https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79425
Bug ID: 79425
Summary: Inline assembly getting clobbered in some case
Product: gcc
Version: 5.4.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: mini at cs dot technion.ac.il
Target Milestone: ---
I am getting inline assembly code clobbered if I #include<algorithm>, it is the
1st code in the .text segment, AND I use -std=c++11
This is the INCORRECT behavior. As you can see, the label reg_in is gone:
$ gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/5/lto-wrapper
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu
5.4.0-6ubuntu1~16.04.4' --with-bugurl=file:///usr/share/doc/gcc-5/README.Bugs
--enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --prefix=/usr
--program-suffix=-5 --enable-shared --enable-linker-build-id
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix
--libdir=/usr/lib --enable-nls --with-sysroot=/ --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 --with-system-zlib
--disable-browser-plugin --enable-java-awt=gtk --enable-gtk-cairo
--with-java-home=/usr/lib/jvm/java-1.5.0-gcj-5-amd64/jre --enable-java-home
--with-jvm-root-dir=/usr/lib/jvm/java-1.5.0-gcj-5-amd64
--with-jvm-jar-dir=/usr/lib/jvm-exports/java-1.5.0-gcj-5-amd64
--with-arch-directory=amd64 --with-ecj-jar=/usr/share/java/eclipse-ecj.jar
--enable-objc-gc --enable-multiarch --disable-werror --with-arch-32=i686
--with-abi=m64 --with-multilib-list=m32,m64,mx32 --enable-multilib
--with-tune=generic --enable-checking=release --build=x86_64-linux-gnu
--host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.4)
$ cat prog.c
#include <algorithm>
//void foo(){}
__asm__ volatile ("reg_in: \n\t"
"pushf\n\t"
"push %rbp\n\t"
"mov %rsp,%rbp\n\t"
"push %rdi\n\t"
"push %rsi\n\t"
"push %rdx\n\t"
"push %rcx\n\t"
"push %r8 \n\t"
"push %r9 \n\t"
"push %rax\n\t"
);
int main(int argc, char *argv[])
{
}
$ g++ -std=c++11 prog.c
$ objdump -d a.out # I put here only partial output
00000000004004b0 <frame_dummy>:
4004b0: bf 20 0e 60 00 mov $0x600e20,%edi
4004b5: 48 83 3f 00 cmpq $0x0,(%rdi)
4004b9: 75 05 jne 4004c0 <frame_dummy+0x10>
4004bb: eb 93 jmp 400450 <register_tm_clones>
4004bd: 0f 1f 00 nopl (%rax)
4004c0: b8 00 00 00 00 mov $0x0,%eax
4004c5: 48 85 c0 test %rax,%rax
4004c8: 74 f1 je 4004bb <frame_dummy+0xb>
4004ca: 55 push %rbp
4004cb: 48 89 e5 mov %rsp,%rbp
4004ce: ff d0 callq *%rax
4004d0: 5d pop %rbp
4004d1: e9 7a ff ff ff jmpq 400450 <register_tm_clones>
00000000004004d6 <main>:
4004d6: 55 push %rbp
4004d7: 48 89 e5 mov %rsp,%rbp
4004da: 89 7d fc mov %edi,-0x4(%rbp)
4004dd: 48 89 75 f0 mov %rsi,-0x10(%rbp)
4004e1: b8 00 00 00 00 mov $0x0,%eax
4004e6: 5d pop %rbp
4004e7: c3 retq
4004e8: 0f 1f 84 00 00 00 00 nopl 0x0(%rax,%rax,1)
4004ef: 00
If I do 1 of the following, I get the correct result
1. uncomment void foo(){}
2. remove #include <algorithm>
The examples below, are of CORRECT behaviors. Here the label reg_in stays:
