All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>,
	Kees Cook <keescook@chromium.org>,
	Randy Dunlap <rdunlap@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Josh Poimboeuf <jpoimboe@redhat.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
	"Tobin C. Harding" <me@tobin.cc>,
	Steven Rostedt <rostedt@goodmis.org>,
	Jonathan Corbet <corbet@lwn.net>, Chris Mason <clm@fb.com>,
	Josef Bacik <jbacik@fb.com>, David Sterba <dsterba@suse.com>,
	"David S. Miller" <davem@davemloft.net>,
	Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>,
	Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Masahiro Yamada <yamada.masahiro@socionext.com>,
	Borislav Petkov <bp@suse.de>, Ian Abbott <abbotti@mev.co.uk>,
	Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
	Petr Mladek <pmladek@suse.com>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Pantelis Antoniou <pantelis.antoniou@konsulko.com>,
	Linux Btrfs <linux-btrfs@vger.kernel.org>,
	Network Development <netdev@vger.kernel.org>,
	Kernel Hardening <kernel-hardening@lists.openwall.com>
Subject: Re: [PATCH v3] kernel.h: Skip single-eval logic on literals in min()/max()
Date: Sun, 11 Mar 2018 12:05:43 +0100	[thread overview]
Message-ID: <20180311110543.pgcnrafloj5qbdjm@gmail.com> (raw)
In-Reply-To: <CA+55aFw+guiiJ3OavrP-RpFV9NRAF4t=oj3-GDfSyuZ9zBRfdg@mail.gmail.com>


* Linus Torvalds <torvalds@linux-foundation.org> wrote:

> So an error message like
> 
>    warning: ISO C90 requires array sizes to be constant-expressions
> 
> would be technically correct and useful from a portability angle. It
> tells you when you're doing something non-portable, and should be
> automatically enabled with "-ansi -pedantic", for example.
> 
> So what's misleading is actually the name of the warning and the
> message, not that it happens. The warning isn't about "variable
> length", it's literally about the rules for what a
> "constant-expression" is.
> 
> And in C, the expression (2,3) has a constant _value_ (namely 3), but
> it isn't a constant-expression as specified by the language.
> 
> Now, the thing is that once you actually do variable length arrays,
> those old front-end rules make no sense any more (outside of the "give
> portability warnings" thing).
> 
> Because once you do variable length arrays, you obviously _parse_
> everything just fine, and you're doing to evaluate much more complex
> expressions than some limited constant-expression rule.

BTW., while I fully agree with everything you said, it's not entirely correct to 
claim that if a C compiler can generate VLA code it is necessarily able to parse 
and evaluate constant array sizes "just fine".

Constant expressions are typically parsed very early on, at the preprocessing 
stage. They can be used with some preprocessor directives as well, such as '#if' 
(with some further limitations on their syntax).

If VLA support is implemented in a later stage, and results in heavy-handed code 
generation that will technically work for constant value expressions as well but 
results in suboptimal code, then a warning should probably be emitted - and it 
wouldn't be pedantic.

The existing warning is still very misleading:

  warning: ISO C90 forbids variable length array ‘array’ [-Wvla]

... and if my above theory is correct then I think a better warning would be 
something like:

  warning: Array declaration is not a C90 constant expression, resulting in VLA code generation

... and note that in this specific case it's not misleading to talk about VLAs in 
the warning text, because the array size, even if it's constant value, results in 
VLA code generation.

I don't know whether GCC has such a limitation, but a quick experiment with GCC 
7.2 suggests that a (2,3) array size expression results in a lot more code being 
generated than with a constant expression.

Thanks,

	Ingo

WARNING: multiple messages have this Message-ID (diff)
From: Ingo Molnar <mingo@kernel.org>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>,
	Kees Cook <keescook@chromium.org>,
	Randy Dunlap <rdunlap@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Josh Poimboeuf <jpoimboe@redhat.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
	"Tobin C. Harding" <me@tobin.cc>,
	Steven Rostedt <rostedt@goodmis.org>,
	Jonathan Corbet <corbet@lwn.net>, Chris Mason <clm@fb.com>,
	Josef Bacik <jbacik@fb.com>, David Sterba <dsterba@suse.com>,
	"David S. Miller" <davem@davemloft.net>,
	Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>,
	Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Masahiro Yamada <yamad
Subject: Re: [PATCH v3] kernel.h: Skip single-eval logic on literals in min()/max()
Date: Sun, 11 Mar 2018 12:05:43 +0100	[thread overview]
Message-ID: <20180311110543.pgcnrafloj5qbdjm@gmail.com> (raw)
In-Reply-To: <CA+55aFw+guiiJ3OavrP-RpFV9NRAF4t=oj3-GDfSyuZ9zBRfdg@mail.gmail.com>


