From e15fafb1d0bed1c832e61ec32ef7efb9fe4cca36 Mon Sep 17 00:00:00 2001 From: Martin Monperrus Date: Fri, 21 Aug 2020 12:29:54 +0000 Subject: [PATCH] doc: add reference doc for AFL --- wasm-fuzzer/README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/wasm-fuzzer/README.md b/wasm-fuzzer/README.md index ab1aff35..47dbdb9d 100644 --- a/wasm-fuzzer/README.md +++ b/wasm-fuzzer/README.md @@ -21,6 +21,8 @@ Right now, we support fuzzing of four data types as function parameter: * float32 * float 64 +Reference documentation in (see part 1, Coverage Measurements): https://github.com/google/AFL/blob/master/docs/technical_details.txt + ## Parsing AFL's fuzzed inputs One major difference between C++ and WASM is that WASM only works with Int32/int, Int64/long, Float32/float and Float64/double. This matters, since AFL most commonly takes a config file as an input parameter, which is then first parsed from chars/strings to the corresponding type by the C++ code. Since a WASM function cannot simply receive a char or string and parse it to the int/long/float/double it needs, we need to do this manually beforehand.