Merge pull request BinaryAnalysisPlatform#61 from ivg/public-API

BAP Public API

resolves #4 and #11

.merlin

@@ -19,6 +19,7 @@ S lib/bap_types
 
 PKG core_kernel
 PKG ocamlbuild
+PKG jsonm
 EXT ounit
 EXT here
 EXT nonrec

.travis-ci.sh

@@ -1,20 +1,25 @@
-OPAM_DEPENDS="core_kernel oasis piqi zarith bitstring utop cmdliner"
-SYS_DEPENDS="libgmp-dev time llvm-3.4-dev"
+OPAM_DEPENDS="core_kernel.111.28.00 oasis zarith bitstring utop cmdliner faillib ezjsonm lwt-zmq uri.1.7.2 re cohttp.0.15.0"
+SYS_DEPENDS="libgmp-dev time llvm-3.4-dev libzmq3-dev aspcud"
 
 case "$OCAML_VERSION,$OPAM_VERSION" in
-4.02.0,1.2.0) ppa=avsm/ocaml42+opam12 ;;
-4.02.0,1.1.0) ppa=avsm/ocaml42+opam11 ;;
+4.02.1,1.2.0) ppa=avsm/ocaml42+opam12 ;;
+4.02.1,1.1.0) ppa=avsm/ocaml42+opam11 ;;
 4.01.0,1.2.0) ppa=avsm/ocaml41+opam12 ;;
 4.01.0,1.1.0) ppa=avsm/ocaml41+opam11 ;;
 *) echo Unknown $OCAML_VERSION,$OPAM_VERSION; exit 1 ;;
 esac
 
-
 install_on_linux () {
-  echo "yes" | sudo add-apt-repository ppa:$ppa
+  echo 'yes' | sudo add-apt-repository ppa:$ppa
+  echo 'yes' | sudo apt-add-repository ppa:chris-lea/zeromq
   sudo apt-get update -qq
   sudo apt-get install -qq ocaml ocaml-native-compilers camlp4-extra opam $SYS_DEPENDS
-  opam init
+
+  if [ $OPAM_VERSION = "1.1.0" ]; then
+      opam init default https://opam.ocaml.org/1.1
+  else
+      opam init
+  fi
 }
 
 install_on_osx () {
@@ -40,13 +45,13 @@ ocaml -version
 echo OPAM versions
 opam --version
 opam --git-version
+opam repository list
 
-opam init
 opam install ${OPAM_DEPENDS}
 eval `opam config env`
 
 oasis setup
-./configure --prefix=$(opam config var prefix) --enable-tests --enable-serialization --with-cxx=`which $CXX`
+./configure --prefix=$(opam config var prefix) --enable-tests --with-cxx=`which $CXX`
 make
 make install
 make test

.travis.yml

@@ -8,6 +8,6 @@ compiler:
 env:
   - OCAML_VERSION=4.01.0 OPAM_VERSION=1.1.0
   - OCAML_VERSION=4.01.0 OPAM_VERSION=1.2.0
-  - OCAML_VERSION=4.02.0 OPAM_VERSION=1.2.0 TEST_TARGETS=arm
-  - OCAML_VERSION=4.02.0 OPAM_VERSION=1.2.0 TEST_TARGETS=x86
-  - OCAML_VERSION=4.02.0 OPAM_VERSION=1.2.0 TEST_TARGETS=x86_64
+  - OCAML_VERSION=4.02.1 OPAM_VERSION=1.2.0 TEST_TARGETS=arm
+  - OCAML_VERSION=4.02.1 OPAM_VERSION=1.2.0 TEST_TARGETS=x86
+  - OCAML_VERSION=4.02.1 OPAM_VERSION=1.2.0 TEST_TARGETS=x86_64

