Re: [RFA] new test for separate debug info

[Elena Zannoni <ezannoni at redhat dot com>]

Michael Elizabeth Chastain writes:
 > Okay, I just ran it on spe191.testdrive.hp.com to see what happens.
 > What happens is:
 > 
 >   (gdb) info source
 >   Current source file is /house/chastain/gdb/s1/gdb/testsuite/gdb.base/sepdebug.c
 >   Located in /house/chastain/gdb/s1/gdb/testsuite/gdb.base/sepdebug.c
 >   Contains 150 lines.
 >   Source language is c.
 >   Compiled with unknown debugging format.

aarrrgh

 >   Does not include preprocessor macro info.
 >   (gdb) UNSUPPORTED: gdb.base/sepdebug.exp: no separate debug info handling
 >   testcase /house/chastain/gdb/s1/gdb/testsuite/gdb.base/sepdebug.exp completed in 0 seconds
 > 
 > This is with HP Ansi C B.11.11.28707.GP.
 > 
 > I'm happy with this outcome, although I am not happy that "info source"
 > prints "compiled with unknown debugging format".  I will investigate and
 > file a PR.
 > 
 > Also, the message does not happen because of "unknown debugging format".
 > It happens because I don't have gnu binutils in my $PATH, and the HP-UX
 > strip barfs on "--strip-debug".

Maybe the two unsupported messages should differ a bit more from each
other: "cannot produce separate debug info file" instead?

 > 
 > I haven't set up a test bed for gcc + binutils on this platform yet,
 > just the native tools.  I did try GNU strip from binutils 2.14 on
 > a hello-world program with "strip --strip-debug hello" and it didn't
 > complain.
 > 

OK, when you do let me know what other marvellous ways the test fails.

elena