[PATCH] D54862: [OpenCL] Add generic AS to 'this' pointer

[John McCall via Phabricator via cfe-commits]

rjmccall added a comment.

In D54862#1328290 <https://reviews.llvm.org/D54862#1328290>, @mikael wrote:

> Seems like my this commit broke: http://lab.llvm.org:8011/builders/lldb-x86_64-ubuntu-14.04-buildserver/builds/33176
>
> But since I don't really know what anything about lldb, I probably won't be able to fix it fast enough.
>
>   /lldb-buildbot/buildServerSlave/lldb-android-buildserver/llvm/tools/lldb/source/Symbol/ClangASTContext.cpp:2208:24: error: no viable overloaded '='
>     proto_info.TypeQuals = type_quals;
>     ~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~
>   /lldb-buildbot/buildServerSlave/lldb-android-buildserver/llvm/tools/clang/include/clang/AST/Type.h:141:7: note: candidate function (the implicit copy assignment operator) not viable: no known conversion from 'unsigned int' to 'const clang::Qualifiers' for 1st argument
>   class Qualifiers {
>         ^
>   /lldb-buildbot/buildServerSlave/lldb-android-buildserver/llvm/tools/clang/include/clang/AST/Type.h:141:7: note: candidate function (the implicit move assignment operator) not viable: no known conversion from 'unsigned int' to 'clang::Qualifiers' for 1st argument
>   1 error generated.
>
>
> I think it can be easily solved by Qualifiers::fromFastMask(type_quals); also updating the variable name since it changed.


I think that's exactly what you need to do, and you'd be welcome to commit that instead of reverting this.



================
Comment at: lib/Index/USRGeneration.cpp:274
+    if (unsigned quals = MD->getTypeQualifiers().getCVRUQualifiers())
       Out << (char)('0' + quals);
     switch (MD->getRefQualifier()) {
----------------
akyrtzi wrote:
> rjmccall wrote:
> > akyrtzi wrote:
> > > rjmccall wrote:
> > > > Paging @akyrtzi here.  The address-space qualifier should be part of the USR but I don't think if there's a defined schema for that.
> > > If I understand correctly from other comments you can't add special mangling and USR generation yet and intend to add FIXME for doing mangling support later ? If yes, could you also add FIXME here and re-visit ?
> > While you're here, can you described the right way for us to extend USR generation here to potentially add an address space?
> I'm not familiar with the mechanism, is address space supposed to be identified via a single integer ? If so, then I think adding '-' character plus the integer as character for the address space, immediately after the type qualifiers, should be a way to go.
Okay.  Some of them are integers; others are identified in the language as more of a keyword and so there's no stable mapping from an integer value.  Should we just come up with some simple schema for that following `-` that's unambiguous with a digit?


Repository:
  rC Clang

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D54862/new/

https://reviews.llvm.org/D54862