LGTM, with the changes you've identified and with a couple inline comments addressed. Thanks!
REPOSITORY rL LLVM ================ Comment at: lib/Headers/altivec.h:797 @@ -749,1 +796,3 @@ +} + static vector signed long long __ATTRS_o_ai ---------------- Just FYI -- the above three are listed in the test case I gave you as "low priority." It's fine that you added them, but for similarly marked ones please defer them until after unmarked ones are complete. (We are not exactly deprecating these, but they make little sense and we will be listing them in a separate table in the vector API -- "not recommended" or something. There is an overuse of vector bool, mixing it with other types where it doesn't make much sense.) ================ Comment at: test/CodeGen/builtins-ppc-p8vector.c:4 @@ -3,3 +3,3 @@ // RUN: %clang_cc1 -faltivec -target-feature +power8-vector -triple powerpc64le-unknown-unknown -emit-llvm %s -o - | FileCheck %s -check-prefix=CHECK-LE -// RUN: not %clang_cc1 -faltivec -triple powerpc64-unknown-unknown -emit-llvm %s -o - 2>&1 | FileCheck %s -check-prefix=CHECK-PPC +// RUN: not %clang_cc1 -faltivec -target-feature +vsx -triple powerpc64-unknown-unknown -emit-llvm %s -o - 2>&1 | FileCheck %s -check-prefix=CHECK-PPC ---------------- nemanjai wrote: > Added -target-feature +vsx to avoid errors about "vector double" and to > generate the correct errors for functions that are only overloaded with > __VSX__ (vec_cmpge, vec_cmple). This should be a comment in the test, not just in the review. http://reviews.llvm.org/D10637 EMAIL PREFERENCES http://reviews.llvm.org/settings/panel/emailpreferences/ _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