_oasis

@@ -16,11 +16,9 @@ BuildDepends:
               comparelib.syntax,
               core_kernel,
               enumerate.syntax,
+              faillib.syntax,
               fieldslib.syntax,
               herelib.syntax,
-              herelib.syntax,
-              pa_bench.syntax,
-              pa_ounit.syntax,
               pa_ounit.syntax,
               sexplib.syntax,
               variantslib.syntax
@@ -53,13 +51,21 @@ Flag benchmarks
   Default: false
 
 Flag disassemblers
-  Description: Builds C++ interface to dissassemblers backend
+  Description: Build C++ interface to dissassemblers backend
+  Default: true
+
+Flag server
+  Description: Build BAP server
   Default: true
 
 Flag llvm
  Description: Build with llvm backend
  Default: true
 
+Flag server
+ Description: Build BAP server
+ Default: true
+
 Library bap
   Path:            lib/bap
   FindLibName:     bap
@@ -129,7 +135,7 @@ Library serialization
   BuildTools: ocamlbuild, piqi
   DataFiles:    *.piqi
   CompiledObject: best
-  BuildDepends: piqirun.pb, piqirun.ext, bap.types
+  BuildDepends: piqirun, bap.types
   Modules:   Bil_piqi,
              Stmt_piqi,
              Stmt_piqi_ext
@@ -250,6 +256,7 @@ Library benchmarks
   Install:        false
   Modules:        Bench_image
 
+
 Library dwarf_test
   Path:           lib_test/bap_dwarf
   Build$:         flag(tests)
@@ -266,6 +273,29 @@ Library disasm_test
   Install:        false
   Modules:        Test_disasm
 
+Library core_lwt
+  Path:           lwt
+  Build$:         flag(server)
+  CompiledObject: best
+  BuildDepends:   lwt, lwt.unix, lwt.log
+  Modules:        Core_lwt,
+                  Core_lwt_basic,
+                  Core_lwt_container,
+                  Core_lwt_container_intf,
+                  Core_lwt_extra,
+                  Core_lwt_or_error,
+                  Core_lwt_pool,
+                  Core_lwt_stream
+
+
+Executable "bap-server"
+  Path:           src/server
+  Build$:         flag(server)
+  CompiledObject: best
+  BuildDepends:   bap, lwt-zmq, ezjsonm, uri, cohttp.lwt, core_lwt
+  Install:        true
+  MainIs:         start_server.ml
+
 Executable readbin
   Path:           src/readbin
   MainIs:         readbin.ml

configure

@@ -22,5 +22,14 @@ for i in "$@"; do
 done
 
 [ -f setup.ml -a setup.ml -nt _oasis ] || oasis -quiet setup
+
+
 ocaml preconfig.ml
+
+case `ocaml -vnum` in
+    4.01.*)
+       sed -i 's/mark_tag_used/ignore/' myocamlbuild.ml
+    ;;
+esac
+
 ocaml setup.ml -configure "$@"

lib/bap_disasm/bap_disasm.ml

@@ -15,6 +15,7 @@ type reg = Reg.t with bin_io, compare, sexp
 type imm = Imm.t with bin_io, compare, sexp
 type fmm = Fmm.t with bin_io, compare, sexp
 type (+'a,+'k) insn
+type kind = Insn.Kind.t with bin_io, compare, sexp
 
 
 (** ARM instruction set  *)

lib/bap_disasm/bap_disasm_arm_env.ml

@@ -6,7 +6,7 @@ module Arm = Bap_disasm_arm
 let (%:) name typ = Var.create name typ
 
