CVE-2021-37692 (CNNVD-202108-1275)
中文标题:
Google TensorFlow 输入验证错误漏洞
英文标题:
Segfault on strings tensors with mistmatched dimensions in TensorFlow
漏洞描述
中文描述:
Google TensorFlow是美国谷歌（Google）公司的一套用于机器学习的端到端开源平台。 Google TensorFlow 存在输入验证错误漏洞，该漏洞源于Go 代码可能会触发字符串释放中的段错误。
英文描述:
TensorFlow is an end-to-end open source platform for machine learning. In affected versions under certain conditions, Go code can trigger a segfault in string deallocation. For string tensors, `C.TF_TString_Dealloc` is called during garbage collection within a finalizer function. However, tensor structure isn't checked until encoding to avoid a performance penalty. The current method for dealloc assumes that encoding succeeded, but segfaults when a string tensor is garbage collected whose encoding failed (e.g., due to mismatched dimensions). To fix this, the call to set the finalizer function is deferred until `NewTensor` returns and, if encoding failed for a string tensor, deallocs are determined based on bytes written. We have patched the issue in GitHub commit 8721ba96e5760c229217b594f6d2ba332beedf22. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, which is the other affected version.
CWE类型:
标签:
受影响产品
| 厂商 | 产品 | 版本 | 版本范围 | 平台 | CPE |
|---|---|---|---|---|---|
| tensorflow | tensorflow | >= 2.5.0, < 2.5.1 | - | - |
cpe:2.3:a:tensorflow:tensorflow:>=_2.5.0,_<_2.5.1:*:*:*:*:*:*:*
|
| tensorflow | * | - | - |
cpe:2.3:a:google:tensorflow:*:*:*:*:*:*:*:*
|
解决方案
中文解决方案:
英文解决方案:
临时解决方案:
CVSS评分详情
3.1 (cna)
MEDIUMCVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
时间信息
利用信息
数据源详情
| 数据源 | 记录ID | 版本 | 提取时间 |
|---|---|---|---|
| CVE | cve_CVE-2021-37692 |
2025-11-11 15:21:02 | 2025-11-11 07:36:58 |
| NVD | nvd_CVE-2021-37692 |
2025-11-11 14:57:40 | 2025-11-11 07:45:16 |
| CNNVD | cnnvd_CNNVD-202108-1275 |
2025-11-11 15:10:42 | 2025-11-11 07:56:51 |
版本与语言
安全公告
变更历史
查看详细变更
- vulnerability_type: 未提取 -> 输入验证错误
- cnnvd_id: 未提取 -> CNNVD-202108-1275
- data_sources: ['cve', 'nvd'] -> ['cnnvd', 'cve', 'nvd']
查看详细变更
- affected_products_count: 1 -> 2
- data_sources: ['cve'] -> ['cve', 'nvd']