$ vi prog.c
$ g++ -std=c++11 prog.c
$ cat prog.c
#include <algorithm>
void foo(){}
__asm__ volatile ("reg_in: \n\t"
"pushf\n\t"
"push %rbp\n\t"
"mov %rsp,%rbp\n\t"
"push %rdi\n\t"
"push %rsi\n\t"
"push %rdx\n\t"
"push %rcx\n\t"
"push %r8 \n\t"
"push %r9 \n\t"
"push %rax\n\t"
);
int main(int argc, char *argv[])
{
}
$ objdump -d a.out # I put here only partial output
00000000004004b0 <frame_dummy>:
4004b0: bf 20 0e 60 00 mov $0x600e20,%edi
4004b5: 48 83 3f 00 cmpq $0x0,(%rdi)
4004b9: 75 05 jne 4004c0 <frame_dummy+0x10>
4004bb: eb 93 jmp 400450 <register_tm_clones>
4004bd: 0f 1f 00 nopl (%rax)
4004c0: b8 00 00 00 00 mov $0x0,%eax
4004c5: 48 85 c0 test %rax,%rax
4004c8: 74 f1 je 4004bb <frame_dummy+0xb>
4004ca: 55 push %rbp
4004cb: 48 89 e5 mov %rsp,%rbp
4004ce: ff d0 callq *%rax
4004d0: 5d pop %rbp
4004d1: e9 7a ff ff ff jmpq 400450 <register_tm_clones>
00000000004004d6 <_Z3foov>:
4004d6: 55 push %rbp
4004d7: 48 89 e5 mov %rsp,%rbp
4004da: 90 nop
4004db: 5d pop %rbp
4004dc: c3 retq
00000000004004dd <reg_in>:
4004dd: 9c pushfq
4004de: 55 push %rbp
4004df: 48 89 e5 mov %rsp,%rbp
4004e2: 57 push %rdi
4004e3: 56 push %rsi
4004e4: 52 push %rdx
4004e5: 51 push %rcx
4004e6: 41 50 push %r8
4004e8: 41 51 push %r9
4004ea: 50 push %rax
00000000004004eb <main>:
4004eb: 55 push %rbp
4004ec: 48 89 e5 mov %rsp,%rbp
4004ef: 89 7d fc mov %edi,-0x4(%rbp)
4004f2: 48 89 75 f0 mov %rsi,-0x10(%rbp)
4004f6: b8 00 00 00 00 mov $0x0,%eax
4004fb: 5d pop %rbp
4004fc: c3 retq
4004fd: 0f 1f 00 nopl (%rax)
$ vi prog.c
$ g++ -std=c++11 prog.c
$ cat prog.c
//#include <algorithm>
//void foo(){}
__asm__ volatile ("reg_in: \n\t"
"pushf\n\t"
"push %rbp\n\t"
"mov %rsp,%rbp\n\t"
"push %rdi\n\t"
"push %rsi\n\t"
"push %rdx\n\t"
"push %rcx\n\t"
"push %r8 \n\t"
"push %r9 \n\t"
"push %rax\n\t"
);
int main(int argc, char *argv[])
{
}
$ objdump -d a.out # I put here only partial output
00000000004004b0 <frame_dummy>:
4004b0: bf 20 0e 60 00 mov $0x600e20,%edi
4004b5: 48 83 3f 00 cmpq $0x0,(%rdi)
4004b9: 75 05 jne 4004c0 <frame_dummy+0x10>
4004bb: eb 93 jmp 400450 <register_tm_clones>
4004bd: 0f 1f 00 nopl (%rax)
4004c0: b8 00 00 00 00 mov $0x0,%eax
4004c5: 48 85 c0 test %rax,%rax
4004c8: 74 f1 je 4004bb <frame_dummy+0xb>
4004ca: 55 push %rbp
4004cb: 48 89 e5 mov %rsp,%rbp
4004ce: ff d0 callq *%rax
4004d0: 5d pop %rbp
4004d1: e9 7a ff ff ff jmpq 400450 <register_tm_clones>
00000000004004d6 <reg_in>:
4004d6: 9c pushfq
4004d7: 55 push %rbp
4004d8: 48 89 e5 mov %rsp,%rbp
4004db: 57 push %rdi
4004dc: 56 push %rsi
4004dd: 52 push %rdx
4004de: 51 push %rcx
4004df: 41 50 push %r8
4004e1: 41 51 push %r9
4004e3: 50 push %rax
00000000004004e4 <main>:
4004e4: 55 push %rbp
4004e5: 48 89 e5 mov %rsp,%rbp
4004e8: 89 7d fc mov %edi,-0x4(%rbp)
4004eb: 48 89 75 f0 mov %rsi,-0x10(%rbp)
4004ef: b8 00 00 00 00 mov $0x0,%eax
4004f4: 5d pop %rbp
4004f5: c3 retq
4004f6: 66 2e 0f 1f 84 00 00 nopw %cs:0x0(%rax,%rax,1)
4004fd: 00 00 00
