https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #50 from Erich Löw ---
Q_1
What would happen if saying (just hypothetically)
--> All prios in range 0 ... 2^^16 - 1 are allowed
--> Imho consequence of removing split in reserved and not-reserved prios.
Q_2
Without usage of [[gnu::i
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #49 from Erich Löw ---
Yep, understood: using priorities below 101 in user code is forbidden.
I was just eager to see if using reserved priority 99 in user code together
with production compiler would trigger the duplicating of symb
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #47 from Erich Löw ---
Created attachment 60620
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=60620&action=edit
The generated assembler file
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #46 from Erich Löw ---
Created attachment 60619
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=60619&action=edit
The make file
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #45 from Erich Löw ---
Created attachment 60618
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=60618&action=edit
The source code
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #38 from Erich Löw ---
Using [[gnu::init_priority(0)]] (I saw it right now per coincidency)
produces error advise as follows:
main.cc:45:39: error: requested ‘init_priority’ 0 is out of range [0, 65535]
45 | [[gnu::init_priority
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #37 from Erich Löw ---
I found in parallel as well that using 98 instead of 99 purges the "as"
generated error.
Not yet created a clean and simplified reproduction
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #25 from Erich Löw ---
With system GCC 15.0.1 I do as below
I created: main.cc
#include
#include
struct A { A() { } };
[[gnu::init_priority(99)]] A a;
[[gnu::init_priority(99)]] A a2;
struct B { B() { } };
[[gnu::init_prio
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #19 from Erich Löw ---
Oki.
I do as you propose.
I'll report in next comment slice the outputs.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #17 from Erich Löw ---
Recompile from scratch with none additional CFLAGS, CCFLAGS et al done
Here the result
libtool: compile: /home/SETUP/GNU/gcc/BUILD/./gcc/xgcc -shared-libgcc
-B/home/SETUP/GNU/gcc/BUILD/./gcc -nostdinc++
-L/h
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #16 from Erich Löw ---
In parallel: how did I come to "CCFLAGS=-pipe -march=native -O2 -fPIC
-fomit-frame-pointer"?
--> They are from linux kernel compiling
--> And I thought 30 Years ago: let compile kernels and local LATEST GNU
omp
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #15 from Erich Löw ---
As proposed I do now:
- I deleted my whole BUILD dir
- I recreated BUILD dir as empty dir
- I unset CFLAGS, CPPFLAGS, CCFLAGS and CXXFLAGS
- I rerun configure steps in BUILD dir
- I say ma
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #11 from Erich Löw ---
The whole env output:
LS_COLORS=no=00:fi=00:di=01;34:ln=00;36:pi=40;33:so=01;35:do=01;35:bd=40;33;01:cd=40;33;01:or=41;33;01:ex=00;32:*.cmd=00;32:*.exe=01;32:*.com=01;32:*.bat=01;32:*.btm=01;32:*.dll=01;32:*.t
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #10 from Erich Löw ---
Envs:
GCCVERSION=15.0.1
CCFLAGS=-pipe -march=native -O2 -fPIC -fomit-frame-pointer
$GCCVERSION is initialized as this:
export GCCVERSION=`gcc -dumpfullversion`
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #8 from Erich Löw ---
I try
1. Version of GCC:
gcc (GCC) 15.0.1 20250219 (experimental)
Copyright (C) 2025 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
--- Comment #5 from Erich Löw ---
I found in "7.7 C++-Specific Variable, Function, and Type Attributes" that
the lowest supported cardinal (indicating highest prio) should be 101.
I tried to replace all occurrences of 99 with 101 and LATEST com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118981
Bug ID: 118981
Summary: tzdb.cc contains 3 times in sequence:
[[gnu::init_priority(99)]]
Product: gcc
Version: 15.0
Status: UNCONFIRMED
Severity: normal
17 matches
Mail list logo