* Linus Torvalds <torvalds@linux-foundation.org> wrote:

> So an error message like
> 
>    warning: ISO C90 requires array sizes to be constant-expressions
> 
> would be technically correct and useful from a portability angle. It
> tells you when you're doing something non-portable, and should be
> automatically enabled with "-ansi -pedantic", for example.
> 
> So what's misleading is actually the name of the warning and the
> message, not that it happens. The warning isn't about "variable
> length", it's literally about the rules for what a
> "constant-expression" is.
> 
> And in C, the expression (2,3) has a constant _value_ (namely 3), but
> it isn't a constant-expression as specified by the language.
> 
> Now, the thing is that once you actually do variable length arrays,
> those old front-end rules make no sense any more (outside of the "give
> portability warnings" thing).
> 
> Because once you do variable length arrays, you obviously _parse_
> everything just fine, and you're doing to evaluate much more complex
> expressions than some limited constant-expression rule.

BTW., while I fully agree with everything you said, it's not entirely correct to 
claim that if a C compiler can generate VLA code it is necessarily able to parse 
and evaluate constant array sizes "just fine".

Constant expressions are typically parsed very early on, at the preprocessing 
stage. They can be used with some preprocessor directives as well, such as '#if' 
(with some further limitations on their syntax).

If VLA support is implemented in a later stage, and results in heavy-handed code 
generation that will technically work for constant value expressions as well but 
results in suboptimal code, then a warning should probably be emitted - and it 
wouldn't be pedantic.

The existing warning is still very misleading:

  warning: ISO C90 forbids variable length array ‘array’ [-Wvla]

... and if my above theory is correct then I think a better warning would be 
something like:

  warning: Array declaration is not a C90 constant expression, resulting in VLA code generation

... and note that in this specific case it's not misleading to talk about VLAs in 
the warning text, because the array size, even if it's constant value, results in 
VLA code generation.

I don't know whether GCC has such a limitation, but a quick experiment with GCC 
7.2 suggests that a (2,3) array size expression results in a lot more code being 
generated than with a constant expression.

Thanks,

	Ingo

  parent reply	other threads:[~2018-03-11 11:05 UTC|newest]

