[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20180111213119.b5e03409d6beb27e3f34fc5a@kernel.org>
Date: Thu, 11 Jan 2018 21:31:19 +0900
From: Masami Hiramatsu <mhiramat@...nel.org>
To: Arnaldo Carvalho de Melo <acme@...nel.org>
Cc: mhiramat@...nel.org,
Ravi Bangoria <ravi.bangoria@...ux.vnet.ibm.com>,
linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org
Subject: Re: [PATCH 1/2] perf-probe: Ensure debuginfo's build-id is correct
Hi Arnaldo,
On Thu, 4 Jan 2018 13:17:28 -0300
Arnaldo Carvalho de Melo <acme@...nel.org> wrote:
> Em Mon, Dec 18, 2017 at 04:29:03PM +0900, Masami Hiramatsu escreveu:
> > Ensure that the build-id of debuginfo is correctly
> > matched to target build-id, if not, it warns user
> > to check the system debuginfo package is correctly
> > installed.
>
> So we look at a variety of files looking for one that has a matching
> build-id, I think the warning message should state that the file with
> the unmatched build-id is simply being skipped, no?
Actually, this patch series came from your request that you faced
buildid mismatch between running binary and debuginfo binary, and
you asked me to show the message to tell user what he needs to
check.
>
> And why do this at 'perf probe -l' time? I.e. at that point whatever
> probes that are in place already have all the needed debug info?
As I pointed, this is not only for perf-probe -l but also perf-probe -a
(or any other operation which uses debuginfo, like -V, -L). If user
see this message at "perf probe -l", he might install debuginfo after
he setup probes. Otherwise, he should see this message when he tries
to setup probes.
BTW, this message is actually shown only if the binary has build-id.
If either file doesn't have build-id, it is just ignored.
Thank you,
>
> I.e. the warning should be done at probe creation time only?
>
> - Arnaldo
>
> > E.g. on such environment, you will see below warning.
> > ======
> > # perf probe -l
> > WARN: There is a build-id mismatch between
> > /usr/lib/debug/usr/lib64/libc-2.25.so.debug
> > and
> > /usr/lib64/libc-2.25.so
> > Please check your system's debuginfo files for mismatches, e.g. check
> > the versions for the target package and debuginfo package.
> > probe_libc:malloc_get_state (on malloc_get_state@...BC_2.2.5 in /us
> > r/lib64/libc-2.25.so)
> > ======
> >
> > Signed-off-by: Masami Hiramatsu <mhiramat@...nel.org>
> > ---
> > tools/perf/util/probe-finder.c | 18 ++++++++++++++++++
> > 1 file changed, 18 insertions(+)
> >
> > diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
> > index a5731de0e5eb..5bb71e056b21 100644
> > --- a/tools/perf/util/probe-finder.c
> > +++ b/tools/perf/util/probe-finder.c
> > @@ -119,9 +119,11 @@ enum dso_binary_type distro_dwarf_types[] = {
> >
> > struct debuginfo *debuginfo__new(const char *path)
> > {
> > + u8 bid[BUILD_ID_SIZE], bid2[BUILD_ID_SIZE];
> > enum dso_binary_type *type;
> > char buf[PATH_MAX], nil = '\0';
> > struct dso *dso;
> > + bool have_build_id = false;
> > struct debuginfo *dinfo = NULL;
> >
> > /* Try to open distro debuginfo files */
> > @@ -129,12 +131,28 @@ struct debuginfo *debuginfo__new(const char *path)
> > if (!dso)
> > goto out;
> >
> > + if (filename__read_build_id(path, bid, BUILD_ID_SIZE) > 0)
> > + have_build_id = true;
> > +
> > for (type = distro_dwarf_types;
> > !dinfo && *type != DSO_BINARY_TYPE__NOT_FOUND;
> > type++) {
> > if (dso__read_binary_type_filename(dso, *type, &nil,
> > buf, PATH_MAX) < 0)
> > continue;
> > +
> > + if (have_build_id) {
> > + /* This can be fail because the file doesn't exist */
> > + if (filename__read_build_id(buf, bid2,
> > + BUILD_ID_SIZE) < 0)
> > + continue;
> > + if (memcmp(bid, bid2, BUILD_ID_SIZE)) {
> > + pr_warning("WARN: There is a build-id mismatch between\n %s\n and\n %s\n"
> > + "Please check your system's debuginfo files for mismatches, e.g. check the "
> > + "versions for the target package and debuginfo package.\n", buf, path);
> > + continue;
> > + }
> > + }
> > dinfo = __debuginfo__new(buf);
> > }
> > dso__put(dso);
--
Masami Hiramatsu <mhiramat@...nel.org>
Powered by blists - more mailing lists