These numbers are measured on SPEC 2006 (C/C++ only) using Clang3.3 (trunk) r179094 (on Google Code) (April 09 2013) on Intel Xeon W3690 @3.47GHz. 2-nd column: clang -O2
3-rd column: clang -O2 -fsanitize=address -fno-omit-frame-pointer
The average slowdown is 1.93, the median slowdown is 1.86.
Older measurements: asan r496 (on Google Code) (clang -O2 -fsanitize=address
) with malloc_context_size=1
(i.e. no unwinding). The baseline is clang -O2
. The average slowdown introduced by full instrumentation (red) is 1.73. When only writes are instrumented (blue), the average slowdown is 1.26.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4