From: Steven Rostedt <rostedt@goodmis.org>
To: Tom Zanussi <tom.zanussi@linux.intel.com>
Cc: tglx@linutronix.de, mhiramat@kernel.org, namhyung@kernel.org,
vedang.patel@intel.com, bigeasy@linutronix.de,
joel.opensrc@gmail.com, joelaf@google.com,
mathieu.desnoyers@efficios.com, baohong.liu@intel.com,
linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org
Subject: Re: [PATCH v2 02/40] tracing: Add support to detect and avoid duplicates
Date: Wed, 6 Sep 2017 14:47:22 -0400 [thread overview]
Message-ID: <20170906144722.6eb71312@gandalf.local.home> (raw)
In-Reply-To: <7f3ca68455bdac47285ed45d608075a4af287362.1504642143.git.tom.zanussi@linux.intel.com>
On Tue, 5 Sep 2017 16:57:14 -0500
Tom Zanussi <tom.zanussi@linux.intel.com> wrote:
> diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c
> index 305039b..437b490 100644
> --- a/kernel/trace/tracing_map.c
> +++ b/kernel/trace/tracing_map.c
> @@ -414,6 +414,7 @@ static inline bool keys_match(void *key, void *test_key, unsigned key_size)
> __tracing_map_insert(struct tracing_map *map, void *key, bool lookup_only)
> {
> u32 idx, key_hash, test_key;
> + int dup_try = 0;
> struct tracing_map_entry *entry;
>
> key_hash = jhash(key, map->key_size, 0);
> @@ -426,10 +427,31 @@ static inline bool keys_match(void *key, void *test_key, unsigned key_size)
> entry = TRACING_MAP_ENTRY(map->map, idx);
> test_key = entry->key;
>
> - if (test_key && test_key == key_hash && entry->val &&
> - keys_match(key, entry->val->key, map->key_size)) {
> - atomic64_inc(&map->hits);
> - return entry->val;
> + if (test_key && test_key == key_hash) {
> + if (entry->val &&
> + keys_match(key, entry->val->key, map->key_size)) {
> + atomic64_inc(&map->hits);
> + return entry->val;
> + } else if (unlikely(!entry->val)) {
I'm thinking we need a READ_ONCE() here.
val = READ_ONCE(entry->val);
then use "val" instead of entry->val. Otherwise, wont it be possible
if two tasks are inserting at the same time, to have this:
(Using reg as when the value is read into a register from memory)
CPU0 CPU1
---- ----
reg = entry->val
(reg == zero)
entry->val = elt;
keys_match(key, reg)
(false)
reg = entry->val
(reg = elt)
if (unlikely(!reg))
Causes the if to fail.
A READ_ONCE(), would make sure the entry->val used to test against key
would also be the same value used to test if it is zero.
-- Steve
> + /*
> + * The key is present. But, val (pointer to elt
> + * struct) is still NULL. which means some other
> + * thread is in the process of inserting an
> + * element.
> + *
> + * On top of that, it's key_hash is same as the
> + * one being inserted right now. So, it's
> + * possible that the element has the same
> + * key as well.
> + */
> +
> + dup_try++;
> + if (dup_try > map->map_size) {
> + atomic64_inc(&map->drops);
> + break;
> + }
> + continue;
> + }
> }
>
> if (!test_key) {
> @@ -451,6 +473,13 @@ static inline bool keys_match(void *key, void *test_key, unsigned key_size)
> atomic64_inc(&map->hits);
>
> return entry->val;
> + } else {
> + /*
> + * cmpxchg() failed. Loop around once
> + * more to check what key was inserted.
