[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250510094638.27aa5f8b@gandalf.local.home>
Date: Sat, 10 May 2025 09:46:38 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc: linux-kernel@...r.kernel.org, linux-trace-kernel@...r.kernel.org,
bpf@...r.kernel.org, x86@...nel.org, Masami Hiramatsu
<mhiramat@...nel.org>, Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
Josh Poimboeuf <jpoimboe@...nel.org>, Peter Zijlstra
<peterz@...radead.org>, Ingo Molnar <mingo@...nel.org>, Jiri Olsa
<jolsa@...nel.org>, Namhyung Kim <namhyung@...nel.org>
Subject: Re: [PATCH v8 15/18] perf: Have get_perf_callchain() return NULL if
crosstask and user are set
On Fri, 9 May 2025 14:53:38 -0700
Andrii Nakryiko <andrii.nakryiko@...il.com> wrote:
> > @@ -224,6 +224,10 @@ get_perf_callchain(struct pt_regs *regs, bool kernel, bool user,
> > struct perf_callchain_entry_ctx ctx;
> > int rctx, start_entry_idx;
> >
> > + /* crosstask is not supported for user stacks */
> > + if (crosstask && user)
> > + return NULL;
>
> I think get_perf_callchain() supports requesting both user and kernel
> stack traces, and if it's crosstask, you can still get kernel (but not
> user) stack, if I'm reading the code correctly.
>
> So by just returning NULL early you will change this behavior, no?
Basically you are saying that one could ask for a kernel/user stack trace
with crosstask enabled and still just get the kernel trace?
If this is the case, then I think it may be best to remove patches 15-18
from this series and work on them in the "perf specific" series, as this
doesn't have anything to do with the unwind infrastructure itself.
Actually, patch 14 doesn't either, so I may move that one too (and keep the
acks to it).
Thanks,
-- Steve
>
> > +
> > entry = get_callchain_entry(&rctx);
> > if (!entry)
> > return NULL;
Powered by blists - more mailing lists