Hello,

Here is the test file for the patch described in
http://sources.redhat.com/ml/gdb-patches/2004-04/msg00463.html .

The patch fixes the case s2-da -- the source was compiled using absolute
name and is not available at that location any more; gdb should be able
to find it in the same directory relative to a directory mentioned in
the argument for the "directory" command (see the link above for a
typescript). This used to work in 5.3 but was broken when fixing the
case b4- -- the binary is specified as dot-slash-basename but does not
exist in the current directory; gdb should not look for it in PATH (see
http://sources.redhat.com/ml/gdb/2004-01/msg00244.html for an example).

I am not sure what we want to do in cases s1-da, s3-da, s4-da and s5-da
(the source was compiled using a non-absolute name; should gdb look for
them under <source path>/$cdir, where <source path> is one of the
arguments to the "directory" command?), so I describe the present
behaviour as FAIL for now.

The file was tested on gdb 6.2 under Debian unstable. I would welcome
any comments about the test and suggestions about the mentioned cases.

With kind regards,
Baurjan.