Building X86-reduce engine
(Note: at the moment, this X86-reduce option is only available in the next branch of our Github repo. It will be integrated into the next release of Capstone)
This documentation introduces how to build the X86 engine of Capstone to be as small as 200KB - about 60% smaller than the diet engine - for embedding purpose.
Later part presents the APIs related to this reduced mode.
1. Building a small engine
Typically, we use Capstone for usual applications, where the library weight does not really matter. Indeed, as of version 2.1-RC1, the whole engine is only 1.9 MB including all architectures, and this size raises no issue to most people.
However, to embed Capstone into special environments, such as OS kernel driver or firmware, the engine size should be as small as possible due to space restriction. To achieve this object, we must compile Capstone using special methods.
To build a tiny engine, consult three documentations below.
For X86 architecture, after applying all of the above techniques, the binary size of Capstone reduces to around 486KB. If you still desire a smaller engine, Capstone has another compile time option called X86-reduce.
2. Building X86-reduce engine
To reduce the X86 engine even further, compile Capstone in X86-reduce mode to remove some exotic non-critical X86 instruction sets. As a result, this downsizes the engine by around 60%, to under 200KB.
Below is the list of instruction sets removed by this option:
-
Floating Point Unit (FPU)
-
MultiMedia eXtension (MMX)
-
Streaming SIMD Extensions (SSE)
-
3DNow
-
Advanced Vector Extensions (AVX)
-
Fused Multiply Add Operations (FMA)
-
eXtended Operations (XOP)
-
Transactional Synchronization Extensions (TSX)
Obviously, the price to pay for this tiny size is that the engine can no longer understand the removed instructions. But in special environments such as OS kernel, where these instructions are never used, this is acceptable.
By default, Capstone for X86 is built with complete instructions. To build and install the X86-reduce engine, do: (demonstration is on *nix systems)
3. Checking X86 engine for “reduce” status
Capstone allows us to check if the engine was compiled in X86-reduce mode with cs_support() API, as follows - sample code in C.
With Python, we can either check the X86-reduce mode via the function cs_support of capstone module, as follows.
Or we can also use the x86_reduce getter of Cs class for the same purpose, as follows.