Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Date:   Mon, 21 Mar 2022 12:22:59 -0400
From:   Steven Rostedt <rostedt@goodmis.org>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Stephen Rothwell <sfr@canb.auug.org.au>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux Next Mailing List <linux-next@vger.kernel.org>,
        mhiramat@kernel.org, ast@kernel.org, hjl.tools@gmail.com,
        rick.p.edgecombe@intel.com, rppt@kernel.org,
        linux-toolchains@vger.kernel.org, Andrew.Cooper3@citrix.com,
        ndesaulniers@google.com
Subject: Re: linux-next: build warnings after merge of the tip tree
Message-ID: <20220321122259.28146a7a@gandalf.local.home>
In-Reply-To: <20220321121549.1c8588c5@gandalf.local.home>
References: <20220321140327.777f9554@canb.auug.org.au>
        <Yjh11UjDZogc3foM@hirez.programming.kicks-ass.net>
        <Yjh3xZuuY3QcZ1Bn@hirez.programming.kicks-ass.net>
        <20220321112805.1393f9b9@gandalf.local.home>
        <YjiiDFHIQg78QwSb@hirez.programming.kicks-ass.net>
        <20220321121209.3b95e406@gandalf.local.home>
        <20220321121549.1c8588c5@gandalf.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Mon, 21 Mar 2022 12:15:49 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:

> And I just thought of another issue, where even my solution wont fix it.
> What happens if we trace funcA but not funcB? How do we get to trace the
> end of funcA?

The only solution I can think of to handle all these cases is if you enable
-mfexit, you have to disable tail calls completely. That's going to cause
a performance impact.

Perhaps we need need compiler help to give us a way to hijack the return
address. But is there a way to do this and still not give up the security
that CET SHSTK gives us?

Or maybe another solution is:

funcA:
	[..]
	jmp funcB
	call __fexit__
	ret

And if funcA is being traced, we change jmp to a call.

	[..]
	call funcB
	call __fexit__
	ret

Such that we only remove the tail calls if we enable tracing on the
function with the tail call.

-- Steve