From: tip-bot for Adrian Hunter <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: acme@redhat.com, linux-kernel@vger.kernel.org,
eranian@google.com, paulus@samba.org, hpa@zytor.com,
mingo@kernel.org, peterz@infradead.org, efault@gmx.de,
namhyung@gmail.com, jolsa@redhat.com, fweisbec@gmail.com,
adrian.hunter@intel.com, dsahern@gmail.com, tglx@linutronix.de
Subject: [tip:perf/core] perf tools: Make it possible to read object code from kernel modules
Date: Mon, 12 Aug 2013 03:21:21 -0700 [thread overview]
Message-ID: <tip-0131c4ec794a7409eafff0c79105309540aaca4d@git.kernel.org> (raw)
In-Reply-To: <1375875537-4509-7-git-send-email-adrian.hunter@intel.com>
Commit-ID: 0131c4ec794a7409eafff0c79105309540aaca4d
Gitweb: http://git.kernel.org/tip/0131c4ec794a7409eafff0c79105309540aaca4d
Author: Adrian Hunter <adrian.hunter@intel.com>
AuthorDate: Wed, 7 Aug 2013 14:38:50 +0300
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 7 Aug 2013 17:35:32 -0300
perf tools: Make it possible to read object code from kernel modules
The new "object code reading" test shows that it is not possible to read
object code from kernel modules. That is because the mappings do not
map to the dsos. This patch fixes that.
This involves identifying and flagging relocatable (ELF type ET_REL)
files (e.g. kernel modules) for symbol adjustment and updating
map__rip_2objdump() accordingly. The kmodule parameter of
dso__load_sym() is taken into use and the module map altered to map to
the dso.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1375875537-4509-7-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/dso.c | 1 +
tools/perf/util/dso.h | 1 +
tools/perf/util/map.c | 14 +++++++++-----
tools/perf/util/symbol-elf.c | 25 ++++++++++++++++++++++++-
tools/perf/util/symbol.c | 11 ++++++++---
5 files changed, 43 insertions(+), 9 deletions(-)
diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
index 121583d..1955804 100644
--- a/tools/perf/util/dso.c
+++ b/tools/perf/util/dso.c
@@ -419,6 +419,7 @@ struct dso *dso__new(const char *name)
dso->symtab_type = DSO_BINARY_TYPE__NOT_FOUND;
dso->data_type = DSO_BINARY_TYPE__NOT_FOUND;
dso->loaded = 0;
+ dso->rel = 0;
dso->sorted_by_name = 0;
dso->has_build_id = 0;
dso->kernel = DSO_TYPE_USER;
diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h
index 02aadaf..735a837 100644
--- a/tools/perf/util/dso.h
+++ b/tools/perf/util/dso.h
@@ -85,6 +85,7 @@ struct dso {
u8 lname_alloc:1;
u8 sorted_by_name;
u8 loaded;
+ u8 rel;
u8 build_id[BUILD_ID_SIZE];
const char *short_name;
char *long_name;
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 5f662a3..4d599fe 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -248,14 +248,18 @@ size_t map__fprintf_dsoname(struct map *map, FILE *fp)
/*
* objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
- * map->dso->adjust_symbols==1 for ET_EXEC-like cases.
+ * map->dso->adjust_symbols==1 for ET_EXEC-like cases except ET_REL which is
+ * relative to section start.
*/
u64 map__rip_2objdump(struct map *map, u64 rip)
{
- u64 addr = map->dso->adjust_symbols ?
- map->unmap_ip(map, rip) : /* RIP -> IP */
- rip;
- return addr;
+ if (!map->dso->adjust_symbols)
+ return rip;
+
+ if (map->dso->rel)
+ return rip - map->pgoff;
+
+ return map->unmap_ip(map, rip);
}
void map_groups__init(struct map_groups *mg)
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index ed6f443..3eaa7b4 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -599,11 +599,13 @@ int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name,
if (dso->kernel == DSO_TYPE_USER) {
GElf_Shdr shdr;
ss->adjust_symbols = (ehdr.e_type == ET_EXEC ||
+ ehdr.e_type == ET_REL ||
elf_section_by_name(elf, &ehdr, &shdr,
".gnu.prelink_undo",
NULL) != NULL);
} else {
- ss->adjust_symbols = ehdr.e_type == ET_EXEC;
+ ss->adjust_symbols = ehdr.e_type == ET_EXEC ||
+ ehdr.e_type == ET_REL;
}
ss->name = strdup(name);
@@ -676,6 +678,14 @@ int dso__load_sym(struct dso *dso, struct map *map,
bool remap_kernel = false, adjust_kernel_syms = false;
dso->symtab_type = syms_ss->type;
+ dso->rel = syms_ss->ehdr.e_type == ET_REL;
+
+ /*
+ * Modules may already have symbols from kallsyms, but those symbols
+ * have the wrong values for the dso maps, so remove them.
