From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Cagney To: Daniel Jacobowitz Cc: gdb-patches@sources.redhat.com Subject: Re: [rfa/mi testsuite] Handle missing inferior I/O for mi-syn-frame.exp Date: Mon, 10 May 2004 18:00:00 -0000 Message-id: <409FC352.50807@gnu.org> References: <20040228182815.GA17243@nevyn.them.org> <406DE30A.3000904@gnu.org> <20040508191439.GA20933@nevyn.them.org> X-SW-Source: 2004-05/msg00292.html Daniel wrote: >The output patterns for this test include things printed by the inferior >program. Obviously, for gdbserver, this output comes out in the wrong >place. This patch allows it to be missing for noinferiorio; I don't think >this compromises the value of the test. OK? Andrew wrote: Given that testing I/O isn't the objective of this test, just removing the output looks to be a better option. Daniel wrote: +dummy_puts (const char *string) +{ + ; +} Er, just _remove_ the output. Done like so. OK? Tested, no regressions on i386-pc-linux-gnu, fixes 8 failures using gdbserver to i386-pc-linux-gnu. >BTW, we have one test that "verifies" (kfail'd) that output is properly >MI-encapsulated, and this test unintentionally "verifies" that it isn't. >Should I just remove recognizing the inferior's printouts unconditionally, >instead? Can you rephrase this? Some test verifies that the output from the inferior is MI encapsulated - I believe it was mi-console.exp, and it's now KFAILed. If we changed MI to successfully encapsulate this sort of output, that would fix mi-console.exp but break tests like this one. That's all I meant. Good, you were agreeing that the output was irrelevant. Andrew