linux-kernel - Re: [PATCH 1/3] perf, tool: Force guest machine definition option

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]

Date:	Sun, 15 Apr 2012 16:05:13 +0200
From:	Jiri Olsa <jolsa@...hat.com>
To:	David Ahern <dsahern@...il.com>
Cc:	acme@...hat.com, a.p.zijlstra@...llo.nl, mingo@...e.hu,
	paulus@...ba.org, cjashfor@...ux.vnet.ibm.com, fweisbec@...il.com,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] perf, tool: Force guest machine definition option

On Fri, Apr 13, 2012 at 08:15:17AM -0600, David Ahern wrote:
> On 4/13/12 6:41 AM, Jiri Olsa wrote:
> >I get same error even on acme's urgent branch.. do you mean
> >below one?  It fixes the guest machine lookup for mmap event.
> >
> >perf kvm: Finding struct machine fails for PERF_RECORD_MMAP
> >commit 7fb0a5ee8889488f7568ffddffeb66ddeb50917e
> >Author: Nikunj A. Dadhania<nikunj@...ux.vnet.ibm.com>
> >Date:   Mon Apr 9 13:52:23 2012 +0530
> 
> That's the one I was referring to.
> 
> >
> >In my case it looks like for some reason the guest buildid DSO
> >is not stored in record phase (no hits maybe?), so report won't
> >create guest machine record at all and get NULL machine.
> 
> I see now -- different problem, but similar in that it's an mmap
> event and the pid is 0.
> 
> I need to take my daughter to school and won't get back to this for
> a while but what I am seeing is that on perf-record mmap events are
> generated with pid set to DEFAULT_GUEST_KERNEL_ID = 0
> (machines__create_guest_kernel_maps).
> 
> On the report side a machine has not been created for pid of 0, so
> the look up in perf_session__find_machine_for_cpumode fails.
> 
> David

how about the patch below? it ensures there's machine record for
the guest. The segfault issue still stays..

jirka

---
Running 'perf kvm record' without any of following options:
 --guestmount
 --guestvmlinux
 --guestkallsyms
 --guestmodules

is causing the guest machine to be ommited from the data file,
and all guest samples are counted in nr_unprocessable_samples.

This patch makes sure the 'perf kvm record' command is not
let through if guest machine isn't defined.

Signed-off-by: Jiri Olsa <jolsa@...hat.com>
---
 tools/perf/builtin-record.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 10b1f1f..0ae4237 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -462,6 +462,14 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 		return -1;
 	}
 
+	if (perf_guest &&
+	    !machines__find(&session->machines, DEFAULT_GUEST_KERNEL_ID)) {
+		ui__warning("You need to define guest with one of guestmount|"
+			    "guestvmlinux|guestkallsyms|guestmodules\n");
+		err = -EINVAL;
+		goto out_delete_session;
+	}
+
 	rec->session = session;
 
 	for (feat = HEADER_FIRST_FEATURE; feat < HEADER_LAST_FEATURE; feat++)
-- 
1.7.7.6

--
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/