For EIP-4844, Ethereum shoppers want the flexibility to compute and confirm KZG commitments. Quite than every consumer rolling their very own crypto, researchers and builders got here collectively to put in writing c-kzg-4844, a comparatively small C library with bindings for higher-level languages. The concept was to create a sturdy and environment friendly cryptographic library that every one shoppers might use. The Protocol Safety Analysis staff on the Ethereum Basis had the chance to evaluation and enhance this library. This weblog publish will talk about some issues we do to make C tasks safer.
Fuzz
Fuzzing is a dynamic code testing approach that includes offering random inputs to find bugs in a program. LibFuzzer and afl++ are two widespread fuzzing frameworks for C tasks. They’re each in-process, coverage-guided, evolutionary fuzzing engines. For c-kzg-4844, we used LibFuzzer since we had been already well-integrated with LLVM mission’s different choices.
Here is the fuzzer for verify_kzg_proof, one in all c-kzg-4844’s features:
#embrace "../base_fuzz.h" static const size_t COMMITMENT_OFFSET = 0; static const size_t Z_OFFSET = COMMITMENT_OFFSET + BYTES_PER_COMMITMENT; static const size_t Y_OFFSET = Z_OFFSET + BYTES_PER_FIELD_ELEMENT; static const size_t PROOF_OFFSET = Y_OFFSET + BYTES_PER_FIELD_ELEMENT; static const size_t INPUT_SIZE = PROOF_OFFSET + BYTES_PER_PROOF; int LLVMFuzzerTestOneInput(const uint8_t* information, size_t measurement) { initialize(); if (measurement == INPUT_SIZE) { bool okay; verify_kzg_proof( &okay, (const Bytes48 *)(information + COMMITMENT_OFFSET), (const Bytes32 *)(information + Z_OFFSET), (const Bytes32 *)(information + Y_OFFSET), (const Bytes48 *)(information + PROOF_OFFSET), &s ); } return 0; }
When executed, that is what the output appears to be like like. If there have been an issue, it could write the enter to disk and cease executing. Ideally, it’s best to have the ability to reproduce the issue.
There’s additionally differential fuzzing, which is a way which fuzzes two or extra implementations of the identical interface and compares the outputs. For a given enter, if the output is completely different, and also you anticipated them to be the identical, you realize one thing is mistaken. This method may be very widespread in Ethereum as a result of we prefer to have a number of implementations of the identical factor. This diversification offers an additional degree of security, realizing that if one implementation had been flawed the others might not have the identical situation.
For KZG libraries, we developed kzg-fuzz which differentially fuzzes c-kzg-4844 (by its Golang bindings) and go-kzg-4844. Up to now, there have not been any variations.
Protection
Subsequent, we used llvm-profdata and llvm-cov to generate a protection report from working the checks. This can be a nice solution to confirm code is executed (“lined”) and examined. See the coverage goal in c-kzg-4844’s Makefile for an instance of learn how to generate this report.
When this goal is run (i.e., make protection) it produces a desk that serves as a high-level overview of how a lot of every perform is executed. The exported features are on the high and the non-exported (static) features are on the underside.
There’s a number of inexperienced within the desk above, however there’s some yellow and purple too. To find out what’s and is not being executed, check with the HTML file (protection.html) that was generated. This webpage reveals your entire supply file and highlights non-executed code in purple. On this mission’s case, many of the non-executed code offers with hard-to-test error instances akin to reminiscence allocation failures. For instance, this is some non-executed code:
At first of this perform, it checks that the trusted setup is sufficiently big to carry out a pairing test. There is not a take a look at case which offers an invalid trusted setup, so this does not get executed. Additionally, as a result of we solely take a look at with the proper trusted setup, the results of is_monomial_form is all the time the identical and does not return the error worth.
Profile
We do not suggest this for all tasks, however since c-kzg-4844 is a efficiency essential library we expect it is necessary to profile its exported features and measure how lengthy they take to execute. This may help determine inefficiencies which might probably DoS nodes. For this, we used gperftools (Google Efficiency Instruments) as a substitute of llvm-xray as a result of we discovered it to be extra feature-rich and simpler to make use of.
The next is a straightforward instance which profiles my_function. Profiling works by checking which instruction is being executed now and again. If a perform is quick sufficient, it will not be seen by the profiler. To scale back the prospect of this, you might have to name your perform a number of instances. On this instance, we name my_function 1000 instances.
