Re: [patch,7.3] Fix JIT clang-lli gdb-7.3 regression Re: [gdb-7.3] Error in gdb-llvm integration: Unable to read JIT descriptor from remote memory!

[Joel Brobecker <brobecker@adacore.com>]

> Another option is to apply a better fix (provided it is deemed safe) ...

Thanks for doing that :)

Your fix certainly looks a lot nicer than Jan's! After careful
reading, I think the patch makes sense, but having never looked
at this code before, it would be better if someone else approved.

Can Doug approve the patch, for instance? The patch is OK for 7.3
if OK for HEAD.

> 2011-07-05  Paul Pluzhnikov  <ppluzhnikov@google.com>
> 
> 	jit.c (jit_breakpoint_re_set_internal): Call jit_inferior_init.

Should we say that you're also adding jit_inferior_init's declaration?

Note that some of us (myself included) prefer moving functions
rather than having to write a declaration which duplicates the
amount of work to do when the function profile changes.  But
others disagree, and would rather use these declarations to allow
them to order the functions in a certain order of their liking...
I think that for the 7.3 branch we'll definitely want your version,
to keep the patch as small and simple as possible, but you're free
to choose whichever style you might prefer.

> testsuite/ChangeLog:
> 
> 2011-07-05  Paul Pluzhnikov  <ppluzhnikov@google.com>
> 
> 	* gdb.base/jit-so.exp: New test.
> 	* gdb.base/jit-dlmain.c: New file.
> 	* gdb.base/jit-main.c: Allow "main" to be elsewhere.

Cheers,
-- 
Joel