Isolating application components is crucial to limit the exposure of sensitive data and code to vulnerabilities in the untrusted components. Process-based isolation is the de facto isolation used in practice, e.g., web browsers. However, it incurs significant performance overhead and is typically infeasible when frequent switches between isolation domains are expected. To address this problem, many intra-process memory isolation techniques have been proposed using novel kernel abstractions, recent CPU extensions (e.g., Intel ® MPK), and software-based fault isolation (e.g., WebAssembly). However, these techniques insufficiently isolate kernel resources, such as file descriptors, or do so by incurring high overheads when resources are accessed. Other work virtualizes the kernel context inside a privileged user space domain, but this is ad-hoc, error-prone, and provides only limited kernel functionalities. We propose μSwitch, an efficient kernel context isolation mechanism with memory protection that addresses these limitations. We use a protected structure, shared by the kernel and the user space, for context switching and propose implicit context switching to improve its performance by deferring the kernel resource switch to the next system call. We apply μSWITCH to isolate libraries in the Firefox web browser and an HTTP server, and reduce the overhead of isolation by 32.7% to 98.4% compared with other isolation techniques.
more »
« less
Isolation without taxation: near-zero-cost transitions for WebAssembly and SFI
Software sandboxing or software-based fault isolation (SFI) is a lightweight approach to building secure systems out of untrusted components. Mozilla, for example, uses SFI to harden the Firefox browser by sandboxing third-party libraries, and companies like Fastly and Cloudflare use SFI to safely co-locate untrusted tenants on their edge clouds. While there have been significant efforts to optimize and verify SFI enforcement, context switching in SFI systems remains largely unexplored: almost all SFI systems use heavyweight transitions that are not only error-prone but incur significant performance overhead from saving, clearing, and restoring registers when context switching. We identify a set of zero-cost conditions that characterize when sandboxed code has sufficient structured to guarantee security via lightweight zero-cost transitions (simple function calls). We modify the Lucet Wasm compiler and its runtime to use zero-cost transitions, eliminating the undue performance tax on systems that rely on Lucet for sandboxing (e.g., we speed up image and font rendering in Firefox by up to 29.7% and 10% respectively). To remove the Lucet compiler and its correct implementation of the Wasm specification from the trusted computing base, we (1) develop a static binary verifier , VeriZero, which (in seconds) checks that binaries produced by Lucet satisfy our zero-cost conditions, and (2) prove the soundness of VeriZero by developing a logical relation that captures when a compiled Wasm function is semantically well-behaved with respect to our zero-cost conditions. Finally, we show that our model is useful beyond Wasm by describing a new, purpose-built SFI system, SegmentZero32, that uses x86 segmentation and LLVM with mostly off-the-shelf passes to enforce our zero-cost conditions; our prototype performs on-par with the state-of-the-art Native Client SFI system.
more »
« less
- NSF-PAR ID:
- 10322268
- Date Published:
- Journal Name:
- Proceedings of the ACM on Programming Languages
- Volume:
- 6
- Issue:
- POPL
- ISSN:
- 2475-1421
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)We describe Swivel, a new compiler framework for hardening WebAssembly (Wasm) against Spectre attacks. Outside the browser, Wasm has become a popular lightweight, in-process sandbox and is, for example, used in production to isolate different clients on edge clouds and function-as-a-service platforms. Unfortunately, Spectre attacks can bypass Wasm's isolation guarantees. Swivel hardens Wasm against this class of attacks by ensuring that potentially malicious code can neither use Spectre attacks to break out of the Wasm sandbox nor coerce victim code—another Wasm client or the embedding process—to leak secret data. We describe two Swivel designs, a software-only approach that can be used on existing CPUs, and a hardware-assisted approach that uses extension available in Intel® 11th generation CPUs. For both, we evaluate a randomized approach that mitigates Spectre and a deterministic approach that eliminates Spectre altogether. Our randomized implementations impose under 10.3% overhead on the Wasm-compatible subset of SPEC 2006, while our deterministic implementations impose overheads between 3.3% and 240.2%. Though high on some benchmarks, Swivel's overhead is still between 9× and 36.3× smaller than existing defenses that rely on pipeline fences.more » « less
-
null (Ed.)WebAssembly is designed to be an alternative to JavaScript that is a safe, portable, and efficient compilation target for a variety of languages. The performance of high-level languages depends not only on the underlying performance of WebAssembly, but also on the quality of the generated WebAssembly code. In this paper, we identify several features of high-level languages that current approaches can only compile to WebAssembly by generating complex and inefficient code. We argue that these problems could be addressed if WebAssembly natively supported first-class continuations. We then present Wasm/k, which extends WebAssembly with delimited continuations. Wasm/k introduces no new value types, and thus does not require significant changes to the WebAssembly type system (validation). Wasm/k is safe, even in the presence of foreign function calls (e.g., to and from JavaScript). Finally, Wasm/k is amenable to efficient implementation: we implement Wasm/k as a local change to Wasmtime, an existing WebAssembly JIT. We evaluate Wasm/k by implementing C/k, which adds delimited continuations to C/C++. C/k uses Emscripten and its implementation serves as a case study on how to use Wasm/k in a compiler that targets WebAssembly. We present several case studies using C/k, and show that on implementing green threads, it can outperform the state-of-the-art approach Asyncify with an 18% improvement in performance and a 30% improvement in code size.more » « less
-
Most programs compiled to WebAssembly (Wasm) today are written in unsafe languages like C and C++. Unfortunately, memory-unsafe C code remains unsafe when compiled to Wasm—and attackers can exploit buffer overflows and use-after-frees in Wasm almost as easily as they can on native platforms. Memory- Safe WebAssembly (MSWasm) proposes to extend Wasm with language-level memory-safety abstractions to precisely address this problem. In this paper, we build on the original MSWasm position paper to realize this vision. We give a precise and formal semantics of MSWasm, and prove that well-typed MSWasm programs are, by construction, robustly memory safe. To this end, we develop a novel, language-independent memory-safety property based on colored memory locations and pointers. This property also lets us reason about the security guarantees of a formal C-to-MSWasm compiler—and prove that it always produces memory-safe programs (and preserves the semantics of safe programs). We use these formal results to then guide several implementations: Two compilers of MSWasm to native code, and a C-to-MSWasm compiler (that extends Clang). Our MSWasm compilers support different enforcement mechanisms, allowing developers to make security-performance trade-offs according to their needs. Our evaluation shows that on the PolyBenchC suite, the overhead of enforcing memory safety in software ranges from 22% (enforcing spatial safety alone) to 198% (enforcing full memory safety), and 51.7% when using hardware memory capabilities for spatial safety and pointer integrity. More importantly, MSWasm’s design makes it easy to swap between enforcement mechanisms; as fast (especially hardware-based) enforcement techniques become available, MSWasm will be able to take advantage of these advances almost for free.more » « less
-
Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should produce identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1145/3498688
