;
> + size_t n;
> + char *sep;
> +
> + module_name[len - 1] = '\0';
> + module_name += 1;
> + snprintf(path, PATH_MAX, "%s/sys/module/%s/sections/.text",
> + machine->root_dir, module_name);
> + file = fopen(path, "r");
> + if (file == NULL)
> + return -1;
> +
> + len = getline(&line, &n, file);
> + if (len < 0) {
> + err = -1;
> + goto out;
> + }
> + line[--len] = '\0'; /* \n */
> + sep = strrchr(line, 'x');
> + if (sep == NULL) {
> + err = -1;
> + goto out;
> + }
> + hex2u64(sep + 1, &text_start);
> +
> + *start = text_start;
> +out:
> + free(line);
> + fclose(file);
> + free(module_name - 1);
> + return err;
> +}
> diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
> index b177218..e5c2721 100644
> --- a/tools/perf/util/machine.c
> +++ b/tools/perf/util/machine.c
> @@ -1091,12 +1091,18 @@ static int machine__set_modules_path(struct machine *machine)
>
> return map_groups__set_modules_path_dir(&machine->kmaps, modules_path, 0);
> }
> +int __weak arch__fix_module_baseaddr(struct machine *machine __maybe_unused,
> + u64 *start __maybe_unused, const char *name __maybe_unused)
> +{
> + return 0;
> +}
>
> static int machine__create_module(void *arg, const char *name, u64 start)
> {
> struct machine *machine = arg;
> struct map *map;
>
> + arch__fix_module_baseaddr(machine, &start, name);
As the description says, I would change the function name to
'arch__fix_module_text_start';
> map = machine__findnew_module_map(machine, start, name);
> if (map == NULL)
> return -1;
> diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
> index 41ac9cf..da7b6c0 100644
> --- a/tools/perf/util/machine.h
> +++ b/tools/perf/util/machine.h
> @@ -216,6 +216,8 @@ struct symbol *machine__find_kernel_function_by_name(struct machine *machine,
>
> struct map *machine__findnew_module_map(struct machine *machine, u64 start,
> const char *filename);
> +int arch__fix_module_baseaddr(struct machine *machine, u64 *start,
> + const char *name);
>
> int __machine__load_kallsyms(struct machine *machine, const char *filename,
> enum map_type type, bool no_kcore, symbol_filter_t filter);
>
--
SongShan Gong
next prev parent reply other threads:[~2016-07-08 2:17 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-07 1:49 [PATCH] [RFC V1]s390/perf: fix 'start' address of module's map Song Shan Gong
2016-07-08 2:17 ` Songshan Gong [this message]
2016-07-08 15:20 ` Jiri Olsa
2016-07-08 15:18 ` Jiri Olsa
2016-07-11 8:11 ` Songshan Gong
2016-07-13 3:32 ` Songshan Gong
2016-07-08 15:20 ` Jiri Olsa
2016-07-08 15:21 ` Jiri Olsa
2016-07-08 15:21 ` Jiri Olsa
2016-07-11 11:06 ` Songshan Gong
2016-07-11 12:01 ` Jiri Olsa
2016-07-13 6:39 ` Songshan Gong
2016-07-13 9:07 ` Jiri Olsa
2016-07-15 7:45 ` Songshan Gong
2016-07-15 8:27 ` Jiri Olsa
2016-07-15 13:23 ` Arnaldo Carvalho de Melo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5c24562a-17ee-d30e-0308-5ec141ff52f1@linux.vnet.ibm.com \
--to=gongss@linux.vnet.ibm.com \
--cc=acme@kernel.org \
--cc=dsahern@gmail.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).