From: Thomas Gleixner <tglx@linutronix.de>
To: Andi Kleen <ak@linux.intel.com>
Cc: Andi Kleen <andi@firstfloor.org>,
x86@kernel.org, Andrew Morton <akpm@linux-foundation.org>,
LKML <linux-kernel@vger.kernel.org>,
Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: [PATCH 02/17] x86, lto: Mark all top level asm statements as .text
Date: Wed, 27 Mar 2019 20:08:23 +0100 (CET) [thread overview]
Message-ID: <alpine.DEB.2.21.1903271958320.1789@nanos.tec.linutronix.de> (raw)
In-Reply-To: <20190327154522.GV18020@tassilo.jf.intel.com>
On Wed, 27 Mar 2019, Andi Kleen wrote:
> > From the GCC manual:
> >
> > "This attribute, attached to a global variable or function, nullifies the
> > effect of the -fw hole-program command-line option, so the object remains
> > visible outside the current compilation unit."
> >
> > Neither the variable nor the data generated are global anymore. This data
> > is only used inside this compilation unit and I don't see why LTO needs a
> > reference outside of it. If so, then I really want to understand WHY
> > exactly.
>
> The LTO code generation doesn't compile file by file, but reorders
> all the functions and other top level statements into "partitions".
> The partitions are ordered by call graph so that inlining and some
> other optimizations work efficiently
>
> Then it finally runs multiple copies of the gcc code generator that
> generate code from these partitions, each generating an own assembler file.
>
> The top level assembler is not part of the call graph that drives
> the partitioning.
There is no top level assembly anymore. It's static data in the .rodata
section which is completely uninteresting for a call graph. That data is
accessed by the C function.
It's static so it's scope is within the file and whatever GCC does with
that C function it has to respect that it accesses static data. If that's
not true then this really needs to be fixed at the compiler side and not in
the kernel.
Thanks,
tglx
next prev parent reply other threads:[~2019-03-27 19:08 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-21 21:59 Fixes and cleanup from LTO tree Andi Kleen
2019-03-21 21:59 ` [PATCH 01/17] kbuild: Disable -Waddress-of-packed-member for gcc 9 Andi Kleen
2019-03-22 13:58 ` Arnd Bergmann
2019-03-22 16:31 ` Andi Kleen
2019-03-22 16:39 ` Peter Zijlstra
2019-03-22 16:57 ` Arnd Bergmann
2019-03-22 16:58 ` Andi Kleen
2019-03-25 16:27 ` David Laight
2019-04-01 1:04 ` Masahiro Yamada
2019-03-21 21:59 ` [PATCH 02/17] x86, lto: Mark all top level asm statements as .text Andi Kleen
2019-03-26 17:03 ` Thomas Gleixner
2019-03-26 21:38 ` Andi Kleen
2019-03-26 22:25 ` Thomas Gleixner
2019-03-27 0:55 ` Andi Kleen
2019-03-27 1:08 ` Andi Kleen
2019-03-27 14:20 ` Thomas Gleixner
2019-03-27 14:39 ` Ingo Molnar
2019-03-27 14:59 ` Andi Kleen
2019-03-27 15:08 ` Thomas Gleixner
2019-03-27 15:45 ` Andi Kleen
2019-03-27 19:08 ` Thomas Gleixner [this message]
2019-03-27 20:40 ` Andi Kleen
2019-03-27 21:55 ` Thomas Gleixner
2019-03-27 22:29 ` Andi Kleen
2019-03-27 22:50 ` Thomas Gleixner
2019-04-02 8:54 ` Thomas Gleixner
2019-03-21 21:59 ` [PATCH 03/17] x86: Don't inline __const_udelay Andi Kleen
2019-03-21 21:59 ` [PATCH 04/17] init: Add __noreorder and mark initcalls __noreorder Andi Kleen
2019-03-21 21:59 ` [PATCH 05/17] Use C version for SYSCALL_ALIAS Andi Kleen
2019-03-21 21:59 ` [PATCH 06/17] locking/core: Mark spinlocks noinline when inline spinlocks are disabled Andi Kleen
2019-03-21 21:59 ` [PATCH 07/17] amdkfd: Fix extern declaration Andi Kleen
2019-03-21 22:00 ` [PATCH 08/17] x86/speculation: Fix __initconst in bugs.c Andi Kleen
2019-03-26 17:24 ` Thomas Gleixner
2019-03-21 22:00 ` [PATCH 09/17] x86/kvm: Make steal_time visible Andi Kleen
2019-03-21 22:00 ` [PATCH 10/17] delta: Fix buffer overrun in delta_ipc_open Andi Kleen
2019-04-01 13:37 ` Hugues FRUCHET
2019-04-01 16:54 ` Andi Kleen
2019-04-02 9:56 ` Hugues FRUCHET
2019-03-21 22:00 ` [PATCH 11/17] x86/kprobes: Make trampoline_handler global and visible Andi Kleen
2019-03-23 9:45 ` Masami Hiramatsu
2019-03-23 14:35 ` Andi Kleen
2019-03-25 1:53 ` Masami Hiramatsu
2019-03-21 22:00 ` [PATCH 12/17] afs: Avoid section confusion in CM_NAME Andi Kleen
2019-03-21 22:00 ` [PATCH 13/17] ASoC: AMD: Fix incorrect extern Andi Kleen
2019-03-21 22:00 ` [PATCH 14/17] crypto: Don't mark AES tables const and cacheline_aligned Andi Kleen
2019-03-26 8:42 ` Rasmus Villemoes
2019-03-26 18:30 ` Andi Kleen
2019-03-21 22:00 ` [PATCH 15/17] x86/hyperv: Make hv_vcpu_is_preempted visible Andi Kleen
2019-03-26 8:15 ` Yi Sun
2019-03-21 22:00 ` [PATCH 16/17] x86/xen: Mark xen_vcpu_stolen as __visible Andi Kleen
2019-03-21 22:00 ` [PATCH 17/17] dm: Fix const confusion in dm Andi Kleen
2019-03-22 8:34 ` [PATCH 12/17] afs: Avoid section confusion in CM_NAME David Howells
2019-03-26 17:00 ` Fixes and cleanup from LTO tree Thomas Gleixner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=alpine.DEB.2.21.1903271958320.1789@nanos.tec.linutronix.de \
--to=tglx@linutronix.de \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).