https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617
Bug ID: 92617
Summary: Invalid loop optimization: no exit condition
Product: gcc
Version: 8.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: piotr.seweryn at gmail dot com
Target Milestone: ---
Created attachment 47322
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47322&action=edit
Compilation process stderr output
Consider the following code:
#include <stdio.h>
int parseList()
{
int i;
for (i = 0; i < 3; i++) {
printf("DUPA: %d %d\n", i, i);
}
}
int main(int argc, char **argv)
{
printf("start\n");
parseList();
}
Obviously there is no return statement in parseList function and "no return
statement in function returning non-void" warning was issued.
Command used to compile:
gcc -Wall -Wextra -v -save-temps -O1 test.cpp -otest -lstdc++ 1> log.stdout 2>
log.stderr
Expected result:
3 loop iterations, 4 messages displayed to stdout
Actual result:
endless loop
If we take a closer look at the assembler code (output of objdump -S):
000000000001145 <_Z9parseListv>:
1145: 48 83 ec 08 sub $0x8,%rsp
1149: ba 00 00 00 00 mov $0x0,%edx
114e: be 00 00 00 00 mov $0x0,%esi
1153: 48 8d 3d aa 0e 00 00 lea 0xeaa(%rip),%rdi # 2004
<_IO_stdin_used+0x4>
115a: b8 00 00 00 00 mov $0x0,%eax
115f: e8 cc fe ff ff callq 1030 <printf@plt>
1164: ba 01 00 00 00 mov $0x1,%edx
1169: be 01 00 00 00 mov $0x1,%esi
116e: 48 8d 3d 8f 0e 00 00 lea 0xe8f(%rip),%rdi # 2004
<_IO_stdin_used+0x4>
1175: b8 00 00 00 00 mov $0x0,%eax
117a: e8 b1 fe ff ff callq 1030 <printf@plt>
117f: ba 02 00 00 00 mov $0x2,%edx
1184: be 02 00 00 00 mov $0x2,%esi
1189: 48 8d 3d 74 0e 00 00 lea 0xe74(%rip),%rdi # 2004
<_IO_stdin_used+0x4>
1190: b8 00 00 00 00 mov $0x0,%eax
1195: e8 96 fe ff ff callq 1030 <printf@plt>
000000000000119a <main>:
119a: 48 83 ec 08 sub $0x8,%rsp
119e: 48 8d 3d 6c 0e 00 00 lea 0xe6c(%rip),%rdi # 2011
<_IO_stdin_used+0x11>
11a5: e8 96 fe ff ff callq 1040 <puts@plt>
11aa: e8 96 ff ff ff callq 1145 <_Z9parseListv>
11af: 90 nop
There is no return instruction at the end of parseList function, so we will
fall-through to main function, then we will call parseList function and
fall-through again to main and so on.
System specification:
gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/8/lto-wrapper
OFFLOAD_TARGET_NAMES=nvptx-none
OFFLOAD_TARGET_DEFAULT=1
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian 8.3.0-6'
--with-bugurl=file:///usr/share/doc/gcc-8/README.Bugs
--enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --prefix=/usr
--with-gcc-major-version-only --program-suffix=-8
--program-prefix=x86_64-linux-gnu- --enable-shared --enable-linker-build-id
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix
--libdir=/usr/lib --enable-nls --enable-bootstrap --enable-clocale=gnu
--enable-libstdcxx-debug --enable-libstdcxx-time=yes
--with-default-libstdcxx-abi=new --enable-gnu-unique-object
--disable-vtable-verify --enable-libmpx --enable-plugin --enable-default-pie
--with-system-zlib --with-target-system-zlib --enable-objc-gc=auto
--enable-multiarch --disable-werror --with-arch-32=i686 --with-abi=m64
--with-multilib-list=m32,m64,mx32 --enable-multilib --with-tune=generic
--enable-offload-targets=nvptx-none --without-cuda-driver
--enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu
--target=x86_64-linux-gnu
Thread model: posix
gcc version 8.3.0 (Debian 8.3.0-6)
Debian 10.1 x86_64, updated to newest version.
Problem occurs when optimization O1 is used.
