public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jonathan Wakely <jwakely@redhat.com>
To: Andrew Pinski <apinski@marvell.com>
Cc: gcc-patches@gcc.gnu.org
Subject: [PATCH v2] GCC-13/changes: Add note about iostream usage
Date: Fri, 28 Apr 2023 13:56:37 +0100	[thread overview]
Message-ID: <ZEvChfzMa0IotL/h@redhat.com> (raw)
In-Reply-To: <ZEl2M+SrYjkUmyub@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 2135 bytes --]

On 26/04/23 20:06 +0100, Jonathan Wakely wrote:
>On 26/04/23 09:53 -0700, Andrew Pinski wrote:
>>This adds a note about iostream usage so it does not catch others
>>in surpise like it has already.
>>
>>OK?
>
>Thanks, I agree we should add something, but have some comments below.
>
>>---
>>htdocs/gcc-13/changes.html | 5 +++++
>>1 file changed, 5 insertions(+)
>>
>>diff --git a/htdocs/gcc-13/changes.html b/htdocs/gcc-13/changes.html
>>index 70732ec0..7c83f7c4 100644
>>--- a/htdocs/gcc-13/changes.html
>>+++ b/htdocs/gcc-13/changes.html
>>@@ -25,6 +25,11 @@ You may also want to check out our
>><!-- .................................................................. -->
>><h2>Caveats</h2>
>><ul>
>>+    <li>libstdc++ uses constructors inside the library to initialize std::cout/std::cin, etc.
>>+     instead of having it done in each source which uses iostream header.
>
>We should use code font for std::cout, std::cin and iostream, and
>style it as <iostream> not just iostream.
>
>>+     This requires you to make sure the dynamic loader to load the new libstdc++v3 library
>>+     (examples of how to do this is to use -Wl,-rpath,... while linking or LD_LIBRARY_PATH
>>+     while running the program).  </li>
>
>I think it would be better to link to https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dynamic_or_shared.html#manual.intro.using.linkage.dynamic
>
>How about:
>
>  <li>For C++, construction of the global iostream objects   
><code>std::cout</code>, <code>std::cin</code> etc. is now done
>  inside the standard library, instead of in every source file that
>  includes the <code>&lt;iostream&gt;</code> header. This change
>  improves the start-up performance of C++ programs, but it means that
>  code compiled with GCC 13.1 will crash if the correct version of
>  <code>libstdc++.so</code> is not used at runtime. See the
>  <a href="https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dynamic_or_shared.html#manual.intro.using.linkage.dynamic">documentation</a>
>  about using the right <code>libstdc++.so</code> at runtime.
>  </li>

Here's a proper patch proposal along those lines.

OK for wwwdocs?



[-- Attachment #2: patch.txt --]
[-- Type: text/plain, Size: 1488 bytes --]

commit cf408a8d7e9ee3c7efd5b4a3fa5697f4a85a036a
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Fri Apr 28 13:47:12 2023 +0100

    Add caveat about C++ iostream init changes (PR108969)
    
    Co-authored-by: Andrew Pinski <apinski@marvell.com>

diff --git a/htdocs/gcc-13/changes.html b/htdocs/gcc-13/changes.html
index 70732ec0..f9533494 100644
--- a/htdocs/gcc-13/changes.html
+++ b/htdocs/gcc-13/changes.html
@@ -56,6 +56,18 @@ You may also want to check out our
       <a href="https://gcc.gnu.org/onlinedocs/gcc-13.1.0/gcc/ARM-iWMMXt-Built-in-Functions.html">
       iWMMXt built-in functions</a>.
     </li>
+    <li>For C++, construction of the global iostream objects
+      <code>std::cout</code>, <code>std::cin</code> etc. is now done
+      inside the standard library, instead of in every source file that
+      includes the <code>&lt;iostream&gt;</code> header. This change
+      improves the start-up performance of C++ programs, but it means that
+      code compiled with GCC 13.1 will crash if the correct version of
+      <code>libstdc++.so</code> is not used at runtime. See the
+      <a href="https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dynamic_or_shared.html#manual.intro.using.linkage.dynamic">documentation</a>
+      about using the right <code>libstdc++.so</code> at runtime.
+      Future GCC releases will mitigate the problem so that the program
+      cannot be run at all with an older <code>libstdc++.so</code>.
+    </li>
 </ul>
 
 

      parent reply	other threads:[~2023-04-28 12:56 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-26 16:53 [PATCH] " Andrew Pinski
2023-04-26 19:06 ` Jonathan Wakely
2023-04-26 19:10   ` Andrew Pinski
2023-04-28 12:56   ` Jonathan Wakely [this message]

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=ZEvChfzMa0IotL/h@redhat.com \
    --to=jwakely@redhat.com \
    --cc=apinski@marvell.com \
    --cc=gcc-patches@gcc.gnu.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 read-only IMAP folder(s) and NNTP newsgroup(s).