 
-let nil = Arm.Reg.to_string `nil %: reg32_t
+let nil = Arm.Reg.to_string `Nil %: reg32_t
 
 let make_register reg ty = Arm.Reg.to_string reg %: ty
 let reg32 reg = make_register reg reg32_t
@@ -80,7 +80,7 @@ let var_of_ccr : Arm.Reg.ccr -> var = function
   | `ITSTATE -> itstate
 
 let of_reg : Arm.Reg.t -> var = function
-  | `nil -> nil
+  | `Nil -> nil
   | #Arm.Reg.gpr as reg -> var_of_gpr reg
   | #Arm.Reg.ccr as reg -> var_of_ccr reg
 

lib/bap_disasm/bap_disasm_arm_lifter.ml

@@ -661,7 +661,7 @@ let lift_mem ops insn =
 
   (* POS_SIGN_BIT *)
   | `LDRSHTi, [|dest1; _unknown; base; Imm imm_off; cond; _|] ->
-    let offset = Mem_shift.mem_offset_reg_or_imm_pos (Op.Reg `nil) imm_off in
+    let offset = Mem_shift.mem_offset_reg_or_imm_pos (Op.Reg `Nil) imm_off in
     let insns =
       Mem_shift.lift_r_exp ~dest1 ~base ~offset PostIndex Signed H Ld
     in

lib/bap_disasm/bap_disasm_arm_types.ml

@@ -36,7 +36,7 @@ with bin_io, compare, sexp, enumerate
 
 
 module Reg = struct
-  type nil = [ `nil ]
+  type nil = [ `Nil ]
   with bin_io, compare, sexp, enumerate
 
   (** General purpose registers  *)

lib/bap_disasm/bap_disasm_basic.ml

@@ -14,7 +14,7 @@ type mem = Mem.t with sexp_of
 type kind = Kind.t with compare, sexp
 
 type pred = [
-  | `valid
+  | `Valid
   |  kind
 ] with sexp,compare
 
@@ -88,7 +88,7 @@ module Reg = struct
     let data =
       let reg_code = C.insn_op_reg_code dis.id ~insn ~oper in
       let reg_name =
-        if reg_code = 0 then lazy "nil"
+        if reg_code = 0 then lazy "Nil"
         else
           let off = C.insn_op_reg_name dis.id ~insn ~oper in
           lazy (Table.lookup dis.reg_table off) in
@@ -215,17 +215,17 @@ with bin_io, compare, sexp
 
 
 let cpred_of_pred : pred -> C.pred = function
-  | `valid -> C.Is_true
-  | `conditional_branch -> C.Is_conditional_branch
-  | `unconditional_branch -> C.Is_unconditional_branch
-  | `indirect_branch -> C.Is_indirect_branch
-  | `return -> C.Is_return
-  | `call -> C.Is_call
-  | `barrier -> C.Is_barrier
-  | `terminator -> C.Is_terminator
-  | `may_affect_control_flow -> C.May_affect_control_flow
-  | `may_store  -> C.May_store
-  | `may_load -> C.May_load
+  | `Valid -> C.Is_true
+  | `Conditional_branch -> C.Is_conditional_branch
+  | `Unconditional_branch -> C.Is_unconditional_branch
+  | `Indirect_branch -> C.Is_indirect_branch
+  | `Return -> C.Is_return
+  | `Call -> C.Is_call
+  | `Barrier -> C.Is_barrier
+  | `Terminator -> C.Is_terminator
+  | `May_affect_control_flow -> C.May_affect_control_flow
+  | `May_store  -> C.May_store
+  | `May_load -> C.May_load
 
 module Insn = struct
   type ins_info = {
@@ -291,6 +291,13 @@ let compare_insn (i1 : ('a,'b) insn) (i2 : ('a,'b) insn) =
   Insn.compare i1 i2
 let sexp_of_insn : ('a,'b) insn -> Sexp.t = Insn.sexp_of_t
 
