[PATCH] D43749: [Attr] Fix alloc_size's diags to report arg idx not value

[Joel E. Denny via Phabricator via cfe-commits]

jdenny added a comment.

In https://reviews.llvm.org/D43749#1018818, @aaron.ballman wrote:

> Aside from a minor testcase nit, this LGTM. Why is this dependent on https://reviews.llvm.org/D43248?


The dependency goes the other way.  Did I get it wrong?



================
Comment at: test/Sema/alloc-size.c:3
 
-void *fail1(int a) __attribute__((alloc_size)); //expected-error{{'alloc_size' attribute takes at least 1 argument}}
+void *fail1(int a) __attribute__((alloc_size));   //expected-error{{'alloc_size' attribute takes at least 1 argument}}
 void *fail2(int a) __attribute__((alloc_size())); //expected-error{{'alloc_size' attribute takes at least 1 argument}}
----------------
aaron.ballman wrote:
> This looks like a whitespace-only change?
clang-format-diff.py suggested that.  Should I drop that?


https://reviews.llvm.org/D43749