Skip argv0-symlink.exp on target argv[0] isn't available

I see the following two fails on arm-none-eabi target, because argv[0]
isn't available.

print argv[0]^M
$1 = 0x1f78 "/dev/null"^M
(gdb) FAIL: gdb.base/argv0-symlink.exp: kept file symbolic link name

print argv[0]^M
$1 = 0x1f78 "/dev/null"^M
(gdb) FAIL: gdb.base/argv0-symlink.exp: kept directory symbolic link name

My first thought is to check [target_info exists noargs], and skip the
test if it returns true.  However, noargs is set in gdbserver board
files, so argv0-symlink.exp will be skipped on gdbserver board file.
The change is too aggressive.

When the program is running with gdbserver, argv[1] to argv[N] aren't
available, but argv[0] is.  Fortunately, argv0-symlink.exp only
requires argv[0].  argv0-symlink.exp can be run with gdbserver board
file, as what we do now.

What we need to check is whether argv[0] is available, so I add a new
proc gdb_has_argv0 to do so by starting a program, and check
argc/argv[0] to see whether argv[0] is available.

Dan fixed the similar problem by checking noargs, which is too strong.
https://sourceware.org/ml/gdb-patches/2010-02/msg00398.html as a
result, the test is skipped on gdbserver.  This patch fixed it too.

Re-run argv0-symlink.exp, scm-value.exp, and py-value.exp on x86-linux
(both native and gdbserver) and arm-none-eabi.  No fail on x86-linux,
and argv0-symlink.exp is skipped on arm-none-eabi target.

gdb/testsuite:

2014-10-16  Yao Qi  <yao@codesourcery.com>

	* gdb.base/argv0-symlink.exp: Skip it if gdb_has_argv0 return
	false.
	* gdb.guile/scm-value.exp (test_value_in_inferior): Don't
	check [target_info exists noargs], check [gdb_has_argv0]
	instead.
	* gdb.python/py-value.exp (test_value_in_inferior): Likewise.
	* lib/gdb.exp (gdb_has_argv0, gdb_has_argv0_1): New
	procedures.
---
 gdb/testsuite/gdb.base/argv0-symlink.exp |  5 ++
 gdb/testsuite/gdb.guile/scm-value.exp    |  5 +-
 gdb/testsuite/gdb.python/py-value.exp    |  5 +-
 gdb/testsuite/lib/gdb.exp                | 79 ++++++++++++++++++++++++++++++++
 4 files changed, 92 insertions(+), 2 deletions(-)

On 10/16/2014 07:25 AM, Yao Qi wrote:

> 
> When the program is running with gdbserver, argv[1] to argv[N] aren't
> available, but argv[0] is.  Fortunately, argv0-symlink.exp only
> requires argv[0].  argv0-symlink.exp can be run with gdbserver board
> file, as what we do now.
> 
> What we need to check is whether argv[0] is available, so I add a new
> proc gdb_has_argv0 to do so by starting a program, and check
> argc/argv[0] to see whether argv[0] is available.

Sounds good.

> --- a/gdb/testsuite/gdb.base/argv0-symlink.exp
> +++ b/gdb/testsuite/gdb.base/argv0-symlink.exp
> @@ -13,6 +13,11 @@
>  # You should have received a copy of the GNU General Public License
>  # along with this program.  If not, see <http://www.gnu.org/licenses/>.
>  
> +if { ![gdb_has_argv0] } {
> +    unsupported "Can't get argv\[0\]."
> +    return
> +}
> +

I think we shouldn't skip the whole test though; only the
printing argv[0] test.  The test file also makes sure
that "info inferiors" shows the symlink name, not the target
of the symlink, and that is host-dependent, not target
dependent.  See git 4856b6bc.

>  
>    # Check that the dereferenced value is sane
> -  if { ! [target_info exists noargs] } {
> +  global has_argv0
> +  if { $has_argv0 } {
>      gdb_test "python print (arg0)" "0x.*$testfile\"" "verify dereferenced value"
>    }
>  
> diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
> index 7b2a402..dd525dc 100644
> --- a/gdb/testsuite/lib/gdb.exp
> +++ b/gdb/testsuite/lib/gdb.exp
> @@ -4227,6 +4227,85 @@ gdb_caching_proc gdb_skip_xml_test {
>      return $xml_missing
>  }
>  
> +# Return true if argv[0] is available.
> +
> +gdb_caching_proc gdb_has_argv0 {
> +    set result 0
> +
> +    # Set up, compile, and execute a test program to check whether
> +    # argv[0] is available.
> +    set src [standard_temp_file has_argv0[pid].c]
> +    set exe [standard_temp_file has_argv0[pid].x]
> +
> +    gdb_produce_source $src {
> +	int main (int argc, char **argv) {
> +	    return 0;
> +	}
> +    }
> +
> +    gdb_compile $src $exe executable {debug}
> +
> +    # Helper proc.
> +    proc gdb_has_argv0_1 { exe } {
> +	global srcdir subdir
> +	global gdb_prompt hex decimal
> +
> +	gdb_exit
> +	gdb_start
> +	gdb_reinitialize_dir $srcdir/$subdir
> +	gdb_load "$exe"
> +
> +	# Set breakpoint on main.
> +	send_gdb "break main\n"
> +	gdb_expect {
> +	    -re "Breakpoint.*${gdb_prompt} $" {
> +	    }
> +	    -re "${gdb_prompt} $" {
> +		return 0
> +	    }
> +	}
> +
> +	# Run to main.
> +	gdb_run_cmd
> +	gdb_expect {
> +	    -re "Breakpoint.*${gdb_prompt} $" {
> +	    }
> +	    -re "${gdb_prompt} $" {
> +		return 0
> +	    }
> +	}
> +
> +	# Check whether argc is 1.
> +	send_gdb "p argc\n"
> +	gdb_expect {
> +	    -re " = 1\r\n${gdb_prompt} $" {
> +
> +		send_gdb "p argv\[0\]\n"
> +		gdb_expect {
> +		    -re " = $hex \".*$exe\"\r\n${gdb_prompt} $" {

I suspect this may break if remote (host|target) testing,
and not sharing the filesystem between build/host/target.
Isn't $exe here a full path on the build?

> +			return 1
> +		    }
> +		    -re "${gdb_prompt} $" {
> +			return 0
> +		    }
> +		}
> +	    }
> +	    -re "${gdb_prompt} $" {
> +		return 0
> +	    }
> +	}
> +	return 0
> +    }
> +

I think these gdb_expect's should be gdb_test_multiple's instead.

Thanks,
Pedro Alves