Vulnerabilities in pkg:pypi/tensorflow-cpu
PACKAGECVE-2022-41889 is a MODERATE (CVSS 5.5) segfault vulnerability (CWE-476: NULL Pointer Dereference) affecting Tensorflow libraries in the PyPI ecosystem. The issue involves segfault via invalid attributes in `pywrap_tfe_src.cc`. Remediation is available by upgrading affected packages to version 2.8.4, 2.9.3, 2.10.1.
pkg:pypi/tensorflow-cpu
Vulnerability advisories affecting package pkg:pypi/tensorflow-cpu. Select an advisory below to view technical details or fixes.
| Advisory ID | CVE / NVD | Severity | Summary |
|---|---|---|---|
| GHSA-xxcj-rhqg-m46g | CVE-2022-41889 | MODERATE | Segfault via invalid attributes in `pywrap_tfe_src.cc` |
| GHSA-6g85-3hm8-83f9 | CVE-2021-29544 | LOW | CHECK-fail in `QuantizeAndDequantizeV4Grad` |
| GHSA-24x6-8c7m-hv3f | CVE-2021-29590 | LOW | Heap OOB read in TFLite's implementation of `Minimum` or `Maximum` |
| GHSA-8r7c-3cm2-3h8f | CVE-2022-23578 | MODERATE | Memory leak in Tensorflow |
| GHSA-8cxv-76p7-jxwr | CVE-2022-23577 | HIGH | Null-dereference in Tensorflow |
| GHSA-wm93-f238-7v37 | CVE-2022-23576 | HIGH | Integer overflow in Tensorflow |
| GHSA-c94w-c95p-phf8 | CVE-2022-23575 | HIGH | Integer overflow in Tensorflow |
| GHSA-77gp-3h4r-6428 | CVE-2022-23574 | HIGH | Out of bounds read and write in Tensorflow |
| GHSA-q85f-69q7-55h2 | CVE-2022-23573 | HIGH | Uninitialized variable access in Tensorflow |
| GHSA-rww7-2gpw-fv6j | CVE-2022-23572 | HIGH | Crash when type cannot be specialized in Tensorflow |
| GHSA-j3mj-fhpq-qqjj | CVE-2022-23571 | HIGH | Reachable Assertion in Tensorflow |
| GHSA-9p77-mmrw-69c7 | CVE-2022-23570 | HIGH | Null-dereference in Tensorflow |
| GHSA-qj5r-f9mv-rffh | CVE-2022-23569 | HIGH | `CHECK`-fails when building invalid tensor shapes in Tensorflow |
| GHSA-6445-fm66-fvq2 | CVE-2022-23568 | HIGH | Integer overflows in Tensorflow |
| GHSA-rrx2-r989-2c43 | CVE-2022-23567 | HIGH | Integer overflows in Tensorflow |
| GHSA-5qw5-89mw-wcg2 | CVE-2022-23566 | HIGH | Out of bounds write in Tensorflow |
| GHSA-4v5p-v5h9-6xjx | CVE-2022-23565 | HIGH | `CHECK`-failures in Tensorflow |
| GHSA-8rcj-c8pj-v3m3 | CVE-2022-23564 | HIGH | Reachable Assertion in Tensorflow |
| GHSA-wc4g-r73w-x8mm | CVE-2022-23563 | HIGH | Insecure temporary file in Tensorflow |
| GHSA-qx3f-p745-w4hr | CVE-2022-23562 | HIGH | Integer overflow in Tensorflow |
| GHSA-9c78-vcq7-7vxq | CVE-2022-23561 | HIGH | Out of bounds write in TFLite |
| GHSA-4hvf-hxvg-f67v | CVE-2022-23560 | HIGH | Read and Write outside of bounds in TensorFlow |
| GHSA-98p5-x8x4-c9m5 | CVE-2022-23559 | HIGH | Integer overflow in TFLite |
| GHSA-9gwq-6cwj-47h3 | CVE-2022-23558 | HIGH | Integer overflow in TFLite array creation |
| GHSA-gf2j-f278-xh4v | CVE-2022-23557 | HIGH | Division by zero in TFLite |
| GHSA-428x-9xc2-m8mj | CVE-2022-21741 | HIGH | Division by zero in TFLite |
| GHSA-44qp-9wwf-734r | CVE-2022-21740 | HIGH | Heap overflow in Tensorflow |
| GHSA-3mw4-6rj6-74g5 | CVE-2022-21739 | HIGH | Null pointer dereference in TensorFlow |
| GHSA-x4qx-4fjv-hmw6 | CVE-2022-21738 | HIGH | Integer overflow leading to crash in Tensorflow |
| GHSA-f2vv-v9cg-qhh7 | CVE-2022-21737 | HIGH | Assertion failure based denial of service in Tensorflow |
| GHSA-pfjj-m3jj-9jc9 | CVE-2022-21736 | HIGH | Undefined behavior in `SparseTensorSliceDataset` |
| GHSA-gcvh-66ff-4mwm | CVE-2022-21734 | HIGH | `CHECK`-failures in Tensorflow |
| GHSA-fpcp-9h7m-ffpx | CVE-2022-23595 | MODERATE | Null pointer dereference in TensorFlow |
| GHSA-gwcx-jrx4-92w2 | CVE-2022-23593 | HIGH | Segfault in `simplifyBroadcast` in Tensorflow |
| GHSA-vq36-27g6-p492 | CVE-2022-23592 | HIGH | Out of bounds read in Tensorflow |
| GHSA-247x-2f9f-5wp7 | CVE-2022-23591 | HIGH | Stack overflow in TensorFlow |
| GHSA-87v6-crgm-2gfj | CVE-2022-21735 | HIGH | Division by zero in Tensorflow |
| GHSA-98j8-c9q4-r38g | CVE-2022-21733 | MODERATE | Memory exhaustion in Tensorflow |
| GHSA-c582-c96p-r5cq | CVE-2022-21732 | MODERATE | Memory exhaustion in Tensorflow |
| GHSA-m4hf-j54p-p353 | CVE-2022-21731 | HIGH | Type confusion leading to segfault in Tensorflow |
| GHSA-vjg4-v33c-ggc4 | CVE-2022-21730 | HIGH | Out of bounds read in Tensorflow |
| GHSA-34f9-hjfq-rr8j | CVE-2022-21729 | HIGH | Overflow and uncaught divide by zero in Tensorflow |
| GHSA-6gmv-pjp9-p8w8 | CVE-2022-21728 | HIGH | Out of bounds read in Tensorflow |
| GHSA-c6fh-56w7-fvjw | CVE-2022-21727 | HIGH | Integer overflow in Tensorflow |
| GHSA-23hm-7w47-xw72 | CVE-2022-21726 | HIGH | Out of bounds read in Tensorflow |
| GHSA-v3f7-j968-4h5f | CVE-2022-21725 | MODERATE | Division by zero in Tensorflow |
| GHSA-3rcw-9p9x-582v | CVE-2021-41228 | MODERATE | Code injection in `saved_model_cli` |
| GHSA-j8c8-67vp-6mx7 | CVE-2021-41227 | MODERATE | Arbitrary memory read in `ImmutableConst` |
| GHSA-374m-jm66-3vj8 | CVE-2021-41226 | MODERATE | Heap OOB in `SparseBinCount` |
| GHSA-7r94-xv9v-63jw | CVE-2021-41225 | MODERATE | A use of uninitialized value vulnerability in Tensorflow |
| GHSA-rg3m-hqc5-344v | CVE-2021-41224 | MODERATE | `SparseFillEmptyRows` heap OOB |
| GHSA-f54p-f6jp-4rhr | CVE-2021-41223 | MODERATE | Heap OOB in `FusedBatchNorm` kernels |
| GHSA-cpf4-wx82-gxp6 | CVE-2021-41222 | MODERATE | Segfault due to negative splits in `SplitV` |
| GHSA-7v94-64hj-m82h | CVE-2021-41207 | MODERATE | FPE in `ParallelConcat` |
| GHSA-pgcq-h79j-2f69 | CVE-2021-41206 | HIGH | Incomplete validation of shapes in multiple TF ops |
| GHSA-49rx-x2rw-pc6f | CVE-2021-41205 | MODERATE | Heap OOB read in all `tf.raw_ops.QuantizeAndDequantizeV*` ops |
| GHSA-786j-5qwq-r36x | CVE-2021-41204 | MODERATE | Segfault while copying constant resource tensor |
| GHSA-7pxj-m4jf-r6h2 | CVE-2021-41203 | HIGH | Missing validation during checkpoint loading |
| GHSA-xrqm-fpgr-6hhx | CVE-2021-41202 | MODERATE | Overflow/crash in `tf.range` |
| GHSA-j86v-p27c-73fm | CVE-2021-41201 | HIGH | Unitialized access in `EinsumHelper::ParseEquation` |
| GHSA-m539-j985-hcr8 | CVE-2021-41196 | MODERATE | Crash in `max_pool3d` when size argument is 0 or negative |
| GHSA-prcg-wp5q-rv7p | CVE-2021-41197 | MODERATE | Crashes due to overflow and `CHECK`-fail in ops with large tensor shapes |