Thread overview: 85+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-09 20:05 [PATCH v3] kernel.h: Skip single-eval logic on literals in min()/max() Kees Cook
2018-03-09 20:05 ` Kees Cook
2018-03-09 21:10 ` Linus Torvalds
2018-03-09 21:10   ` Linus Torvalds
2018-03-09 21:47   ` Kees Cook
2018-03-09 21:47     ` Kees Cook
2018-03-11 22:46   ` Tobin C. Harding
2018-03-11 22:46     ` Tobin C. Harding
2018-03-11 22:46     ` Tobin C. Harding
2018-03-13 13:31   ` David Laight
2018-03-13 13:31     ` David Laight
2018-03-10  0:07 ` Andrew Morton
2018-03-10  0:07   ` Andrew Morton
2018-03-10  0:28   ` Linus Torvalds
2018-03-10  0:28     ` Linus Torvalds
2018-03-10  0:32     ` Andrew Morton
2018-03-10  0:32       ` Andrew Morton
2018-03-10  0:38       ` Linus Torvalds
2018-03-10  0:38         ` Linus Torvalds
2018-03-10  1:30         ` Kees Cook
2018-03-10  1:30           ` Kees Cook
2018-03-10  1:31           ` Kees Cook
2018-03-10  1:31             ` Kees Cook
2018-03-10  2:37             ` Linus Torvalds
2018-03-10  2:37               ` Linus Torvalds
2018-03-12 22:55           ` Andrew Morton
2018-03-12 22:55             ` Andrew Morton
2018-03-12 23:57             ` Linus Torvalds
2018-03-12 23:57               ` Linus Torvalds
2018-03-13  4:28               ` Kees Cook
2018-03-13  4:28                 ` Kees Cook
2018-03-13 21:02                 ` Andrew Morton
2018-03-13 21:02                   ` Andrew Morton
2018-03-13 22:14                   ` Kees Cook
2018-03-13 22:14                     ` Kees Cook
2018-03-14 11:35                     ` David Laight
2018-03-14 11:35                       ` David Laight
2018-03-10  3:11   ` Randy Dunlap
2018-03-10  3:11     ` Randy Dunlap
2018-03-10  6:10     ` Miguel Ojeda
2018-03-10  6:10       ` Miguel Ojeda
2018-03-10  7:03       ` Miguel Ojeda
2018-03-10  7:03         ` Miguel Ojeda
2018-03-10 16:04         ` Linus Torvalds
2018-03-10 16:04           ` Linus Torvalds
2018-03-10 15:33       ` Kees Cook
2018-03-10 15:33         ` Kees Cook
2018-03-10 16:11         ` Linus Torvalds
2018-03-10 16:11           ` Linus Torvalds
2018-03-10 16:30         ` Linus Torvalds
2018-03-10 16:30           ` Linus Torvalds
2018-03-10 17:34           ` Miguel Ojeda
2018-03-10 17:34             ` Miguel Ojeda
2018-03-10 17:51             ` Linus Torvalds
2018-03-10 17:51               ` Linus Torvalds
2018-03-10 19:08               ` Miguel Ojeda
2018-03-10 19:08                 ` Miguel Ojeda
2018-03-11 11:05               ` Ingo Molnar [this message]
2018-03-11 11:05                 ` Ingo Molnar
2018-03-11 18:23                 ` Linus Torvalds
2018-03-11 18:23                   ` Linus Torvalds
2018-03-10  2:34 [PATCH 0/3] tracing: Rewrite the function filter code Steven Rostedt
2018-03-10  2:34 ` [PATCH 1/3] tracing: Combine enum and arrays into single macro in " Steven Rostedt
2018-03-12 10:31   ` Masami Hiramatsu
2018-03-10  2:34 ` [PATCH 2/3] tracing: Clean up and document pred_funcs_##type creation and use Steven Rostedt
2018-03-12 13:42   ` Masami Hiramatsu
2018-03-10  2:34 ` [PATCH 3/3] tracing: Rewrite filter logic to be simpler and faster Steven Rostedt
2018-03-10  3:10   ` Steven Rostedt
2018-03-10  3:10     ` Steven Rostedt
2018-03-10  3:15     ` Steven Rostedt
2018-03-10  3:15       ` Steven Rostedt
2018-03-10  3:22       ` Steven Rostedt
2018-03-10  3:22         ` Steven Rostedt
2018-03-10  3:18   ` Steven Rostedt
2018-03-12 12:42   ` Jiri Olsa
2018-03-12 18:38     ` Steven Rostedt
2018-03-12 15:10   ` Jiri Olsa
2018-03-12 18:40     ` Steven Rostedt
2018-03-12 18:54       ` Jiri Olsa
2018-03-12 19:10         ` Steven Rostedt
2018-03-12 23:52         ` Steven Rostedt
2018-03-13 10:14           ` Jiri Olsa
2018-03-13 14:12             ` Steven Rostedt
2018-03-13 14:27               ` Jiri Olsa
2018-03-11 19:54 ` [PATCH 0/3] tracing: Rewrite the function filter code Jiri Olsa

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=20180311110543.pgcnrafloj5qbdjm@gmail.com \
    --to=mingo@kernel.org \
    --cc=abbotti@mev.co.uk \
    --cc=akpm@linux-foundation.org \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=bp@suse.de \
    --cc=clm@fb.com \
    --cc=corbet@lwn.net \
    --cc=davem@davemloft.net \
    --cc=dsterba@suse.com \
    --cc=gustavo@embeddedor.com \
    --cc=jbacik@fb.com \
    --cc=jpoimboe@redhat.com \
    --cc=keescook@chromium.org \
    --cc=kernel-hardening@lists.openwall.com \
    --cc=kuznet@ms2.inr.ac.ru \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=me@tobin.cc \
    --cc=miguel.ojeda.sandonis@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=pantelis.antoniou@konsulko.com \
    --cc=peterz@infradead.org \
    --cc=pmladek@suse.com \
    --cc=rdunlap@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=sergey.senozhatsky.work@gmail.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=yamada.masahiro@socionext.com \
    --cc=yoshfuji@linux-ipv6.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.