@@ -48,7 +48,7 @@ We will indicate the bug fix in the release of PaddlePaddle, and publish the vul
### What is a vulnerability?
In the process of computation graphs in PaddlePaddle, models can perform arbitrary computations , including reading and writing files, communicating with the network, etc. It may cause memory exhaustion, deadlock, etc., which will lead to unexpected behavior of PaddlePaddle. We consider these behavior to be security vulnerabilities only if they are out of the intention of the operation involved.
In the process of computation graphs in PaddlePaddle, models can perform arbitrary computations , including reading and writing files, communicating with the network, etc. It may cause memory exhaustion, deadlock, etc., which will lead to unexpected behavior of PaddlePaddle. We consider these behavior to be security vulnerabilities only if they are out of the intention of the operation involved.
...
...
@@ -60,4 +60,4 @@ If malicious input can trigger memory corruption or non-clean exit, such bug is
@@ -4,9 +4,9 @@ We regularly publish security advisories about using PaddlePaddle.
*Note*: In conjunction with these security advisories, we strongly encourage PaddlePaddle users to read and understand PaddlePaddle's security model as outlined in [SECURITY.md](https://github.com/PaddlePaddle/Paddle/blob/develop/SECURITY.md).
*Note*: In conjunction with these security advisories, we strongly encourage PaddlePaddle users to read and understand PaddlePaddle's security model as outlined in [SECURITY.md](../SECURITY.md).
| Advisory Number | Type | Versions affected | Reported by | Additional Information|
The [implementation](https://github.com/PaddlePaddle/Paddle/blob/release/2.3/paddle/phi/kernels/cpu/gather_tree_kernel.cc#L31-L33) of GatherTreeKernel does not validate the ids_dims size which would result in a memory out-of-bounds read if the ids shape is invalid.
```c++
template<typenameT,typenameContext>
voidGatherTreeKernel(constContext&dev_ctx,
constDenseTensor&ids,
constDenseTensor&parents,
DenseTensor*out){
constauto*ids_data=ids.data<T>();
constauto*parents_data=parents.data<T>();
T*out_data=dev_ctx.templateAlloc<T>(out);
auto&ids_dims=ids.dims();
automax_length=ids_dims[0];
autobatch_size=ids_dims[1];
autobeam_size=ids_dims[2];//[1]
```
### Patches
We have patched the issue in commit [6712e262fc6734873cc6d5ca4f45973339a88697](https://github.com/PaddlePaddle/Paddle/commit/6712e262fc6734873cc6d5ca4f45973339a88697).
The fix will be included in PaddlePaddle 2.4.
### For more information
Please consult [our security guide](../../SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.
### Attribution
This vulnerability has been reported by Wang Xuan(王旋) of Qihoo 360 AIVul Team.