#embraceint task_a(int n) { if (n <= 1) return 1; return task_a(n - 1) * n; } int task_b(int n) { if (n <= 1) return 1; return task_b(n - 2) + n; } void my_function(void) { for (int i = 0; i < 500; i++) { if (i % 2 == 0) { task_a(i); } else { task_b(i); } } } int predominant(void) { ProfilerStart("instance.prof"); for (int i = 0; i < 1000; i++) { my_function(); } ProfilerStop(); return 0; }
Use ProfilerStart(“
Right here is the graph generated from the command above:
Here is a much bigger instance from one in all c-kzg-4844’s features. The next picture is the profiling graph for compute_blob_kzg_proof. As you possibly can see, 80% of this perform’s time is spent performing Montgomery multiplications. That is anticipated.
Reverse
Subsequent, view your binary in a software program reverse engineering (SRE) instrument akin to Ghidra or IDA. These instruments may help you perceive how high-level constructs are translated into low-level machine code. We predict it helps to evaluation your code this manner; like how studying a paper in a special font will pressure your mind to interpret sentences in a different way. It is also helpful to see what kind of optimizations your compiler makes. It is uncommon, however generally the compiler will optimize out one thing which it deemed pointless. Hold a watch out for this, one thing like this really occurred in c-kzg-4844, some of the tests were being optimized out.
Whenever you view a decompiled perform, it won’t have variable names, complicated varieties, or feedback. When compiled, this info is not included within the binary. It is going to be as much as you to reverse engineer this. You will typically see features are inlined right into a single perform, a number of variables declared in code are optimized right into a single buffer, and the order of checks are completely different. These are simply compiler optimizations and are typically high-quality. It could assist to construct your binary with DWARF debugging info; most SREs can analyze this part to supply higher outcomes.
For instance, that is what blob_to_kzg_commitment initially appears to be like like in Ghidra:
With a bit of work, you possibly can rename variables and add feedback to make it simpler to learn. Here is what it might seem like after a couple of minutes:
Static Evaluation
Clang comes built-in with the Clang Static Analyzer, which is a wonderful static evaluation instrument that may determine many issues that the compiler will miss. Because the title “static” suggests, it examines code with out executing it. That is slower than the compiler, however loads quicker than “dynamic” evaluation instruments which execute code.
Here is a easy instance which forgets to free arr (and has one other downside however we are going to discuss extra about that later). The compiler won’t determine this, even with all warnings enabled as a result of technically that is utterly legitimate code.
#embraceint predominant(void) { int* arr = malloc(5 * sizeof(int)); arr[5] = 42; return 0; }
The unix.Malloc checker will determine that arr wasn’t freed. The road within the warning message is a bit deceptive, nevertheless it is sensible if you consider it; the analyzer reached the return assertion and seen that the reminiscence hadn’t been freed.
Not all the findings are that straightforward although. Here is a discovering that Clang Static Analyzer present in c-kzg-4844 when initially launched to the mission:
Given an sudden enter, it was potential to shift this worth by 32 bits which is undefined habits. The answer was to limit the enter with CHECK(log2_pow2(n) != 0) in order that this was not possible. Good job, Clang Static Analyzer!
Sanitize
Santizers are dynamic evaluation instruments which instrument (add directions) to packages which may level out points throughout execution. These are significantly helpful at discovering widespread errors related to reminiscence dealing with. Clang comes built-in with a number of sanitizers; listed here are the 4 we discover most helpful and simple to make use of.
Deal with
AddressSanitizer (ASan) is a quick reminiscence error detector which may determine out-of-bounds accesses, use-after-free, use-after-return, use-after-scope, double-free, and reminiscence leaks.
Right here is similar instance from earlier. It forgets to free arr and it’ll set the sixth aspect in a 5 aspect array. This can be a easy instance of a heap-buffer-overflow:
#embraceint predominant(void) { int* arr = malloc(5 * sizeof(int)); arr[5] = 42; return 0; }
When compiled with -fsanitize=handle and executed, it can output the next error message. This factors you in path (a 4-byte write in predominant). This binary may very well be considered in a disassembler to determine precisely which instruction (at predominant+0x84) is inflicting the issue.
Equally, this is an instance the place it finds a heap-use-after-free:
#embraceint predominant(void) { int *arr = malloc(5 * sizeof(int)); free(arr); return arr[2]; }
It tells you that there is a 4-byte learn of freed reminiscence at predominant+0x8c.
