[Lldb-commits] [PATCH] Fix type detection for 'char' when it is unsigned

Tamas Berghammer tberghammer at google.com
Thu Mar 26 14:55:22 PDT 2015


What is your opinion about setting up CharTy based on the target triple with the same logic what clang use in Driver/Tools.cpp::isSignedCharDefault and then when we hit a case where the type of the char in the dwarf isn't match with our setting then we fall back to the fully qualified "signed char" or "unsigned char" based on the information in the dwarf.

This solution will work perfectly when the inferior use the same char type as the default for the architecture and fall back to an other type what is at least as good as the current implementation if the inferior have non-default char type.


http://reviews.llvm.org/D8636

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/






More information about the lldb-commits mailing list