[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200310115724.25152-1-sjpark@amazon.com>
Date: Tue, 10 Mar 2020 12:57:24 +0100
From: SeongJae Park <sjpark@...zon.com>
To: Jonathan Cameron <Jonathan.Cameron@...wei.com>
CC: SeongJae Park <sjpark@...zon.com>, <akpm@...ux-foundation.org>,
"SeongJae Park" <sjpark@...zon.de>, <aarcange@...hat.com>,
<yang.shi@...ux.alibaba.com>, <acme@...nel.org>,
<alexander.shishkin@...ux.intel.com>, <amit@...nel.org>,
<brendan.d.gregg@...il.com>, <brendanhiggins@...gle.com>,
<cai@....pw>, <colin.king@...onical.com>, <corbet@....net>,
<dwmw@...zon.com>, <jolsa@...hat.com>, <kirill@...temov.name>,
<mark.rutland@....com>, <mgorman@...e.de>, <minchan@...nel.org>,
<mingo@...hat.com>, <namhyung@...nel.org>, <peterz@...radead.org>,
<rdunlap@...radead.org>, <rientjes@...gle.com>,
<rostedt@...dmis.org>, <shuah@...nel.org>, <sj38.park@...il.com>,
<vbabka@...e.cz>, <vdavydov.dev@...il.com>, <linux-mm@...ck.org>,
<linux-doc@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: Re: [PATCH v6 09/14] mm/damon: Add a tracepoint for result writing
On Tue, 10 Mar 2020 09:03:31 +0000 Jonathan Cameron <Jonathan.Cameron@...wei.com> wrote:
> On Mon, 24 Feb 2020 13:30:42 +0100
> SeongJae Park <sjpark@...zon.com> wrote:
>
> > From: SeongJae Park <sjpark@...zon.de>
> >
> > This commit adds a tracepoint for DAMON's result buffer writing. It is
> > called for each writing of the DAMON results and print the result data.
> > Therefore, it would be used to easily integrated with other tracepoint
> > supporting tracers such as perf.
> >
> > Signed-off-by: SeongJae Park <sjpark@...zon.de>
>
> I'm curious, why at the flush of rbuf rather than using a more structured trace
> point for each of the writes into rbuf?
>
> Seems it would make more sense to have a tracepoint for each record write out.
> Probably at the level of each task, though might be more elegant to do it at the
> level of each region within a task and duplicate the header stuff.
I was worried if the format changes, but agree your suggestion is the right
way. Will change so in next spin.
Thanks,
SeongJae Park
>
> > ---
> > include/trace/events/damon.h | 32 ++++++++++++++++++++++++++++++++
> > mm/damon.c | 4 ++++
> > 2 files changed, 36 insertions(+)
> > create mode 100644 include/trace/events/damon.h
> >
> > diff --git a/include/trace/events/damon.h b/include/trace/events/damon.h
> > new file mode 100644
> > index 000000000000..fb33993620ce
> > --- /dev/null
> > +++ b/include/trace/events/damon.h
> > @@ -0,0 +1,32 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +#undef TRACE_SYSTEM
> > +#define TRACE_SYSTEM damon
> > +
> > +#if !defined(_TRACE_DAMON_H) || defined(TRACE_HEADER_MULTI_READ)
> > +#define _TRACE_DAMON_H
> > +
> > +#include <linux/types.h>
> > +#include <linux/tracepoint.h>
> > +
> > +TRACE_EVENT(damon_write_rbuf,
> > +
> > + TP_PROTO(void *buf, const ssize_t sz),
> > +
> > + TP_ARGS(buf, sz),
> > +
> > + TP_STRUCT__entry(
> > + __dynamic_array(char, buf, sz)
> > + ),
> > +
> > + TP_fast_assign(
> > + memcpy(__get_dynamic_array(buf), buf, sz);
> > + ),
> > +
> > + TP_printk("dat=%s", __print_hex(__get_dynamic_array(buf),
> > + __get_dynamic_array_len(buf)))
> > +);
> > +
> > +#endif /* _TRACE_DAMON_H */
> > +
> > +/* This part must be outside protection */
> > +#include <trace/define_trace.h>
> > diff --git a/mm/damon.c b/mm/damon.c
> > index facb1d7f121b..8faf3879f99e 100644
> > --- a/mm/damon.c
> > +++ b/mm/damon.c
> > @@ -9,6 +9,8 @@
> >
> > #define pr_fmt(fmt) "damon: " fmt
> >
> > +#define CREATE_TRACE_POINTS
> > +
> > #include <linux/damon.h>
> > #include <linux/debugfs.h>
> > #include <linux/delay.h>
> > @@ -20,6 +22,7 @@
> > #include <linux/sched/mm.h>
> > #include <linux/sched/task.h>
> > #include <linux/slab.h>
> > +#include <trace/events/damon.h>
> >
> > #define damon_get_task_struct(t) \
> > (get_pid_task(find_vpid(t->pid), PIDTYPE_PID))
> > @@ -553,6 +556,7 @@ static void damon_flush_rbuffer(struct damon_ctx *ctx)
> > */
> > static void damon_write_rbuf(struct damon_ctx *ctx, void *data, ssize_t size)
> > {
> > + trace_damon_write_rbuf(data, size);
> > if (!ctx->rbuf_len || !ctx->rbuf)
> > return;
> > if (ctx->rbuf_offset + size > ctx->rbuf_len)
>
Powered by blists - more mailing lists