This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH 1/2] GDB test suite: Add helper for locating core files


Hi Maciej,

On 10/17/2017 07:19 PM, Maciej W. Rozycki wrote:
> On Tue, 17 Oct 2017, Pedro Alves wrote:

>> Wouldn't an UNTESTED or UNSUPPORTED be better?  It's what
>> we tend to do with other cases of unsupported/untested tests.
> 
>  I think UNSUPPORTED is the right status; UNTESTED is meant for missing 
> tests really; see:
> 
> UNTESTED
>     A test was not run.  This is a placeholder, used when there is no real 
>     test case yet.
> 
> vs:
> 
> UNSUPPORTED
>     There is no support for the tested case. This may mean that a 
>     conditional feature of an operating system, or of a compiler, is not 
>     implemented.  DejaGnu also uses this message when a testing 
>     environment (often a "bare board" target) lacks basic support for 
>     compiling or running the test case.  For example, a test for the 
>     system subroutine gethostname would never work on a target board 
>     running only a boot monitor.
> 

IMHO, this "placeholder" status of UNTESTED is pretty useless, at
least for GDB.  I don't recall ever adding such a placeholder
testcase, and I don't think GDB uses it like that.

I think a more useful distinction would be:

- UNSUPPORTED to indicate that the feature isn't supported by GDB or
  the remote stub.  E.g., in this case we'd use it if the GDB port
  does not support debugging core dumps at all.

- UNTESTED to indicate that the feature is supported by GDB but
  the test wasn't run because it's not possible to run it,
  or we choose to not run it, in the current test environment, e.g.,
  because of a board limitation.  In this case, we skip core tests
  when the host or target boards are remote, even though
  the feature is supported by GDB (even with "target remote" and
  cross debugging, at least on some ports).

We use UNTESTED in many places already meaning something like
the above (instead of placeholder status), though I'm not aware
of rationale written down anywhere.  The above is just my
intuition, and most of these untested/unsupported calls predate
my involvement with GDB.

Thanks,
Pedro Alves


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]