| GHSA-2p25-55c9-h58q | CVE-2021-41198 | MODERATE | Overflow/crash in `tf.tile` when tiling tensor is large |
| GHSA-gh8h-7j2j-qv4f | CVE-2021-41200 | MODERATE | Incomplete validation in `tf.summary.create_file_writer` |
| GHSA-cq76-mxrc-vchh | CVE-2021-41195 | MODERATE | Crash in `tf.math.segment_*` operations |
| GHSA-cmgw-8vpc-rc59 | CVE-2021-37692 | MODERATE | Segfault on strings tensors with mistmatched dimensions, due to Go code |
| GHSA-27qf-jwm8-g7f3 | CVE-2021-37691 | MODERATE | FPE in LSH in TFLite |
| GHSA-3hxh-8cp2-g4hg | CVE-2021-37690 | MODERATE | Use after free and segfault in shape inference functions |
| GHSA-wf5p-c75w-w3wh | CVE-2021-37689 | HIGH | Null pointer dereference in TFLite MLIR optimizations |
| GHSA-vcjj-9vg7-vf68 | CVE-2021-37688 | HIGH | Null pointer dereference in TFLite |
| GHSA-jwf9-w5xm-f437 | CVE-2021-37687 | MODERATE | Heap OOB in TFLite's `Gather*` implementations |
| GHSA-mhhc-q96p-mfm9 | CVE-2021-37686 | MODERATE | Infinite loop in TFLite |
| GHSA-c545-c4f9-rf6v | CVE-2021-37685 | MODERATE | Heap OOB in TFLite |
| GHSA-q7f7-544h-67h9 | CVE-2021-37684 | MODERATE | FPE in TFLite pooling operations |
| GHSA-rhrq-64mq-hf9h | CVE-2021-37683 | MODERATE | FPE in TFLite division operations |
| GHSA-4c4g-crqm-xrxw | CVE-2021-37682 | MODERATE | Use of unitialized value in TFLite |
| GHSA-7xwj-5r4v-429p | CVE-2021-37681 | HIGH | NPE in TFLite |
| GHSA-cfpj-3q4c-jhvr | CVE-2021-37680 | MODERATE | Division by zero in TFLite |
| GHSA-g8wg-cjwc-xhhp | CVE-2021-37679 | HIGH | Heap OOB in nested `tf.map_fn` with `RaggedTensor`s |
| GHSA-r6jx-9g48-2r5r | CVE-2021-37678 | HIGH | Arbitrary code execution due to YAML deserialization |
| GHSA-qfpc-5pjr-mh26 | CVE-2021-37677 | MODERATE | Missing validation in shape inference for `Dequantize` |
| GHSA-v768-w7m9-2vmm | CVE-2021-37676 | HIGH | Reference binding to nullptr in shape inference |
| GHSA-9c8h-2mv3-49ww | CVE-2021-37675 | MODERATE | Division by 0 in most convolution operators |
| GHSA-7ghq-fvr3-pj2x | CVE-2021-37674 | MODERATE | Incomplete validation in `MaxPoolGrad` |
| GHSA-278g-rq84-9hmg | CVE-2021-37673 | MODERATE | `CHECK`-fail in `MapStage` |
| GHSA-5hj3-vjjf-f5m7 | CVE-2021-37672 | MODERATE | Heap OOB in `SdcaOptimizerV2` |
| GHSA-qr82-2c78-4m8h | CVE-2021-37671 | HIGH | Reference binding to nullptr in map operations |
| GHSA-9697-98pf-4rw7 | CVE-2021-37670 | MODERATE | Heap OOB in `UpperBound` and `LowerBound` |
| GHSA-vmjw-c2vp-p33c | CVE-2021-37669 | MODERATE | Crash in NMS ops caused by integer conversion to unsigned |
| GHSA-2wmv-37vq-52g5 | CVE-2021-37668 | MODERATE | FPE in `tf.raw_ops.UnravelIndex` |
| GHSA-w74j-v8xh-3w5h | CVE-2021-37667 | HIGH | Reference binding to nullptr in unicode encoding |
| GHSA-w4xf-2pqw-5mq7 | CVE-2021-37666 | HIGH | Reference binding to nullptr in `RaggedTensorToVariant` |
| GHSA-v82p-hv3v-p6qp | CVE-2021-37665 | HIGH | Incomplete validation in MKL requantization |
| GHSA-r4c4-5fpq-56wg | CVE-2021-37664 | HIGH | Heap OOB in boosted trees |
| GHSA-g25h-jr74-qp5j | CVE-2021-37663 | HIGH | Incomplete validation in `QuantizeV2` |
| GHSA-f5cx-5wr3-5qrc | CVE-2021-37662 | MODERATE | Reference binding to nullptr in boosted trees |
| GHSA-gf88-j2mg-cc82 | CVE-2021-37661 | MODERATE | Crash caused by integer conversion to unsigned |
| GHSA-cm5x-837x-jf3c | CVE-2021-37660 | MODERATE | Division by 0 in inplace operations |
| GHSA-q3g3-h9r4-prrc | CVE-2021-37659 | HIGH | Reference binding to nullptr and heap OOB in binary cwise ops |
| GHSA-6p5r-g9mq-ggh2 | CVE-2021-37658 | MODERATE | Reference binding to nullptr in `MatrixSetDiagV*` ops |
| GHSA-5xwc-mrhx-5g3m | CVE-2021-37657 | MODERATE | Reference binding to nullptr in `MatrixDiagV*` ops |
| GHSA-4xfp-4pfp-89wg | CVE-2021-37656 | MODERATE | Reference binding to nullptr in `RaggedTensorToSparse` |
| GHSA-7fvx-3jfc-2cpc | CVE-2021-37655 | HIGH | Heap OOB in `ResourceScatterUpdate` |
| GHSA-2r8p-fg3c-wcj4 | CVE-2021-37654 | HIGH | Heap OOB and CHECK fail in `ResourceGather` |
| GHSA-qjj8-32p7-h289 | CVE-2021-37653 | MODERATE | Division by 0 in `ResourceGather` |
| GHSA-m7fm-4jfh-jrg6 | CVE-2021-37652 | HIGH | Use after free in boosted trees creation |
| GHSA-hpv4-7p9c-mvfr | CVE-2021-37651 | HIGH | Heap buffer overflow in `FractionalAvgPoolGrad` |
| GHSA-6gv8-p3vj-pxvr | CVE-2021-37649 | HIGH | Null pointer dereference in `UncompressElement` |
| GHSA-f8h4-7rgh-q2gm | CVE-2021-37650 | HIGH | Segfault and heap buffer overflow in `{Experimental,}DatasetToTFRecord` |
| GHSA-wp77-4gmm-7cq8 | CVE-2021-37648 | HIGH | Incorrect validation of `SaveV2` inputs |
| GHSA-c5x2-p679-95wc | CVE-2021-37647 | HIGH | Null pointer dereference in `SparseTensorSliceDataset` |
| GHSA-h6jh-7gv5-28vg | CVE-2021-37646 | MODERATE | Bad alloc in `StringNGrams` caused by integer conversion |
| GHSA-ch4f-829c-v5pw | CVE-2021-37642 | MODERATE | Division by 0 in `ResourceScatterDiv` |
| GHSA-cgfm-62j4-v4rf | CVE-2021-37635 | HIGH | Heap out of bounds access in sparse reduction operations |
| GHSA-hp4c-x6r7-6555 | CVE-2021-37636 | MODERATE | Floating point exception in `SparseDenseCwiseDiv` |
| GHSA-c9qf-r67m-p7cg | CVE-2021-37637 | HIGH | Null pointer dereference in `CompressElement` |
| GHSA-hwr7-8gxx-fj5p | CVE-2021-37638 | HIGH | Null pointer dereference in `RaggedTensorToTensor` |
| GHSA-gh6x-4whr-2qv4 | CVE-2021-37639 | HIGH | Null pointer dereference and heap OOB read in operations restoring tensors |
| GHSA-95xm-g58g-3p88 | CVE-2021-37640 | MODERATE | Integer division by 0 in sparse reshaping |
| GHSA-wvjw-p9f5-vq28 | CVE-2021-29619 | LOW | Segfault in `tf.raw_ops.SparseCountSparseOutput` |
| GHSA-xqfj-cr6q-pc8w | CVE-2021-29618 | LOW | Crash in `tf.transpose` with complex inputs |
| GHSA-mmq6-q8r3-48fm | CVE-2021-29617 | LOW | Crash in `tf.strings.substr` due to `CHECK`-fail |
| GHSA-4hvv-7x94-7vq8 | CVE-2021-29616 | LOW | Null dereference in Grappler's `TrySimplify` |
| GHSA-qw5h-7f53-xrp6 | CVE-2021-29615 | LOW | Stack overflow in `ParseAttrValue` with nested tensors |
| GHSA-8pmx-p244-g88h | CVE-2021-29614 | MODERATE | Interpreter crash from `tf.io.decode_raw` |
