On 12 February 2008 19:15, Hans-Peter Nilsson wrote: >> From: "Dave Korn" <[EMAIL PROTECTED]> >> Date: Tue, 12 Feb 2008 18:52:27 -0000 > >> On 12 February 2008 18:19, Hans-Peter Nilsson wrote: > Dave wrote: >>>> # The simulator isn't really remote. >>>> set_board_info isremote 0 >>>> >>>> I think you'll need that as well. >>> >>> Why? >> >> Because the simulator is not remote and does not need to be accessed by >> rsh/telnet/whatever? > > I don't think that's what the isremote means. To me it seems to > mean "can execute on the local host just like any program".
"isremote" means that? Or "isremote 0" (which to me means "!isremote") means that? > By this definition a simulator is remote. I really don't understand this statement; ISTM that a simulator "can execute on the local host just like any program". >>> No, don't do that; don't add random things for no known reason. >> >> But if the simulator really *is* not remote, wouldn't /not/ setting it be >> just plain wrong? > > No; consider that no other sim testboard file sets isremote 0. That's a reasonable point, but I wouldn't know how to draw any inferences from it without further study. >>> If you "need it", it's because something else is wrong and you're >>> papering over it. >> >> Well, if you don't set it, there are a whole bunch of "if [is_remote >> host] {" clauses in the gcc testsuite .exp scripts that are going to be >> activated and attempt to do such things as upload test files or download >> target headers or gcov results from the hypothetical remote board. > > We're talking about "[is_remote target]". Oh, yes, I see that now. Well, in that case, "isremote" is AFAICT not used *anywhere* except in a line of verbose output, where its value is reported. And I want my output to claim that the simulator is not remote. I'll try a testrun with and without it and see if it makes any difference. cheers, DaveK -- Can't think of a witty .sigline today.... _______________________________________________ DejaGnu mailing list DejaGnu@gnu.org http://lists.gnu.org/mailman/listinfo/dejagnu