From: Eric B Munson <emunson@mgebm.net>
To: a.p.zijlstra@chello.nl
Cc: mingo@elte.hu, borislav.petkov@amd.com, bblum@andrew.cmu.edu,
linux-kernel@vger.kernel.org, mhack@us.ibm.com,
eranian@google.com, Eric B Munson <emunson@mgebm.net>
Subject: [PATCH V2] events: Ensure that timers are updated without requiring read() call
Date: Fri, 24 Jun 2011 12:26:26 -0400 [thread overview]
Message-ID: <1308932786-5111-1-git-send-email-emunson@mgebm.net> (raw)
The event tracing infrastructure exposes two timers which should be updated
each time the value of the counter is updated. Currently, these counters are
only updated when userspace calls read() on the fd associated with an event.
This means that counters which are read via the mmap'd page exclusively never
have their timers updated. This patch adds ensures that the timers are updated
each time the values in the mmap'd page are updated.
Signed-off-by: Eric B Munson <emunson@mgebm.net>
---
Changes from V1:
Use local variables to hold enabled and running times instead of writing to
the event to avoid possible race. The event updates will have to be done in
the arch code
kernel/events/core.c | 15 +++++++++++++--
1 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 9e9a7fa..a6ae668 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3380,8 +3380,19 @@ void perf_event_update_userpage(struct perf_event *event)
{
struct perf_event_mmap_page *userpg;
struct perf_buffer *buffer;
+ u64 enabled, running;
rcu_read_lock();
+ /*
+ * compute total_time_enabled, total_time_running
+ * based on snapshot values taken when the event
+ * was last scheduled in.
+ *
+ * we cannot simply called update_context_time()
+ * because of locking issue as we can be called in
+ * NMI context
+ */
+ calc_timer_values(event, &enabled, &running);
buffer = rcu_dereference(event->buffer);
if (!buffer)
goto unlock;
@@ -3400,10 +3411,10 @@ void perf_event_update_userpage(struct perf_event *event)
if (event->state == PERF_EVENT_STATE_ACTIVE)
userpg->offset -= local64_read(&event->hw.prev_count);
- userpg->time_enabled = event->total_time_enabled +
+ userpg->time_enabled = enabled +
atomic64_read(&event->child_total_time_enabled);
- userpg->time_running = event->total_time_running +
+ userpg->time_running = running +
atomic64_read(&event->child_total_time_running);
barrier();
--
1.7.4.1
next reply other threads:[~2011-06-24 16:26 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-24 16:26 Eric B Munson [this message]
2011-07-01 15:20 ` [tip:perf/core] " tip-bot for Eric B Munson
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=1308932786-5111-1-git-send-email-emunson@mgebm.net \
--to=emunson@mgebm.net \
--cc=a.p.zijlstra@chello.nl \
--cc=bblum@andrew.cmu.edu \
--cc=borislav.petkov@amd.com \
--cc=eranian@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhack@us.ibm.com \
--cc=mingo@elte.hu \
/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
all inboxes | Powered by JetHome®