Reminiscence
MemorySanitizer (MSan) is a detector of uninitialized reads. Here is a easy instance which reads (and returns) an uninitialized worth:
int predominant(void) { int information[2]; return information[0]; }
When compiled with -fsanitize=reminiscence and executed, it can output the next error message:
Undefined Conduct
UndefinedBehaviorSanitizer (UBSan) detects undefined habits, which refers back to the scenario the place a program’s habits is unpredictable and never specified by the langauge customary. Some widespread examples of this are accessing out-of-bounds reminiscence, dereferencing an invalid pointer, studying uninitialized variables, and overflow of a signed integer. For instance, right here we increment INT_MAX which is undefined habits.
#embraceint predominant(void) { int a = INT_MAX; return a + 1; }
When compiled with -fsanitize=undefined and executed, it can output the next error message which tells us precisely the place the issue is and what the circumstances are:
Thread
ThreadSanitizer (TSan) detects information races, which may happen in multi-threaded packages when two or extra threads entry a shared reminiscence location on the identical time. This case introduces unpredictability and may result in undefined habits. Here is an instance wherein two threads increment a world counter variable. There are no locks or semaphores, so it is totally potential that these two threads will increment the variable on the identical time.
#embraceint counter = 0; void *increment(void *arg) { (void)arg; for (int i = 0; i < 1000000; i++) counter++; return NULL; } int predominant(void) { pthread_t thread1, thread2; pthread_create(&thread1, NULL, increment, NULL); pthread_create(&thread2, NULL, increment, NULL); pthread_join(thread1, NULL); pthread_join(thread2, NULL); return 0; }
When compiled with -fsanitize=thread and executed, it can output the next error message:
This error message tells us that there is a information race. In two threads, the increment perform is writing to the identical 4 bytes on the identical time. It even tells us that the reminiscence is counter.
Valgrind
Valgrind is a robust instrumentation framework for constructing dynamic evaluation instruments, however its finest recognized for figuring out reminiscence errors and leaks with its built-in Memcheck instrument.
The next picture reveals the output from working c-kzg-4844’s checks with Valgrind. Within the purple field is a sound discovering for a “conditional soar or transfer [that] depends upon uninitialized worth(s).”
This identified an edge case in expand_root_of_unity. If the mistaken root of unity or width had been supplied, it was potential that the loop will break earlier than out[width] was initialized. On this scenario, the ultimate test would depend upon an uninitialized worth.
static C_KZG_RET expand_root_of_unity( fr_t *out, const fr_t *root, uint64_t width ) { out[0] = FR_ONE; out[1] = *root; for (uint64_t i = 2; !fr_is_one(&out[i - 1]); i++) { CHECK(i <= width); blst_fr_mul(&out[i], &out[i - 1], root); } CHECK(fr_is_one(&out[width])); return C_KZG_OK; }
Safety Assessment
After improvement stabilizes, it has been completely examined, and your staff has manually reviewed the codebase themselves a number of instances, it is time to get a safety evaluation by a good safety group. This may not be a stamp of approval, nevertheless it reveals that your mission is at the least considerably safe. Take into account there is no such thing as a such factor as excellent safety. There’ll all the time be the chance of vulnerabilities.
For c-kzg-4844 and go-kzg-4844, the Ethereum Basis contracted Sigma Prime to conduct a safety evaluation. They produced this report with 8 findings. It incorporates one essential vulnerability in go-kzg-4844 that was a very good discover. The BLS12-381 library that go-kzg-4844 makes use of, gnark-crypto, had a bug which allowed invalid G1 and G2 factors to be sucessfully decoded. Had this not been fastened, this might have resulted in a consensus bug (a disagreement between implementations) in Ethereum.
Bug Bounty
If a vulnerability in your mission may very well be exploited for features, like it’s for Ethereum, take into account organising a bug bounty program. This permits safety researchers, or anybody actually, to submit vulnerability reviews in change for cash. Typically, that is particularly for findings which may show that an exploit is feasible. If the bug bounty payouts are cheap, bug finders will notify you of the bug moderately than exploiting it or promoting it to a different social gathering. We suggest beginning your bug bounty program after the findings from the primary safety evaluation are resolved; ideally, the safety evaluation would value lower than the bug bounty payouts.
Conclusion
The event of strong C tasks, particularly within the essential area of blockchain and cryptocurrencies, requires a multi-faceted strategy. Given the inherent vulnerabilities related to the C language, a mix of finest practices and instruments is important for producing resilient software program. We hope our experiences and findings from our work with c-kzg-4844 present precious insights and finest practices for others embarking on comparable tasks.