> + */
> + dup_try++;
> + continue;
> }
> }
>
next prev parent reply other threads:[~2017-09-06 18:47 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-05 21:57 [PATCH v2 00/40] tracing: Inter-event (e.g. latency) support Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 01/40] tracing: Exclude 'generic fields' from histograms Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 02/40] tracing: Add support to detect and avoid duplicates Tom Zanussi
2017-09-06 18:32 ` Steven Rostedt
2017-09-06 18:47 ` Steven Rostedt [this message]
2017-09-06 20:58 ` Patel, Vedang
2017-09-05 21:57 ` [PATCH v2 03/40] tracing: Remove code which merges duplicates Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 04/40] tracing: Add hist_field_name() accessor Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 05/40] tracing: Reimplement log2 Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 06/40] ring-buffer: Add interface for setting absolute time stamps Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 07/40] tracing: Apply absolute timestamps to instance max buffer Tom Zanussi
2017-09-06 19:57 ` Steven Rostedt
2017-09-07 0:49 ` Steven Rostedt
2017-09-07 1:15 ` Liu, Baohong
2017-09-05 21:57 ` [PATCH v2 08/40] ring-buffer: Redefine the unimplemented RINGBUF_TIME_TIME_STAMP Tom Zanussi
2017-09-07 14:35 ` Steven Rostedt
2017-09-07 15:05 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 09/40] tracing: Give event triggers access to ring_buffer_event Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 10/40] tracing: Add ring buffer event param to hist field functions Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 11/40] tracing: Increase tracing map KEYS_MAX size Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 12/40] tracing: Break out hist trigger assignment parsing Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 13/40] tracing: Make traceprobe parsing code reusable Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 14/40] tracing: Add hist trigger timestamp support Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 15/40] tracing: Add per-element variable support to tracing_map Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 16/40] tracing: Add hist_data member to hist_field Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 17/40] tracing: Add usecs modifier for hist trigger timestamps Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 18/40] tracing: Add variable support to hist triggers Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 19/40] tracing: Account for variables in named trigger compatibility Tom Zanussi
2017-09-07 16:40 ` Steven Rostedt
2017-09-07 17:00 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 20/40] tracing: Add simple expression support to hist triggers Tom Zanussi
2017-09-07 16:46 ` Steven Rostedt
2017-09-07 17:01 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 21/40] tracing: Generalize per-element hist trigger data Tom Zanussi
2017-09-07 17:56 ` Steven Rostedt
2017-09-07 18:14 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 22/40] tracing: Pass tracing_map_elt to hist_field accessor functions Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 23/40] tracing: Add hist_field 'type' field Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 24/40] tracing: Add variable reference handling to hist triggers Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 25/40] tracing: Add support for dynamic tracepoints Tom Zanussi
2017-09-05 23:29 ` Mathieu Desnoyers
2017-09-06 2:35 ` Tom Zanussi
2017-09-07 22:02 ` Steven Rostedt
2017-09-08 14:18 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 26/40] tracing: Add hist trigger action hook Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 27/40] tracing: Add support for 'synthetic' events Tom Zanussi
2017-09-07 23:40 ` Steven Rostedt
2017-09-08 14:30 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 28/40] tracing: Add support for 'field variables' Tom Zanussi
2017-09-07 23:43 ` Steven Rostedt
2017-09-08 15:37 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 29/40] tracing: Add 'onmatch' hist trigger action support Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 30/40] tracing: Add 'onmax' " Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 31/40] tracing: Allow whitespace to surround hist trigger filter Tom Zanussi
2017-09-08 18:50 ` Steven Rostedt
2017-09-08 19:08 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 32/40] tracing: Add cpu field for hist triggers Tom Zanussi
2017-09-08 19:08 ` Steven Rostedt
2017-09-08 19:35 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 33/40] tracing: Add hist trigger support for variable reference aliases Tom Zanussi
2017-09-08 19:09 ` Steven Rostedt
2017-09-08 19:41 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 34/40] tracing: Add 'last error' error facility for hist triggers Tom Zanussi
2017-09-08 19:25 ` Steven Rostedt
2017-09-08 19:44 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 35/40] tracing: Reverse the order event_mutex/trace_types_lock are taken Tom Zanussi
2017-09-08 19:31 ` Steven Rostedt
2017-09-08 19:41 ` Steven Rostedt
2017-09-08 20:00 ` Steven Rostedt
2017-09-05 21:57 ` [PATCH v2 36/40] tracing: Remove lookups from tracing_map hitcount Tom Zanussi
2017-09-12 2:16 ` Masami Hiramatsu
2017-09-12 14:16 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 37/40] tracing: Add inter-event hist trigger Documentation Tom Zanussi
2017-09-20 14:44 ` Julia Cartwright
2017-09-20 17:15 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 38/40] tracing: Make tracing_set_clock() non-static Tom Zanussi
2017-09-12 2:18 ` Masami Hiramatsu
2017-09-12 14:18 ` Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 39/40] tracing: Add a clock attribute for hist triggers Tom Zanussi
2017-09-05 21:57 ` [PATCH v2 40/40] tracing: Add trace_event_buffer_reserve() variant that allows recursion Tom Zanussi
2017-09-07 22:29 ` kbuild test robot
2017-09-07 22:35 ` kbuild test robot
2017-09-08 20:27 ` Steven Rostedt
2017-09-08 20:41 ` Tom Zanussi
2017-09-12 1:50 ` [PATCH v2 00/40] tracing: Inter-event (e.g. latency) support Masami Hiramatsu
2017-09-12 14:14 ` Tom Zanussi
2017-09-19 16:31 ` Steven Rostedt
2017-09-19 18:44 ` Tom Zanussi
2017-09-21 20:20 ` Steven Rostedt
2017-09-21 21:11 ` Tom Zanussi
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=20170906144722.6eb71312@gandalf.local.home \
--to=rostedt@goodmis.org \
--cc=baohong.liu@intel.com \
--cc=bigeasy@linutronix.de \
--cc=joel.opensrc@gmail.com \
--cc=joelaf@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mhiramat@kernel.org \
--cc=namhyung@kernel.org \
--cc=tglx@linutronix.de \
--cc=tom.zanussi@linux.intel.com \
--cc=vedang.patel@intel.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).