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 3/3] Add a test case for the jit-reader interface.


On 10/17/2012 05:27 PM, Tom Tromey wrote:
>>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
> 
> Pedro> This means that shlib tests are never skipped on native, but
> Pedro> it doesn't prevent it from running elsewhere.
> 
> Oh, duh.
> Sanjoy, your test has to check isnative.
> 
> Pedro> Should the test be native-only?  If so, why?  Doesn't the jit
> Pedro> interface work when debugging with remote targets (e.g.,
> Pedro> gdbserver)?
> 
> It does, but this tests the 'jit-reader-load' functionality, which means
> it must build a .so that gdb itself loads.  'isnative' isn't maybe
> exactly the right test -- you might have a build-x-host toolchain -- but
> it seems simplest.

Ah, I see.  We're talking about different "native's".  Dejagnu's isnative
only compares host vs target triplets.  Indeed that sounds simplest.

Bailing if !isnative will still run the tests if testing with a gdbserver
of the triplet as the host (of which native gdbserver testing is a special
case), so I'm happy with that.

(We should still add a gdb_load_shlibs call.)

-- 
Pedro Alves


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