| GHSA-vvg4-vgrv-xfr7 | CVE-2021-29613 | MODERATE | Incomplete validation in `tf.raw_ops.CTCLoss` |
| GHSA-9c8h-vvrj-w2p8 | CVE-2021-37641 | MODERATE | Heap OOB in `RaggedGather` |
| GHSA-mq5c-prh3-3f3h | CVE-2021-29610 | LOW | Invalid validation in `QuantizeAndDequantizeV2` |
| GHSA-9rpc-5v9q-5r7f | CVE-2021-29611 | LOW | Incomplete validation in `SparseReshape` |
| GHSA-fcwc-p4fc-c5cc | CVE-2021-37643 | HIGH | Null pointer dereference in `MatrixDiagPartOp` |
| GHSA-cjc7-49v2-jp64 | CVE-2021-29609 | MODERATE | Incomplete validation in `SparseAdd` |
| GHSA-rgvq-pcvf-hx75 | CVE-2021-29608 | MODERATE | Heap OOB and null pointer dereference in `RaggedTensorToTensor` |
| GHSA-27j5-4p9v-pp67 | CVE-2021-37644 | MODERATE | `std::abort` raised from `TensorListReserve` |
| GHSA-gv26-jpj9-c8gq | CVE-2021-29607 | MODERATE | Incomplete validation in `SparseSparseMinimum` |
| GHSA-h4pc-gx2w-f2xv | CVE-2021-29606 | MODERATE | Heap OOB read in TFLite |
| GHSA-jf7h-7m85-w2v2 | CVE-2021-29605 | MODERATE | Integer overflow in TFLite memory allocation |
| GHSA-8rm6-75mf-7r7r | CVE-2021-29604 | LOW | Division by zero in TFLite's implementation of hashtable lookup |
| GHSA-crch-j389-5f84 | CVE-2021-29603 | LOW | Heap OOB write in TFLite |
| GHSA-rf3h-xgv5-2q39 | CVE-2021-29602 | LOW | Division by zero in TFLite's implementation of `DepthwiseConv` |
| GHSA-9c84-4hx6-xmm4 | CVE-2021-29601 | MODERATE | Integer overflow in TFLite concatentation |
| GHSA-j8qh-3xrq-c825 | CVE-2021-29600 | LOW | Division by zero in TFLite's implementation of `OneHot` |
| GHSA-97wf-p777-86jq | CVE-2021-29599 | LOW | Division by zero in TFLite's implementation of Split |
| GHSA-pmpr-55fj-r229 | CVE-2021-29598 | LOW | Division by zero in TFLite's implementation of `SVDF` |
| GHSA-v52p-hfjf-wg88 | CVE-2021-29597 | LOW | Division by zero in TFLite's implementation of `SpaceToBatchNd` |
| GHSA-4vrf-ff7v-hpgr | CVE-2021-29596 | LOW | Division by zero in TFLite's implementation of `EmbeddingLookup` |
| GHSA-9w2p-5mgw-p94c | CVE-2021-37645 | MODERATE | Integer overflow due to conversion to unsigned |
| GHSA-vf94-36g5-69v8 | CVE-2021-29595 | LOW | Division by zero in TFLite's implementation of `DepthToSpace` |
| GHSA-2xgj-xhgf-ggjv | CVE-2021-29612 | LOW | Heap buffer overflow in `BandedTriangularSolve` |
| GHSA-j7rm-8ww4-xx2g | CVE-2021-29587 | LOW | Division by zero in TFLite's implementation of `SpaceToDepth` |
| GHSA-26j7-6w8w-7922 | CVE-2021-29586 | LOW | Division by zero in optimized pooling implementations in TFLite |
| GHSA-mv78-g7wq-mhp4 | CVE-2021-29585 | LOW | Division by zero in padding computation in TFLite |
| GHSA-xvjm-fvxx-q3hv | CVE-2021-29584 | LOW | CHECK-fail due to integer overflow |
| GHSA-9xh4-23q4-v6wr | CVE-2021-29583 | LOW | Heap buffer overflow and undefined behavior in `FusedBatchNorm` |
| GHSA-pqrv-8r2f-7278 | CVE-2022-23590 | MODERATE | Crash due to erroneous `StatusOr` in TensorFlow |
| GHSA-9px9-73fg-3fqp | CVE-2022-23589 | MODERATE | Null pointer dereference in Grappler's `IsConstant` |
| GHSA-fx5c-h9f6-rv7c | CVE-2022-23588 | MODERATE | `CHECK`-fails due to attempting to build a reference tensor |
| GHSA-8jj7-5vxc-pg2q | CVE-2022-23587 | HIGH | Integer overflow in TensorFlow |
| GHSA-43jf-985q-588j | CVE-2022-23586 | MODERATE | Multiple `CHECK`-fails in `function.cc` in TensowFlow |
| GHSA-fq6p-6334-8gr4 | CVE-2022-23585 | MODERATE | Memory leak in decoding PNG images |
| GHSA-24x4-6qmh-88qg | CVE-2022-23584 | HIGH | Use after free in `DecodePng` kernel |
| GHSA-gjqc-q9g6-q2j3 | CVE-2022-23583 | MODERATE | `CHECK`-failures in binary ops in Tensorflow |
| GHSA-4j82-5ccr-4r8v | CVE-2022-23582 | MODERATE | `CHECK`-failures in `TensorByteSize` in Tensorflow |
| GHSA-fq86-3f29-px2c | CVE-2022-23581 | MODERATE | `CHECK`-failures during Grappler's `IsSimplifiableReshape` in Tensorflow |
| GHSA-627q-g293-49q7 | CVE-2022-23580 | MODERATE | Abort caused by allocating a vector that is too large in Tensorflow |
| GHSA-5f2r-qp73-37mr | CVE-2022-23579 | MODERATE | `CHECK`-failures during Grappler's `SafeToRemoveIdentity` in Tensorflow |
| GHSA-57wx-m983-2f88 | CVE-2021-41208 | CRITICAL | Incomplete validation in boosted trees code |
| GHSA-6hpv-v2rx-c5g6 | CVE-2021-41209 | MODERATE | FPE in convolutions with zero size filters |
| GHSA-m342-ff57-4jcc | CVE-2021-41210 | HIGH | Heap OOB read in `tf.raw_ops.SparseCountSparseOutput` |
| GHSA-cvgx-3v3q-m36c | CVE-2021-41211 | HIGH | Heap OOB in shape inference for `QuantizeV2` |
| GHSA-fr77-rrx3-cp7g | CVE-2021-41212 | HIGH | Heap OOB read in `tf.ragged.cross` |
| GHSA-h67m-xg8f-fxcf | CVE-2021-41213 | MODERATE | Deadlock in mutually recursive `tf.function` objects |
| GHSA-vwhq-49r4-gj9v | CVE-2021-41214 | HIGH | Reference binding to `nullptr` in `tf.ragged.cross` |
| GHSA-x3v8-c8qx-3j3r | CVE-2021-41215 | MODERATE | Null pointer exception in `DeserializeSparse` |
| GHSA-3ff2-r28g-w7h9 | CVE-2021-41216 | MODERATE | Heap buffer overflow in `Transpose` |
| GHSA-5crj-c72x-m7gq | CVE-2021-41217 | MODERATE | Null pointer exception when `Exit` node is not preceded by `Enter` op |
| GHSA-9crf-c6qr-r273 | CVE-2021-41218 | MODERATE | Integer division by 0 in `tf.raw_ops.AllToAll` |
| GHSA-4f99-p9c2-3j8x | CVE-2021-41219 | HIGH | Undefined behavior via `nullptr` reference binding in sparse matrix multiplication |
| GHSA-gpfh-jvf9-7wg5 | CVE-2021-41220 | HIGH | Use after free / memory leak in `CollectiveReduceV2` |
| GHSA-cqv6-3phm-hcwx | CVE-2021-41221 | HIGH | Access to invalid memory during shape inference in `Cudnn*` ops |
| GHSA-5hx2-qx8j-qjqm | CVE-2021-41199 | MODERATE | Overflow/crash in `tf.image.resize` when size is large |
| GHSA-c45w-2wxr-pp53 | CVE-2021-29582 | LOW | Heap OOB read in `tf.raw_ops.Dequantize` |
| GHSA-vq2r-5xvm-3hc3 | CVE-2021-29581 | LOW | Segfault in `CTCBeamSearchDecoder` |
| GHSA-x8h6-xgqx-jqgp | CVE-2021-29580 | LOW | Undefined behavior and `CHECK`-fail in `FractionalMaxPoolGrad` |
| GHSA-79fv-9865-4qcv | CVE-2021-29579 | LOW | Heap buffer overflow in `MaxPoolGrad` |
| GHSA-6f89-8j54-29xf | CVE-2021-29578 | LOW | Heap buffer overflow in `FractionalAvgPoolGrad` |
| GHSA-v6r6-84gr-92rm | CVE-2021-29577 | LOW | Heap buffer overflow in `AvgPool3DGrad` |
