{"id":"GHSA-6976-qm5m-7mcj","aliases":["RUSTSEC-2026-0236"],"status":"published","title":"A `BigInt` division panics, and two neighbouring operations answer wrongly in silence","description":"`viperjs` is a JavaScript engine intended to run untrusted script inside a host application, so\nscript text is data rather than a trusted caller and a panic reachable from script is a denial of\nservice in the embedder's process.\n\nA divisor whose magnitude lands exactly on the engine's internal limb ceiling reaches an\nout-of-bounds index. On every released version up to and including 0.2.1:\n\n```js\nconst d = (1n \u003c\u003c 33554399n) * 2n;\n1n / d;   // panic: index out of bounds\n```\n\nTwo further operations on the same value returned wrong results without raising anything, which\nis the more dangerous half for an embedder that acts on the answer:\n\n```js\nd % 7n;      // 0n  — the true remainder is 1n\nString(d);   // \"0\"\n```\n\n## Cause\n\nThe left-shift helper reserved one limb for the bits a shift may push past the top of the\nmagnitude, measured *that* width against the size ceiling, and then trimmed the reserved limb away\nagain — so a magnitude landing exactly on the ceiling was refused on account of room it does not\nkeep. The division treated that refusal as unreachable and discarded it with `unwrap_or_default`,\nleaving an empty divisor magnitude; the subsequent `divisor[n - 1]` is then an index of\n`usize::MAX`.\n\nThe crate is `#![forbid(unsafe_code)]`, so this is a panic and not memory unsafety.\n\n## Remediation\n\nUpgrade to 0.2.2, in which all three behaviours are fixed: both divisions now produce correct\nresults, and `String()` of a magnitude beyond what the engine can divide raises a `RangeError`\nrather than producing `\"0\"` — ECMA-262 §6.1.4 requires an implementation that imposes a limit to\nthrow rather than answer something else.\n\nThere is no workaround short of upgrading; the values are reachable from any script the embedder\nevaluates.\n\nReported by [@Zniece](https://github.com/Zniece).","severity":{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high","from":"osv"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"published":"2026-08-06T12:00:00Z","modified":"2026-10-12T18:00:14Z","fixes":[{"ecosystem":"crates.io","package":"viperjs","fixed_in":["0.2.2"],"sources":["osv"]}],"affected":[{"source":"osv","ecosystem":"crates.io","package":"viperjs","ranges":[{"type":"SEMVER","events":[{"introduced":"0.0.0-0"},{"fixed":"0.2.2"}]}]}],"references":[{"url":"https://crates.io/crates/viperjs","tags":["package"]},{"url":"https://rustsec.org/advisories/RUSTSEC-2026-0236.html","tags":["advisory"]},{"url":"https://github.com/MerlijnW70/viperjs/commit/92988456c44fe7744c13cba1d891f127b8437f95","tags":["web"]},{"url":"https://github.com/MerlijnW70/viperjs/releases/tag/v0.2.2","tags":["web"]}],"sources":{"osv":[{"id":"RUSTSEC-2026-0236","title":"A `BigInt` division panics, and two neighbouring operations answer wrongly in silence","description":"`viperjs` is a JavaScript engine intended to run untrusted script inside a host application, so\nscript text is data rather than a trusted caller and a panic reachable from script is a denial of\nservice in the embedder's process.\n\nA divisor whose magnitude lands exactly on the engine's internal limb ceiling reaches an\nout-of-bounds index. On every released version up to and including 0.2.1:\n\n```js\nconst d = (1n \u003c\u003c 33554399n) * 2n;\n1n / d;   // panic: index out of bounds\n```\n\nTwo further operations on the same value returned wrong results without raising anything, which\nis the more dangerous half for an embedder that acts on the answer:\n\n```js\nd % 7n;      // 0n  — the true remainder is 1n\nString(d);   // \"0\"\n```\n\n## Cause\n\nThe left-shift helper reserved one limb for the bits a shift may push past the top of the\nmagnitude, measured *that* width against the size ceiling, and then trimmed the reserved limb away\nagain — so a magnitude landing exactly on the ceiling was refused on account of room it does not\nkeep. The division treated that refusal as unreachable and discarded it with `unwrap_or_default`,\nleaving an empty divisor magnitude; the subsequent `divisor[n - 1]` is then an index of\n`usize::MAX`.\n\nThe crate is `#![forbid(unsafe_code)]`, so this is a panic and not memory unsafety.\n\n## Remediation\n\nUpgrade to 0.2.2, in which all three behaviours are fixed: both divisions now produce correct\nresults, and `String()` of a magnitude beyond what the engine can divide raises a `RangeError`\nrather than producing `\"0\"` — ECMA-262 §6.1.4 requires an implementation that imposes a limit to\nthrow rather than answer something else.\n\nThere is no workaround short of upgrading; the values are reachable from any script the embedder\nevaluates.\n\nReported by [@Zniece](https://github.com/Zniece).","status":"published","published":"2026-08-06T12:00:00Z","modified":"2026-10-12T18:00:14.309974197Z","cvss":[{"version":"3.1","score":7.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","level":"high"}],"updated_at":"2026-11-02T15:53:40.430679Z"}]}}