From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Oleg Nesterov <oleg@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Steven Rostedt <rostedt@goodmis.org>,
Anton Arapov <anton@redhat.com>, Frank Eigler <fche@redhat.com>,
Jiri Olsa <jolsa@redhat.com>, Josh Stone <jistone@redhat.com>,
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
"Suzuki K. Poulose" <suzuki@in.ibm.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 4/7] uprobes/perf: Teach trace_uprobe/perf code to track the active perf_event's
Date: Mon, 11 Feb 2013 15:15:08 +0530 [thread overview]
Message-ID: <20130211094508.GF525@linux.vnet.ibm.com> (raw)
In-Reply-To: <20130204190254.GA10875@redhat.com>
* Oleg Nesterov <oleg@redhat.com> [2013-02-04 20:02:54]:
> Introduce "struct trace_uprobe_filter" which records the "active"
> perf_event's attached to ftrace_event_call. For the start we simply
> use list_head, we can optimize this later if needed. For example, we
> do not really need to record an event with ->parent != NULL, we can
> rely on parent->child_list. And we can certainly do some optimizations
> for the case when 2 events have the same ->tp_target or tp_target->mm.
>
> Change trace_uprobe_register() to process TRACE_REG_PERF_OPEN/CLOSE
> and add/del this perf_event to the list.
>
> We can probably avoid any locking, but lets start with the "obvioulsy
> correct" trace_uprobe_filter->rwlock which protects everything.
>
> Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
> ---
> kernel/trace/trace_uprobe.c | 55 +++++++++++++++++++++++++++++++++++++++++++
> 1 files changed, 55 insertions(+), 0 deletions(-)
>
> diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
> index 0a9a8de..f05ec32 100644
> --- a/kernel/trace/trace_uprobe.c
> +++ b/kernel/trace/trace_uprobe.c
> @@ -28,6 +28,12 @@
>
> #define UPROBE_EVENT_SYSTEM "uprobes"
>
> +struct trace_uprobe_filter {
> + rwlock_t rwlock;
> + int nr_systemwide;
> + struct list_head perf_events;
> +};
> +
> /*
> * uprobe event core functions
> */
> @@ -35,6 +41,7 @@ struct trace_uprobe {
> struct list_head list;
> struct ftrace_event_class class;
> struct ftrace_event_call call;
> + struct trace_uprobe_filter filter;
> struct uprobe_consumer consumer;
> struct inode *inode;
> char *filename;
> @@ -58,6 +65,18 @@ static LIST_HEAD(uprobe_list);
>
> static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs);
>
> +static inline void init_trace_uprobe_filter(struct trace_uprobe_filter *filter)
> +{
> + rwlock_init(&filter->rwlock);
> + filter->nr_systemwide = 0;
> + INIT_LIST_HEAD(&filter->perf_events);
> +}
> +
> +static inline bool uprobe_filter_is_empty(struct trace_uprobe_filter *filter)
> +{
> + return !filter->nr_systemwide && list_empty(&filter->perf_events);
> +}
> +
> /*
> * Allocate new trace_uprobe and initialize it (including uprobes).
