Date:   Mon, 20 Dec 2021 10:54:55 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Song Liu <songliubraving@fb.com>
Cc:     Namhyung Kim <namhyung@kernel.org>, Ingo Molnar <mingo@kernel.org>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        Jiri Olsa <jolsa@redhat.com>,
        Mark Rutland <mark.rutland@arm.com>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Stephane Eranian <eranian@google.com>,
        Andi Kleen <ak@linux.intel.com>,
        Ian Rogers <irogers@google.com>
Subject: Re: [PATCH v3] perf/core: Set event shadow time for inactive events
 too
Message-ID: <YcBS75PpjOujBTUe@hirez.programming.kicks-ass.net>
References: <20211205224843.1503081-1-namhyung@kernel.org>
 <YbHn6JaaOo3b5GLO@hirez.programming.kicks-ass.net>
 <CAM9d7ciJTJB1rumzmxGeJrAdeE9R4eXhtJRUQGj9y6DBN-ovig@mail.gmail.com>
 <20211210103341.GS16608@worktop.programming.kicks-ass.net>
 <Yby8Su+fVA1lqVjT@hirez.programming.kicks-ass.net>
 <7A415BC0-E6F2-4ED2-8996-8F5871ED8001@fb.com>
 <YcBNOsMG3aGVpnWK@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <YcBNOsMG3aGVpnWK@hirez.programming.kicks-ass.net>
Precedence: bulk

On Mon, Dec 20, 2021 at 10:30:34AM +0100, Peter Zijlstra wrote:
> On Sat, Dec 18, 2021 at 09:09:05AM +0000, Song Liu wrote:

> > Unfortunately, this change bring the bug back. For time_enabled in rdpmc
> > case to work properly, we have to touch all the enabled but not running 
> > events, right?
> 
> Ohh.. argh. I think I see why, it looses the context time enable edge,
> and because this is all strictly per-event in the uapi (there is no ctx
> representation) it can't be cured by improving ctx time handling :/
> 
> Bah, I so hate this.

For now I've added this then ...

+/*
+ * Because the userpage is strictly per-event (there is no concept of context,
+ * so there cannot be a context indirection), every userpage must be updated
+ * when context time starts :-(
+ *
+ * IOW, we must not miss EVENT_TIME edges.
+ */
 static inline bool event_update_userpage(struct perf_event *event)


But that same is not true for perf_event_read_local(), that *has* access
to the context and so must be able to DTRT.