Update perfparser with important performance optimization

We now build a sorted symtab for much faster lookups, yielding up to 10x performance improvements to before - esp. with large DSOs with many symbols. Fixes: #192
KDAB · Jul 1, 2020 · fc12536 · fc12536
1 parent 600b667
commit fc12536
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 1 deletion.
diff --git a/3rdparty/perfparser b/3rdparty/perfparser
diff --git a/3rdparty/perfparser.cmake b/3rdparty/perfparser.cmake
@@ -69,6 +69,7 @@ ecm_add_test(
 ecm_add_test(
     perfparser/tests/auto/addresscache/tst_addresscache.cpp
     perfparser/app/perfelfmap.cpp
+    perfparser/app/perfdwarfdiecache.cpp
     perfparser/app/perfaddresscache.cpp
     LINK_LIBRARIES
         Qt5::Core
+46 −13		app/perfaddresscache.cpp
+8 −3		app/perfaddresscache.h
+44 −8		app/perfsymboltable.cpp
+10 −26		tests/auto/addresscache/tst_addresscache.cpp
+215 −215		tests/auto/perfdata/fork_static_gcc/perf.data.zstd.expected.txt
+1 −1		tests/auto/perfdata/vector_static_clang/perf.data.expected.txt
+2 −2		tests/auto/perfdata/vector_static_gcc/perf.data.expected.txt
+1 −1		tests/auto/perfdata/vector_static_gcc/perf.data.zstd.expected.txt