> */
> @@ -87,6 +106,7 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs)
>
> INIT_LIST_HEAD(&tu->list);
> tu->consumer.handler = uprobe_dispatcher;
> + init_trace_uprobe_filter(&tu->filter);
> return tu;
>
> error:
> @@ -541,6 +561,8 @@ static int probe_event_enable(struct trace_uprobe *tu, int flag)
> if (is_trace_uprobe_enabled(tu))
> return -EINTR;
>
> + WARN_ON(!uprobe_filter_is_empty(&tu->filter));
> +
> tu->flags |= flag;
> ret = uprobe_register(tu->inode, tu->offset, &tu->consumer);
> if (ret)
> @@ -554,6 +576,8 @@ static void probe_event_disable(struct trace_uprobe *tu, int flag)
> if (!is_trace_uprobe_enabled(tu))
> return;
>
> + WARN_ON(!uprobe_filter_is_empty(&tu->filter));
> +
> uprobe_unregister(tu->inode, tu->offset, &tu->consumer);
> tu->flags &= ~flag;
> }
> @@ -629,6 +653,30 @@ static int set_print_fmt(struct trace_uprobe *tu)
> }
>
> #ifdef CONFIG_PERF_EVENTS
> +static int uprobe_perf_open(struct trace_uprobe *tu, struct perf_event *event)
> +{
> + write_lock(&tu->filter.rwlock);
> + if (event->hw.tp_target)
> + list_add(&event->hw.tp_list, &tu->filter.perf_events);
> + else
> + tu->filter.nr_systemwide++;
> + write_unlock(&tu->filter.rwlock);
> +
> + return 0;
> +}
> +
> +static int uprobe_perf_close(struct trace_uprobe *tu, struct perf_event *event)
> +{
> + write_lock(&tu->filter.rwlock);
> + if (event->hw.tp_target)
> + list_del(&event->hw.tp_list);
> + else
> + tu->filter.nr_systemwide--;
> + write_unlock(&tu->filter.rwlock);
> +
> + return 0;
> +}
> +
> /* uprobe profile handler */
> static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
> {
> @@ -684,6 +732,13 @@ int trace_uprobe_register(struct ftrace_event_call *event, enum trace_reg type,
> case TRACE_REG_PERF_UNREGISTER:
> probe_event_disable(tu, TP_FLAG_PROFILE);
> return 0;
> +
> + case TRACE_REG_PERF_OPEN:
> + return uprobe_perf_open(tu, data);
> +
> + case TRACE_REG_PERF_CLOSE:
> + return uprobe_perf_close(tu, data);
> +
> #endif
> default:
> return 0;
> --
> 1.5.5.1
>
--
Thanks and Regards
Srikar Dronamraju
next prev parent reply other threads:[~2013-02-11 9:47 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-04 19:02 [PATCH 0/7] uprobes/perf: pre-filtering Oleg Nesterov
2013-02-04 19:02 ` [PATCH 1/7] perf: Ensure we do not free event->parent before event Oleg Nesterov
2013-03-20 13:35 ` Jiri Olsa
2013-02-04 19:02 ` [PATCH 2/7] perf: Introduce hw_perf_event->tp_target and ->tp_list Oleg Nesterov
2013-02-11 9:44 ` Srikar Dronamraju
2013-02-04 19:02 ` [PATCH 3/7] uprobes: Introduce uprobe_apply() Oleg Nesterov
2013-02-11 9:43 ` Srikar Dronamraju
2013-02-04 19:02 ` [PATCH 4/7] uprobes/perf: Teach trace_uprobe/perf code to track the active perf_event's Oleg Nesterov
2013-02-11 9:45 ` Srikar Dronamraju [this message]
2013-02-04 19:02 ` [PATCH 5/7] uprobes/perf: Teach trace_uprobe/perf code to pre-filter Oleg Nesterov
2013-02-11 9:46 ` Srikar Dronamraju
2013-02-04 19:03 ` [PATCH 6/7] uprobes/perf: Teach trace_uprobe/perf code to use UPROBE_HANDLER_REMOVE Oleg Nesterov
2013-02-11 9:54 ` Srikar Dronamraju
2013-02-04 19:03 ` [PATCH 7/7] uprobes/perf: Avoid uprobe_apply() whenever possible Oleg Nesterov
2013-02-11 9:55 ` Srikar Dronamraju
2013-02-06 18:10 ` [PATCH 0/7] uprobes/perf: pre-filtering Oleg Nesterov
2013-02-06 19:42 ` [PATCH 0/1] (Was uprobes/perf: pre-filtering) Oleg Nesterov
2013-02-06 19:42 ` [PATCH 1/1] perf/tools: Fix "perf record -C... workload" behaviour Oleg Nesterov
2013-02-25 9:58 ` Jiri Olsa
2013-02-07 6:01 ` [PATCH 0/1] (Was uprobes/perf: pre-filtering) Namhyung Kim
2013-02-07 15:22 ` Oleg Nesterov
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=20130211094508.GF525@linux.vnet.ibm.com \
--to=srikar@linux.vnet.ibm.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@ghostprotocols.net \
--cc=anton@redhat.com \
--cc=fche@redhat.com \
--cc=jistone@redhat.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mingo@elte.hu \
--cc=oleg@redhat.com \
--cc=rostedt@goodmis.org \
--cc=suzuki@in.ibm.com \
/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).