| GHSA-7cqx-92hp-x6wh | CVE-2021-29576 | LOW | Heap buffer overflow in `MaxPool3DGradGrad` |
| GHSA-6qgm-fv6v-rfpv | CVE-2021-29575 | LOW | Overflow/denial of service in `tf.raw_ops.ReverseSequence` |
| GHSA-9vpm-rcf4-9wqw | CVE-2021-29573 | LOW | Division by 0 in `MaxPoolGradWithArgmax` |
| GHSA-545v-42p7-98fq | CVE-2021-29570 | LOW | Heap out of bounds read in `MaxPoolGradWithArgmax` |
| GHSA-3h8m-483j-7xxm | CVE-2021-29569 | LOW | Heap out of bounds read in `RequantizationRange` |
| GHSA-4p4p-www8-8fv9 | CVE-2021-29568 | LOW | Reference binding to null in `ParameterizedTruncatedNormal` |
| GHSA-wp3c-xw9g-gpcg | CVE-2021-29567 | LOW | Lack of validation in `SparseDenseCwiseMul` |
| GHSA-pvrc-hg3f-58r6 | CVE-2021-29566 | LOW | Heap OOB access in `Dilation2DBackpropInput` |
| GHSA-r6pg-pjwc-j585 | CVE-2021-29565 | LOW | Null pointer dereference in `SparseFillEmptyRows` |
| GHSA-8gv3-57p6-g35r | CVE-2021-29560 | LOW | Heap buffer overflow in `RaggedTensorToTensor` |
| GHSA-gvm4-h8j3-rjrq | CVE-2021-29561 | LOW | CHECK-fail in `LoadAndRemapMatrix` |
| GHSA-36vm-xw34-x4pj | CVE-2021-29562 | LOW | CHECK-fail in `tf.raw_ops.IRFFT` |
| GHSA-ph87-fvjr-v33w | CVE-2021-29563 | LOW | CHECK-fail in `tf.raw_ops.RFFT` |
| GHSA-75f6-78jr-4656 | CVE-2021-29564 | LOW | Null pointer dereference in `EditDistance` |
| GHSA-vfr4-x8j2-3rf9 | CVE-2021-29588 | LOW | Division by zero in TFLite's implementation of `TransposeConv` |
| GHSA-jjr8-m8g8-p6wv | CVE-2021-29592 | MODERATE | Null pointer dereference in TFLite's `Reshape` operator |
| GHSA-cfx7-2xpc-8w4h | CVE-2021-29593 | LOW | Division by zero in TFLite's implementation of `BatchToSpaceNd` |
| GHSA-3qgw-p4fm-x7gf | CVE-2021-29594 | LOW | Division by zero in TFLite's convolution code |
| GHSA-cwv3-863g-39vx | CVE-2021-29591 | HIGH | Stack overflow due to looping TFLite subgraph |
| GHSA-whr9-vfh2-7hm6 | CVE-2021-29571 | LOW | Memory corruption in `DrawBoundingBoxesV2` |
| GHSA-5gqf-456p-4836 | CVE-2021-29572 | LOW | Reference binding to nullptr in `SdcaOptimizer` |
| GHSA-3w67-q784-6w7c | CVE-2021-29589 | LOW | Division by zero in TFLite's implementation of `GatherNd` |
| GHSA-828x-qc2p-wprq | CVE-2021-29574 | LOW | Undefined behavior in `MaxPool3DGradGrad` |
| GHSA-xw93-v57j-fcgh | CVE-2021-29557 | LOW | Division by 0 in `SparseMatMul` |
| GHSA-fxqh-cfjm-fp93 | CVE-2021-29556 | LOW | Division by 0 in `Reverse` |
| GHSA-r35g-4525-29fq | CVE-2021-29555 | LOW | Division by 0 in `FusedBatchNorm` |
| GHSA-mqh2-9wrp-vx84 | CVE-2021-29558 | LOW | Heap buffer overflow in `SparseSplit` |
| GHSA-59q2-x2qc-4c97 | CVE-2021-29559 | LOW | Heap OOB access in unicode ops |
| GHSA-qg48-85hg-mqc5 | CVE-2021-29554 | LOW | Division by 0 in `DenseCountSparseOutput` |
| GHSA-h9px-9vqg-222h | CVE-2021-29553 | LOW | Heap OOB in `QuantizeAndDequantizeV3` |
| GHSA-jhq9-wm9m-cf89 | CVE-2021-29552 | LOW | CHECK-failure in `UnsortedSegmentJoin` |
| GHSA-vqw6-72r7-fgw7 | CVE-2021-29551 | LOW | OOB read in `MatrixTriangularSolve` |
| GHSA-x83m-p7pv-ch8v | CVE-2021-29549 | LOW | Division by 0 in `QuantizedAdd` |
| GHSA-p45v-v4pw-77jr | CVE-2021-29548 | LOW | Division by 0 in `QuantizedBatchNormWithGlobalNormalization` |
| GHSA-4fg4-p75j-w5xj | CVE-2021-29547 | LOW | Heap out of bounds in `QuantizedBatchNormWithGlobalNormalization` |
| GHSA-fphq-gw9m-ghrv | CVE-2021-29543 | LOW | CHECK-fail in `CTCGreedyDecoder` |
| GHSA-4hrh-9vmp-2jgg | CVE-2021-29542 | LOW | Heap buffer overflow in `StringNGrams` |
| GHSA-xqfj-35wv-m3cr | CVE-2021-29541 | LOW | Null pointer dereference in `StringNGrams` |
| GHSA-xgc3-m89p-vr3x | CVE-2021-29540 | LOW | Heap buffer overflow in `Conv2DBackpropFilter` |
| GHSA-g4h2-gqm3-c9wq | CVE-2021-29539 | LOW | Segfault in tf.raw_ops.ImmutableConst |
| GHSA-j8qc-5fqr-52fp | CVE-2021-29538 | LOW | Division by zero in `Conv2DBackpropFilter` |
| GHSA-8c89-2vwr-chcq | CVE-2021-29537 | LOW | Heap buffer overflow in `QuantizedResizeBilinear` |
| GHSA-2gfx-95x2-5v3x | CVE-2021-29536 | LOW | Heap buffer overflow in `QuantizedReshape` |
| GHSA-m3f9-w3p3-p669 | CVE-2021-29535 | LOW | Heap buffer overflow in `QuantizedMul` |
| GHSA-6j9c-grc6-5m6g | CVE-2021-29534 | LOW | CHECK-fail in SparseConcat |
| GHSA-393f-2jr3-cp69 | CVE-2021-29533 | LOW | CHECK-fail in DrawBoundingBoxes |
| GHSA-j47f-4232-hvv8 | CVE-2021-29532 | LOW | Heap out of bounds read in `RaggedCross` |
| GHSA-3qxp-qjq7-w4hf | CVE-2021-29531 | LOW | CHECK-fail in tf.raw_ops.EncodePng |
| GHSA-xcwj-wfcm-m23c | CVE-2021-29530 | LOW | Invalid validation in `SparseMatrixSparseCholesky` |
| GHSA-jfp7-4j67-8r3q | CVE-2021-29529 | LOW | Heap buffer overflow caused by rounding |
| GHSA-6f84-42vf-ppwp | CVE-2021-29528 | LOW | Division by 0 in `QuantizedMul` |
| GHSA-x4g7-fvjj-prg8 | CVE-2021-29527 | LOW | Division by 0 in `QuantizedConv2D` |
| GHSA-r4pj-74mg-8868 | CVE-2021-29524 | LOW | Division by 0 in `Conv2DBackpropFilter` |
| GHSA-f78g-q7r4-9wcv | CVE-2021-29550 | LOW | Division by 0 in `FractionalAvgPool` |
| GHSA-2cpx-427x-q2c6 | CVE-2021-29523 | LOW | CHECK-fail in AddManySparseToTensorsMap |
| GHSA-452g-f7fp-9jf7 | CVE-2021-29513 | LOW | Type confusion during tensor casts lead to dereferencing null pointers |
| GHSA-62gx-355r-9fhg | CVE-2021-29518 | LOW | Session operations in eager mode lead to null pointer dereferences |
| GHSA-wcv5-qrj6-9pfm | CVE-2021-29520 | LOW | Heap buffer overflow in `Conv3DBackprop*` |
| GHSA-c968-pq7h-7fxv | CVE-2021-29522 | LOW | Division by 0 in `Conv3DBackprop*` |
| GHSA-m34j-p8rj-wjxq | CVE-2021-29546 | LOW | Division by 0 in `QuantizedBiasAdd` |
| GHSA-xm2v-8rrw-w9pm | CVE-2021-29525 | LOW | Division by 0 in `Conv2DBackpropInput` |
| GHSA-4vf2-4xcg-65cx | CVE-2021-29526 | LOW | Division by 0 in `Conv2D` |
| GHSA-hmg3-c7xj-6qwm | CVE-2021-29545 | LOW | Heap buffer overflow in `SparseTensorToCSRSparseMatrix` |
| GHSA-hc6c-75p4-hmq4 | CVE-2021-29515 | LOW | Reference binding to null pointer in `MatrixDiag*` ops |
| GHSA-8h46-5m9h-7553 | CVE-2021-29514 | LOW | Heap out of bounds write in `RaggedBinCount` |
| GHSA-4278-2v5v-65r4 | CVE-2021-29512 | LOW | Heap buffer overflow in `RaggedBinCount` |
