Re: [RFA] Add shared object relocation tests

[Raoul Gough <RaoulGough at yahoo dot co dot uk>]

Daniel Jacobowitz <drow@mvista.com> writes:

> On Tue, Jun 03, 2003 at 11:14:44AM +0100, Raoul Gough wrote:
[snip]
>> Should I go ahead and add the new files?
>
> First of all, even new files get a changelog entry.  It will go in
> gdb/testeuite/ChangeLog.

Thanks for pointing this out - I'd forgotten this entirely. The
src/gdb/testsuite/ChangeLog entry would be:

2003-06-03  Raoul Gough  <RaoulGough@yahoo.co.uk>

	* gdb.base/shreloc.exp: New file, check symbols obtained from shared
        objects that required relocation at load time (gdb PR/1132).
	* gdb.base/shreloc.c, gdb.base/shreloc1.c, gdb.base/shreloc2.c:
        as above, part of the shared object relocation test.

>
>> foreach module [list "shreloc" "shreloc1" "shreloc2"] {
>>     if {[gdb_compile "${srcdir}/${subdir}/${module}.c" "${workdir}/${module}.o" object {debug}] != ""} {
>>  gdb_suppress_entire_file "${module}.c compile failed, so all tests in this file will automatically fail."
>>  return -1
>>     }
>> }

[snip similar cases]
>
> Please just use return, not gdb_suppress_entire_file.  In particular,
> using both causes the _next_ test to fail, I think.

Hmmm.... gdb_suppress_entire_file seems to be the usual response to
testcase compilation failures, so it would probably make more sense to
remove the return statement instead. I don't really know what the pros
and cons are in this case - is there some reason to prefer the return
option (what about logging the error in that case?)

>
> Other than those two details, this is OK.  Give other people a day or
> so to comment and then check it in.

Thanks for the feedback, and I'll hold off for a couple of days on the
update.

-- 
Raoul Gough
"Let there be one measure for wine throughout our kingdom, and one
measure for ale, and one measure for corn" - Magna Carta