Summary
There is an error in the stack management when compiling the IR for sha3_64. Concretely, the height variable is miscalculated.
The vulnerability can't be triggered without writing the IR by hand. That is, it cannot be triggered from regular vyper code, it can only be triggered by using the fang binary directly (this binary used to be called vyper-ir prior to v0.3.4).
Details
To compile sha3_64, the arg[0] and arg[1] have to be compiled:
|
o = _compile_to_assembly(code.args[0], withargs, existing_labels, break_dest, height) |
|
o.extend(_compile_to_assembly(code.args[1], withargs, existing_labels, break_dest, height)) |
As can be seen, after compiling the 0th arg, the height variable isn't increased. If new withargs are defined in the inner scope, they are manipulated correctly, because both their height is off and also the global height is off and thus their placement on the stack is computed correctly.
sha3_64 is used for retrieval in mappings. No flow that would cache the key was found, the issue shouldn't be possible to trigger when compiling the compiler-generated IR.
PoC
Suppose the following hand-written IR:
(with _loc
(with val 1
(with key 2
(sha3_64 val key)))
(seq
(sstore _loc
(with x (sload _loc)
(with ans (add x 1) (seq (assert (ge ans x)) ans))))))
after compilation:
the generated bytecode: 6001600281806020525f5260405f2090509050805460018101818110610026579050815550005b5f80fd
0000 60 PUSH1 0x01
0002 60 PUSH1 0x02
0004 81 DUP2
0005 80 DUP1 *********** bad code here!!!!!!
0006 60 PUSH1 0x20
0008 52 MSTORE
It can be seen that the second DUP will dup the item on the top of the stack which is incorrect.
Patches
Patched in #4063.
Impact
Versions v0.2.0-v0.3.10 were evaluated, and access of the variable with the invalid height is not reachable from IR generated by the vyper front-end. Because the issue isn't triggered during normal compilation of vyper code, the impact is considered low.
cyberthirst Coordinator
kuroi8 Reporter
Summary
There is an error in the stack management when compiling the
IRforsha3_64. Concretely, theheightvariable is miscalculated.The vulnerability can't be triggered without writing the
IRby hand. That is, it cannot be triggered from regular vyper code, it can only be triggered by using thefangbinary directly (this binary used to be calledvyper-irprior to v0.3.4).Details
To compile
sha3_64, thearg[0]andarg[1]have to be compiled:vyper/vyper/ir/compile_ir.py
Lines 585 to 586 in c150fc4
As can be seen, after compiling the 0th arg, the
heightvariable isn't increased. If newwithargsare defined in the inner scope, they are manipulated correctly, because both theirheightis off and also the globalheightis off and thus their placement on the stack is computed correctly.sha3_64is used for retrieval in mappings. No flow that would cache thekeywas found, the issue shouldn't be possible to trigger when compiling the compiler-generatedIR.PoC
Suppose the following hand-written IR:
after compilation:
It can be seen that the second
DUPwill dup the item on the top of the stack which is incorrect.Patches
Patched in #4063.
Impact
Versions v0.2.0-v0.3.10 were evaluated, and access of the variable with the invalid height is not reachable from IR generated by the vyper front-end. Because the issue isn't triggered during normal compilation of vyper code, the impact is considered low.