| GHSA-977j-xj7q-2jr9 | CVE-2020-5215 | LOW | Segmentation faultin TensorFlow when converting a Python string to `tf.float16` |
| GHSA-q263-fvxm-m5mw | CVE-2020-26271 | MODERATE | Heap out of bounds access in MakeEdge in TensorFlow |
| GHSA-9jjw-hf72-3mxw | CVE-2020-26269 | HIGH | TensorFlow vulnerable to heap out of bounds read in filesystem glob matching |
| GHSA-rrfp-j2mp-hq9c | CVE-2020-15265 | HIGH | Segfault in `tf.quantization.quantize_and_dequantize` |
| GHSA-hx2x-85gr-wrpq | CVE-2020-15212 | CRITICAL | Out of bounds access in tensorflow-lite |
| GHSA-mxjj-953w-2c2v | CVE-2020-15208 | HIGH | Data corruption in tensorflow-lite |
| GHSA-q4qf-3fc6-8x34 | CVE-2020-15207 | MODERATE | Segfault and data corruption in tensorflow-lite |
| GHSA-g7p5-5759-qv46 | CVE-2020-15205 | MODERATE | Data leak in Tensorflow |
| GHSA-84mw-34w6-2q43 | CVE-2021-29516 | LOW | Null pointer dereference via invalid Ragged Tensors |
| GHSA-772p-x54p-hjrv | CVE-2021-29517 | LOW | Division by zero in `Conv3D` |
| GHSA-xmq7-7fxm-rr79 | CVE-2020-15203 | HIGH | Denial of Service in Tensorflow |
| GHSA-772j-h9xw-ffp5 | CVE-2021-29519 | LOW | CHECK-fail in SparseCross due to type confusion |
| GHSA-h6fg-mjxg-hqq4 | CVE-2020-15202 | HIGH | Integer truncation in Shard API usage |
| GHSA-hr84-fqvp-48mm | CVE-2021-29521 | LOW | Segfault in SparseCountSparseOutput |
| GHSA-p5f8-gfw5-33w4 | CVE-2020-15201 | MODERATE | Heap buffer overflow in Tensorflow |
| GHSA-x7rp-74x2-mjf3 | CVE-2020-15200 | HIGH | Segfault in Tensorflow |
| GHSA-qc53-44cj-vfvx | CVE-2020-15197 | MODERATE | Denial of Service in Tensorflow |
| GHSA-63xm-rx5p-xvqr | CVE-2020-15195 | MODERATE | Heap buffer overflow in Tensorflow |
| GHSA-rjjg-hgv6-h69v | CVE-2020-15193 | HIGH | Memory corruption in Tensorflow |
| GHSA-pg59-2f92-5cph | CVE-2020-15196 | MODERATE | Heap buffer overflow in Tensorflow |
| GHSA-m648-33qf-v3gp | CVE-2020-26270 | MODERATE | CHECK-fail in LSTM with zero-length input in TensorFlow |
| GHSA-8fxw-76px-3rxv | CVE-2020-15192 | MODERATE | Memory leak in Tensorflow |
| GHSA-4g9f-63rx-5cw4 | CVE-2020-15190 | MODERATE | Segfault in Tensorflow |
| GHSA-hhvc-g5hv-48c6 | CVE-2020-26268 | MODERATE | Write to immutable memory region in TensorFlow |
| GHSA-9mqp-7v2h-2382 | CVE-2020-15194 | MODERATE | Denial of Service in Tensorflow |
| GHSA-c9f3-9wfr-wgh7 | CVE-2020-26267 | LOW | Lack of validation in data format attributes in TensorFlow |
| GHSA-jc87-6vpp-7ff3 | CVE-2020-15198 | MODERATE | Heap buffer overflow in Tensorflow |
| GHSA-qhxx-j73r-qpm2 | CVE-2020-26266 | MODERATE | Uninitialized memory access in TensorFlow |
| GHSA-p2cq-cprg-frvm | CVE-2020-15214 | CRITICAL | Out of bounds write in tensorflow-lite |
| GHSA-hjmq-236j-8m87 | CVE-2020-15213 | MODERATE | Denial of service in tensorflow-lite |
| GHSA-cvpc-8phh-8f45 | CVE-2020-15211 | MODERATE | Out of bounds access in tensorflow-lite |
| GHSA-x9j7-x98r-r4w2 | CVE-2020-15210 | HIGH | Segmentation fault in tensorflow-lite |
| GHSA-w5gh-2wr2-pm6g | CVE-2020-15206 | HIGH | Denial of Service in Tensorflow |
| GHSA-q8qj-fc9q-cphr | CVE-2020-15191 | MODERATE | Undefined behavior in Tensorflow |
| GHSA-x5cp-9pcf-pp3h | CVE-2020-15199 | HIGH | Denial of Service in Tensorflow |
| GHSA-qh32-6jjc-qprm | CVE-2020-15209 | HIGH | Null pointer dereference in tensorflow-lite |
| GHSA-q8gv-q7wr-9jf8 | CVE-2020-15204 | MODERATE | Segfault in Tensorflow |
| GHSA-xwhf-g6j5-j5gc | CVE-2020-15266 | MODERATE | Float cast overflow undefined behavior |
| GHSA-844w-j86r-4x2j | CVE-2019-16778 | LOW | Heap buffer overflow in `UnsortedSegmentSum` in TensorFlow |
| GHSA-gjh7-xx4r-x345 | CVE-2023-33976 | HIGH | TensorFlow has segfault in array_ops.upper_bound |
| GHSA-fxgc-95xx-grvq | CVE-2023-25661 | MODERATE | TensorFlow Denial of Service vulnerability |
| GHSA-7x4v-9gxg-9hwj | CVE-2023-25675 | HIGH | TensorFlow has Segfault in Bincount with XLA |
| GHSA-j5w9-hmfh-4cr6 | CVE-2023-25671 | HIGH | TensorFlow has segmentation fault in tfg-translate |
| GHSA-5w96-866f-6rm8 | CVE-2023-27579 | HIGH | TensorFlow has Floating Point Exception in TFLite in conv kernel |
| GHSA-gf97-q72m-7579 | CVE-2023-25674 | HIGH | TensorFlow has Null Pointer Error in RandomShuffle with XLA enable |
| GHSA-49rq-hwc3-x77w | CVE-2023-25670 | HIGH | TensorFlow has Null Pointer Error in QuantizedMatMulWithBiasAndDequantize |
| GHSA-rcf8-g8jv-vg6p | CVE-2023-25669 | HIGH | TensorFlow has Floating Point Exception in AvgPoolGrad with XLA |
| GHSA-647v-r7qq-24fh | CVE-2023-25673 | HIGH | TensorFlow has Floating Point Exception in TensorListSplit with XLA |
| GHSA-94mm-g2mv-8p7r | CVE-2023-25672 | HIGH | TensorFlow has Null Pointer Error in LookupTableImportV2 |
| GHSA-gw97-ff7c-9v96 | CVE-2023-25668 | CRITICAL | TensorFlow has a heap out-of-buffer read vulnerability in the QuantizeAndDequantize operation |
| GHSA-fqm2-gh8w-gr68 | CVE-2023-25667 | MODERATE | TensorFlow vulnerable to segfault when opening multiframe gif |
| GHSA-f637-vh3r-vfh2 | CVE-2023-25666 | HIGH | TensorFlow has Floating Point Exception in AudioSpectrogram |
| GHSA-558h-mq8x-7q9g | CVE-2023-25665 | HIGH | TensorFlow has Null Pointer Error in SparseSparseMaximum |
| GHSA-6hg6-5c2q-7rcr | CVE-2023-25664 | HIGH | TensorFlow has Heap-buffer-overflow in AvgPoolGrad |
| GHSA-64jg-wjww-7c5w | CVE-2023-25663 | HIGH | TensorFlow has Null Pointer Error in TensorArrayConcatV2 |
| GHSA-7jvm-xxmr-v5cw | CVE-2023-25662 | HIGH | TensorFlow vulnerable to integer overflow in EditDistance |
| GHSA-qjqc-vqcf-5qvj | CVE-2023-25660 | HIGH | TensorFlow vulnerable to seg fault in `tf.raw_ops.Print` |
| GHSA-93vr-9q9m-pj8p | CVE-2023-25659 | HIGH | TensorFlow vulnerable to Out-of-Bounds Read in DynamicStitch |
| GHSA-68v3-g9cm-rmm6 | CVE-2023-25658 | HIGH | TensorFlow vulnerable to Out-of-Bounds Read in GRUBlockCellGrad |
| GHSA-6wfh-89q8-44jq | CVE-2023-25676 | HIGH | TensorFlow has null dereference on ParallelConcat with XLA |
| GHSA-f49c-87jh-g47q | CVE-2023-25801 | HIGH | TensorFlow has double free in Fractional(Max/Avg)Pool |
| GHSA-8w5g-3wcv-9g2j | CVE-2022-41880 | MODERATE | Tensorflow vulnerable to Out-of-Bounds Read |
| GHSA-cqvq-fvhr-v6hc | - | LOW | `CHECK` failure in `SobolSample` via missing validation |
