On Mon, Mar 20, 2017 at 11:19:37AM +0000, Andrew Cooper wrote: > This series is a couple of bugfixes, but is mainly a set of performance > improvements to using the AFL test harness. This series increases the > performance as reported by afl-fuzz from 6k execs/s to 38k execs/s, or more > than 6 times faster. > > Andrew Cooper (8): > tools/fuzz: Remove .d files in clean > tools/fuzz: Use $(CC) for linking the harnesses > tools/fuzz: Include LLVMFuzzerTestOneInput() in the generated .a > tools/insn-fuzz: Use shorter filenames > tools/insn-fuzz: Use getopt() for parsing the command line > tools/insn-fuzz: Accept fuzzing input on stdin > tools/insn-fuzz: Make use of LLVMFuzzerInitialize() > tools/insn-fuzz: Support AFL's afl-clang-fast mode
Reviewed-by: Wei Liu <wei.l...@citrix.com> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
