[Bug dyninst/23605] New: Intermittent errors when running a simplest stap script against a simplest target program

[Bug dyninst/23605] New: Intermittent errors when running a simplest stap script against a simplest target program

[agentzh at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=23605

            Bug ID: 23605
           Summary: Intermittent errors when running a simplest stap
                    script against a simplest target program
           Product: systemtap
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: dyninst
          Assignee: systemtap at sourceware dot org
          Reporter: agentzh at gmail dot com
  Target Milestone: ---

When running a trivial stap script against a trivial C program process, I
occasionally see the following error:

```
$ stap --runtime=dyninst -c ./a.out a.stp
IRPC on terminated process, ret false!
--SERIOUS-- #124: `�3<�^?
stapdyn: ERROR: Couldn't load
/tmp/stapsBBcUd/stap_c7fd539785fa2c0bcc7ddfb6321bcd6f_1605.so into the target
process
WARNING: /opt/stap-plus/bin/stapdyn exited with status: 1
Pass 5: run failed.  [man error::pass5]
```

The a.stp file looks like this:

```
probe process.function("main") {
    println(@var("a"));
}
```

And the C program is like this:

```
int a = -32;

int main(void) { a++; return 0; }
```

The C program is compiled with `gcc -g`.

What could go wrong here please?

I'm not Fedora 27 x86_64 (kernel 4.16.16):

```
$ uname -r
4.16.16-200.fc27.x86_64
```

I'm using the current master branch of systemtap.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug dyninst/23605] Intermittent errors when running a simplest stap script against a simplest target program

[agentzh at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=23605

--- Comment #1 from agentzh <agentzh at gmail dot com> ---
I'm using the dyninst-devel package from Fedora 27's official yum repo:

```
$ dnf info dyninst-devel
Fedora 27 - x86_64 - Updates                                                   
                                                                               
                                                                               
                                              3.0 MB/s |  26 MB     00:08
Fedora 27 - x86_64                                                             
                                                                               
                                                                               
                                              6.2 MB/s |  58 MB     00:09
OpenResty Proprietary Repository                                               
                                                                               
                                                                               
                                               34 MB/s |  92 kB     00:00
Installed Packages
Name         : dyninst-devel
Version      : 9.3.2
Release      : 7.fc27
Arch         : x86_64
Size         : 1.2 M
Source       : dyninst-9.3.2-7.fc27.src.rpm
Repo         : @System
From repo    : updates
Summary      : Header files for the compiling programs with Dyninst
URL          : http://www.dyninst.org
License      : LGPLv2+
Description  : dyninst-devel includes the C header files that specify the
Dyninst user-space
             : libraries and interfaces. This is required for rebuilding any
program
             : that uses Dyninst.
```

Extra verbosity is a bit tricky since it is hard to reproduce and only
reproducible when running my test suite.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug dyninst/23605] Intermittent errors when running a simplest stap script against a simplest target program

[agentzh at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=23605

--- Comment #2 from agentzh <agentzh at gmail dot com> ---
Just got a different error for the exactly same example:

```
stapdyn:
/builddir/build/BUILD/dyninst-9.3.2/dyninst-9.3.2/dyninstAPI/src/emit-x86.C:1866:
virtual Register EmitterAMD64::emitCall(opCode, codeGen&, const
std::vector<boost::shared_ptr<AstNode> >&, bool, func_instance*): Assertion `0'
failed.
WARNING: /opt/stap-plus/bin/stapdyn exited with signal: 6 (Aborted)
Pass 5: run failed.  [man error::pass5]
```

Hope it helps.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug dyninst/23605] Intermittent errors when running a simplest stap script against a simplest target program

[agentzh at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=23605

agentzh <agentzh at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |agentzh at gmail dot com

--- Comment #3 from agentzh <agentzh at gmail dot com> ---
Created attachment 11229
  --> https://sourceware.org/bugzilla/attachment.cgi?id=11229&action=edit
stap -vvvv output for the assertion failure

It's not from the same test case, but a similar one.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug dyninst/23605] Intermittent errors when running a simplest stap script against a simplest target program

[scox at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=23605

Stan Cox <scox at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |scox at redhat dot com

--- Comment #4 from Stan Cox <scox at redhat dot com> ---
Have not seen that particular behavior but:
stap --runtime=dyninst -c ./a.out a.stp -vv --dyninst -k
will keep around the so file so it can be reproduced by running stapdyn
standalone, e.g.

% stap -e 'probe process("/usr/bin/eu-nm").statement("main@*:*")
{println(pp());}'  -c 'eu-nm' -vv --dyninst -k
% stapdyn -v -c eu-nm /tmp/stapy8u46n/stap_6292.so

-- 
You are receiving this mail because:
You are the assignee for the bug.