| GHSA-xf83-q765-xm6m | - | LOW | `CHECK` fail in `TensorListScatter` and `TensorListScatterV2` in eager mode |
| GHSA-pf36-r9c6-h97j | CVE-2022-41911 | MODERATE | Invalid char to bool conversion when printing a tensor |
| GHSA-frqp-wp83-qggv | CVE-2022-41910 | MODERATE | Heap overflow in `QuantizeAndDequantizeV2` |
| GHSA-rjx6-v474-2ch9 | CVE-2022-41909 | MODERATE | Segfault in `CompositeTensorVariantToComponents` |
| GHSA-mv77-9g28-cwg3 | CVE-2022-41908 | MODERATE | `CHECK` fail via inputs in `PyFunc` |
| GHSA-368v-7v32-52fx | CVE-2022-41907 | MODERATE | Overflow in `ResizeNearestNeighborGrad` |
| GHSA-cg88-rpvp-cjv5 | CVE-2022-41902 | HIGH | Out of bounds write in grappler in Tensorflow |
| GHSA-g9fm-r5mm-rf9f | CVE-2022-41901 | MODERATE | `CHECK_EQ` fail via input in `SparseMatrixNNZ` |
| GHSA-xvwp-h6jv-7472 | CVE-2022-41900 | HIGH | FractionalMaxPool and FractionalAVGPool heap out-of-bounds acess |
| GHSA-27rc-728f-x5w2 | CVE-2022-41899 | MODERATE | `CHECK` fail via inputs in `SdcaOptimizer` |
| GHSA-hq7g-wwwp-q46h | CVE-2022-41898 | MODERATE | `CHECK` fail via inputs in `SparseFillEmptyRowsGrad` |
| GHSA-f2w8-jw48-fr7j | CVE-2022-41897 | MODERATE | `FractionalMaxPoolGrad` Heap out of bounds read |
| GHSA-rmg2-f698-wq35 | CVE-2022-41896 | MODERATE | `tf.raw_ops.Mfcc` crashes |
| GHSA-gq2j-cr96-gvqx | CVE-2022-41895 | MODERATE | `MirrorPadGrad` heap out of bounds read |
| GHSA-67pf-62xr-q35m | CVE-2022-41893 | MODERATE | `CHECK_EQ` fail in `tf.raw_ops.TensorListResize` |
| GHSA-66vq-54fq-6jvv | CVE-2022-41891 | MODERATE | Segfault in `tf.raw_ops.TensorListConcat` |
| GHSA-h246-cgh4-7475 | CVE-2022-41890 | MODERATE | `CHECK` fail in `BCast` overflow |
| GHSA-6x99-gv2v-q76v | CVE-2022-41888 | MODERATE | FPE in `tf.image.generate_bounding_box_proposals` |
| GHSA-8fvv-46hw-vpg3 | CVE-2022-41887 | MODERATE | Overflow in `tf.keras.losses.poisson` |
| GHSA-54pp-c6pp-7fpx | CVE-2022-41886 | MODERATE | Overflow in `ImageProjectiveTransformV2` |
| GHSA-762h-vpvw-3rcx | CVE-2022-41885 | MODERATE | Overflow in `FusedResizeAndPadConv2D` |
| GHSA-jq6x-99hj-q636 | CVE-2022-41884 | MODERATE | Seg fault in `ndarray_tensor_bridge` due to zero and large inputs |
| GHSA-w58w-79xv-6vcj | CVE-2022-41883 | MODERATE | Out of bounds segmentation fault due to unequal op inputs in Tensorflow |
| GHSA-fqxc-pvf8-2w9v | CVE-2022-36000 | MODERATE | TensorFlow vulnerable to null dereference on MLIR on empty function attributes |
| GHSA-wqmc-pm8c-2jhc | CVE-2022-36017 | MODERATE | TensorFlow vulnerable to segfault in `Requantize` |
| GHSA-ffjm-4qwc-7cmf | CVE-2022-35939 | HIGH | TensorFlow vulnerable to OOB write in `scatter_nd` in TF Lite |
| GHSA-x989-q2pq-4q5x | CVE-2022-35940 | MODERATE | TensorFlow vulnerable to Int overflow in `RaggedRangeOp` |
| GHSA-mgmh-g2v6-mqw5 | CVE-2022-35941 | MODERATE | TensorFlow vulnerable to `CHECK` failure in `AvgPoolOp` |
| GHSA-vm7x-4qhj-rrcq | CVE-2022-35991 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `TensorListScatter` and `TensorListScatterV2` |
| GHSA-9v8w-xmr4-wgxp | CVE-2022-35992 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `TensorListFromTensor` |
| GHSA-wq6q-6m32-9rv9 | CVE-2022-35993 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `SetSize` |
| GHSA-h7ff-cfc9-wmmh | CVE-2022-35990 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `FakeQuantWithMinMaxVarsPerChannelGradient` |
| GHSA-9cr2-8pwr-fhfq | CVE-2022-36026 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `QuantizeAndDequantizeV3` |
| GHSA-9j4v-pp28-mxv7 | CVE-2022-36019 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `FakeQuantWithMinMaxVarsPerChannel` |
| GHSA-m6cv-4fmf-66xf | CVE-2022-36018 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `RaggedTensorToVariant` |
| GHSA-f4w6-h4f5-wx45 | CVE-2022-35934 | MODERATE | TensorFlow vulnerable to `CHECK` failure in tf.reshape via overflows |
| GHSA-97p7-w86h-vcf9 | CVE-2022-35935 | MODERATE | TensorFlow vulnerable to `CHECK` failure in `SobolSample` via missing validation |
| GHSA-h5vq-gw2c-pq47 | CVE-2022-35952 | MODERATE | TensorFlow vulnerable to `CHECK` failures in `UnbatchGradOp` |
| GHSA-pxrw-j2fv-hx3h | CVE-2022-35937 | HIGH | TensorFlow vulnerable to OOB read in `Gather_nd` in TF Lite |
| GHSA-397c-5g2j-qxpv | CVE-2022-35982 | MODERATE | TensorFlow vulnerable to segfault in `SparseBincount` |
| GHSA-m6vp-8q9j-whx4 | CVE-2022-35983 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `Save` and `SaveSlices` |
| GHSA-p2xf-8hgm-hpw5 | CVE-2022-35984 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `ParameterizedTruncatedNormal` |
| GHSA-9942-r22v-78cp | CVE-2022-35985 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `LRNGrad` |
| GHSA-w62h-8xjm-fv49 | CVE-2022-35987 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `DenseBincount` |
| GHSA-9vqj-64pv-w55c | CVE-2022-35988 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `tf.linalg.matrix_rank` |
| GHSA-j43h-pgmg-5hjq | CVE-2022-35989 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `MaxPool` |
| GHSA-9fpg-838v-wpv7 | CVE-2022-35971 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `FakeQuantWithMinMaxVars` |
| GHSA-wxjj-cgcx-r3vq | CVE-2022-35959 | MODERATE | TensorFlow vulnerable to `CHECK` failures in `AvgPool3DGrad` |
| GHSA-4pc4-m9mj-v2r9 | CVE-2022-35972 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedBiasAdd` |
| GHSA-v5xg-3q2c-c2r4 | CVE-2022-35960 | MODERATE | TensorFlow vulnerable to `CHECK` failure in `TensorListReserve` via missing validation |
| GHSA-689c-r7h2-fv9v | CVE-2022-35973 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedMatMul` |
| GHSA-vgvh-2pf4-jr2x | CVE-2022-35974 | MODERATE | TensorFlow vulnerable to segfault in `QuantizeDownAndShrinkRange` |
| GHSA-v7vw-577f-vp8x | CVE-2022-35979 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedRelu` and `QuantizedRelu6` |
| GHSA-vxv8-r8q2-63xw | CVE-2022-35981 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `FractionalMaxPoolGrad` |
| GHSA-q2c3-jpmc-gfjx | CVE-2022-35969 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `Conv2DBackpropInput` |
| GHSA-84jm-4cf3-9jfm | CVE-2022-35963 | MODERATE | TensorFlow vulnerable to `CHECK` failures in `FractionalAvgPoolGrad` |
