Bah, I missed your post and put up a poll at almost the same time to cfe-dev on
this issue.
Oops :o)
From: Richard Smith
Sent: Monday, July 09, 2012 11:08 PM
To: Jordan Rose
Cc: Andy Gibbs ; cfe-commits cfe
Subject: Re: [cfe-commits] [Patch 4 of 7] -verify fixes and enhancement
On Mon, Jul 9, 2012 at 12:01 PM, Jordan Rose <[email protected]> wrote:
On Jul 4, 2012, at 8:41 , Andy Gibbs <[email protected]> wrote:
> So, in conclusion, we have the choice now: use the above patch and have
fatal
> errors shown but all other diagnostics suppressed, or I can fix the patch
from
> last time with your comment above and we can use -verify to even check fatal
> errors.
>
> I am personally happy with either route. I can see advantages both ways.
I prefer this approach, but I'm hoping someone else can comment here. It's
something that really affects the behavior of -verify.
-verify already works when checking fatal errors, and is currently used for
that by several tests. What doesn't work currently is printing out the -verify
summary after a fatal error occurs, if there are mismatches between
expectations and diagnostics. So let's just fix that.
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
