C++ 用GDB在Fortran中调试MPI程序
我阅读并到达，所以现在我想我应该（如果不是，请告诉我）重写代码C++ 用GDB在Fortran中调试MPI程序,c++,debugging,fortran,gdb,mpi,C++,Debugging,Fortran,Gdb,Mpi,我阅读并到达，所以现在我想我应该（如果不是，请告诉我）重写代码 { int i = 0; char hostname[256]; gethostname(hostname, sizeof(hostname)); printf("PID %d on %s ready for attach\n", getpid(), hostname); fflush(stdout); while (0 == i) sleep(5); } 用Fort
{
int i = 0;
char hostname[256];
gethostname(hostname, sizeof(hostname));
printf("PID %d on %s ready for attach\n", getpid(), hostname);
fflush(stdout);
while (0 == i)
sleep(5);
}
MPI\u Get\u processor\u namegethostnameflushMPI_Init-v-da-QmpiforttopPID PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1642 me 20 0 167328 7716 5816 R 100.0 0.047 0:25.02 a.out
1644 me 20 0 167328 7656 5756 R 100.0 0.047 0:25.04 a.out
1645 me 20 0 167328 7700 5792 R 100.0 0.047 0:24.97 a.out
1646 me 20 0 167328 7736 5836 R 100.0 0.047 0:25.00 a.out
1641 me 20 0 167328 7572 5668 R 99.67 0.046 0:24.95 a.out
gdb-pidbt您的代码： 我认为在Fortran中不需要刷新。我认为至少在我使用的编译器中，Fortran
写和打印刷新是必要的
但是您肯定可以使用flush
语句
use iso_fortran_env
flush(output_unit)
只需在打印主机名和pid的write
之后刷新即可。但正如我所说的，我将只从印刷开始
您需要做的是登录到该节点，并使用类似于
gdb -pid 12345
use mpi
implicit none
character(MPI_MAX_PROCESSOR_NAME) :: hostname
integer :: rank, ie, pid, hostname_len
integer, volatile :: i
call MPI_Init(ie)
call MPI_Get_processor_name(hostname, hostname_len, ie)
!non-standard extension
pid = getpid()
call MPI_Comm_rank(MPI_COMM_WORLD, rank, ie)
write(*,*) "PID ", pid, " on ", trim(hostname), " ready for attach is world rank ", rank
!this serves to block the execution at a specific place until you unblock it in GDB by setting i=0
i = 1
do
!non-standard extension
call sleep(1)
if (i==0) exit
end do
end
MAIN__ () at mpi_gdb.f90:26
26 if (i==0) exit
(gdb) info locals
hostname = 'linux.site', ' ' <repeats 246 times>
hostname_len = 10
i = 1
ie = 0
pid = 2457
rank = 0
(gdb) set var i = 0
(gdb) cont
Continuing.
[Inferior 1 (process 2355) exited normally]
> mpif90 prog.f90 -O0 -g -o prog.exe
Rank 0 has escaped!
对于睡眠，您可以使用许多编译器中提供的非标准sleep
内部子例程，也可以编写自己的子例程
是在MPI_Init之前还是之后？如果要打印排名，必须在后面。同样，如果要使用MPI\u Get\u processor\u name
，它必须在后面。通常建议在程序中尽早调用MPI_Init
代码是这样的
gdb -pid 12345
use mpi
implicit none
character(MPI_MAX_PROCESSOR_NAME) :: hostname
integer :: rank, ie, pid, hostname_len
integer, volatile :: i
call MPI_Init(ie)
call MPI_Get_processor_name(hostname, hostname_len, ie)
!non-standard extension
pid = getpid()
call MPI_Comm_rank(MPI_COMM_WORLD, rank, ie)
write(*,*) "PID ", pid, " on ", trim(hostname), " ready for attach is world rank ", rank
!this serves to block the execution at a specific place until you unblock it in GDB by setting i=0
i = 1
do
!non-standard extension
call sleep(1)
if (i==0) exit
end do
end
MAIN__ () at mpi_gdb.f90:26
26 if (i==0) exit
(gdb) info locals
hostname = 'linux.site', ' ' <repeats 246 times>
hostname_len = 10
i = 1
ie = 0
pid = 2457
rank = 0
(gdb) set var i = 0
(gdb) cont
Continuing.
[Inferior 1 (process 2355) exited normally]
> mpif90 prog.f90 -O0 -g -o prog.exe
Rank 0 has escaped!
重要提示：如果使用优化进行编译，编译器会发现i==0
永远不会为真，并且会完全删除检查。您必须降低优化或将i
声明为volatile
。Volatile意味着值可以随时更改，编译器必须从内存重新加载其值以进行检查。这需要Fortran 2003
附加正确的流程：
上面的代码将打印，例如
> mpif90 -ggdb mpi_gdb.f90
> mpirun -n 4 ./a.out
PID 2356 on linux.site ready for attach is world rank 1
PID 2357 on linux.site ready for attach is world rank 2
PID 2358 on linux.site ready for attach is world rank 3
PID 2355 on linux.site ready for attach is world rank 0
上面看起来像
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2355 me 20 0 167328 7452 5564 R 100.0 0.045 1:42.55 a.out
2356 me 20 0 167328 7428 5548 R 100.0 0.045 1:42.54 a.out
2357 me 20 0 167328 7384 5500 R 100.0 0.045 1:42.54 a.out
2358 me 20 0 167328 7388 5512 R 100.0 0.045 1:42.51 a.out
你只需选择你想要的排名并执行
gdb -pid 2355
附加秩0，依此类推。当然，在另一个终端窗口中
然后你会得到这样的结果
gdb -pid 12345
use mpi
implicit none
character(MPI_MAX_PROCESSOR_NAME) :: hostname
integer :: rank, ie, pid, hostname_len
integer, volatile :: i
call MPI_Init(ie)
call MPI_Get_processor_name(hostname, hostname_len, ie)
!non-standard extension
pid = getpid()
call MPI_Comm_rank(MPI_COMM_WORLD, rank, ie)
write(*,*) "PID ", pid, " on ", trim(hostname), " ready for attach is world rank ", rank
!this serves to block the execution at a specific place until you unblock it in GDB by setting i=0
i = 1
do
!non-standard extension
call sleep(1)
if (i==0) exit
end do
end
MAIN__ () at mpi_gdb.f90:26
26 if (i==0) exit
(gdb) info locals
hostname = 'linux.site', ' ' <repeats 246 times>
hostname_len = 10
i = 1
ie = 0
pid = 2457
rank = 0
(gdb) set var i = 0
(gdb) cont
Continuing.