| GHSA-f7r5-q7cx-h668 | CVE-2022-35964 | MODERATE | TensorFlow vulnerable to segfault in `BlockLSTMGradV2` |
| GHSA-qxpx-j395-pw36 | CVE-2022-35965 | MODERATE | TensorFlow vulnerable to segfault in `LowerBound` and `UpperBound` |
| GHSA-4w68-4x85-mjj9 | CVE-2022-35966 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedAvgPool` |
| GHSA-v6h3-348g-6h5x | CVE-2022-35967 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedAdd` |
| GHSA-2475-53vw-vp25 | CVE-2022-35968 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `AvgPoolGrad` |
| GHSA-g35r-369w-3fqp | CVE-2022-35970 | MODERATE | TensorFlow vulnerable to segfault in `QuantizedInstanceNorm` |
| GHSA-fhfc-2q7x-929f | CVE-2022-35994 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `CollectiveGather` |
| GHSA-g9h5-vr8m-x2h4 | CVE-2022-35995 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `AudioSummaryV2` |
| GHSA-q5jv-m6qw-5g37 | CVE-2022-35996 | MODERATE | TensorFlow vulnerable to floating point exception in `Conv2D` |
| GHSA-p7hr-f446-x6qf | CVE-2022-35997 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `tf.sparse.cross` |
| GHSA-cv2p-32v3-vhwq | CVE-2022-36003 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `RandomPoissonV2` |
| GHSA-37jf-mjv6-xfqw | CVE-2022-35999 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `Conv2DBackpropInput` |
| GHSA-jqm7-m5q7-3hm5 | CVE-2022-36001 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `DrawBoundingBoxes` |
| GHSA-mh3m-62v7-68xg | CVE-2022-36002 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `Unbatch` |
| GHSA-fv43-93gv-vm8f | CVE-2022-36011 | MODERATE | TensorFlow vulnerable to null dereference on MLIR on empty function attributes |
| GHSA-jvhc-5hhr-w3v5 | CVE-2022-36012 | MODERATE | TensorFlow vulnerable to assertion fail on MLIR empty edge names |
| GHSA-mv8m-8x97-937q | CVE-2022-36004 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `tf.random.gamma` |
| GHSA-828c-5j5q-vrjq | CVE-2022-36013 | MODERATE | TensorFlow vulnerable to null-dereference in `mlir::tfg::GraphDefImporter::ConvertNodeDef` |
| GHSA-r26c-679w-mrjm | CVE-2022-36005 | MODERATE | TensorFlow vulnerable to `CHECK` fail in `FakeQuantWithMinMaxVarsGradient` |
| GHSA-7j3m-8g3c-9qqq | CVE-2022-36014 | MODERATE | TensorFlow vulnerable to null-dereference in `mlir::tfg::TFOp::nameAttr` |
| GHSA-79h2-q768-fpxr | CVE-2022-36027 | MODERATE | TensorFlow segfault TFLite converter on per-channel quantized transposed convolutions |
| GHSA-g468-qj8g-vcjc | CVE-2022-36016 | LOW | TensorFlow vulnerable to `CHECK`-fail in `tensorflow::full_type::SubstituteFromAttrs` |
| GHSA-rh87-q4vg-m45j | CVE-2022-36015 | LOW | TensorFlow vulnerable to integer overflow in math ops |
| GHSA-5889-7v45-q28m | CVE-2022-29213 | MODERATE | Incomplete validation in signal ops leads to crashes in TensorFlow |
| GHSA-2r2f-g8mw-9gvr | CVE-2022-29208 | HIGH | Segfault and OOB write due to incomplete validation in `EditDistance` in TensorFlow |
| GHSA-f4rr-5m7v-wxcw | CVE-2022-29209 | MODERATE | Type confusion leading to `CHECK`-failure based denial of service in TensorFlow |
| GHSA-hc2f-7r5r-r2hg | CVE-2022-29210 | MODERATE | Heap buffer overflow due to incorrect hash function in TensorFlow |
| GHSA-75c9-jrh4-79mc | CVE-2022-29216 | HIGH | Code injection in `saved_model_cli` in TensorFlow |
| GHSA-xrp2-fhq4-4q3w | CVE-2022-29211 | MODERATE | Segfault if `tf.histogram_fixed_width` is called with NaN values in TensorFlow |
| GHSA-8wwm-6264-x792 | CVE-2022-29212 | MODERATE | Core dump when loading TFLite models with quantization in TensorFlow |
| GHSA-mw6j-hh29-h379 | - | MODERATE | `CHECK` failure in depthwise ops via overflows |
| GHSA-5wpj-c6f7-24x8 | CVE-2022-29207 | MODERATE | Undefined behavior when users supply invalid resource handles |
| GHSA-rc9w-5c64-9vqq | CVE-2022-29206 | MODERATE | Missing validation results in undefined behavior in `SparseTensorDenseAdd |
| GHSA-54ch-gjq5-4976 | CVE-2022-29205 | MODERATE | Segfault due to missing support for quantized types |
| GHSA-hx9q-2mx4-m4pg | CVE-2022-29204 | MODERATE | Missing validation causes denial of service via `Conv3DBackpropFilterV2` |
| GHSA-jjm6-4vf7-cjh4 | CVE-2022-29203 | MODERATE | Integer overflow in `SpaceToBatchND` |
| GHSA-cwpm-f78v-7m5c | CVE-2022-29202 | MODERATE | Denial of service in `tf.ragged.constant` due to lack of validation |
| GHSA-pqhm-4wvf-2jg8 | CVE-2022-29201 | MODERATE | Missing validation results in undefined behavior in `QuantizedConv2D` |
| GHSA-2vv3-56qg-g2cf | CVE-2022-29200 | MODERATE | Missing validation causes denial of service via `LSTMBlockCell` |
| GHSA-p9rc-rmr5-529j | CVE-2022-29199 | MODERATE | Missing validation causes denial of service via `LoadAndRemapMatrix` |
| GHSA-mg66-qvc5-rm93 | CVE-2022-29198 | MODERATE | Missing validation causes denial of service via `SparseTensorToCSRSparseMatrix` |
| GHSA-hrg5-737c-2p56 | CVE-2022-29197 | MODERATE | Missing validation causes denial of service via `UnsortedSegmentJoin` |
| GHSA-5v77-j66x-4c4g | CVE-2022-29196 | MODERATE | Missing validation causes denial of service via `Conv3DBackpropFilterV2` |
| GHSA-h48f-q7rw-hvr7 | CVE-2022-29195 | MODERATE | Missing validation causes denial of service via `StagePeek` |
| GHSA-h5g4-ppwx-48q2 | CVE-2022-29194 | MODERATE | Missing validation causes denial of service via `DeleteSessionTensor` |
| GHSA-h2wq-prv9-2f56 | CVE-2022-29192 | MODERATE | Missing validation crashes `QuantizeAndDequantizeV4Grad` |
| GHSA-fv25-wrff-wf86 | CVE-2022-29191 | MODERATE | Missing validation causes denial of service via `GetSessionTensor` |
| GHSA-2p9q-h29j-3f5v | CVE-2022-29193 | MODERATE | Missing validation causes `TensorSummaryV2` to crash |
| GHSA-9x52-887g-fhc2 | CVE-2022-23594 | HIGH | Out of bounds read in Tensorflow |
| GHSA-wcv5-vrvr-3rx2 | - | MODERATE | Integer Overflow or Wraparound in TensorFlow |
| GHSA-43q8-3fv7-pr5x | - | HIGH | Improper Validation of Integrity Check Value in TensorFlow |
| GHSA-h6gw-r52c-724r | - | CRITICAL | NULL Pointer Dereference and Access of Uninitialized Pointer in TensorFlow |
CVE-2022-41889 at a Glance
Security Impact
CVSS: 5.5
The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H).
