Skip to content

Latest commit

 

History

History
53 lines (44 loc) · 2.17 KB

tfsa-2020-014.md

File metadata and controls

53 lines (44 loc) · 2.17 KB
Raw

TFSA-2020-014: Integer truncation in Shard API usage

CVE Number

CVE-2020-15202

Impact

The Shard API in TensorFlow expects the last argument to be a function taking two int64 (i.e., long long) arguments:

void Shard(int max_parallelism, thread::ThreadPool* workers, int64 total,
           int64 cost_per_unit, std::function<void(int64, int64)> work);

However, there are several places in TensorFlow where a lambda taking int or int32 arguments is being used:

    auto DoWork = [samples_per_alpha, num_alphas, &rng, samples_flat,
                   alpha_flat](int start_output, int limit_output) {...};
    Shard(worker_threads.num_threads, worker_threads.workers,
          num_alphas * samples_per_alpha, kElementCost, DoWork);

In these cases, if the amount of work to be parallelized is large enough, integer truncation occurs. Depending on how the two arguments of the lambda are used, this can result in segfaults, read/write outside of heap allocated arrays, stack overflows, or data corruption.

Vulnerable Versions

TensorFlow 1.15.0, 1.15.1, 1.15.2, 1.15.3, 2.0.0, 2.0.1, 2.0.2, 2.1.0, 2.1.1, 2.2.0, 2.3.0.

Patches

We have patched the issue in 27b417360cbd671ef55915e4bb6bb06af8b8a832 and ca8c013b5e97b1373b3bb1c97ea655e69f31a575. We will release patch releases for all versions between 1.15 and 2.3.

We recommend users to upgrade to TensorFlow 1.15.4, 2.0.3, 2.1.2, 2.2.1, or 2.3.1.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by members of the Aivul Team from Qihoo 360.