[Inferior 1 (process 2355) exited normally]
> mpif90 prog.f90 -O0 -g -o prog.exe
Rank 0 has escaped!
mpi_gdb.f90:26处的主菜单（MAIN）
26如果（i==0）退出
（gdb）本地信息
主机名='linux.site'，'
主机名\u len=10
i=1
ie=0
pid=2457
排名=0
（gdb）设置变量i=0
（gdb）续
持续的。
[次1（进程2355）正常退出]
发布的代码基本上只是一个无限循环，用于在附加调试器时“暂停”执行。然后可以使用调试器控件退出此循环，程序将继续。您可以用fortran编写一个等价的循环，因此如果您愿意从另一个方法获取主机名和pid（请参阅VladimirF在回答中提到的mpi_get_processor_name，如果您愿意使用编译器扩展，gnu和intel编译器都提供了getpid
扩展），您可以使用以下内容（感谢您提供的睡眠示例）
现在，使用类似
gdb -pid 12345
use mpi
implicit none
character(MPI_MAX_PROCESSOR_NAME) :: hostname
integer :: rank, ie, pid, hostname_len
integer, volatile :: i
call MPI_Init(ie)
call MPI_Get_processor_name(hostname, hostname_len, ie)
!non-standard extension
pid = getpid()
call MPI_Comm_rank(MPI_COMM_WORLD, rank, ie)
write(*,*) "PID ", pid, " on ", trim(hostname), " ready for attach is world rank ", rank
!this serves to block the execution at a specific place until you unblock it in GDB by setting i=0
i = 1
do
!non-standard extension
call sleep(1)
if (i==0) exit
end do
end
MAIN__ () at mpi_gdb.f90:26
26 if (i==0) exit
(gdb) info locals
hostname = 'linux.site', ' ' <repeats 246 times>
hostname_len = 10
i = 1
ie = 0
pid = 2457
rank = 0
(gdb) set var i = 0
(gdb) cont
Continuing.
[Inferior 1 (process 2355) exited normally]
> mpif90 prog.f90 -O0 -g -o prog.exe
Rank 0 has escaped!
如果我现在在本地机器的两个核心上使用
> mpirun -np 2 ./prog.exe
在屏幕上我看到的只是
rank = 0
rank = 1
现在在另一个终端中，我连接到相关的机器，并使用
ps -ef | grep prog.exe
这给了我几个对应于不同等级的进程id值
gdb --pid <pidFromPSCmd> ./prog.exe
在我们最初的终端，我们会看到
gdb -pid 12345
use mpi
implicit none
character(MPI_MAX_PROCESSOR_NAME) :: hostname
integer :: rank, ie, pid, hostname_len
integer, volatile :: i
call MPI_Init(ie)
call MPI_Get_processor_name(hostname, hostname_len, ie)
!non-standard extension
pid = getpid()
call MPI_Comm_rank(MPI_COMM_WORLD, rank, ie)
write(*,*) "PID ", pid, " on ", trim(hostname), " ready for attach is world rank ", rank
!this serves to block the execution at a specific place until you unblock it in GDB by setting i=0
i = 1
do
!non-standard extension
call sleep(1)
if (i==0) exit
end do
end
MAIN__ () at mpi_gdb.f90:26
26 if (i==0) exit
(gdb) info locals
hostname = 'linux.site', ' ' <repeats 246 times>
hostname_len = 10
i = 1
ie = 0
pid = 2457
rank = 0
(gdb) set var i = 0
(gdb) cont
Continuing.
[Inferior 1 (process 2355) exited normally]
> mpif90 prog.f90 -O0 -g -o prog.exe
Rank 0 has escaped!
目前程序没有挂起；我只是遇到了一个分段错误的问题，所以当我启动可执行文件时，执行会以一个错误终止。那么我该怎么做呢？gdb-pid？你从_getpid
得到的_pid？。程序打印的值或者你在top
中看到的值gelis，但我从未发现GDB对于segfaults如此常见。我使用编译器检查-fcheck=all
或清理。或者只在正确的位置打印变量。如果您有回溯，但您应该能够在没有GDB的情况下通过类似（-fbacktrace
或-traceback
）的方式获得一个.getpid？如果“没有上面的任何代码”呢？可能我遗漏了什么。在SEGFULT之后，执行被终止，我想没有什么可以看的了。这是我希望遵循一些变量的方式，以便了解错误（可能我太沉迷于MATLAB的调试）.打印变量？我认为这是一种肮脏的方式。但如果你这样做，也许没那么糟糕！@EnricoMariaDeAngelis我的意思是使用答案中的代码，打印PID！并阅读如何使用它。另一个答案也将帮助你理解其用法。未绑定循环正好放在你需要停止代码并附加调试器的地方。您必须知道要检查代码的哪一部分。在top
中，您只需使用