+ */
+ if (kmodule && syms_ss->symtab)
+ symbols__delete(&dso->symbols[map->type]);
if (!syms_ss->symtab) {
syms_ss->symtab = syms_ss->dynsym;
@@ -828,11 +838,24 @@ int dso__load_sym(struct dso *dso, struct map *map,
map_groups__insert(kmap->kmaps, map);
}
+ /*
+ * The initial module mapping is based on
+ * /proc/modules mapped to offset zero.
+ * Overwrite it to map to the module dso.
+ */
+ if (remap_kernel && kmodule) {
+ remap_kernel = false;
+ map->pgoff = shdr.sh_offset;
+ }
+
curr_map = map;
curr_dso = dso;
goto new_symbol;
}
+ if (!kmap)
+ goto new_symbol;
+
snprintf(dso_name, sizeof(dso_name),
"%s%s", dso->short_name, section_name);
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 04300dd..b9056a8 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -854,10 +854,15 @@ int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter)
if (!runtime_ss && syms_ss)
runtime_ss = syms_ss;
- if (syms_ss)
- ret = dso__load_sym(dso, map, syms_ss, runtime_ss, filter, 0);
- else
+ if (syms_ss) {
+ int km;
+
+ km = dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE ||
+ dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE;
+ ret = dso__load_sym(dso, map, syms_ss, runtime_ss, filter, km);
+ } else {
ret = -1;
+ }
if (ret > 0) {
int nr_plt;
next prev parent reply other threads:[~2013-08-12 10:22 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-07 11:38 [PATCH V4 00/13] perf tools: add support for reading object code Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 01/13] perf tools: add test " Adrian Hunter
2013-08-12 10:20 ` [tip:perf/core] perf tests: Add " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 02/13] perf tools: load kernel maps before using Adrian Hunter
2013-08-12 10:20 ` [tip:perf/core] perf symbols: Load " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 03/13] perf tools: make it possible to read object code from vmlinux Adrian Hunter
2013-08-12 10:21 ` [tip:perf/core] perf tools: Make " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 04/13] perf tools: adjust the vmlinux symtab matches kallsyms test Adrian Hunter
2013-08-12 10:21 ` [tip:perf/core] perf tests: Adjust " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 05/13] perf tools: avoid SyS kernel syscall aliases Adrian Hunter
2013-08-12 10:20 ` [tip:perf/core] perf symbols: " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 06/13] perf tools: make it possible to read object code from kernel modules Adrian Hunter
2013-08-12 10:21 ` tip-bot for Adrian Hunter [this message]
2013-08-07 11:38 ` [PATCH V4 07/13] perf tools: add support for reading from /proc/kcore Adrian Hunter
2013-08-12 10:21 ` [tip:perf/core] perf symbols: Add support for reading from /proc/ kcore tip-bot for Adrian Hunter
2013-09-12 13:13 ` Ingo Molnar
2013-09-12 13:16 ` [PATCH] Fix old GCC build error in perf/util/trace-event-parse.c:parse_proc_kallsyms() Ingo Molnar
2013-09-20 9:56 ` [tip:perf/urgent] perf tools: Fix old GCC build error in trace-event-parse.c:parse_proc_kallsyms() tip-bot for Ingo Molnar
2013-08-07 11:38 ` [PATCH V4 08/13] perf tools: adjust the vmlinux symtab matches kallsyms test again Adrian Hunter
2013-08-12 10:21 ` [tip:perf/core] perf tests: Adjust " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 09/13] perf tools: add kcore to the object code reading test Adrian Hunter
2013-08-12 10:21 ` [tip:perf/core] perf tests: Add " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 10/13] perf tools: allow annotation using /proc/kcore Adrian Hunter
2013-08-12 10:22 ` [tip:perf/core] perf annotate: Allow disassembly using /proc/ kcore tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 11/13] perf tools: put dso name in symbol annotation title Adrian Hunter
2013-08-12 10:22 ` [tip:perf/core] perf annotate: Put " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 12/13] perf tools: remove nop at end of annotation Adrian Hunter
2013-08-12 10:22 ` [tip:perf/core] perf annotate: Remove " tip-bot for Adrian Hunter
2013-08-07 11:38 ` [PATCH V4 13/13] perf tools: add annotation call target name if it is missing Adrian Hunter
2013-08-12 10:22 ` [tip:perf/core] perf annotate: Add " tip-bot for Adrian Hunter
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=tip-0131c4ec794a7409eafff0c79105309540aaca4d@git.kernel.org \
--to=tipbot@zytor.com \
--cc=acme@redhat.com \
--cc=adrian.hunter@intel.com \
--cc=dsahern@gmail.com \
--cc=efault@gmx.de \
--cc=eranian@google.com \
--cc=fweisbec@gmail.com \
--cc=hpa@zytor.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@gmail.com \
--cc=paulus@samba.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/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).