unable to debug malloc_shared with gdb

[guoyejun]

Hi,

I'm trying the hello world code of DPCPP book, and I tried the code at https://github.com/Apress/data-parallel-CPP/blob/main/samples/Ch01_intro/fig_1_1_hello.cpp, but met crash when debug the program with gdb, see below. I also tried with .../intel/oneapi/debugger/10.1.2/gdb/intel64/bin/gdb-oneapi, the same issue.

$ dpcpp -g fig_1_1_hello.cpp
$ gdb ./a.out
GNU gdb (Ubuntu 9.1-0ubuntu1) 9.1
Copyright (C) 2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
    <http://www.gnu.org/software/gdb/documentation/>.

For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./a.out...
(gdb) r
Starting program: /home/yjguo/work/ooo/dpcpp/a.out
warning: File "/home/yjguo/intel/oneapi/compiler/2021.3.0/linux/lib/libsycl.so.5.2.0-gdb.py" auto-loading has been declined by your `auto-load safe-path' set to "$debugdir:$datadir/auto-load".
To enable execution of this file add
        add-auto-load-safe-path /home/yjguo/intel/oneapi/compiler/2021.3.0/linux/lib/libsycl.so.5.2.0-gdb.py
line to your configuration file "/home/yjguo/.gdbinit".
To completely disable this security protection add
        set auto-load safe-path /
line to your configuration file "/home/yjguo/.gdbinit".
For more information about this security protection see the
"Auto-loading safe path" section in the GDB manual.  E.g., run from the shell:
        info "(gdb)Auto-loading safe path"
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7ffff31e3700 (LWP 64327)]
[New Thread 0x7ffff1bd9700 (LWP 64328)]

Thread 1 "a.out" received signal SIGSEGV, Segmentation fault.
__strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:65
65      ../sysdeps/x86_64/multiarch/strlen-avx2.S: No such file or directory.
(gdb) bt
#0  __strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:65
#1  0x00007ffff7f02b1e in std::basic_ostream<char, std::char_traits<char> >& std::operator<< <std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&, char const*) () from /lib/x86_64-linux-gnu/libstdc++.so.6
#2  0x00000000004037ef in main () at fig_1_1_hello.cpp:25
(gdb) f 2
#2  0x00000000004037ef in main () at fig_1_1_hello.cpp:25
25        std::cout << result << "\n";
(gdb) p result
$1 = 0x7ffff0c00000 "Ifmmp-!xpsme\"\nJ(n!tpssz-!Ebwf/!J(n!bgsbje!J!dbo(u!ep!uibu/!.!IBM\001"
(gdb)

btw, Intel gpu is selected by the queue as default.

[alexbatashev]

@guoyejun this sounds like a bug to me, so I transferred this discussion to issues: #4331

[abijaz]

For the Seg fault issue Applications that use unified shared memory (USM) may appear as raising a SIGSEGV when a USM-allocated memory is being accessed. This is a mechanism used by the runtime to trigger memory migration. In such cases, sending the signal back to the application resumes the program. For this, use GDB's signal SIGSEGV command which should resume the program.
Also in case of issue please provide gdb-oneapi logs.

[guoyejun]

could you let me know the detail for 'GDB's signal SIGSEGV command'? thanks.

[abijaz]

as mentioned above it is "signal SIGSEGV"

[guoyejun]

I tried below, anything I missed? thanks.

$ gdb ./a.out
(gdb) signal SIGSEGV
The program is not being run.

(gdb) b main
Breakpoint 1 at 0x4036ad: file fig_1_1_hello.cpp, line 16.

(gdb) r
...
Breakpoint 1, main () at fig_1_1_hello.cpp:16
16        queue Q;

(gdb) signal SIGSEGV
Continuing with signal SIGSEGV.

Program terminated with signal SIGSEGV, Segmentation fault.
The program no longer exists.

(gdb) c
The program is not being run.

(gdb) r
Starting program: /tmp/dpcpp/a.out
warning: File "/home/gta/compiler1123/linux_prod/compiler/linux/lib/libsycl.so.5.4.0-gdb.py" auto-loading has been declined by your `auto-load safe-path' set to "$debugdir:$datadir/auto-load".
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".

Breakpoint 1, main () at fig_1_1_hello.cpp:16
16        queue Q;

(gdb) c
Continuing.
[New Thread 0x7fffe688f700 (LWP 33181)]

Thread 1 "a.out" received signal SIGSEGV, Segmentation fault.
__strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:65
65      ../sysdeps/x86_64/multiarch/strlen-avx2.S: No such file or directory.

(gdb) bt
#0  __strlen_avx2 () at ../sysdeps/x86_64/multiarch/strlen-avx2.S:65
#1  0x00007ffff7f00bce in std::basic_ostream<char, std::char_traits<char> >& std::operator<< <std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char> >&, char const*) ()
   from /lib/x86_64-linux-gnu/libstdc++.so.6
#2  0x00000000004037df in main () at fig_1_1_hello.cpp:25

(gdb) f 2
#2  0x00000000004037df in main () at fig_1_1_hello.cpp:25
25        std::cout << result << "\n";

[abijaz]

above where you give (gdb) bt before this can you please signal command

[guoyejun]

thanks!