[cfe-commits] [patch] rewrite-includes crash support

[Eli Friedman]

On Tue, Jun 26, 2012 at 1:54 PM, David Blaikie <dblaikie at gmail.com> wrote:
> On Fri, Jun 22, 2012 at 4:50 PM, Chandler Carruth <chandlerc at google.com> wrote:
>> Doh, sorry for missing this last patch.
>>
>> Looks pretty good. As discussed in person, you may want to sink the arg
>> mutation below the PrintJob... Not sure.
>
> We also discussed the possibility of avoiding mutating the original
> args - did you/we end up dismissing that since we're mutating the jobs
> anyway, so there's no real immutability invariant that we'd be
> mantaining?
>
> As for moving it, if it's all the same to you, I think I'll just leave
> it where it is next to the point at which we raise the CCCIsCPP flag
> as they seem related. (I could move all of that down below PrintJob I
> suppose, though... )
>
>> I'm concerned by the lack of test case modifications. If we don't have a
>> test case yet for this, lets get one and check the filename. If we do, we
>> should tighten it up to check the filename or figure out whats going on.
>
> Agreed, but how exactly would I do that? The existing mechanism we
> have for crashing clang is "#pragma clang __debug crash" which crashes
> in the preprocessor. That means it crashes when crash recovery
> attempts to produce the preprocessed source file. As we discussed
> offline there's probably a couple of options here:
>
> 1) Transform the pragma into a special token then look for that token
> in the parser & crash there. But how do we do this without adding a
> test to a very hot part of the parser?

We don't necessarily need to crash anywhere we see the token in
question; we could say it's only "legal" in places where a top-level
declaration would be allowed.

-Eli