Sync from internal Google development.

Many improvements, too many to mention.  One significant
perf regression warrants investigation:

omitfp.parsetoproto2_googlemessage1.upb_jit: 343 -> 252 (-26.53)
plain.parsetoproto2_googlemessage1.upb_jit: 334 -> 251 (-24.85)

25% regression for this benchmark is bad, but since I don't think
there's any fundamental design issue that caused it I'm going to
go ahead with the commit anyway.  Can investigate and fix later.

Other benchmarks were neutral or showed slight improvement.

1 files changed, 12 insertions, 0 deletions

diff --git a/bindings/cpp/upb/pb/glue.hpp b/bindings/cpp/upb/pb/glue.hpp
index be072a7..d43baeb 100644
--- a/bindings/cpp/upb/pb/glue.hpp
+++ b/bindings/cpp/upb/pb/glue.hpp
@@ -13,11 +13,23 @@
 
 namespace upb {
 
+// All routines that load descriptors expect the descriptor to be a
+// FileDescriptorSet.
 bool LoadDescriptorFileIntoSymtab(SymbolTable* s, const char *fname,
                                   Status* status) {
   return upb_load_descriptor_file_into_symtab(s, fname, status);
 }
 
+bool LoadDescriptorIntoSymtab(SymbolTable* s, const char* str,
+                              size_t len, Status* status) {
+  return upb_load_descriptor_into_symtab(s, str, len, status);
+}
+
+template <typename T>
+bool LoadDescriptorIntoSymtab(SymbolTable* s, const T& desc, Status* status) {
+  return upb_load_descriptor_into_symtab(s, desc.c_str(), desc.size(), status);
+}
+
 }  // namespace upb
 
 #endif