linux-kernel - Re: [PATCH 2/3] arm64: perf: Improve compat perf_callchain

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <47a95789-ca75-70a5-9d65-a2d3e9c651bc@arm.com>
Date:   Mon, 7 Jun 2021 12:14:20 +0300
From:   James Clark <james.clark@....com>
To:     Douglas Anderson <dianders@...omium.org>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will@...nel.org>
Cc:     Nick Desaulniers <ndesaulniers@...gle.com>,
        Seth LaForge <sethml@...gle.com>,
        Ricky Liang <jcliang@...omium.org>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Ingo Molnar <mingo@...hat.com>, Jiri Olsa <jolsa@...hat.com>,
        Mark Rutland <mark.rutland@....com>,
        Namhyung Kim <namhyung@...nel.org>,
        Nathan Chancellor <nathan@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        clang-built-linux@...glegroups.com,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        linux-perf-users@...r.kernel.org,
        Alexandre Truong <alexandre.truong@....com>,
        Wilco Dijkstra <wilco.dijkstra@....com>,
        Al Grant <Al.Grant@....com>
Subject: Re: [PATCH 2/3] arm64: perf: Improve compat perf_callchain_user() for
 clang leaf functions



On 07/05/2021 23:55, Douglas Anderson wrote:
> It turns out that even when you compile code with clang with
> "-fno-omit-frame-pointer" that it won't generate a frame pointer for
> leaf functions (those that don't call any sub-functions). Presumably
> clang does this to reduce the overhead of frame pointers. In a leaf
> function you don't really need frame pointers since the Link Register
> (LR) is guaranteed to always point to the caller> 
[...]
> 
>  arch/arm64/kernel/perf_callchain.c | 14 ++++++++++++++
>  1 file changed, 14 insertions(+)
> 
> diff --git a/arch/arm64/kernel/perf_callchain.c b/arch/arm64/kernel/perf_callchain.c
> index e5ce5f7965d1..b3cd9f371469 100644
> --- a/arch/arm64/kernel/perf_callchain.c
> +++ b/arch/arm64/kernel/perf_callchain.c
> @@ -326,6 +326,20 @@ static void compat_perf_callchain_user(struct perf_callchain_entry_ctx *entry,
>  	while ((entry->nr < entry->max_stack) && fp && !(fp & 0x3)) {
>  		err = compat_perf_trace_1(&fp, &pc, leaf_lr);
>  
> +		/*
> +		 * If this is the first trace and it didn't find the LR then
> +		 * let's throw it in the trace first. This isn't perfect but
> +		 * is the best we can do for handling clang leaf functions (or
> +		 * the case where we're right at the start of the function
> +		 * before the new frame has been pushed). In the worst case
> +		 * this can cause us to throw an extra entry that will be some
> +		 * location in the same function as the PC. That's not
> +		 * amazing but shouldn't really hurt. It seems better than
> +		 * throwing away the LR.
> +		 */

Hi Douglas,

I think the behaviour with GCC is also similar. We were working on this change
(https://lore.kernel.org/lkml/20210304163255.10363-4-alexandre.truong@arm.com/)
in userspace Perf which addresses the same issue.

The basic concept of our version is to record only the link register
(as in --user-regs=lr). Then use the existing dwarf based unwind
to determine if the link register is valid for that frame, and then if
it is and it doesn't already exist on the stack then insert it.

You mention that your version isn't perfect, do you think that saving the
LR and using something like libunwind in a post process could be better?

Thanks
James