public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "redi at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libstdc++/112882] [14 Regression] std::clamp no longer usable in header only mode
Date: Wed, 06 Dec 2023 17:07:35 +0000	[thread overview]
Message-ID: <bug-112882-4-OUEG2hq3ME@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-112882-4@http.gcc.gnu.org/bugzilla/>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112882

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
That change was broken anyway: when _GLIBCXX_ASSERTIONS was not defined, the
condition in the assertion is if constexpr (is_constant_evaluated()) which is
always true, even when not actually doing constant eval.

I'll test this:

--- a/libstdc++-v3/include/bits/c++config
+++ b/libstdc++-v3/include/bits/c++config
@@ -538,6 +538,7 @@ namespace std
   // This can be used without checking if the compiler supports the feature.
   // The macro _GLIBCXX_HAVE_IS_CONSTANT_EVALUATED can be used to check if
   // the compiler support is present to make this function work as expected.
+  __attribute__((__always_inline__))
   _GLIBCXX_CONSTEXPR inline bool
   __is_constant_evaluated() _GLIBCXX_NOEXCEPT
   {
@@ -589,19 +590,28 @@ namespace std
 #endif

 #if defined(_GLIBCXX_ASSERTIONS)
-# define _GLIBCXX_DO_ASSERT true
-#elif _GLIBCXX_HAVE_IS_CONSTANT_EVALUATED
-# define _GLIBCXX_DO_ASSERT std::__is_constant_evaluated()
-#else
-# define _GLIBCXX_DO_ASSERT false
-#endif
-
 # define __glibcxx_assert(cond)                                               
\
   do {                                                                 \
-    if _GLIBCXX17_CONSTEXPR (_GLIBCXX_DO_ASSERT)                       \
-      if (__builtin_expect(!bool(cond), false))                               
\
-       _GLIBCXX_ASSERT_FAIL(cond);                                     \
+    if (__builtin_expect(!bool(cond), false))                          \
+      _GLIBCXX_ASSERT_FAIL(cond);                                      \
   } while (false)
+#elif _GLIBCXX_HAVE_IS_CONSTANT_EVALUATED
+namespace std
+{
+  __attribute__((__always_inline__,visibility("default")))
+  inline void
+  __glibcxx_compile_time_assert_fail()
+  { }
+}
+# define __glibcxx_assert(cond)                                               
\
+  do {                                                                 \
+    if (std::__is_constant_evaluated())                                       
\
+      if (__builtin_expect(!bool(cond), false))                               
\
+       __glibcxx_compile_time_assert_fail();                           \
+  } while (false)
+#else
+# define __glibcxx_assert(cond)
+#endif

 // Macro indicating that TSAN is in use.
 #if __SANITIZE_THREAD__

  parent reply	other threads:[~2023-12-06 17:07 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-06 15:20 [Bug libstdc++/112882] New: " tnfchris at gcc dot gnu.org
2023-12-06 16:47 ` [Bug libstdc++/112882] " redi at gcc dot gnu.org
2023-12-06 16:55 ` redi at gcc dot gnu.org
2023-12-06 17:07 ` redi at gcc dot gnu.org [this message]
2023-12-06 17:09 ` redi at gcc dot gnu.org
2023-12-06 17:20 ` redi at gcc dot gnu.org
2023-12-07  8:23 ` tnfchris at gcc dot gnu.org
2023-12-07  9:49 ` redi at gcc dot gnu.org
2023-12-07 20:55 ` cvs-commit at gcc dot gnu.org
2023-12-07 21:03 ` redi at gcc dot gnu.org

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=bug-112882-4-OUEG2hq3ME@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@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).