Message-ID: <Y3ZGAf2rdV1Ia5P3@hirez.programming.kicks-ass.net>
Date:   Thu, 17 Nov 2022 15:32:33 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Richard Biener <rguenther@...e.de>
Cc:     Ard Biesheuvel <ardb@...nel.org>,
        "Jiri Slaby (SUSE)" <jirislaby@...nel.org>,
        linux-kernel@...r.kernel.org,
        Alexander Potapenko <glider@...gle.com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Alexei Starovoitov <ast@...nel.org>,
        Alexey Makhalov <amakhalov@...are.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Andrey Konovalov <andreyknvl@...il.com>,
        Andrey Ryabinin <ryabinin.a.a@...il.com>,
        Andrii Nakryiko <andrii@...nel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Ben Segall <bsegall@...gle.com>,
        Borislav Petkov <bp@...en8.de>,
        Daniel Borkmann <daniel@...earbox.net>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Don Zickus <dzickus@...hat.com>, Hao Luo <haoluo@...gle.com>,
        "H . J . Lu" <hjl.tools@...il.com>,
        "H. Peter Anvin" <hpa@...or.com>, Huang Rui <ray.huang@....com>,
        Ingo Molnar <mingo@...hat.com>, Jan Hubicka <jh@...e.de>,
        Jason Baron <jbaron@...mai.com>,
        Jiri Kosina <jikos@...nel.org>, Jiri Olsa <jolsa@...nel.org>,
        Joe Lawrence <joe.lawrence@...hat.com>,
        John Fastabend <john.fastabend@...il.com>,
        Josh Poimboeuf <jpoimboe@...nel.org>,
        Juergen Gross <jgross@...e.com>,
        Juri Lelli <juri.lelli@...hat.com>,
        KP Singh <kpsingh@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Martin KaFai Lau <martin.lau@...ux.dev>,
        Martin Liska <mliska@...e.cz>,
        Masahiro Yamada <masahiroy@...nel.org>,
        Mel Gorman <mgorman@...e.de>, Miguel Ojeda <ojeda@...nel.org>,
        Michal Marek <michal.lkml@...kovi.net>,
        Miroslav Benes <mbenes@...e.cz>,
        Namhyung Kim <namhyung@...nel.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Oleksandr Tyshchenko <oleksandr_tyshchenko@...m.com>,
        Petr Mladek <pmladek@...e.com>,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        Sedat Dilek <sedat.dilek@...il.com>,
        Song Liu <song@...nel.org>,
        Stanislav Fomichev <sdf@...gle.com>,
        Stefano Stabellini <sstabellini@...nel.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Valentin Schneider <vschneid@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Vincenzo Frascino <vincenzo.frascino@....com>,
        Viresh Kumar <viresh.kumar@...aro.org>,
        VMware PV-Drivers Reviewers <pv-drivers@...are.com>,
        Yonghong Song <yhs@...com>
Subject: Re: [PATCH 00/46] gcc-LTO support for the kernel

On Thu, Nov 17, 2022 at 01:55:07PM +0000, Richard Biener wrote:

> > > I'm not sure what you're on about; only symbols annotated with
> > > EXPORT_SYMBOL*() are accessible from modules (aka DSOs) and those will
> > > have their address taken. You can feely eliminate any unused symbol.
> 
> But IIRC that's not reflected on the ELF level by making EXPORT_SYMBOL*()
> symbols public and the rest hidden - instead all symbols global in the C TUs
> will become public and the module dynamic loader details are hidden from
> GCCs view of the kernel image as ELF relocatable object.

It is reflected by keeping their address in __ksymtab_$foo sections, as
such their address 'escapes'.

> > We have an __ADDRESSABLE() macro and asmlinkage modifier to annotate
> > symbols that may appear to the compiler as though they are never
> > referenced.
> > 
> > Would it be possible to repurpose those so that the LTO code knows
> > which symbols it must not remove?
> 
> I find
> 
> /*
>  * Force the compiler to emit 'sym' as a symbol, so that we can reference
>  * it from inline assembler. Necessary in case 'sym' could be inlined
>  * otherwise, or eliminated entirely due to lack of references that are
>  * visible to the compiler.
>  */
> #define ___ADDRESSABLE(sym, __attrs) \
> 	static void * __used __attrs \
> 		__UNIQUE_ID(__PASTE(__addressable_,sym)) = (void *)&sym;
> #define __ADDRESSABLE(sym) \
> 	___ADDRESSABLE(sym, __section(".discard.addressable"))
> 
> that should be enough to force LTO keeping 'sym' - unless there's
> a linker script that discards .discard.addressable which I fear LTO
> will notice, losing the effect.

The initial LTO link pass will not discard .discard sections in order to
generate a regular ELF object file. This object file is then fed to
objtool and the kallsyms tool and eventually linked with the linker
script in a multi-stage link pass.

Also see scripts/link-vmlinux.sh for all the horrible details.

> The folks who worked on LTO enablement of the kernel should know the
> real issue better - I understand asm()s are a pain because GCC
> refuses to parse the assembler string heuristically for used
> symbols (but it can never be more than heuristics). 

I don't understand why it can't be more than heuristics; eventually the
asm() contents end up in a real assembler and it has to make sense.

Might as well parse it directly -- isn't that what clang-ias does?

> The issue with asm()s is not so much elimination (__used solves that)
> but that GCC can end up moving the asm() and the refered to symbols to
> different link-time units causing unresolved symbols for non-global
> symbols.  -fno-toplevel-reorder should fix that at some cost.

I thought the whole point of LTO was that there was only a single link
time unit, translate all the tus into intermadiate gunk and then collect
the whole lot in one go.

Powered by blists - more mailing lists