+
+type full_insn = (asm,kinds) insn
+
+let sexp_of_full_insn = sexp_of_insn
+
+
+
 type (+'a,+'k) insns = (mem * ('a,'k) insn option) list
 
 module Pred = struct
@@ -473,7 +480,7 @@ let insn_of_mem dis mem =
     if Mem.(max_addr mem' = max_addr mem) then Ok `finished
     else Mem.view mem ~from:Addr.(Mem.max_addr mem' ++ 1)
       >>| fun r -> `left r in
-  run ~stop_on:[`valid] dis mem ~return ~init
+  run ~stop_on:[`Valid] dis mem ~return ~init
     ~hit:(fun s mem' insn _ ->
         split mem' >>= fun r -> stop s (mem',Some insn,r))
     ~invalid:(fun s mem' _ ->

lib/bap_disasm/bap_disasm_basic.mli

@@ -12,7 +12,7 @@ type kind = Bap_insn_kind.t with compare, sexp
 
 (** predicate to drive the disassembler *)
 type pred = [
-  | `valid  (** stop on first valid insn  *)
+  | `Valid  (** stop on first valid insn  *)
   |  kind   (** stop on first insn of the specified kind *)
 ] with sexp
 
@@ -22,6 +22,11 @@ type imm  with bin_io, compare, sexp
 type fmm  with bin_io, compare, sexp
 type (+'a,+'k) insn
 type (+'a,+'k) insns = (mem * ('a,'k) insn option) list
+type empty     (** set when information is not stored                *)
+type asm       (** set when assembler information is stored        *)
+type kinds     (** set when instruction kind information is stored *)
+
+type full_insn = (asm,kinds) insn with sexp_of
 
 
 
@@ -43,13 +48,10 @@ type (+'a,+'k) insns = (mem * ('a,'k) insn option) list
     store extra information about instruction kind.
 
     Note: at some points you can have an access to this information
-    even if you don't enable it explicitely.
+    even if you don't enable it explicitly.
 *)
 type ('a,'k) t
 
-type empty     (** set when information is not stored                *)
-type asm       (** set when assembler information is stored        *)
-type kinds     (** set when instruction kind information is stored *)
 
 
 (** Disassembler state.
@@ -109,15 +111,14 @@ val store_kinds : ('a,_) t -> ('a,kinds) t
     monad, like [Or_error], or [Lwt]. Otherwise, just use [ident]
     function and assume that ['s == 'r].
 
-    In a process of disassembly three user provided callbacks are
-    invoked by the engine. To each callback at least two parameters
-    are passed: [state] and [user_data]. [user_data] is arbitrary data
-    of type ['s] with which the folding over the memory is actually
+    In a process of disassembly user provided callbacks are invoked by
+    the engine. To each callback at least two parameters are passed:
+    [state] and [user_data]. [user_data] is arbitrary data of type ['s]
+    with which the folding over the memory is actually
     performed. [state] incapsulates the current state of the
     disassembler, and provides continuation functions, namely [stop],
-    [next] and [back], that drives the process of
-    disassembly. This functions are used to pass control back to the
-    disassembler.
+    [next] and [back], that drives the process of disassembly. This
+    functions are used to pass control back to the disassembler.
 
     [stopped state user_data] is called when there is no more data to
     disassemble. This handler is optional and defaults to [stop].

lib/bap_disasm/bap_insn_kind.ml

@@ -1,24 +1,24 @@
 type branch = [
-  | `conditional_branch
-  | `unconditional_branch
-  | `indirect_branch
-] with compare, enumerate, sexp
+  | `Conditional_branch
+  | `Unconditional_branch
+  | `Indirect_branch
+] with bin_io, compare, enumerate, sexp
 
 type affecting_control = [
   | branch
-  | `return
-  | `call
-  | `barrier
-  | `terminator
-  | `may_affect_control_flow
-] with compare, enumerate, sexp
+  | `Return
+  | `Call
+  | `Barrier
+  | `Terminator
+  | `May_affect_control_flow
+] with bin_io, compare, enumerate, sexp
 
 type having_side_effect = [
-  | `may_load
-  | `may_store
-] with compare, enumerate, sexp
+  | `May_load
+  | `May_store
+] with bin_io, compare, enumerate, sexp
 
 type t = [
   | affecting_control
   | having_side_effect
-] with compare, enumerate, sexp
+] with bin_io, compare, enumerate, sexp