How to Fix CVE-2022-41889
To remediate this vulnerability, upgrade all affected Tensorflow dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.
pip install tensorflow>=2.8.4
pip install tensorflow>=2.9.3
pip install tensorflow>=2.10.1
pip install tensorflow-cpu>=2.8.4
pip install tensorflow-gpu>=2.8.4
pip install tensorflow-cpu>=2.9.3
pip install tensorflow-gpu>=2.9.3
pip install tensorflow-cpu>=2.10.1
pip install tensorflow-gpu>=2.10.1
Run the following command in your project directory to verify that all resolved dependencies use the patched release:
pip show tensorflow
How to Check if You Are Affected
To determine whether your Python environment is affected by CVE-2022-41889, inspect your installed packages using pip:
pip show tensorflow
Analyze the terminal output to locate any version of Tensorflow in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 2.8.4), your application is affected.
Note: The pip show tensorflow command checks currently installed packages in your active Python environment.
Technical Details
Vulnerability Mechanism & Technical Analysis
This vulnerability is classified under CWE-476: NULL Pointer Dereference, where the product dereferences a pointer that it expects to be valid but is null. In Tensorflow, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.
Source Advisory Statement
CVE-2022-41889 Affected Tensorflow Versions
9 component(s)| Product / Package | Ecosystem | Affected Versions | Fixed Version |
|---|---|---|---|
|
tensorflow
|
PyPI |
< 2.8.4
|
2.8.4
|
|
tensorflow
|
PyPI |
≥ 2.9.0
< 2.9.3
|
2.9.3
|
|
tensorflow
|
PyPI |
≥ 2.10.0
< 2.10.1
|
2.10.1
|
|
tensorflow-cpu
|
PyPI |
< 2.8.4
|
2.8.4
|
|
tensorflow-gpu
|
PyPI |
< 2.8.4
|
2.8.4
|
|
tensorflow-cpu
|
PyPI |
≥ 2.9.0
< 2.9.3
|
2.9.3
|
|
tensorflow-gpu
|
PyPI |
≥ 2.9.0
< 2.9.3
|
2.9.3
|
|
tensorflow-cpu
|
PyPI |
≥ 2.10.0
< 2.10.1
|
2.10.1
|
|
tensorflow-gpu
|
PyPI |
≥ 2.10.0
< 2.10.1
|
2.10.1
|
References
Frequently Asked Questions
What is CVE-2022-41889?
Which Tensorflow versions are affected?
What version fixes CVE-2022-41889?
Is tensorflow affected?
Is tensorflow-cpu affected?
What is the CVSS score of CVE-2022-41889?
What CWE is associated with CVE-2022-41889?
How can I check PyPI dependencies for CVE-2022-41889?
Machine-readable OSV JSON
OSV v1.5 Schema
{
"schema_version" : "1.4.0",
"id" : "GHSA-xxcj-rhqg-m46g",
"modified" : "2025-01-15T16:58:58Z",
"published" : "2022-11-21T20:42:00Z",
"aliases" : [ "CVE-2022-41889" ],
"summary" : "Segfault via invalid attributes in `pywrap_tfe_src.cc`",
"details" : "### Impact\nIf a list of quantized tensors is assigned to an attribute, the pywrap code fails to parse the tensor and returns a `nullptr`, which is not caught. An example can be seen in [`tf.compat.v1.extract_volume_patches`](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/image/generate_box_proposals_op.cu.cc) by passing in quantized tensors as input `ksizes`.\n```python\nimport numpy as np\nimport tensorflow as tf\n\na_input = np.array([1, -1], dtype= np.int32)\na_ksizes = a_strides = tf.constant(dtype=tf.dtypes.qint16, value=[[1, 4], [5, 2]])\n\n\ntf.compat.v1.extract_volume_patches(input=a_input,ksizes=a_ksizes,strides=a_strides,padding='VALID')\n```\n\n### Patches\nWe have patched the issue in GitHub commit [e9e95553e5411834d215e6770c81a83a3d0866ce](https://github.com/tensorflow/tensorflow/commit/e9e95553e5411834d215e6770c81a83a3d0866ce).\n\nThe fix will be included in TensorFlow 2.11. We will also cherrypick this commit on TensorFlow 2.10.1, 2.9.3, and TensorFlow 2.8.4, as these are also affected and still in supported range.\n\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.\n\n\n### Attribution\nThis vulnerability has been reported by Pattarakrit Rattankul.\n",
"severity" : [ {
"type" : "CVSS_V3",
"score" : "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H"
} ],
"affected" : [ {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0"
}, {
"fixed" : "2.8.4"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.9.0"
}, {
"fixed" : "2.9.3"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.10.0"
}, {
"fixed" : "2.10.1"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-cpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0"
}, {
"fixed" : "2.8.4"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-gpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0"
}, {
"fixed" : "2.8.4"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-cpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.9.0"
}, {
"fixed" : "2.9.3"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-gpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.9.0"
}, {
"fixed" : "2.9.3"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-cpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.10.0"
}, {
"fixed" : "2.10.1"
} ]
} ]
}, {
"package" : {
"ecosystem" : "PyPI",
"name" : "tensorflow-gpu"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "2.10.0"
}, {
"fixed" : "2.10.1"
} ]
} ]
} ],
"references" : [ {
"type" : "WEB",
"url" : "https://github.com/tensorflow/tensorflow/security/advisories/GHSA-xxcj-rhqg-m46g"
}, {
"type" : "ADVISORY",
"url" : "https://nvd.nist.gov/vuln/detail/CVE-2022-41889"
}, {
"type" : "WEB",
"url" : "https://github.com/tensorflow/tensorflow/commit/e9e95553e5411834d215e6770c81a83a3d0866ce"
}, {
"type" : "PACKAGE",
"url" : "https://github.com/tensorflow/tensorflow"
}, {
"type" : "WEB",
"url" : "https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/image/generate_box_proposals_op.cu.cc"
} ],
"database_specific" : {
"cwe_ids" : [ "CWE-476" ],
"severity" : "MODERATE",
"github_reviewed" : true,
"github_reviewed_at" : "2022-11-21T20:42:00Z",
"nvd_published_at" : "2022-11-18T22:15:00Z"
}
}