JDevlieghere added inline comments.
================
Comment at: source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp:430
const size_t num_ranges =
- die->GetAttributeAddressRanges(dwarf, this, ranges, false);
+ die->GetAttributeAddressRanges(dwarf, this, ranges, true);
if (num_ranges > 0) {
----------------
Please give this bool a name, either by assigning it to a named variable or by
prefixing it with a comment like `/* check_hi_lo_pc */`. I think the LLDB code
base prefer the former?
================
Comment at: source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp:786
+ assert(debug_map == m_debug_map_symfile &&
+ "It's a nested instance derived from SymbolFile, "
+ "NOT SymbolFileDWARF");
----------------
What is "it" here?
================
Comment at: source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp:800
+ // FIXME: This must never happen!
+ assert(false && "Need more info to find the correct CU");
+ cu_sp = debug_map->GetCompileUnit(this);
----------------
You can use `llvm_unreachable` here.
================
Comment at: source/Plugins/SymbolFile/DWARF/SymbolFileDWARFDebugMap.cpp:1338
+ dw_offset_t cu_offset) {
+ // TODO: Can we assume partial order in offsets and bsearch?
+ const uint32_t cu_count = GetNumCompileUnits();
----------------
Yes, otherwise we have invalid DWARF, which I don't think it's worth supporting
something like that here.
https://reviews.llvm.org/D52375
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
